https://www.geotashfeen.

tech/

DESIGNERS’ GUIDES TO THE EUROCODES

DESIGNERS’ GUIDE TO EUROCODE 7:

GEOTECHNICAL DESIGN

DESIGNERS’ GUIDE TO EN 1997-1

EUROCODE 7: GEOTECHNICAL DESIGN – GENERAL RULES

R. FRANK, C. BAUDUIN, R. DRISCOLL, M. KAVVADAS,

N. KREBS OVESEN, T. ORR and B. SCHUPPENER

Series editor
H. Gulvanessian

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

Published by ICE Publishing, One Great George Street, Westminster, London SW1P 3AA.

Full details of ICE Publishing sales representatives and distributors can be found at:
www.icevirtuallibrary.com/info/printbooksales

First published 2005

Reprinted 2009, 2013

Eurocodes Expert

Structural Eurocodes offer the opportunity of harmonized design standards for the European
construction market and the rest of the world. To achieve this, the construction industry needs to
become acquainted with the Eurocodes so that the maximum advantage can be taken of these
opportunities

Eurocodes Expert is a new ICE and Thomas Telford initiative set up to assist in creating a greater
awareness of the impact and implementation of the Eurocodes within the UK construction industry

Eurocodes Expert provides a range of products and services to aid and support the transition to
Eurocodes. For comprehensive and useful information on the adoption of the Eurocodes and their
implementation process please visit our website or email [email protected]

A catalogue record for this book is available from the British Library

ISBN: 978-0-7277-3154-8

© The authors and Thomas Telford Limited 2005

All rights, including translation, reserved. Except as permitted by the Copyright, Designs and Patents
Act 1988, no part of this publication may be reproduced, stored in a retrieval system or transmitted in
any form or by any means, electronic, mechanical, photocopying or otherwise, without the prior
written permission of the Publisher, ICE Publishing, One Great George Street, Westminster, London
SW1P 3AA.

This book is published on the understanding that the authors are solely responsible for the statements
made and opinions expressed in it and that its publication does not necessarily imply that such
statements and/or opinions are or reflect the views or opinions of the publishers. While every effort
has been made to ensure that the statements made and the opinions expressed in this publication
provide a safe and accurate guide, no liability or responsibility can be accepted in this respect by the
authors or publishers.

Typeset by Helius, Brighton and Rochester

Printed and bound in Great Britain by CPI Group (UK) Ltd, Croydon CR0 4YY

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

Preface

EN 1997-1, Eurocode 7: Geotechnical Design, Part 1: General Rules, is the document in the
Eurocode suite concerned with the general geotechnical aspects of the design of structures.
It applies the principles of EN 1990, Eurocode: Basis of Structural Design, by setting the
rules for determining the geotechnical actions and for checking the acceptability of the
geotechnical resistances.

Aims and objectives of this guide

The principal aim of this guide is to provide guidance on the use and interpretation of
EN 1997-1.
Eurocode 7 assumes that the user has an adequate knowledge and understanding of soil
mechanics and geotechnical engineering. The reader of this guide is also expected to be a
geotechnical engineer or to be familiar with conventional geotechnical design.
Throughout this guide emphasis is placed on everyday practice, avoiding complicated
geotechnical design cases, in order to ease the understanding of the new concepts and rules
for geotechnical design appearing in EN 1997-1. Comment is made only on material in
EN 1997-1 that is felt to differ from traditional practice.
For many aspects, this guide aims to be a self-sufficient document but, as the clauses of
EN 1997-1 are repeated only when strictly necessary, the reader should read the guide in
conjunction with the code itself.

Layout of this guide

EN 1997-1 has a Foreword and 12 sections together with nine annexes; this guide has the
same structure, with the chapters corresponding to the sections in the code. Annex A of
EN 1997-1 gives the partial factors and their recommended values for checking ultimate
limit states in persistent and transient situations. All the other annexes of EN 1997-1 relate
to a specific section, and are thus dealt with in the corresponding chapters of this guide.
Each chapter of the guide follows the order of its corresponding section of EN 1997-1
unless this is found to be unhelpful for providing guidance on the use and interpretation of
EN 1997-1 (this is particularly the case for Section 6 and, to some extent, for Section 8).
Consequently, the section numbering in this guide does not necessarily match that in
EN 1997-1: the correspondence between the numbering is indicated in the contents list
beginning each chapter in this guide.
Worked examples are given for the determination of characteristic values (Chapter 2), for
spread foundations (Chapter 6), for pile foundations (Chapter 7), for anchorages (Chapter

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

8), for retaining structures (Chapter 9) and for overall stability (Chapter 11). These examples
are intended to highlight issues relevant to the application of EN 1997-1.
All cross-references in this guide to sections, clauses, subclauses, paragraphs, annexes,
figures, tables and expressions of EN 1997-1-1 are in italic type, which is also used where text
from EN 1997-1-1 has been directly reproduced (conversely, quotations from other sources,
including other Eurocodes, and cross-references to sections, etc., of this guide, are in roman
type). Expressions repeated from EN 1997-1-1 retain their numbering; other expressions
have numbers prefixed by D (for Designers’ Guide), e.g. equation (D2.1) in Chapter 2. Bold
type is used for textual emphasis.

Acknowledgements
This book would not have been possible without the successful completion of Eurocode 7 –
Part 1. Those involved in this process included:
• the project team for converting ENV 1997-1 into EN 1997-1
• the working group for converting ENV 1997-1 into EN 1997-1
• the project team for ENV 1997-1 (1994)
• the chairman and members of the ad hoc group of the European Commission, who in
1978 drafted the first model code for Eurocode 7.
The important contributions of the following in the development of Eurocode 7 – Part 1
are also acknowledged:
• the national geotechnical societies of the EC countries (who are members of the
International Society for Soil Mechanics and Geotechnical Engineering, ISSMGE) for
their support, especially in the early years of the development of Eurocode 7
• national delegations to CEN/TC 250/SC7, and their national technical contacts, for their
valuable and constructive comments
• members of the project team for EN 1990, Eurocode: Basis of Structural Design, for their
contributions to the clauses in EN 1997-1 relating to soil–structure interaction.
This guide is dedicated by its authors to their colleagues mentioned above. The authors
also wish to thank:
• Their wives, Vassilia Frank, Bénédicte Bauduin, Liz Driscoll, Kitty Kavvadas, Hanne
Krebs Ovesen, Diane Orr and Jutta Schuppener, for their support and patience.
• Their employers, CERMES (ENPC-LCPC), Paris; BESIX, Brussels; BRE, Garston;
NTUA, Athens; GEO-Danish Geotechnical Institute, Lyngby; Trinity College, Dublin;
and BAW, Karlsruhe.

R. Frank
C. Bauduin
R. Driscoll
M. Kavvadas
N. Krebs Ovesen
T. Orr
B. Schuppener

vi

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

Contents

Preface v
Aims and objectives of this guide v
Layout of this guide v
Acknowledgements vi

Foreword 1
The Eurocode programme 1
The development of Eurocode 7 2
The content of Eurocode 7 3
The three Design Approaches 3
National implementation of Eurocode 7 5
Application of informative annexes 7
The schedule 8
Packages of EN Eurocode parts 8
National tasks for implementation 9

Chapter 1. General 11
1.1. Scope 11
1.1.1. Scope of Eurocode 7 – Part 1 11
1.1.2. Designs not fully covered by Eurocode 7 – Part 1 12
1.1.3. Contents and organization of Eurocode 7 – Part 1 12
1.1.4. Eurocode 7 – Part 2 13
1.2. References 13
1.3. Assumptions 14
1.4. Distinction between Principles and Application Rules 16
1.5. Definitions 16
1.5.1. Definitions common to all Eurocodes 16
1.5.2. Definitions specific to Eurocode 7 16
1.6. Symbols 17

Chapter 2. Basis of geotechnical design 19

2.1. Design requirements 19
2.2. Design situations 20
2.3. Durability 21
2.4. Geotechnical design by calculation 23
2.4.1. General 23
2.4.2. Actions 24

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

2.4.3. Ground properties 24

2.4.4. Characteristic values of geotechnical parameters 24
2.4.5. Ultimate limit states 30
2.4.6. Serviceability limit state design 37
2.5. Design by prescriptive measures 39
2.6. Observational method 39
2.7. Geotechnical Design Report 39
Example 2.1: selection of a characteristic value using statistical
methods 41
Appendix: an example of the use of statistical methods to assess
characteristic values 44

Chapter 3. Geotechnical data 53

3.1. Introduction 53
3.2. Geotechnical investigations 53
3.3. Evaluation of geotechnical parameters 54
3.3.1. General 54
3.3.2. Characterization of soil and rock type 55
3.3.3. Procedure for evaluating geotechnical parameters 55
3.3.4. Characteristic values 58
3.4. Ground Investigation Report 58

Chapter 4. Supervision of construction, monitoring and maintenance 61

4.1. Introduction 61
4.2. Supervision 62
4.3. Checking ground conditions 63
4.4. Checking construction 63
4.5. Monitoring 64

Chapter 5. Fill, dewatering, ground improvement and reinforcement 65

5.1. General 65
5.2. Fundamental requirements 65
5.3. Fill construction 66
5.4. Dewatering 66
5.5. Ground improvement and reinforcement 66

Chapter 6. Spread foundations 69

6.1. Design methods 70
6.2. Overall stability 70
6.3. Direct method: ULS design 72
6.3.1. Bearing resistance 72
6.3.2. Sliding resistance 78
6.3.3. Loads with large eccentricities 80
6.3.4. Structural failure due to foundation movement 81
6.4. Direct method: SLS design by settlement calculations 82
6.5. Indirect method: simplified SLS method 84
6.5.1. General 84
6.5.2. Indirect method based on limiting the mobilization
of bearing resistance 84
6.6. Prescriptive method 85
6.7. Structural design 85
Example 6.1: square pad foundation on soft clay 86

viii

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CONTENTS

Example 6.2: ULS design of spread foundation for a tower 93

Example 6.3: design based on the indirect method using
pressuremeter test results 99

Chapter 7. Pile foundations 101

7.1. General 101
7.2. Limit states 102
7.3. Actions and design situations 102
7.4. Design methods and design considerations 103
7.5. Pile load tests 104
7.6. Axially loaded piles 105
7.6.1. General 105
7.6.2. Compressive ground resistance (ULS) 106
7.6.3. Ground tensile resistance 112
7.6.4. Vertical displacements of pile foundations 114
7.7. Transversely loaded piles 114
7.8. Structural design of piles 114
7.9. Supervision of construction 115
Example 7.1: design of a pile in compression from static load
test results 115
Example 7.2: design of a pile in compression from in situ test
results 118
Example 7.3: design of a pile in compression from laboratory
test results 121
Example 7.4: design of a pile subject to downdrag 125
Example 7.5: uplift of piled structures 128

Chapter 8. Anchorages 133

8.1. General 133
8.2. Ultimate limit state design 134
8.2.1. Design of the anchorage 134
8.2.2. Design value of the anchorage load 134
8.2.3. Design value of the anchorage resistance 138
8.3. Structural design of anchorages 139
8.4. Load testing of ground anchorages 140
8.4.1. Acceptance tests 140
8.4.2. Suitability tests 140
8.4.3. Investigation tests 141
8.4.4. Proof load as an action to the structure 141
Example 8.1: assessment of proof load for suitability and
acceptance tests 141

Chapter 9. Retaining structures 145

9.1. General 146
9.2. Limit states 146
9.3. Actions, geometrical data and design situations 147
9.3.1. Actions 147
9.3.2. Geometrical data 148
9.4. Design and construction considerations 148
9.5. Determination of earth pressures 149
9.6. Water pressures 151
9.7. Ultimate limit state design 151
9.8. Serviceability limit state design 160

ix

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

9.8.1. General 160

9.8.2. Displacements 161
Example 9.1: ULS design of a stem (gravity) wall 161
Example 9.2: ULS and SLS design of an embedded sheet pile wall 173

Chapter 10. Hydraulic failure 185

10.1. General 185
10.2. Failure by uplift (UPL) 186
10.2.1. General 186
10.2.2. Submerged structures 187
10.2.3. Design against uplift of an impermeable layer 188
10.2.4. Worked example of a design against uplift 188
10.3. Failure by heave (HYD) 189
10.3.1. General 189
10.3.2. Design using total stresses 189
10.3.3. Design using submerged weight 190
10.3.4. Determination of the relevant pore water pressure 190
10.3.5. Worked example of a design against failure by heave 191
10.3.6. Discussion on failure by uplift and failure by heave 191
10.4. Internal erosion 191
10.4.1. Filter criteria and hydraulic criteria 191
10.4.2. Effects of material transport 192
10.5. Failure by piping 192
10.5.1. General 192
10.5.2. Design against failure by piping 192

Chapter 11. Overall stability 195

11.1. General 195
11.2. Limit states 195
11.3. Actions and design situations 196
11.4. Design and construction considerations 196
11.5. Ultimate limit state design 196
11.6. Serviceability limit state design 201
11.7. Monitoring 202
Example 11.1: overall stability of a cutting in stiff clay 202

Chapter 12. Embankments 207

12.1. General 207
12.2. Limit states 207
12.3. Actions and design situations 208
12.4. Design and construction considerations 208
12.5. Ultimate limit state design 208
12.6. Serviceability limit state design 208
12.7. Supervision and monitoring 208

References 211

Index 213

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 1

General

This chapter is concerned with the general aspects of EN 1997-1. The structure of the
chapter follows that of Section 1:

1.1. Scope Clause 1.1

1.2. References Clause 1.2
1.3. Assumptions Clause 1.3
1.4. Distinction between Principles and Application Rules Clause 1.4
1.5. Definitions Clause 1.5
1.6. Symbols Clause 1.6

1.1. Scope
1.1.1. Scope of Eurocode 7 – Part 1
EN 1997-1 gives the general principles and requirements, as well as the general application Clause 1.1.1(2)
rules, relevant to the geotechnical aspects of the design of buildings and civil engineering Clause 1.1.2(1)
works. It has to be used in conjunction with EN 1990, Eurocode: Basis of Structural Design, Clause 1.1.1(1)
which is the head document in the Eurocode suite and thus establishes, for all the structural
Eurocodes, the principles and requirements for safety, serviceability and durability of
structures; it further describes the basis of design and verification and provides guidelines for
related aspects of structural reliability.
EN 1990 gives, in particular, the rules for calculating the combinations of the actions on
buildings and civil engineering works. The numerical values of the structural actions are Clause 1.1.1(4)
given in EN 1991, Eurocode 1: Actions on Structures, and in the corresponding National
Annex for a particular country.
The provisions for the design of a structure in a particular material (e.g. concrete or
steel), specifically its strength and resistance, are the subject of the ‘material’ Eurocodes
(Eurocodes 2 to 6 and 9). Eurocode 7 (on geotechnical design) and Eurocode 8 (on
earthquake resistance) are relevant to all types of structures, whatever the construction
material.
EN 1997-1 describes the requirements for geotechnical design, in order to ensure safety Clause 1.1.1(3)
(strength and stability), serviceability and durability of supported structures, i.e. of buildings Clause 1.1.1(4)
and civil engineering works, founded on soil and rock. In particular, it deals with the
calculation of geotechnical actions, of their effects on structures and of geotechnical
resistances.
For geotechnical design under seismic conditions, the design rules of EN 1997-1 should be Clause 1.1.1(7)
complemented by the rules of EN 1998-5, Eurocode 8 – Part 5: Design of Structures for
Earthquake Resistance. Foundations, Retaining Structures and Geotechnical Aspects.

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

1.1.2. Designs not fully covered by Eurocode 7 – Part 1

As already noted in the Foreword to this guide, Eurocode 7 can also serve as a reference
document for the geotechnical aspects of dams and tunnels, of slope stabilization, and of
foundations for special construction works (e.g. nuclear power plants); additional provisions
to those provided by EN 1997-1 will probably be necessary (see clause 1.1(2) in EN 1990 and
Clause 2.1(21) clause 2.1(21) in EN 1997-1).

1.1.3. Contents and organization of Eurocode 7 – Part 1

Clause 1.1.2(2) The subjects covered in the different sections of EN 1997-1 are as follows:
• Section 1: general
• Section 2: basis of geotechnical design
• Section 3: geotechnical data
• Section 4: supervision of construction, monitoring and maintenance
• Section 5: fill, dewatering, ground improvement and reinforcement
• Section 6: spread foundations
• Section 7: pile foundations
• Section 8: anchorages
• Section 9: retaining structures
• Section 10: hydraulic failure
• Section 11: site stability
• Section 12: embankments.
The sections of EN 1997-1 can be described as follows:
• Section 1 gives the general assumptions and definitions, the symbols, etc.
• Sections 2, 3, 4, 10 and 11 are applicable to all types of geotechnical structures.
• Sections 6, 7 and 9 are specific to particular categories of geotechnical works (shallow or
spread foundations, deep or pile foundations and retaining structures, respectively).
• Section 8 on anchorages is intended to be used for the design of temporary and
permanent anchorages used to support retaining structures, to stabilize slopes, cuts or
tunnels, and to resist uplift forces on structures.
• Sections 5 and 12 cover geotechnical works of a more general nature.
The chapters in this guide and their contents correspond to the sections of Eurocode 7.
Clause 1.1.2(3) The following annexes are included in EN 1997-1:
• Annex A (normative): partial and correlation factors for ultimate limit states and
recommended values
• Annex B (informative): background information on partial factors for Design Approaches
1, 2 and 3
• Annex C (informative): sample procedures to determine limit values of earth pressures
on vertical walls
• Annex D (informative): a sample analytical method for bearing resistance calculation
• Annex E (informative): a sample semi-empirical method for bearing resistance estimation
• Annex F (informative): sample methods for settlement evaluation
• Annex G (informative): a sample method for deriving presumed bearing resistance for
spread foundations on rock
• Annex H (informative): limiting values of structural deformation and foundation movement
• Annex J (informative): checklist for construction supervision and performance monitoring.
Annex A is to be used with Sections 6 to 12, as it gives the relevant partial and correlation
factors for ultimate limit state design. Annex A is normative, which means that it is an integral
part of the standard and must be applied. However, the values of the partial and correlation
factors, given in informative notes, are recommended values and therefore may be modified
in the National Annex for each country.

12

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 1. GENERAL

Links between Eurocodes

Structural safety,
EN 1990 serviceability and
durability

EN 1991 Actions on structures

EN 1992 EN 1993 EN 1994

Design and detailing

EN 1995 EN 1996 EN 1999

EN 1997 EN 1998 Geotechnical and

seismic design

Fig. 1.1. Scope of the 10 Eurocodes and the links between them

Annex B gives some background information on partial factors for applying the three
possible Design Approaches permitted by EN 1990 and by EN 1997-1 (for ultimate limit
states in persistent and transient situations).
Annexes C to J are informative, which means that in a given country, a choice can be made
in the National Annex whether or not to apply them in that country.
Annexes C to G are examples of internationally recognized calculation methods relevant
to the design of foundations or retaining structures.
Annex H deals with limiting movements of foundations, and Annex J is a proposed
checklist for construction supervision and performance monitoring.
The contents of each annex are discussed in this guide in the chapter that corresponds to
the appropriate section of EN 1997-1.

1.1.4. Eurocode 7 – Part 2

EN 1997-1 will be supplemented by a second part: EN 1997-2, Eurocode 7: Geotechnical Clause 1.1.3(1)
Design, Part 2: Ground Investigation and Testing. Part 2 will give the general requirements
and rules for the performance and evaluation of laboratory and field testing for use in
geotechnical design. Note that Part 2 is the result of the merger of the two pre-standards
ENV 1997-2 and ENV 1997-3.

1.2. References
There are references in clause 1.2(1) to the other Eurocodes and to other standards that are Clause 1.2(1)
relevant for geotechnical designs to EN 1997-1. The list of the 10 Eurocodes is given in the
Foreword to this guide. Figure 1.1 illustrates the scope of these Eurocodes and the links
between them.
The execution (or construction) of geotechnical works is covered by Eurocode 7 only to Clause 1.1.1(6)
the extent necessary to comply with the assumptions in the design rules. A series of Clause 1.1.1(5)
European standards on the execution of special geotechnical works is presently (July 2004)
being developed under the auspices of CEN Technical Committee 288 (CEN/TC 288); a list Clause 1.2(1)
of these standards is provided in Eurocode 7 and is reproduced in Table 1.1. Reference to
corresponding CEN/TC 288 standards is given in the various sections of EN 1997-1, where
relevant.

13

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

European standards for the execution of many geotechnical tests are being drafted under
the auspices of CEN/TC 341 on geotechnical investigation and testing. The present list of
expected test standards and technical specifications is given in Table 1.2.

1.3. Assumptions
Clause 1.3(2) The assumptions on which the provisions of EN 1997-1 are based, and with which the users of
the code must strive to comply, are that:
(1) data required for design are collected, recorded and interpreted by appropriately
qualified personnel
(2) structures are designed by appropriately qualified and experienced personnel
(3) adequate continuity and communication exist between the personnel involved in data-
collection, design and construction
(4) adequate supervision and quality control are provided in factories, in plants, and on site
(5) execution is carried out according to the relevant standards and specifications by
personnel having the appropriate skill and experience
(6) construction materials and products are used as specified in EN 1997-1 or in the relevant
material or product specifications
(7) the structure will be adequately maintained to ensure its safety and serviceability for the
designed service life
(8) the structure will be used for the purpose defined for the design.
Clause 1.3(3) These assumptions need to be considered both by the designer and the client. To prevent
uncertainty, compliance with them should be documented, e.g. in the geotechnical design
report.
The seventh and eighth assumptions relate to the responsibilities of the client (owner/
user), who needs to be aware of his/her responsibilities regarding a maintenance regime for
the structure and needs to ensure that neither overloading nor change in the local or
surrounding geotechnical conditions takes place. The designer of the structure should
recommend a maintenance regime, and should clearly state to the owner the limits of
intended use in terms of loads, as well as the ground conditions assumed in the design (i.e.
water levels and other relevant conditions).

Table 1.1. Work programme of CEN/TC 288 on the execution of special geotechnical works

Document Title Status at July 2004 Future progress

EN 1536: 1999 Bored Piles ¸

EN 1537: 1999 Ground Anchors Ô
EN 1538: 2000 Diaphragm Walls Ô
Ô EN published:
EN 12063: 1999 Sheet Piling ˝ see year in title NA
EN 12699: 2000 Displacement Piles Ô
EN 12715: 2000 Grouting Ô
Ô
EN 12716:2001 Jet Grouting ˛
prEN 14199 Micro Piling prEN dated April 1998 Conversion to EN in progress
prEN 14475 Reinforcement of Fills ¸
prEN 14490 Soil Nailing ˝ prEN dated March 2002 CEN enquiry stage
˛
prEN 288011 Deep Mixing CEN enquiry stage
Deep Vibration ¸
Deep Drainage ˝ Drafting Working groups in progress
˛

14

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 1. GENERAL

Table 1.2. Work programme of CEN/TC 341 on geotechnical investigation and testing

Title Status at July 2004

TC 341 Testing Standards

Drilling and Sampling Methods, and Groundwater Part 1 nearing completion ready for public
Measurements: enquiry in 2004; Parts 2 and 3 to follow in
Part 1: Sampling – Principles spring 2004
Part 2: Sampling – Qualification Criteria
Part 3: Sampling – Conformity Assessment
Cone and Piezocone Penetration Tests: Part 1: to enquiry in mid-2004; Part 2 to
Part 1: Electrical Cone and Piezocone enquiry late 2004
Part 2: Mechanical Cone
Dynamic Probing and Standard Penetration Test Public enquiry completed; publication in 2004
of the two standards
Vane Testing Drafting underway; target date for enquiry is
2005
Borehole Expansion Tests: Drafts on the Ménard pressuremeter, the
Ménard Pressuremeter flexible dilatometer and the borehole jack tests
Flexible Dilatometer are well advanced
Self-boring Pressuremeter
Borehole Jack
Full Displacement Pressuremeter
Borehole Shear Test
Plate Load Test Drafting yet to commence
Pumping Tests Drafting yet to commence
Testing of Geotechnical Structures: Drafting of pile load test documents now
Pile Load Test – Static Axially Loaded underway, as is the document on testing of
Compression Test anchorages
Pile Load Test – Static Axially Loaded
Tension Test
Pile Load Test – Static Transversally
Loaded Tension Test
Pile Load Test – Dynamic Axially
Loaded Compression Test
Testing of Anchorages
Testing of Nailing
Testing of Reinforced Fill

TC 341 Technical Specifications

Water Content All out for editorial comment
Density of Fine Grained Soils
Density of Solid Particles
Particle Size Distribution
Oedometer Test
Fall Cone Test
Compression Test
Unconsolidated Triaxial Test
Consolidated Triaxial Test
Direct Shear Test
Permeability Test
Laboratory Tests on Rock

15

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

1.4. Distinction between Principles and Application Rules

Clause 1.4(1) In Eurocode 7, as in all other Eurocodes, a distinction is made between Principles and
Application Rules, depending on the character of the individual clauses. Eurocode 7 states
that:
Clause 1.4(2) • The Principles comprise:
– general statements and definitions for which there is no alternative
– requirements and analytical models for which no alternative is permitted unless
specifically stated.
Clause 1.4(3) • The Principle clauses are preceded by the letter P following the paragraph number.
• Application Rule clauses are identified by the paragraph number only.
Clause 1.4(4) • The Application Rules are examples of generally recognized rules which follow the
Principles and satisfy their requirements.
Clause 1.4(5) • It is permissible to use alternatives to the Application Rules provided it is shown that the
alternative rules accord with the relevant Principles.
The word ‘shall’ is always used in Principle clauses. The word ‘should’ is normally used
for Application Rule clauses; the word ‘may’ is also used, for example in an alternative
Application Rule. The words ‘is’ and ‘can’ are used for a definitive statement or as an
‘assumption’.
Clause 1.4(5) With regard to alternatives to the Application Rules, clause 1.4(5) and the note to
the clause (both reproduced from EN 1990) add that the alternatives should at least
demonstrate equivalent levels of structural safety, serviceability and durability to those
expected when using the Eurocode. Furthermore, if an alternative rule is substituted for an
Application Rule, the resulting design cannot be claimed to be wholly in accordance with
EN 1997-1 although the design may remain in accordance with the Principles of EN 1997-1.
It has already mentioned in the Foreword to this guide that, in implementing Eurocode 7
through its National Annex, a member state has special dispensation to refer to
‘supplementary rules/standards’; these are meant to provide application rules that confirm
to the Principles of the code but which are not provided in it. Therefore, as mentioned above,
these ‘supplementary rules/standards’ are not ‘alternatives’ to any application rules that are
provided in the code.

1.5. Definitions
1.5.1. Definitions common to all Eurocodes
Clause 1.5.1(1) Much of the limit state design terminology is defined in EN 1990 (see also the companion
title to this guide, the Designer’s Guide to EN 1990, Eurocode: Basis of Structural Design) and
is not repeated in Eurocode 7. In fact, repetition of any kind is avoided as far as possible.
Therefore, users of Eurocode 7 are advised to have EN 1990 available.
It is important to note that in all the Eurocodes an ‘action’ is defined as a load or an
imposed deformation, e.g. a temperature effect or settlement (clause 1.5.3.1 of EN 1990).
Examples of actions in geotechnical design are given in clause 2.4.2, and comments on
geotechnical actions are given in Chapter 2 of this guide.

1.5.2. Definitions specific to Eurocode 7

Terms which are specific to Eurocode 7, or are repeated (and adapted) from EN 1990, are
defined as follows:
Clause 1.5.2.1 • Geotechnical action: action transmitted to the structure by the ground, fill, standing
water or groundwater (definition adapted from clause 1.5.3.7 of EN 1990). Examples of
geotechnical actions are earth pressures on retaining walls and downdrag on piles.
Clause 1.5.2.2 • Comparable experience: documented or other clearly established information related to
the ground being considered in design, involving the same types of soil and rock and for

16

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 1. GENERAL

which similar geotechnical behaviour is expected, and involving similar structures.

Information gained locally is considered to be particularly relevant.
• Ground: soil, rock and fill in place prior to the execution of the construction works. Clause 1.5.2.3
• Structure: an organized combination of connected parts, including fill placed during Clause 1.5.2.4
execution of the construction works, designed to carry loads and provide adequate
rigidity (definition adapted from EN 1990).
• Derived value: value of a geotechnical parameter obtained by theory, correlation or Clause 1.5.2.5
empiricism from test results.
• Stiffness: material resistance against deformation. Clause 1.5.2.6
• Resistance: capacity of a component, or cross-section of a component, of a structure to Clause 1.5.2.7
withstand actions without mechanical failure, e.g. resistance of the ground, bending
resistance, buckling resistance and tensile resistance (definition adapted from EN 1990).
The verbs ‘consider’, ‘assess’, ‘account’ and ‘evaluate’ are used frequently throughout
Eurocode 7, for example in Clause 3.3, but are not defined in Eurocode 7. The following
definitions for these verbs, based on Orr and Farrell (1999) and Simpson and Driscoll (1998),
are offered:
• To consider is to think carefully and rationally about all relevant factors affecting the
design and to decide, on the basis of the available information, what effects they are
likely to have. If it is decided that one (or more) of them affects the design, then it must
be included in the design, while if it is decided that the factor is not significant for the
design, then it may be ignored. The verb ‘consider’ often does not imply the need to
include the factors in a calculation, although this may be appropriate in some cases. It is
recommended that, in a geotechnical design, checklists are prepared of the items to be
considered and that the designer should put a mark against the items on the checklist
once they have been considered.
• To assess is to use a process involving some combination of calculation, measurement
and comparable experience, including consideration of all relevant factors, to obtain the
numerical value of a parameter or check if certain criteria are satisfied.
• To take into account is to include the influence of an aspect of the design process. In
Eurocode 7 this phrase generally has a stronger meaning than ‘to consider’, and implies
that the influence of the aspect is included in the design calculation.
• To evaluate is to determine the numerical value of a parameter, taking account of all
relevant factors affecting its value.

1.6. Symbols
Many symbols used in limit state design are defined in EN 1990 and are not repeated in
Eurocode 7.
All the symbols unique to Eurocode 7 are listed in clause 1.6(1). They are in accordance Clause 1.6(1)
with ISO 3898, as well as with the recommendations of the International Society for Soil
Mechanics and Geotechnical Engineering (ISSMGE, 1981).
Characteristic values of parameters are identified by the subscript ‘k’, while design values
are identified by the subscript ‘d’. The subscript ‘dst’ indicates a destabilizing action while the
subscript ‘stb’ indicates a stabilizing one.
In this guide the same symbols and subscripts are used as in EN 1997-1.
The ‘Système International’ (SI) units should be used in geotechnical designs to Eurocode 7.
These units are defined in ISO 1000.
The units most commonly used in geotechnical calculations are presented in Eurocode 7
in clause 1.6(2). Clause 1.6(2)

17

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 2

Basis of geotechnical design

In this chapter the basic philosophy and concepts of EN 1997-1 are presented. The chapter
describes the material covered by Section 2 of EN 1997-1, together with Annex A, for partial
factors, and Annex B for background information on Design Approaches 1, 2 and 3.
The structure of the chapter follows that of Section 2:
2.1. Design requirements Clause 2.1
2.2. Design situations Clause 2.2
2.3. Durability Clause 2.3
2.4. Geotechnical design by calculation Clause 2.4
2.5. Design by prescriptive methods Clause 2.5
2.6. Observational method Clause 2.7
2.7. Geotechnical Design Report Clause 2.8
An appendix presents information on the use of statistical methods for the quantitative
assessment of characteristic values.

2.1. Design requirements

EN 1990 defines limit states as ‘states beyond which the structure no longer fulfils the Clause 2.1(1)P
relevant design criteria’. The aim of limit state design is to check that no limit state is exceeded
when the relevant design values of actions, of material or product resistance properties and
of geometrical properties are used in appropriate calculation models. In order to simplify the
design procedures, two fundamentally different types of limit state are generally recognized,
each of them having its own relevant design criteria (see the Designers’ Guide to EN 1990,
pp. 36–40, for further discussion on limit states (Gulvanessian et al., 2002)):
• ultimate limit states (ULS) defined in EN 1990 as ‘states associated with collapse or
with other similar forms of structural failure’ (e.g. failure of the foundation due to
insufficient bearing resistance);
• serviceability limit states (SLS) defined in EN 1990 as ‘states that correspond to
conditions beyond which specified service requirements for a structure or structural
member are no longer met’ (e.g. excessive settlement related to the intended use of the
structure).
Ultimate limit states corresponding to full ‘collapse’ of geotechnical structures are Clause 2.1(3)
extremely rare; instead, ultimate states usually develop from such large displacements that
the safety requirements of the supported structure are no longer fulfilled. Therefore, the Clause 2.4.7.1(1)P
code requires a check to be made that ultimate limit states cannot occur through failure of
the ground, or through failure of the supported structure itself; the avoidance of an ultimate

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

limit state in the supported structure due to very large (excessive) deformations in the
ground should be also checked.
Clause 2.1(4) The avoidance of limit states should be checked by one or a combination of following:

• use of calculations (described in clause 2.4)

• the adoption of prescriptive measures (described in clause 2.5), in which a well
established and proven design is adopted without calculation under well defined ground
and loading conditions
• tests on models or full scale tests (described in clause 2.6), which are particularly useful
in the design of piles and anchors
• the observational method (described in clause 2.7).

Clauses 2.1(8) To establish geotechnical design requirements EN 1997-1 recommends the classification
to 2.1(28) of structures into Geotechnical Categories 1, 2 or 3 according to the complexity of the
structure, of the ground conditions and of the loading, and the level of risk that is acceptable
for the purposes of the structure; however, this categorization is not mandatory. Geotechnical
Categories are used in the code to establish the extent of site investigation required and the
amount of effort to be expended in the checking of the design. In Fig. 2.1 a flow diagram
illustrates the stages of geotechnical design according to the principles and rules of EN 1997-1.
It is important to note that the Geotechnical Category should be checked at each stage of the
design and construction processes.
Clauses 2.1(14) Simple structures with negligible risk and where the requirements can be satisfied on the
to 2.1(21) basis of local experience will fall into Category 1. Most structures will be in Category 2, whilst
complex problems fall into Category 3.
Clause 2.1(19) EN 1997-1 concentrates on structures in Geotechnical Category 2, and lists examples of
typical design problems.
Figure 2.2 is a flow chart to assist in the assignment of a problem to an appropriate
geotechnical category.

2.2. Design situations

The geotechnical design must be checked for the relevant ‘design situations’. These should
be selected so as to encompass all conditions which are reasonably foreseeable as likely to
occur during the construction and use of the structure. The different design situations
for ultimate and serviceability limit states are defined in EN 1990, and discussed in the
Designers’ Guide to EN 1990 (pp. 35–36). EN 1997-1 deals with ultimate limit states in
persistent and transient situations and in accidental situations, and with serviceability limit
states.
Clause 2.2(1) Where the mass permeability of saturated ground is relatively low (i.e. the time required
for the dissipation of excess positive or negative pore water pressures generated by
construction activities is large compared with the time of construction), both drained and
undrained situations have to be considered in the checking of the ultimate limit state; that is,
the undrained condition with excess pore water pressures and the drained condition when
the pore water pressures have dissipated. Undrained conditions are likely to be critical when
fine-grained soils are loaded and where pore water pressure dissipation with time causes an
increase in the soil strength. Typically, such conditions exist during the loading of soft clays
(e.g. in soft clays beneath dams). Drained conditions are likely to be critical in fine-grained
soils where negative pore water pressure dissipation with time causes a decrease in the soil
strength. Typically, such conditions exist during the unloading of stiff clays, e.g. after
excavating a cutting.
Clause 2.2(2) A list is presented in clause 2.2(2) for consideration of items which can be important when
specifying the design situations.
The probability of occurrence and the consequences of the various design situations
may be different. The safety requirements may thus also be different. For example, for

20

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 2. BASIS OF GEOTECHNICAL DESIGN

Establish preliminary Geotechnical Category

of the structure (2.1(10))

Preliminary ground investigations (3.2.2) and

check of Geotechnical Category

Design investigations (3.2.3)

No

Sufficient information?

Yes

Ground investigation report (3.4) and check

of Geotechnical Category

Design by calculations (2.4), prescriptive

measures (2.5), load or model tests (2.6)
or observational method (2.7)

Geotechnical design report (2.8) and

reassessment of Geotechnical Category

Supervision of the execution of the work (4)

and reassessment of Geotechnical Category

Fig. 2.1. The design process of EN 1997-1 (the numbers in parentheses refer to the relevant
section and clause in EN 1997-1). (After Simpson and Driscoll, 1998)

an accidental situation a structure may be required merely not to collapse, with the
serviceability condition being irrelevant (for further details see p. 30).
Seismic design situations are not treated in EN 1997-1; the reader is referred to EN 1998-5,
Eurocode 8: Design of Structures for Earthquake Resistance – Part 5: Foundations, Retaining
Structures and Geotechnical Aspects.

2.3. Durability
Durability is the ability of the structure to remain fit for use during its design life, given Clause 2.3(1)P
appropriate maintenance. For geotechnical structures, maintenance is often difficult or
impossible. In this case, the design should take into account the degradation of materials
over time due to any aggressiveness of the environment (ground, groundwater chemistry) by
providing adequately resistant materials or protection for them.

21

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

22

DESIGNERS’ GUIDE TO EN 1997-1

No
Is the structure small and relatively simple? Is the structure very large or unusual?
Yes
No
Yes

Does it involve abnormal risks?

Are ground conditions known from comparable Yes
local experience to be sufficiently straightforward No No
that routine methods may be used for foundation
design and construction?
Is there unusual or exceptionally difficult ground?
Yes
Yes No

Are there unusual or exceptional loading conditions?

If excavation below the water table is involved, No Yes
does comparable local experience indicate that No
it will be straightforward?

Is the structure in a highly seismic area?

Yes Yes
No

Is there negligible risk in terms of overall stability No

or ground movements? Is the structure in an area of probable site
instability or persistent ground movements? Yes

Yes No

Category 2
Category 1 Category 3
For example, spread foundations, raft foundations,
pile foundations, walls and other structures Structures or parts of structures which do
retaining or supporting soil or water, excavations, not fall within the limits of Geotechnical
bridge piers and abutments, embankments and Categories 1 and 2
earthworks, ground anchorages and other
tieback systems, tunnels in hard, non-fractured
rock not subjected to special water tightness or
other requirements

Fig. 2.2. Flow chart for geotechnical categorization. (After Simpson and Driscoll, 1998)

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 2. BASIS OF GEOTECHNICAL DESIGN

2.4. Geotechnical design by calculation

2.4.1. General
EN 1990 defines the actions that have to be considered in the calculations. The values of Clause 2.4.1(1)P
structural actions must be taken from EN 1991, whereas EN 1997-1 deals with

• geotechnical actions
• geotechnical resistance.

Design by calculation is the most commonly applied procedure for checking the avoidance
of limit states. It is therefore the main subject of EN 1997-1.
The limit state design procedure involves:

• establishing actions, which may be either imposed loads or imposed displacements

• establishing ground properties and properties of the structural materials
• defining limiting values of deformation, crack width, vibrations, etc.
• setting up calculation models for the relevant ultimate and serviceability limit states
which predict the effect of actions, the resistance and/or the deformations of the ground
and in which the various design situations are considered
• showing that the limit states will not be exceeded in the design situations by using
appropriate calculation models.

The design values of actions and material resistances, as well as the load (action)
combinations, are different for the persistent and transient design situations, for the
accidental design situations and for the serviceability limit states.
Although design by calculation is the most commonly used method of geotechnical design, Clause 2.4.1(2)
the designer should always be aware

that knowledge of the ground conditions depends on the extent and quality of the geotechnical
investigations. Such knowledge and the control of workmanship are usually more significant to
fulfilling the fundamental requirements than is precision in the calculation models and partial
factors.

The calculation model may consist of an analytical model, a semi-empirical rule or a Clauses 2.4.1(3)P
numerical model. EN 1997-1 does not prescribe calculation models for the limit states, but to 2.4.1(5)
some sample models are given in informative annexes. Several examples of analytical models
and semi-empirical calculation rules are illustrated in the examples of this guide. Note that
not only analytical and semi-empirical models are recognized by EN 1997-1, but also
numerical models (the finite element method, finite differences, etc.), although they are not
discussed further in the code.
When no reliable calculation model is available for a specific limit state, EN 1997-1 Clause 2.4.1(4)
permits the analysis of another limit state, using factors to ensure that the specific limit state
is sufficiently improbable. This approach is commonly used in geotechnical design for
checking serviceability limit states in a simplified way, when no values of deformations are
required to be known, by using ultimate limit state models (e.g. bearing capacity models)
with rather large ‘safety factors’ on loads (see also Section 2.4.6 in this guide). This method is
applied for example in Section 6 by the ‘indirect method’ for checking the design of spread
foundations.
Calculation models often include simplifications, the results of which should err on Clauses 2.4.1(6)
the side of safety. It may happen that the calculation model includes a systematic error to 2.4.1(9)
or that it presents a range of uncertainty. The results of calculations based on such
models may be modified, if needed, by a model factor to ensure that the results are either
accurate or err on the safe side. Model factors may be applied to the effects of actions or
to resistances. The practical use of model factors is illustrated in several chapters of this
guide.

23

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

2.4.2. Actions
Clause 2.4.2(1)P The characteristic values of actions must be derived using the principles of EN 1990. The
values of the actions from the structure must be taken from EN 1991. EN 1997-1 is devoted
to geotechnical actions on structures and to geotechnical resistances.
Actions may be loads (forces) applied to the structure or to the soil and displacements or
accelerations that are imposed by the soil on the structure, or by the structure on the soil.
Loads may be permanent (e.g. self-weight of structures or soil), variable (e.g. imposed loads
on building floors) or accidental (e.g. impact loads).
It is necessary to distinguish between actions imposed by the structure on the ground and
geotechnical actions imposed by the ground because, in some Design Approaches, partial
factors are applied to each of them differently (see the section on Design Approaches on
p. 31 of this guide).
Clause 2.4.2(9)P An important principle when dealing with actions is the ‘single-source principle’ (see note
3 in Table A.1.2(B) of Annex A.1 in EN 1990). This principle states that, if permanent actions
arising from the same physical source act simultaneously both favourably and unfavourably,
a single factor may be applied to the sum of these actions or to the effect of them. A typical
example is water pressure acting on both sides of a retaining wall when the water is from the
same hydrogeological formation; the effect of the water pressure on the active and passive
sides of the retaining structure is therefore calculated using the same partial factor for both
sides (see Example 9.2).

2.4.3. Ground properties

Clause 2.4.3(1)P EN 1997-1 stresses that ground properties must be obtained from results of tests or from
other relevant data. Such data might be, for example, back-calculations of settlement
measurements or of failures of foundations or slopes.
Clause 2.4.3(3)P When assessing geotechnical parameters from test results, account must be taken of the
Clause 2.4.3(4) possible difference between the properties obtained from the tests and those governing the
behaviour of the ground mass and/or the geotechnical structure. A checklist is given of those
factors that might cause these differences. One of the most important features to be checked
is whether the ground shows marked strain-softening behaviour or brittleness. When the
peak strength is exceeded locally, there is a dramatic loss of resistance, and redistribution
of stresses might lead to further exceeding of the resistance of the ground, which may
eventually lead to progressive failure.
Section 2.4.4 of this guide provides further explanation.

2.4.4. Characteristic values of geotechnical parameters

General
Clause 2.4.5.2(1)P The process of selecting, from laboratory and/or field measurements, characteristic values
for the geotechnical parameters relevant for design can usually be divided into two main
steps (Fig. 2.3):
• Step 1: establish the values of the appropriate ground properties
Clause 2.4.5.2(2)P • Step 2: from these, select the characteristic value as a cautious estimate of the value
affecting the occurrence of the limit state, including all relevant, complementary
information.
All aspects concerned with step 1 are discussed in Section 3.3.3 of this guide. This chapter
deals with step 2.
Clause 2.4.5.2(2)P EN 1997-1 defines the characteristic value as being ‘selected as a cautious estimate of the
value affecting the occurrence of the limit state’. Each word and phrase in this clause is
important:
• selected – emphasizes the importance of engineering judgement
• cautious estimate – some conservatism is required
• limit state – the selected value must relate to the limit state (this is discussed further in
Chapter 3).

24

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 2. BASIS OF GEOTECHNICAL DESIGN

Step 1
Measured values
Covered by:
EN 1997-1, clauses 2.4.3, 3.3
Test related correction, independent of any
and EN 1997-2
further analysis

Test results
Results of field tests at particular points in the ground or
locations on a site or laboratory tests on particular
specimens

Selection of relevant test results, e.g. peak or

constant volume strengths

Theory, empirical relationships or correlations

to obtain
Derived values
• Choice of a very cautious correlation when
using standard tables relating parameters
to test results

Assessment of influence of test and design

conditions on parameter value. Calibration
and correction factors applied to relate the
parameter to actual design situation and to
account for correlations used to obtain
derived values from test results, e.g.
• factor to convert from axisymmetric to
plane strain conditions
• Correction factor to derive appropriate cu
values obtained from cfv measured values
in a field vane test

Relevant published data and local and

general experience

Geotechnical parameter values

Quantified for design calculations

Step 2 Cautious estimate of geotechnical parameter

value taking account of:
Covered by: • number of test results
EN 1997-1, clause 2.4.5.2 • variability of the ground
• the scatter of the test results, e.g. application
of x factors to pile test results
• particular limit state and volume of ground
involved
• nature of the structure, its stiffness and
ability to redistribute loads

Characteristic parameter value

Fig. 2.3. General procedure for determining characteristic values from measured values

25

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Clause2.4.5.2(4)P There are two major aspects to consider when selecting the characteristic value:
(1) the amount of, and degree of confidence in, knowledge of the parameter values
(2) the soil volume involved in the limit state considered and the ability of the structure to
transfer loads from weak to strong zones in the ground.

Amount and degree of confidence in the information

The cautiousness with which a characteristic value will be selected depends on, among other
things, the confidence the geotechnical engineer has in his or her knowledge of the ground.
This is determined by:
(1) the amount of information (local test results and other relevant information)
(2) the scatter (variability) of the results.
Clearly, the larger the number of tests performed at the site and the greater the amount of
other relevant information, the better the determination can be expected to be of the
characteristic value governing the occurrence of a limit state in the ground. Any other
relevant background information may include tests in the neighbourhood and regional or
geological database information. This is especially important for simple projects, where
normally only a small number of test results are available. A cautious margin between the
selected characteristic value and, for example, the mean value of the test results will be larger
if only a small number of test results is available.
Clearly also, the larger the scatter of the results, the greater the uncertainty about the
value governing the limit state in the ground. The cautious margin between the selected
characteristic value and, for example, the mean value of the test results will be larger if the
test results show a large scatter.
It should be noted that a cautious estimate of the mean property value in a soil layer may
sometimes be misleading as it does not reveal, say, weak zones which may govern the
occurrence of the limit state. Examples of such weak zones which should be detected in the
ground investigations are:
• previously developed failure surfaces
• a kinematically admissible slip surface through a ‘chain’ of weak points.

Soil volume involved and ability of the structure to transfer loads

Clause 2.4.5.2(7) The values of test results of ground parameters fluctuate at random (stochastically) around a
mean value or a mean trend. In situ or laboratory tests involve small volumes of soil. The
volume of the soil involved in a limit state in the ground is much larger than the volume of a
test sample. Therefore, the test results have to be averaged over the volume of soil involved
in the limit state considered. Consequently, a value very close to the mean value of the soil
parameter governs the limit state when:
• a ‘large’ soil volume within the homogeneous layers is involved, allowing for compensation
of weaker areas by stronger areas or
Clause 2.4.5.2(9) • the structure is sufficiently stiff and strong to transfer forces from ‘weaker’ foundation
points to ‘stronger’ foundation points.
It should be noted that piled foundations are an example where advantage may be taken of
the ability of the structure to redistribute loads between the piles (see Chapter 7, ξ values). In
this case, the stiffness of the structure must be sufficient to allow transfer of load from ‘softer’
to ‘harder’ piles.
Clause 2.4.5.2(8) On the other hand, a value close to the (randomly occurring) lowest values of the soil
parameter may govern the limit state when:
• a ‘small’ volume of ground is involved and the failure surface may develop mainly within
the volume of weak soil and/or
• the structure fails before transfer of forces from the ‘weak’ to the ‘strong’ areas occurs,
because it is not sufficiently strong and stiff.

26

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 2. BASIS OF GEOTECHNICAL DESIGN

Undrained shear strength cu

z1

Test result: cu

Mean value of the test results of the undrained

shear strength cu over the length of the shaft

Characteristic value of undrained shear

strength cu over the length of the shaft

z3

More cautious characteristic value of undrained

z2 shear strength cu around the pile base

z4

z depth

Fig. 2.4. Characteristic values of undrained shear strength cu for the determination of shaft and
base resistance of a pile

In such cases the selected characteristic value should be close to the lowest test result, or the
mean value of the test results in the relevant (small) volume of soil.
Figure 2.4 illustrates the items above and shows the test results for undrained shear
strength cu as a function of depth. The pile shaft resistance, which averages the strengths over
the length of the shaft, should be calculated from a characteristic value which is a cautious
estimate of the mean of the test results of undrained shear strength along the shaft between
the depth z1 and z2. The base resistance, which is determined by a small volume of ground
around the pile base, should be calculated using a characteristic value close to the lowest test
result between depth z1 and z4 if there are no test results in the small volume which is relevant
for the behaviour of the base. If there are such test results, as is the case in Fig. 2.4, a cautious
estimate of the mean of the test results between depths z3 and z4 should be taken. The
characteristic value shown in Fig. 2.4 is a very cautious estimate of the mean, with greater
emphasis placed on the lowest value as very few test results are available between depths z3
and z4.
Characteristic values are usually values lower than the most probable value (when lower Clause 2.4.5.2(5)
values of ground parameters yield more conservative results, e.g. for bearing capacity
problems). In some situations, when higher values of ground parameters yield more conservative
results, e.g. downdrag, the characteristic values should be greater than the most probable
value.
Some limit states may be governed more by the difference between the highest and lowest Clause 2.4.5.2(6)P
values, rather than by the mean values themselves. This is especially relevant for serviceability
limit states, where differential settlements may be more detrimental than overall settlements.
Differential settlements are governed by the difference between ‘lower’ and ‘higher’ mean
values of soil compressibility parameters. In this case, the determination of characteristic
values should focus on the differences between weaker and stronger zones and on the
extent of these zones, in relation to the stiffness of the supported structure. Where the
parameters are independent, the most adverse combination of upper and lower values
should be used.

27

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Use of statistical methods

Clause 2.4.5.2(10) Statistical methods may be used when selecting characteristic values of geotechnical parameters,
but they are not mandatory. The statistical techniques aim to calculate the ‘characteristic’
parameter value from the sample parameters (mean value, standard deviation) and a priori
knowledge. The characteristic value is selected such that there is only a small probability
that the value governing the limit state in the ground will be less favourable than the
characteristic value.
Clause 2.4.5.2(11) The use of statistical methods implies that there is a sufficiently large number of test
results (these test results may include data from previous experience).
When statistical methods are used, the code recommends that the calculated probability
of a worse value governing the occurrence of the limit state considered should not be greater
than 5%. The note of clause 2.4.5.2(11) raises the difference between a situation where a
cautious estimate of the ‘mean value of the limited set of geotechnical parameter values’
becomes relevant and a situation ‘where local failure is concerned’.
When the mean value of a soil parameter governs a limit state (e.g. when the limit state is
governed by a large soil volume and when redistribution can occur), the characteristic value
Xc, mean should be selected as a cautious estimate of the (unknown) mean value. The statistical
methods need to deliver an estimate of Xc, mean, the unknown mean value of the parameter
governing the limit state in the ground, with a given confidence level (e.g. 95%) that this
value will be more favourable than the characteristic value Xc, mean (see Fig. 2.5).
When a cautious estimate of the local low value is sought (e.g. if a small soil volume is
involved in the limit state and there are no test results in the small soil volume), the
characteristic value Xlow should be selected such that there is only a 5% chance that
somewhere in the ground a value is less favourable than the characteristic value. In such
cases the characteristic value Xlow should be selected as a 5% fractile (see Fig. 2.5).
In many cases the 5% fractile will give a very low characteristic value Xlow, which may
result in a very conservative design. In such cases it is advisable to intensify the ground
investigation and determine the local mean parameter values at those locations where they
are relevant for design (see Fig. 2.4).
Clause 2.4.5.2(10) The statistical formulae to determine the 95% reliable mean value or the 5% fractile
depend on the type of population, the type of samples and the amount and reliability of a
priori knowledge. Populations without trend and populations showing significant trend
should be distinguished. In a homogeneous population without trend, the fluctuations of the
values of the parameter are purely random around the mean value. There is no relationship
between the value of the parameter and location. In a population with trend, the parameter
values are randomly distributed around a clearly distinguishable variation as a function of
another parameter. The sample data, and any other relevant information (e.g. geological
information), can be used to decide whether the population has a significant trend or is
homogeneous. Examples of trends are undrained shear strength increasing with depth, and
drained shear strength increasing with normal stress. The statistical formulae are different
for populations with and without trend.
Parameter values are gathered in statistical populations of ‘samples’. Different types of
statistical population are distinguished depending on the way the populations are built up. In
a local population, the sample test results or the derived values are obtained from tests at the
site of or very close to the geotechnical structure being designed. In the case of regional
populations, the sample test results are obtained from tests on the same ground formation
extending over a large area and collected, for example in a data bank. If a sufficiently large
local population is available, it will be used primarily to select the characteristic value of the
parameter considered; however, if no or only little local information is available, the
selection of the characteristic value may be mainly based on results of regional sampling or
other relevant experience. For structures of Geotechnical Categories 2 and 3, regional
sampling should only be used for preliminary design. The assumed characteristic value
should in a later stage be confirmed by local sampling. When a limited number of local

28

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 2. BASIS OF GEOTECHNICAL DESIGN

Number of test results, n

Mean of test results, Xmean

*
Possible distribution best guess
* through n tests results
*
* *

* sX sX

*
*
*
*
*
*

Xlow = 5% fractile Xc, mean Xmean

Value of parameter

Xmeankn, fractileVx Xmeankn, meanVx

Fig. 2.5. Cautious estimate of the mean value Xc, mean and cautious estimate of the local low value
Xlow by the 5% fractile from the sample parameters Xmean and sX in the case ‘VX unknown’

test results are available, but an important regional population exists, both sources of
information can be combined when selecting the characteristic value.
When selecting a characteristic value, any complementary information and a priori knowledge
should be introduced. This can be done through Bayesian techniques. Discussion of Bayesian
techniques is, however, outside the scope of this guide, and they are not practical for routine
problems.
Another way to introduce a priori knowledge is by assuming the coefficient of variation VX
of the property is known. The concept of the case called ‘VX known’ is introduced in EN 1990
(see EN 1990, clause D7.1(5)). Within a soil layer, the coefficient of variation does not vary
much. Knowledge of the coefficient of variation can thus often be assumed when selecting
the characteristic value. The statistical formula to be used will give a characteristic value
which will be closer to the sample mean value, compared with the one to be used when the
coefficient of variation of the property is not known from a priori knowledge (a case called
‘VX unknown’) and has to be established from the sample data alone.
A simple approach to select the characteristic value Xk is to apply the equation (D2.1),
given in the appendix to this chapter:
Xk = Xmean(1 – knVX)
where Xmean is the arithmetical mean value of the parameter values; VX is the coefficient of
variation; and kn is a statistical coefficient which depends on the number n of test results, on
the ‘type’ of characteristic value (‘mean’ or ‘fractile’) and a priori knowledge about the
coefficient of variation (case ‘VX unknown’ or ‘VX known’). For further information on the
application see the appendix to this chapter and Example 2.1.
Figure 2.5 illustrates the determination of characteristic values in the simple case of a
sample consisting of n parameter values Xi in a homogeneous soil layer without trend. It is

29

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

assumed that the parameter has a normal distribution and there is no complementary
information available (case ‘VX unknown’). The sample parameters are its mean value Xmean
and standard deviation sX. The characteristic value Xc, mean is determined using equation
(D2.1) so that there is a probability of 95% that the mean value governing the occurrence of
a limit state in the ground is larger than the characteristic value. From Table 2.5 in the
appendix to this chapter the coefficient kn is then taken as the value of kn, mean for ‘VX
unknown’.
This simple case illustrates that:
• the characteristic value Xc, mean becomes closer to the sample mean Xmean as the number n
of test results increases
• as the standard deviation sX increases, the ‘distance’ between the sample mean Xmean and
the characteristic value Xc, mean increases.
Figure 2.5 also indicates the 5% fractile of the low value Xlow, for comparison. Xlow can be
calculated using equation (D2.1) in which the value of kn is taken as kn, fractile from Table 2.7 in
the appendix. It should be noted that Table 2.7 distinguishes between the cases ‘VX unknown’
and ‘VX known’ for which the values of kn, fractile are different. kn, fractile is considerably greater
than kn, mean for a 95% reliable mean value. Therefore, the value Xlow of the fractile is
considerably lower than the 95% reliable estimate Xc, mean of the mean value.
Example 2.1 on the statistical evaluation of test results illustrates the difference between
local low and mean values and shows the effect of the knowledge of the coefficient of
variation, case ‘VX known’.
Further discussion and details on statistical methods and the formulae to be applied to
determine characteristic values are given in the appendix to this chapter and in Appendix C
of the Designers’ Guide to EN 1990.

2.4.5. Ultimate limit states

General
Although EN 1997-1 deals with the design of different types of foundation, retaining
structure and other geotechnical structures, the code does not specify which soil mechanics
theories or soil behaviour models to use to determine, for example, the earth pressure acting
on a retaining structure or the stability of a slope. But EN 1997-1 does state which design
criteria are to be used in the calculations, and it makes mandatory the format of checking
using partial factors. The values of the partial factors in Annex A are recommendations, and
can be altered in the National Annex. (The general concepts of the method of checking by
partial factors, the definition of the different partial factors and the uncertainties they cover
are described in EN 1990 (see Section 6 and Annex C9).)
Clause 2.4.7.1(1)P EN 1997-1 distinguishes between five different types of ultimate limit state, and uses
abbreviations for them that are defined in EN 1990:
• ‘loss of equilibrium of the structure or the ground, considered as a rigid body, in which the
strengths of structural materials and the ground are insignificant in providing resistance
(EQU)’, e.g. tilting of a retaining structure on rock
• ‘internal failure or excessive deformation of the structure or structural elements, including
footings, piles, basement walls, etc, in which the strength of structural materials is significant
in providing resistance (STR)’
• ‘failure or excessive deformation of the ground, in which the strength of soil or rock is
significant in providing resistance (GEO)’, e.g. overall stability, bearing resistance of
spread foundations or pile foundations
• ‘loss of equilibrium of the structure or the ground due to uplift by water pressure (buoyancy)
or other vertical actions (UPL)’
• ‘hydraulic heave, internal erosion and piping in the ground caused by hydraulic gradients
(HYD)’.
Formulae for checking these limit states are given in clauses 2.4.7.2 to 2.4.7.5.

30

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 2. BASIS OF GEOTECHNICAL DESIGN

The avoidance of ultimate limit states as dealt with in EN 1997-1 mainly applies to Clause 2.4.7.1(2)P
persistent and transient situations, and the partial factors proposed in Annex A are only valid
for these situations.
In accidental situations, all values of partial factors should normally be taken as equal to Clause 2.4.7.1(3)
1.0. Requirements and recommendations for seismic design are given in EN 1998-5.
In cases of abnormal risk or exceptionally difficult ground conditions, more severe values Clause 2.4.7.1(4)
for the partial factors than those given in Annex A should be used.
Less severe values may be used for temporary structures or transient design situations Clause 2.4.7.1(5)
where the likely conditions justify it.
In design situations where the ground strength acts in an unfavourable manner (e.g.
downdrag or heave on piles), the design value of the unfavourable action may be obtained by
either of the following methods:
(1) Applying the inverse of the M2 set of partial material factors as partial action factors to
the characteristic unfavourable action (see Example 7.4).
(2) Applying the inverse of the M2 set of partial material factors to the characteristic
material strengths to obtain the design strengths, and hence the design unfavourable
action. When adopting this approach, if the soil strength parameters are used to
calculate another component of the unfavourable action (e.g. the lateral earth pressure
on piles), it is important to ensure that the application of this inverse partial factor to the
characteristic soil strength to determine this component of the action does not, because
of compensating effects, lead to reduced (unconservative) unfavourable actions, and
hence reduced safety. A similar situation of possible compensating effects can arise
when the ground strength acts in a favourable manner (as in the case of uplift; see
Section 10.2.2, equation (D10.4)).

Clause 2.4.7.1(6) mentions cases where a model factor is applied to the effect of the Clause 2.4.7.1(6)
actions instead of applying it at source to the characteristic value of the action.

Checking for static equilibrium

Checking for static equilibrium (EQU) presumes that the strength of the ground and the Clause 2.4.7.2(1)P
structure are insignificant in providing stability. Static equilibrium is mainly relevant in
structural design. In geotechnical design, checking the avoidance of the EQU limit state
applies only in rare cases such as a foundation, bearing on rock, which can tilt about an edge.
The inequality for static equilibrium (inequality (2.4)) requires that the design value of the
destabilizing action (e.g. the overturning moment from earth or water pressures) is not
greater than the sum of the design value of the stabilizing action (e.g. restoring moment due
to the weight of the structure) and the design value of any minor shearing resistance (e.g. on
the side of the structure). Partial factors to apply for EQU checking in persistent and
transient design situations are given in EN 1997-1, in Tables A.1 and A.2 of Annex A.
Table A.2 is meant to apply to any minor shearing resistance. (This resistance should not
be treated as a stabilizing action and therefore should not be factored using the partial factor
values given in Table A.1.)

Checking against failure in the ground (GEO) and in the structure (STR)
When checking for limit states of failure or excessive deformation in the ground and in the Clause
structure, the following inequality must be satisfied: 2.4.7.3.1(1)P
Ed £ Rd (2.5)
where Ed is the design value of the effects of all the actions, and Rd is the design value of the
corresponding resistance of the ground and/or structure.
In contrast to the checking of structural designs, geotechnical actions from and resistances
of the ground cannot be separated: geotechnical actions sometimes depend on the ground
resistance, e.g. active earth pressure, and ground resistance sometimes depends on actions,

31

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

e.g. the bearing resistance of a shallow foundation depends on the actions on the foundation.
There are different and equivalent ways to account for this coupling between the geotechnical
actions and resistances. Therefore, EN 1997-1 proposes three Design Approaches for
checking the avoidance of failure in the ground (GEO) and in the structure (STR).

Design effects of actions

Clause Effects of actions are functions of the action itself, of ground properties and of geometrical
2.4.7.3.2(1)P data. Expressions (2.6a) and (2.6b) express the calculation of the design values of the effects
of actions in different mathematical forms, depending on the manner in which the partial
factors are applied.
Partial factors for actions can be applied:
• either to the representative values Frep of the action,
Ed = E{γFFrep, Xk/γM, ad} (2.6a)
• or to their effect (E),
Ed = γEE{Frep, Xk/γM, ad} (2.6b)
where γF is the partial factor for an action, γM is the partial factor for a material property, γE is
the partial factor for the effect of actions and ad is the design value of the geometrical data.
EN 1990 (clause 6.3.1) and Annex B of EN 1997-1 give some explanation of when to use
either expressions (2.6a) or (2.6b). The term Xk/γM introduces into the calculation the effects
of geotechnical actions such as earth pressures.
The following methods can be used for calculating the design value of the effect of a
geotechnical action (see Annex B.2 of EN 1997-1):
• by using design values of the strength parameters and by applying a partial factor γM
larger than 1.0 to the source of the action
• by using design values that are equal to the characteristic values of the strength parameters,
i.e. by applying a partial factor γM equal to 1.0.
Then, expressions (2.6a) and (2.6b) become
Ed = E{γFFrep, Xk, ad} (expression (B.3.2) in Annex B.2)
Ed = γEE{Frep, Xk, ad} (expression (B.3.1) in Annex B.2)
Clause Recommended values of partial factors for use in persistent and transient situations are
2.4.7.3.2(3)P given in Tables A.3 and A.4. Alternative values may be given in the National Annex. Rules for
the combining of actions and values of the combination factor y are given in EN 1990.

Design resistances
Clause 2.4.7.3.3 The resistance in the ground is a function of the ground strength, Xk, sometimes of the
actions, Frep (when the value of the resistance is affected by the action, e.g. spread foundations
subjected to inclined loads), and of the geometrical data. To obtain the design value of the
resistance, Rd, partial factors can be applied either to ground properties (X) or to resistance
(R), or to both, as follows:
Rd = R{γFFrep, Xk/γM, ad} (2.7a)
Rd = R{γFFrep, Xk, ad}/γR (2.7b)
Rd = R{γFFrep, Xk/γM, ad}/γR (2.7c)
where γR is the partial factor for the resistance of the ground.
In expression (2.7a) the design value of the resistance is obtained by applying the partial
factor γM > 1.0 to the characteristic values of the ground strength parameters ck¢
and tan ϕk¢ or cu, k, etc. If actions play a role in the resistance, design values of actions

32

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 2. BASIS OF GEOTECHNICAL DESIGN

(γF × Frep) are introduced in the calculation of Rd (see Design Approaches 1 and 3 and Figs
2.6 and 2.8).
In expression (2.7b) the design value of the resistance is obtained by applying the partial
factor γR > 1.0 to the resistance obtained using design values equal to characteristic values of
the ground strength parameters. If actions play a role in the resistance, design values of
actions (γFFrep) are introduced in the calculation of Rd (see Design Approach 2 and Fig. 2.7).
If the effects of actions are factored (see Annex B.3(6) of EN 1997-1), γF = 1.0, and
expression (2.7b) becomes
Rd = R{Frep, Xk, ad}/γR (expression (B.6.2.2) in Annex B.3)
Expression (2.7c) is similar to expression (2.7a), but a complementary resistance factor
γR > 1.0 can be applied to obtain the design value of the resistance. This factor then acts
merely as a model factor.
The values of partial factors to be used in the expressions above should be selected from
Annex A; alternative values may be set in the National Annex.

Design Approaches
Expressions (2.6) and (2.7) differ in the way that they distribute the partial factors between Clause
actions, ground properties and resistances. Different combinations of expressions (2.6) and 2.4.7.3.4.1(1)P
(2.7), and thus different ways to introduce the partial factors in the terms E and R of
the fundamental inequality (2.5), have led to the three Design Approaches permitted in
EN 1997-1. The choice of the Design Approach is for national determination, and should be
stated in the National Annex. Different design problems may be treated by different Design
Approaches. The values of the partial factors to apply with a chosen Design Approach are
also left to national determination, as indicated in the National Annex.
The ways of combining sets of partial factors to obtain the design values of the effects of
actions and of the resistance in inequality (2.5) are indicated in a symbolic way, e.g.
A1 ‘+’ M1 ‘+’ R1
The meaning of the above expression is as follows:
(1) the partial factors for the actions (γF) or the effects of actions (γE) are represented by the
symbol A and are taken from set A1 in Table A.3 of Annex A of EN 1997-1; the symbol ‘+’
signifies that they are used in combination with
(2) the partial factors (γM) for strength (material) parameters of the ground (symbol M),
which are taken from set M1 in Table A.4, and
(3) the partial factors for the resistance (γR) (symbol R), which are taken from set R1 in
Tables A.5 to A.6.
The procedure for combining partial factors described in a symbolic way above implies
that a geotechnical action, or the effect of an action including a geotechnical action,
will involve two sets of partial factors: An ‘+’ Mn. Likewise, a geotechnical resistance will
always involve two sets of partial factors: Mn ‘+’ Rn. However, in a number of cases, the factor
values in these sets will be equal to unity; this applies, for example, to sets M1, R1 and R3.
Making use of the partial factor values of set M1 implies that the design values of the soil
parameters are equal to their characteristic values. As a consequence, design geotechnical
actions, design values of the effects of actions and design resistances, calculated using the
M1 set of partial factors, may sometimes, for simplicity, be stated as calculated ‘from
characteristic values’, when the true meaning is that they are calculated ‘from design soil
parameters with partial factors equal to unity’. Examples of such simplifications may be
found in this guide, especially in the examples.

Design Approach 1
The design should be separately checked for failure in the soil and in the structure using two Clause
combinations of sets of partial factors. 2.4.7.3.4.2(1)P

33

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

The partial factors are applied at the source, i.e. to the representative values of the actions
and to the characteristic values of the ground strength parameters (such as c¢ and tan ϕ¢ or
cu) using expression (2.7a). However, an exception to this is made for the design of pile and
anchorage resistances, which are obtained by applying partial resistance factors to measured
or calculated resistances using expression (2.7b).
The partial factors are usually applied directly to the representative values of the actions
(expression (2.6a)), except when doing so leads to physically impossible situations (e.g. in the
Clause case of factoring the known depth of a free water surface. In such cases, partial factors are
2.4.7.3.2(2) applied to the effects of the action; thus, expression (2.6b) is used (clause 2.4.7.3.2(2)).

Combination 1. The combination of partial factors used is A1 ‘+’ M1 ‘+’ R1. For most
readers familiar with ENV 1997-1, Combination 1 is the old ‘Case B’ used in that ‘trial’
version of the code. It aims to provide safe design against unfavourable deviations of the
actions, or their effects, from their characteristic values, while the design values of ground
properties are equal to their characteristic values. So, for unfavourable actions (or their
effects) the calculations for Combination 1 are performed using set A1 of Table A.3 of
Annex A (Fig. 2.6a: recommended values γG = 1.35 and γQ = 1.5); for favourable actions,
recommended values are γG = 1.0 and γQ = 0.0. For ground resistances, the calculations are
performed using set M1 of Table A.4, and R1 of Tables A.5 to A.8 and A.12 to A.14 (Fig. 2.6a:
γϕ¢ = γc¢ = γcu = 1.0 and γR, v = 1.0).

Combination 2. The combination of partial factors used is A2 ‘+’ M2 ‘+’ R1. Combination
2 was called ‘Case C’ in ENV 1997-1. It aims to provide safe design against unfavourable
deviations of the ground strength properties from their characteristic value and against
uncertainties in the calculation model, while it is assumed that the permanent actions are
very close to their expected representative values and the variable actions from the structure
may deviate slightly in an unfavourable way. So, for actions (or their effects) the calculations
for Combination 2 are performed using set A2 of Table A.3 of Annex A (Fig. 2.6b:

Qd = gQQk = 1.5Qk Qd = gQQk = 1.3Qk

Gd = gGGk = 1.35Gk Gd = gGGk = 1.0Gk

qd = gQqk = 1.5qk qd = gQqk = 1.3qk

gj¢ = gc = 1.0 tan j¢d = tan j¢k /gj¢

j¢d = j¢k, c¢d = c¢k = tan j¢k /1.25
c¢d = c¢k/gc = c¢k /1.25

EQ, d = EQ(j¢d, c¢d, qd) EQ, d = EQ(j¢d, c¢d, qd)

EG, d = gGEG(j¢d, c ¢d) EG, d = gGEG(j¢d, c ¢d) = 1.0EG(j¢d, c¢d)
= 1.35EG(j¢k, c¢k)
Vd = ÂVG, d + ÂV Q, d Vd = ÂVG, d + ÂV Q, d

Rv, d = Rv(V d, H d, j¢d, c ¢d)/gRv Rv, d = Rv(V d, H d, j¢d, c ¢d)/gRv

= Rv(V d, H d, j¢k, c¢k)/1.0 = Rv(V d, H d, j¢d, c ¢d)/1.0

(a) (b)

Fig. 2.6. Design Approach 1: introduction of partial factors (recommended values) in the checking
of ground bearing capacity using (a) Combination 1 and (b) Combination 2. For simplicity, only
vertical equilibrium is considered and only unfavourable actions are shown

34

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 2. BASIS OF GEOTECHNICAL DESIGN

Qd = gQQk = 1.5Qk Qk

Gd = gGGk = 1.35Gk Gk

qd = gQqk = 1.5qk qk

gj¢ = gc = 1.0 gj = gc = 1.0

j¢d = j¢k, c ¢d = c¢k j¢d = j¢k, c ¢d = c¢k

EQ, d = EQ(j¢d, c ¢d, qd) EQ, k = EQ(j¢k, c¢k, qk)

EG, d = gGEG(j¢d, c ¢d) = 1.35EG(j¢k, c¢k) EG, k = EG(j¢k, c ¢k)

Vd = ÂVG, d + ÂV Q, d Vd = gG ÂVG, k + gQ ÂV Q, k
= 1.35 ÂVG, k + 1.5 ÂV Q, k

Rv, d = Rv(V d, H d, j¢d, c ¢d)/gRv Rv, d = Rv(V k, H k, j¢k, c¢k)/gRv

= Rv(V d, H d, j¢k, c¢k)/1.4 = Rv(V k, H k, j¢k, c ¢k)/1.4

(a) (b)

Fig. 2.7. Introduction of partial factors (recommended values) in the checking of ground bearing
capacity using Design Approach 2. (a) Factoring actions at the source, Design Approach DA-2. (b)
Factoring effects of actions, Design Approach DA-2*. For simplicity, only vertical equilibrium is
considered and only unfavourable actions are shown

recommended values γG = 1.0, for both favourable and unfavourable actions, and γQ = 1.3
(unfavourable) and γQ = 0.0 (favourable)). For ground resistances, the calculations are
performed using set M2 of Table A.4 and set R1 of Tables A.5 to A.8 and A.12 to A.14 (Fig.
2.6b: recommended values γϕ¢ = γc¢ = 1.25, γcu = 1.4; γR, v = 1.0).

For the design of piles and anchorages, the design value of their resistance is calculated Clause
using set M1 of Table A.4 (γϕ = γc = γcu = 1.0) and a partial factor (γR > 1.0) from set R4 of 2.4.7.3.4.2(2)P
Tables A.6 to A.8 or A.12. The design values of the unfavourable actions on the piles or
anchorages are calculated using the partial factors of sets A2 and M2 (see also Chapters 7
and 8).
Where it is obvious that one combination of sets of partial factors governs the design,
it is not necessary to perform full calculations for the other combination. Usually, the
geotechnical ‘sizing’ is governed by Combination 2, and the structural design is governed by
Combination 1. So, it is often obvious in a first step to determine the size of the geotechnical
element using Combination 2 and then simply to check in a second step that this size of
element is acceptable using Combination 1; similarly, it is often obvious to determine the
structural strength of the resulting element using Combination 2 and, when relevant, to
check it for Combination 1.
For further details of Design Approach 1, see Simpson (2000).

Design Approach 2
In Design Approach 2 a single combination of the sets of partial factors is applied to the Clause 2.4.7.3.4.3
calculations for the checking of each relevant ultimate limit state in the ground and in the
structure. The combination of partial factors A1 ’+’ M1’ +’ R2 is used. The same values of
the partial factors are applied to geotechnical actions and to actions on/from the structure.
Partial factors are applied to the ground resistance and either to the actions (referred to as
‘DA-2’) or to the effects of actions (referred to as ‘DA-2*’) (Fig. 2.7).

35

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

The practical application of and the results for the two ways of applying the partial factors
are different.
For the procedure where the actions are factored at their source (DA-2), the factors of set
A1 of Table A.3, set M1 of Table A.4 and set R2 of Tables A.5 to A.8 and A.12 to A.14 are used
(Fig 2.7a: recommended values γG = 1.35 and γQ = 1.5; set M1, γϕ¢ = γc¢ = γcu = 1.0; set R2,
γR, v =1.4 applied to the bearing resistance).
For the procedure where the partial factors are applied to the effect of actions (DA-2*),
the same factors are used but the calculations of E and R are performed with characteristic
values of actions and characteristic values of ground strength parameters. The partial
factors (sets A1 and R2) are applied, at the end, to the resulting effects of permanent
and variable characteristic actions (Fig 2.7b: Vd = 1.35 × ÂVG, k + 1.5 × ÂVQ, k) and to the
resistance calculated using the characteristic values of the ground properties (Fig. 2.7b:
Rv, d = Rv(ϕk¢, ck¢)/1.4). In this procedure, expressions (B.3.1) and (B.6.2.2) of Annex B in
EN 1997-1 are used, so that a direct relationship with the traditional overall factor of safety
η = Rk/Ek can be established thus:
Ed £ Rd
is written as
γEE{Frep, Xk, ad} £ R{Frep, Xk, ad}/γR
so that
η = γEγR
It should be noted that γE is a compound factor, where the value depends on the
proportions of the permanent and variable parts of the action. The product γEγR is dependent
on this proportion, whereas the overall factor of safety η is usually independent of it.
For further details of Design Approach 2 see Schuppener et al. (1998).

Design Approach 3
Clause 2.4.7.3.4.4 In Design Approach 3, a single combination of the sets of partial factors is applied to the
calculations for the checking of each relevant ultimate limit state in the ground and in the
structure. The combination of partial factors (A1 or A2) ‘+’ M2 ‘+’ R3 is used. The
characteristic values of actions coming from the structure (structural actions) are multiplied
by the factors of set A1 in Table A.3 (Fig. 2.8: recommended values γG = 1.35 and γQ = 1.5)
to produce their design values. Design values of actions arising from the ground or
transferred through it (geotechnical actions) are assessed using the partial factors on ground
strength of set M2 and the partial action factors of set A2 (Fig. 2.8: recommended values
γG = 1.0 and γQ = 1.3). Design values of soil strength parameters are obtained by applying
the factors of set M2 in Table A.4 (Fig. 2.8: recommended values γϕ¢ = γc¢ = 1.25; γcu = 1.4).
Design values of the soil resistance are obtained by applying the partial factors from set M2
in Table A.4 to the ground strength parameters and the partial resistance factors of set R3 in
Tables A.5 to A.8 and A.12 to A.14 of Annex A (Fig. 2.8: set M2, recommended values
γϕ¢ = γc¢ = 1.25, γcu = 1.4 and γR, v =1.0).
In all Design Approaches the design values of structural material strength properties are
taken from the relevant material code.

Checking procedure and partial factors for the avoidance of uplift

Clause 2.4.7.4(1)P The inequality for the checking of resistance to failure by uplift (UPL) requires that the
design value of the destabilizing permanent and variable vertical actions, Vdst, d, is less than or
equal to the sum of the design value of the stabilizing permanent vertical actions, Gstb, d, and
of any additional resistance to uplift, Rd (see inequality (2.8)). The recommended values for
partial factors to be applied in the avoidance of UPL in persistent and transient design
situations are given in Tables A.15 and A.16 of Annex A.

36

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 2. BASIS OF GEOTECHNICAL DESIGN

Qd = gQQk = 1.5Qk

Gd = gGGk = 1.35Gk

qd = gQqk = 1.3qk

tan j¢d = tan j¢k /gj¢ = tan j¢k /1.25

c ¢d = c¢k /gc = c¢k /1.25

EQ, d = EQ(j¢d, c¢d, qd)

EG, d = gGEG(j¢d, c¢d) = 1.0EG(j¢d, c¢d)

Vd = ÂVG, d + ÂV Q, d

R v, d = Rv(V d, H d, j¢d, c¢d)/gRv

= Rv(V d, H d, j¢d, c¢d)/1.0

Fig. 2.8. Introduction of partial factors (recommended values) in the checking of ground bearing
capacity using Design Approach 3. For simplicity, only vertical equilibrium is considered and only
unfavourable actions are shown

Table A.15 is applied to the stabilizing and destabilizing actions. Stabilizing permanent
actions are, for example, the weight of the structure and/or of the ground. Water pressures
under the structure and any other upward or pull-out forces are destabilizing actions.
Table A.16 is applied to any additional resistance, Rd, to uplift provided by shearing
resistance (where this resistance is calculated using shear strength parameters), pile tensile
resistance or anchorage resistance.
Alternatively, this resistance may be treated as a stabilizing permanent vertical action Clause 2.4.7.4(2)
Gstb, d, and Table A.15 is applied to it (note that, in this case, model factors are also generally
introduced).
Some details of UPL avoidance and measures to resist failure by uplift forces are
described in Chapter 10 of this guide (on hydraulic failure) and in Example 7.5, which
comprises the UPL checking of a pile foundation for a structure subjected to uplift water
pressures.

Checking resistance to failure by heave due to seepage of water in the ground

Checking the resistance to failure by heave due to seepage of water in the ground (HYD) is Clause 2.4.7.5(1)P
performed using either stresses or forces as the variables. Considering a column of soil in the
ground, through which water percolates vertically upwards, inequalities (2.9a) and (2.9b)
require that the design value of the destabilizing total pore pressure at the bottom of the
column or the seepage force in the column should not be greater than the stabilizing total
vertical stress at the bottom of the column or the buoyant weight of the column, respectively.
The recommended values for the partial factors to be applied in HYD checking in
persistent and transient design situations are given in Table A.17 of Annex A.
Details of HYD checking and measures to resist failure by heave due to seepage of water
are described in Chapter 10 of this guide and in Example 9.2.

2.4.6. Serviceability limit state design

Limit state design requires that the occurrence of serviceability limit states is sufficiently
improbable. Serviceability limit states may be checked in two ways:

37

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Table 2.1. Combinations of actions for checking serviceability limit states

Combination Use according to EN 1990

Characteristic Irreversible limit states

Frequent Reversible limit states
Quasi-permanent Long-term effect and appearance

Clause 2.4.8(1)P • by calculating the design values of the effects of the actions Ed (deformations,
differential settlements, vibrations, etc.) and comparing them with limiting values, Cd,
using inequality (2.10)
Clause 2.4.8(4) • by a simplified method, based on comparable experience.

Clause 2.4.8(2) Design values of actions and of material properties for checking the avoidance of
serviceability limit states will normally be equal to their characteristic values. In cases where
differential settlements are calculated, a combination of upper and lower characteristic
values of deformation moduli should be considered, to account for any local variations in the
ground properties.
Clause 2.4.9(1)P Ideally, the limiting values of deformations should be specified as design requirements for
Clause 2.4.9(3)P each supported structure, and the code lists a series of items to take into account when
establishing limiting values of movements. It is important that limiting values are established
as realistically as possible and in close cooperation with the designer of the structure.
Unnecessarily severe values usually lead to uneconomic design. Annex H gives some indications
of limiting values of differential settlements which can be used as guidelines in the absence of
specified limiting values of structural deformations.
Clause 2.4.8(4) As an alternative to serviceability checks by calculation, it may be shown in a simplified
method that a sufficiently low fraction of the ground strength is mobilized to keep
deformations within the required serviceability limit (see also clause 2.4.1(4)). This
simplified method requires the existence of comparable experience with similar soil and
structure. These requirements clearly restrict the circumstances in which the simplified
method may be applied to conventional structures and foundations in familiar ground
conditions.
Clause 6.4(5) The simplified method is applied in EN 1997-1 to spread foundations through the indirect
Clause 7.6.4.2(1) method, to pile foundations and to retaining structures. EN 1997-1 gives no indication of
Clauses 9.8.2(2)P what is a ‘sufficiently low fraction of ground strength’. Considerable experience exists of
to 9.8.2(4)P applying to the characteristic resistance global safety factors that are sufficiently high to
deal implicitly with the avoidance of serviceability limit states. It is probable that similar
experience with partial factors will increasingly develop in the future.
EN 1990 (clause 6.5.3) defines three combinations of actions for checking serviceability
limit states, and gives a note on their use (Table 2.1).
The combinations differ by the value of the y factors (see EN 1990, Annex A (normative),
Table A1.1: ‘Recommended values of y factors for buildings’).
EN 1990 refers to the relevant parts of EN 1991 to EN 1997 for more information on rules
of application (see the note to clause 6.5.3(4)). Unfortunately, EN 1997-1 does not give this
information. It is intended for EN 1990 and EN 1997-1 to apply the combinations for SLS
checking as follows:

• When serviceability limit states are checked by applying inequality (2.10) of clause
2.4.8(1), the frequent or quasi-permanent combinations are appropriate.
• When serviceability limit states are checked by applying the alternative method given
in clause 2.4.8(4), the characteristic combination is appropriate, as the comparable

38

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 2. BASIS OF GEOTECHNICAL DESIGN

experience required for the use of clause 2.4.8(4) has been built up with combinations of
actions close to the characteristic combination.

2.5. Design by prescriptive measures

Prescriptive measures should be applied in the context defined in EN 1997-1; an example is Clause 2.5
given in Annex G. The partial factors of Annex A are not intended to be applied when using Clause 1.5.2.2
prescriptive measures. It should be noted that charts and tables established from comparable
experience implicitly contain their own safety factor. A definition of comparable experience
is given in EN 1997-1. Very often, the concept of ‘allowable stress on the soil’ is used in
such charts or tables. Here, the comparable experience is related to serviceability limit
states, and ultimate limit states are considered to be implicitly covered by the serviceability
limit states.
Prescriptive measures can be applied in cases where calculation models are not available
or not appropriate. Examples of such prescriptive measures are often related to durability
(e.g. excess thickness to prevent the effects of corrosion loss) or in rules of good practice (e.g.
depth of footings below the frost penetration depth).

2.6. Observational method

EN 1997-1 introduces design by the ‘observational method’, in which the design is reviewed in Clause 2.7
a planned manner during the course of the construction and in response to the monitored
performance of the structure. The essence of the method is a precise plan of monitoring and Clauses 2.7(2)P
of the actions to be taken as a result of the observations. The minimum requirements to be to 2.7(4)P
met before and during construction are indicated. The advantage of this method is that it
facilitates design where a precise prediction of the geotechnical behaviour is difficult, e.g.
where the ground conditions are complex or not sufficiently well known. The observational
method allows ‘optimistic’ or ‘pessimistic’ assumptions to be checked by monitoring the
actual behaviour.
EN 1997-1 leaves open the manner in which safety is introduced in the supporting
calculations. This might be done through reduced values of partial factors or through a less
cautious selection of the characteristic values of the soil properties. The way to introduce
safety into design when using the observational method is best evaluated for each individual
project, depending on the perceived reasons for using factors of safety (uncertainty or
displacement control) and on the consequences of ‘failure’.
The observational method cannot be applied where a sudden collapse could occur
without warning, such as when ground and the interaction between ground and structure
are not sufficiently ductile (brittle behaviour).

2.7. Geotechnical Design Report

The design assumptions, data, methods of calculation and results of the checking of safety Clause 2.8(1)P
and serviceability should be recorded in a Geotechnical Design Report (GDR) for all Clause 2.8(2)
geotechnical designs, including small and relatively simple structures in straightforward Clauses 2.8(3)
ground conditions. The level of detail may vary greatly. For simple designs, a single-sheet Clause 2.8(4)P
report may be sufficient (Fig. 2.9). A checklist of items which should normally be included
in the GDR is given in EN 1997-1. The two most important parts of the GDR are:

• the Ground Investigation Report (see Chapter 3 of this guide)

• a plan for appropriate supervision and monitoring (see Chapter 4 of this guide).

A checklist is given of the items to be covered in the plan for supervision and monitoring. Clause 2.8(5)

39

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Fig. 2.9. Single-sheet GDR. (After Simpson and Driscoll, 1998)

40

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 2. BASIS OF GEOTECHNICAL DESIGN

Example 2.1: selection of a characteristic value using statistical methods

Description of the problem
This example illustrates how the characteristic value of the drained shear strength
parameters ϕk¢ and ck¢ can be selected from the sets of values obtained from triaxial tests.
The ultimate limit state of an embedded retaining wall is considered. This ultimate limit
state involves a large volume of soil. The characteristic value of the shear strength
parameters is then a cautious estimate of their mean value. In this example, it is assumed
that the variations of the shear strength are random, without significant local weaker
zones and without significant trend with depth.
The calculations are performed using the statistical formulae given in the appendix to
this chapter; different cases are treated with the corresponding statistical method and
their results are compared:
(1) There is no information available other than the results obtained from the triaxial test
results on four local samples from boreholes located on the site (local sampling).
Complementary to the results, it is assumed that there is reliable knowledge of the
variability of c¢ and tan ϕ¢ through their coefficients of variation.
(2) Instead of analysing c¢ and tan ϕ¢ separately, an analysis is performed on the shear
resistance using t as a linear function of s¢: the characteristic mean value at the 95%
confidence level for a linear trend between t and s¢ is determined, from which the
values of ck¢ and tan ϕk¢ can be obtained.
Note:
s¢ = (σv¢ + σh¢)/2 t = (σv¢ – σh¢)/2 sin ϕk¢ = ∆t/∆s¢ c¢ = t(0)/cos ϕ¢

Evaluation using results of ϕ and c from tests with local samples only (case ‘VX
unknown’)
As only the information obtained from the results of the triaxial test is used in this first
calculation, the method outlined in the appendix to this chapter is applied for
homogeneous soil without significant trend and the case ‘VX unknown’. Table 2.2 gives the
mean value and the standard deviation of the values of c¢ and tan ϕ¢ estimated from the
four triaxial test results.
From Table 2.2 the characteristic value of tan ϕ¢ is calculated using equation (D2.1).
For a characteristic value having a reliability of 95%, Table 2.5 in the appendix (on the
basis of four tests, n = 4, and ‘VX unknown’) gives kn, mean = 1.18:
tan ϕk¢ = (tan ϕ¢)mean(1 – kn, meanVtan ϕ)
tan ϕk¢ = 0.603 × (1 – 1.18 × 0.118) = 0.519
tan ϕk¢ = 27.5°

Table 2.2. Mean values of c¢ and tan ϕ¢, their standard deviation and coefficient of variation
obtained from four triaxial test results

c¢ (kPa) ϕ¢ (°) tan ϕ¢ (–)

Borehole/test
BH 1/1 3 31 0.601
BH 1/2 4 30 0.577
BH 2/1 1 35 0.700
BH 2/2 7 28 0.532
Statistical results
Mean value cmean = 3.75 (tan ϕ¢)mean = 0.603
Standard deviation σc = 2.50 σϕ = 0.071
Coefficient of variation Vc = 0.667 Vtan ϕ = 0.118

41

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

The calculation for the characteristic value of ck¢ gives

ck¢ = cmean(1 – kn, meanVc)
ck¢ = 3.75 × (1 – 1.18 × 0.667)
ck¢ = 0.8 kPa
In Fig. 2.10 the corresponding Mohr’s envelope is presented together with the results of
the triaxial tests (values of t and s¢ at failure).

Evaluation using results of ϕ and c assuming ‘VX known’

It should be noted that the use of ‘VX known’ is new and has not yet gained general
acceptance in the geotechnical engineering profession. The method is presented here
because it may be of some help to the engineer; but it should be used with caution.
Schneider (1999), using results from Rethati (1998) and Lumb (1974), presents typical
values for the coefficient of variation VX for soil properties, which are generally valid and
of which the magnitudes have been confirmed by many researchers world-wide. Typical
values of VX for the tangent of the angle of shearing resistance and for the cohesion range
from 0.05 to 0.15 and 0.3 to 0.5, respectively. Schneider (1999) recommends average
values of VX = 0.1 for the angle of internal friction and VX = 0.4 for cohesion. These
values are used to select the characteristic mean values with a reliability of 95% for tan ϕ¢
and c¢ from the four tests results, applying the case ‘VX known’.
When four tests results are available (n = 4) Table 2.6 in the appendix gives:
kn, mean = 0.82 for ‘VX known’. Using equation (D2.1) the characteristic value tan ϕk¢ is
tan ϕk¢ = (tan ϕ¢)mean(1 – kn, meanVtan ϕ)
tan ϕk¢ = 0.603 × (1 – 0.82 × 0.10) = 0.554
tan ϕk¢ = 29°
Similarly, the characteristic value ck¢ can be calculated using equation (D2.1):
ck¢ = cmean(1 – kn, meanVc)
ck¢ = 3.75 × (1 – 0.82 × 0.4)
ck¢ = 2.5 kPa

350

300

250
t = ( ¢v –s ¢h)/2
s

200

150

100
s ¢–t test results
s ¢–t characteristic
50 j¢k and c ¢k 'Vx known'
j¢k and c ¢k 'Vx unknown'
0
0 200 400 600 800
s ¢ = (s ¢v + s ¢h)/2

Fig. 2.10. Results of triaxial tests and their statistical evaluations with respect to Mohr´s
envelopes for the characteristic shear parameters

42

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 2. BASIS OF GEOTECHNICAL DESIGN

Table 2.3. Statistical analysis of the values of t as a function of s¢ according to equations (D2.2)
and (D2.3) in the appendix to this chapter

Test results Statistical analysis

t*(s¢) from linear s1 t (0n.95– 2 ) s1 t k = t * – t (0n.95– 2 ) s1

s¢ (kPa) t (kPa) regression (kPa) (kPa) (kPa) (kPa)

70.0 40.0 40.8 4.5 8.2 32.6

90.0 52.0 51.0 4.2 7.7 43.3
100.0 55.0 56.1 4.1 7.5 48.6
120.0 64.0 66.3 3.9 7.0 59.3
200.0 105.0 107.0 3.1 5.6 101.4
225.0 130.0 119.8 3.0 5.4 114.4
240.0 121.0 127.4 2.9 5.3 122.1
250.0 134.0 132.5 2.9 5.3 127.2
400.0 231.0 208.9 3.8 6.9 202.0
420.0 201.0 219.1 4.0 7.3 211.8
450.0 229.0 234.3 4.4 8.0 226.3
600.0 312.0 310.7 6.7 12.1 298.6

Coefficients of linear regression:

Intercept: c¢= 5.2 kPa
Slope: sin ϕ¢= 0.51, ϕ¢= 30.5°

In Fig. 2.10 the corresponding Mohr’s envelope is presented together with the results of
the triaxial tests (values of t and s¢ at failure).

Evaluation using the values of s and t at failure

In the analysis above, c¢ and tan ϕ¢ have been analysed separately. However, one observes
that they are negatively correlated, i.e. the larger c¢ values correspond to the smaller ϕ¢,
and vice versa. Analysing c¢ and tan ϕ¢ separately is conservative. Starting from the
assumption of a linear relation between effective normal stress s¢ and shear strength t, one
can, alternatively, plot all s¢– t points obtained from the triaxial tests (see Fig. 2.10) and
determine the characteristic value of the angle of shearing resistance, ϕk¢, by using the
method for local sampling, with a linear trend of the shear strength t with s¢.
Equation (D2.3) permits the determination of the characteristic value tk as a function of
s¢, and this is applied in Table 2.3, where the parameter z stands for s¢ and parameter Xk
stands for tk. The factor t (0n.95– 2 ) is determined using Table 2.6 of the appendix, which,
for n = 12 test results and r = n – 2 = 10, gives t (0n.95– 2 ) = 1.812. Table 2.3 indicates the
parameter values for linear regression through the measured s¢ and t values (zero
intercept and slope of the linear regression of t* for s¢), the value of the term s1 (applying
equation (D2.2)) and the characteristic value tk as a function of s¢. It should be noted that
the relationship between s¢ and tk is slightly hyperbolic due to the non-linear term s1. The
distance between the linear regression and the characteristic value is smallest in the
middle of the stress interval and increases slightly towards its beginning and end.
The characteristic values of ck¢ and tan ϕk¢ may be deduced by linearizing the relation
s¢–tk in the stress interval relevant for the problem. As sin ϕ¢ = ∆t/∆s¢ and
c¢ = t(s¢ = 0)/cos ϕ¢, for the stress interval s¢ between 200 and 500 kPa (see Fig. 2.10):
tan ϕk¢ = 0.5
tan ϕk¢ = 30°

tan ck¢ = 1 kPa

43

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Table 2.4. Summary of statistical evaluations of the results of the triaxial tests

Characteristic values of shear parameters

Basis and method of statistical evaluation ϕk¢ (°) ck¢ (kPa)

ϕ¢ and c¢ of four tests for the case ‘VX unknown’ 27.5 0.8
ϕ¢ and c¢ of four tests for the case ‘VX known’ 29.0 2.5
Schneider (1999) 29.5 2.5
Evaluation of values of s¢ and t of 12 tests 30.0 1.0

In the present example, the characteristic value is close to the linear regression line
through all the s¢–t values. This is due to the fact that the results show quite small
variability.

Discussion
A summary of the statistical evaluations of the characteristic values of the shear strength
parameters ϕk¢ and ck¢ is presented in Fig 2.10 and Table 2.4. The characteristic values
based on local test results for ϕ¢ and c¢, without any further knowledge (case ‘VX
unknown’), are in this example close to the smallest values of the test results (see Table
2.2). This is due to the small number of samples and the rather large variability of the tests
results, especially for c¢.
The introduction of knowledge of the coefficient of variation VX (case ‘VX known’)
has significant influence on the calculated characteristic value. The introduction of
complementary information is especially relevant when few test results are available or
when the variability is rather large.
Schneider (1999) proposed the following simplified equation, based on the assumption
that there is always certain knowledge of the coefficient of variation of the soil
parameters:
Xk = Xmean – 0.5s
where Xmean and the standard deviation s are deduced from values for the local samples.
Application of Schneider’s equation to the values of Table 2.2 yields results that are in
close agreement with the values obtained for the case ‘VX known’.
The analysis of c¢ and tan ϕ¢ as independent variables is generally over-conservative for
evaluating the characteristic shear resistance of the soil. When c¢ and tan ϕ¢ are not
correlated, advantage can be taken from an analysis where all results are treated together
in an s¢–t linear relation.

Appendix: an example of the use of statistical methods to

assess characteristic values
General
The statistical determination of a material property is introduced in Annex D (informative)
of EN 1990, ‘Basis of structural design’. Statistical terms and techniques are explained in
Appendix C of the Designers’ Guide to EN 1990, and are not repeated here. Appendix C also
gives detailed information on the way characteristic values can be estimated as a fractile,
according to the recommendations of EN 1990.
The main difference between a characteristic value of a property of a structural material
and of the ground is that, in the former, the characteristic value is often defined as a (5%)
fractile while in the latter the characteristic value is usually an estimate of the mean value

44

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 2. BASIS OF GEOTECHNICAL DESIGN

Geotechnical
parameters or
derived values

• Soil volume
• Structural
stiffness

Cautious mean value Mean value Low local values

Local low value

Significant No
trend

No Useful
Significant No
experience
trend
(regional)
Useful
No
experience Yes
(regional)

Yes Yes Full knowledge

Yes of variation
coefficient
Full knowledge No
of variation
Yes No
coefficient

Yes
Mean value Mean value at Bayesian Mean value 5% fractile, 5% fractile, Bayesian 5% fractile,
at 95% 95% confidence analysis at 95% linear V known analysis V unknown
confidence level, V unknown confidence variation
level, linear level,
variation V unknown

Fig. 2.11. Flow chart to formulae for the assessment of characteristic value of a ground property,
starting from a local sampling programme.

with a probability of 95% that the mean value governing the occurrence of a limit state in the
ground is more favourable than the characteristic value. This appendix can be considered as
an extension of Annex D (informative) of EN 1990 for the selection of characteristic ground
property values for geotechnical design; this appendix also complements Chapter 10 of this
guide and Appendix C of the Designers’ Guide to EN 1990.
Statistical methods to assess characteristic values of ground parameters should take into
account:

• the type (local and regional) and extent (number of results) of the sample population
and the related statistical uncertainty
• the variability of the sample test results, the rate of value fluctuation in relation to the
ground volume sampled and the ability of the structure to redistribute loads
• significant trends in the sample test results
• the prior, statistically quantified parameter knowledge, if available
• the required level of confidence in the characteristic value.

The flow chart in Fig. 2.11 brings together all the major items affecting the selection of the
characteristic value of a ground parameter, starting from a local sampling programme (i.e.
the test results are obtained from tests at the site considered) in homogeneous soil,
supplemented by complementary information, if any. The path leads to appropriate
statistical formulations for the assessment of characteristic values relevant for current daily
practice. Statistical formulae for some simple cases, corresponding to some of the bottom

45

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

blocks of the flow chart, will be given in the following sections. Bayesian analysis is not
treated further.
The statistical formulae are valid for ‘homogeneous’ soil. The formulae assume a normal
distribution of the value of the geotechnical parameter. They can also be applied to
log-normally distributed parameters by applying the formulae after transforming the
parameter value X to its logarithm in Y = ln X. For processes governed by extreme values of
soil properties, analysis of a sample consisting of only the extreme values is much more
appropriate than an analysis based on all the values of test results.

Homogeneous soil with local sampling and without a significant trend in the
ground parameter
A normal distribution of the value of the geotechnical parameter can be assumed, although
this assumption is not always valid; in some cases, a log-normal distribution is preferred,
especially when the coefficient of variation is large.
For soil layers where the geotechnical parameter does not show a significant, systematic
trend in either the horizontal direction or with depth, the characteristic value Xk of the
parameter can be assessed from a set of individual parameter values according to
Xk = Xmean(1 – knVX) (D2.1)
where
Xmean is the arithmetical mean value of the individual sample parameter values
XA = Â X i /n
VX is the coefficient of variation of the parameter X
kn is a statistical coefficient taking account of the following:
– the number of test results (n) of the sample;
– the volume of ground involved in the limit state in question, related to length
of the fluctuations (often denoted ‘auto-correlation length’) of the material
property;
– the type of sample population: local sample only or local sample together with
relevant experience in the soil layer considered;
– the statistical level of confidence required for the assessed characteristic
value.
Two extreme situations may be considered for the coefficient of variation VX:
(1) VX is not known a priori but must be estimated from the n test results of the local sample
only. This case is denoted as ‘VX unknown’. The value of VX to be introduced in equation
(D2.1) is estimated as:
VX = sX/Xmean
where sX is the standard deviation of the n sample test results:
1
sX2 =
n-1
 ( X i - X mean )2
(2) The coefficient of variation VX is known a priori. This case is denoted as ‘VX known’. A
priori knowledge can be found from the evaluation of previous tests, from databases or
from published tables of coefficients of variations of ground properties in comparable
situations; a ‘comparable situation’ is determined by engineering judgement (see also
clause 1.5.2.2). In this case, the value of VX in equation (D2.1) is the a priori known value,
and does not relate to the sample. Xmean is the sample mean value.
In practice, there is often some prior knowledge of the coefficient of variation; it is
recommended (see also EN 1990) to use ‘VX known’ together with a conservative upper
estimate of VX rather than applying the rules given for ‘VX unknown’. As the number of

46

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 2. BASIS OF GEOTECHNICAL DESIGN

Table 2.5. Values of the coefficient kn, mean for the assessment of a
characteristic value as a 95% reliable mean value

n VX unknown VX known

3 1.69 0.95
4 1.18 0.82
5 0.95 0.74
6 0.82 0.67
8 0.67 0.58
10 0.58 0.52
20 0.39 0.37
30 0.31 0.30
• 0 0
1 1
kn k n, mean = tn0.95
-1 k n, mean = 1.64
n n

Table 2.6. Values of the t factor of Student’s distribution

tn – 1 and tn – 2

r p = 95% p = 90%

2 2.920 1.886
3 2.353 1.638
4 2.132 1.533
5 2.015 1.476
6 1.943 1.440
7 1.895 1.415
8 1.860 1.397
9 1.833 1.383
10 1.812 1.372
12 1.782 1.356
15 1.753 1.341
20 1.725 1.325
25 1.708 1.316
30 1.686 1.310
• 1.645 1.282

test results is usually small in geotechnical design, advantage should be taken of a priori
information whenever possible.
When the soil volume involved in the limit state considered is very large compared with
the length of the fluctuations (auto-correlation length) of the soil property, and when it may
therefore be assumed that the behaviour is governed by the mean value of the ground
parameter, the characteristic value should be a cautious estimate of the mean value. In this
case the characteristic value Xc, mean represents an estimated value corresponding to a 95%
confidence level that the (unknown) mean value of the population governing the occurrence
of a limit state in the ground is more favourable than the calculated characteristic value Xk
derived from the sample and previous knowledge.
The basic equations for the cases ‘VX unknown’ and ‘VX known’ are equation (D21) and
equation (D20), respectively, in Appendix C of the Designers’ Guide to EN 1990. The
corresponding values of kn, mean to be used in equation (D2.1) are given in Table 2.5 for the
case ‘VX unknown’ (second column) and for the case ‘VX known’ (third column). The last row
in Table 2.5 gives the full equation for kn.
The factor t n0 .–951 is the t factor of Student’s distribution with n – 1 degrees of freedom and
confidence of 95%, tabulated in Table 2.6. The values tn – 1 and tn – 2 can be read from Table

47

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Table 2.7. Values of the coefficient kn, low for the assessment of a
characteristic value as a 5% fractile

n VX unknown VX known

3 3.37 1.89
4 2.63 1.83
5 2.33 1.80
6 2.18 1.77
8 2.00 1.74
10 1.92 1.72
20 1.76 1.68
30 1.73 1.67
• 1.64 1.64
1 1
kn k n, low = tn0.95
-1 +1 k n, low = 1.64 +1
n n

Table 2.8. Overview of cases and the values of the coefficient kn to apply

Coefficient of variation

Estimated from results Known from

Characteristic value of the sample only: published data:
looked for Statistical formulation case ‘VX unknown’ case ‘VX known’

Cautious estimate of 95% reliable estimate kn, mean kn, mean

the mean value of the mean value
Cautious estimate of 5% fractile kn, low kn, low
the low local value in
the ground

2.6, putting r = n – 1 and r = n – 2, respectively. The values are given for p = 95% (for 95%
reliable mean value and 5% fractile) and for p = 90% (for 90% reliable mean value and 10%
fractile).
When the soil volume involved in the limit state considered is very small compared with
the length of the fluctuations of the soil property, or when it may be assumed that the
behaviour is governed by the local low value, the characteristic value Xlow should be selected
as a 5% fractile. In this case, there will be only 5% probability that somewhere in the layer
considered there is an element of soil having property values lower than the characteristic
value. The values of kn, low to be used in equation (D2.1) are given in Table 2.7 for the case
where the coefficient of variation is unknown (second column) and for the case where
the coefficient of variation is known (third column). (Tables from EN 1990, Annex D
(informative), Table D1; see also Designers’ Guide to EN 1990, Appendix C, equations (D37)
and (D38) and Tables 2 and 3).
Table 2.8 summarizes the different cases and indicates which value of the coefficient kn to
apply.
‘Large’ and ‘small’ soil volumes are related to the auto-correlation length of the parameter;
this is the length over which the value of the parameter does not vary significantly (the rate of
fluctuation). When the auto-correlation length is small compared with the dimensions of the
soil volume, low and high local values compensate. When the auto-correlation length is large
compared with the dimension of the soil volume involved in the limit state, a large part of the
soil is possibly located in the weaker part; hence a ‘mean’ value might be too optimistic. A
characteristic value somewhere between the 5% fractile and the 95% reliable mean value
may be selected.

48

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 2. BASIS OF GEOTECHNICAL DESIGN

When ground shows marked strain softening behaviour or brittle failure, statistics should
be used cautiously when selecting the characteristic value of the strength: when the strength
is exceeded locally there is a local loss of any resistance. Due to the absence of ductility,
‘averaging’ of weak and strong spots is limited or even does not occur, and the characteristic
value is usually determined to be close to the lowest values of the test results.

Comment on introducing ‘previous knowledge’ by applying the formulae for ‘VX known’
The formulae for the case ‘VX known’ assume that there is no prior knowledge about the
value of the mean and that there is full knowledge of the coefficient of variation. These
assumptions are usually valid for geotechnical parameters because:
• the mean value of a soil property may fluctuate significantly from one site to another
• the value of the coefficient of variation of a soil property has been found by many
researchers world-wide to lie within narrow ranges (e.g. see Schneider, 1999) – which
means that the coefficient of variation can be estimated from published tables,
databases, etc., and can be applied to the parameter values found at the site.
Thus, it is assumed that, even if the mean value and the standard deviations are unknown,
the coefficient of variation may be known. This implies that:
• the characteristic mean value (or fractile) at a given site is merely governed by the
estimate of the mean value on that site through the sample mean, and not by some
overall ‘known mean value’ from, for example, a database
• the (estimate of) the standard deviation sX of the sample at a given site is not introduced
in the calculation of the characteristic value, but is assumed to be related to the estimate
of the mean value on that site through the ‘overall’ coefficient of variation, of which the
value is known from previous knowledge.
When analysing the local geotechnical data, care should be taken if there is strong and
clear evidence that the standard deviation sX of the parameter X on the site differs
significantly from the standard deviation (VX, knownXmean) obtained with the a priori known
‘overall’ coefficient of variation VX, known and the mean value Xmean.

Comment on characteristic values of ground parameters when a high value is unfavourable

In some design situations, high values of the ground parameters may be unfavourable. In
such cases, the characteristic values can be assessed by using equation (D2.1) in which the
term (1 – knVX) is replaced by the term (1 + knVX):
Xk = Xmean(1 + knVX)

Comment on log-normal distributions

The very common normal distribution may often be used to approximate many symmetrical
bell-shaped distributions. Adopting a log-normal distribution (if a parameter X is log-
normally distributed then Y = ln X becomes normally distributed) has the advantage that no
negative value can occur, which is physically correct. If a log-normal distribution is adopted,
the formulae above may be used, with Y = ln X being introduced instead of X. More details
can be found in EN 1990, Annex D (clause D7.2(3), note 2), and in the Designers’ Guide to
EN 1990, Chapter 4 (Appendix 2) and Appendix C. The kn values are those of Table 2.5 or
Table 2.7.

Homogeneous soil, with local sampling and a linear trend in the ground
parameters
The characteristic mean value Xk at the 95% confidence level for a linear trend can be
derived from the best estimate (x) of the ground parameter at a depth (z) following Student’s
t distribution with (n – 2) degrees of freedom, a mean value equal to the true mean of the
ground parameter at this depth, and a standard deviation from

49

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Ê ˆ
1 Á 1 (z - z) ˜ n
2
s1 = Á + n ˜ Â [( xi - x ) - b( zi - z )]2 (D2.2)
n-2 Á n 2 ˜ i =1
Á
Ë
 ( zi - z ) ˜¯
i =1

Thus, the characteristic of the mean value of (X) at depth (z) is

X k = [ x + b( z - z )] - tn0.95
- 2 s1 (D2.3)

x* = x + b( z - z ): linear regression
where
1
x= ( x1 + x2 + º + xn )
n
1
z= ( z1 + z2 + º + zn )
n
n

Â( x i - x )( zi - z )
i =1
b= n

Â(z i - z )2
i =1

The calculated characteristic values are no longer linear functions of depth but hyperbolic
functions, due to the term tn0.95
- 2 s1 . For tn – 2 values, see Table 2.6.
The distance between the linear regression and the characteristic value is smallest at the
centre of gravity of the measurements x and increases towards the beginning and the end of
the interval of the measurements. This indicates the advantage of performing tests in the
relevant problem interval and slightly outside it.
As many calculation methods and computer programs use linear relationships, one has to
linearize the hyperbolic relation for Xk in the relevant problem stress interval. Errors
involved with that linearization are usually small.
For the local low value, the 5% fractile can be derived from the difference between the
local value of the ground parameter and its best estimate at a depth (z), following Student’s
t distribution with (n – 2) degrees of freedom, mean value equal to zero and standard
deviation:

Ê ˆ
1 Á 1 (z - z) ˜ n
2
s2 = Á1 + + n ˜ Â [( xi - x ) - b( zi - z )]2 (D2.4)
n-2 Á n 2 ˜ i =1
Á
Ë
 ( zi - z ) ˜¯
i =1

Thus, the characteristic of the local value of (X) at depth (z) is

X k = [ x + b( z - z )] - tn0.95
- 2 s2

In some situations, one can apply statistical formulae directly to the parameters of the
regression line (zero intercept or slope, or both). Thus, instead of applying equations (D2.2)
or (D2.4) and (D2.3) above, the linear regression line can be treated statistically as a whole.
These methods allow the imposition of some conditions to the linear regression line; such
conditions can, for example, result from previous experience (e.g. condition on the slope,
or condition on the zero intercept). Formulations can be found in Van Aalboom and
Mengé (1999).
The example on p. 41 of this guide illustrates the application of equations (D2.3) and
(D2.2) to the results of triaxial tests.

50

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 2. BASIS OF GEOTECHNICAL DESIGN

Regional sampling
Characteristic values can also be deduced from regional samples. This is beyond the scope of
this guide, but an introduction can be found in Bauduin (2002a).

51

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 3

Geotechnical data

This chapter is concerned with the acquisition, evaluation, assessment and reporting of
geotechnical data. The structure of the chapter follows that of Section 3 of EN 1997-1:
3.1. Introduction Clause 3.1
3.2. Geotechnical investigations Clause 3.2
3.3. Evaluation of geotechnical parameters Clause 3.3
3.4. Ground Investigation Report Clause 3.4

3.1. Introduction
The reliability of every geotechnical design depends on the quality of the geotechnical
investigations and the correct interpretation of the data obtained. Hence, Section 3
of EN 1997-1, which provides the general requirements for planning geotechnical Clause 3.1(1)P
investigations and assessing geotechnical parameter values, emphasizes that the careful
collection, recording and interpretation of geotechnical information must always be
undertaken. The specific requirements for field and laboratory tests and for the
determination of derived geotechnical parameter values from these are provided in Clause 3.1(4)
EN 1997-2. The concept of derived geotechnical parameter values, which is introduced in
Part 2 of the code, is explained in this chapter.
As EN 1997-1 is a standard for geotechnical design, it does not provide standards for field
or laboratory tests. Standards for the most common geotechnical field and laboratory tests,
for the identification and classification of soils and for the identification and description
of rock, are being prepared by CEN Technical Committee 341 and ISO Technical
Committee 182, as noted in Section 1.2 of this guide. Designs to EN 1997-1 should be based
on geotechnical data obtained from tests carried out and reported generally in accordance Clause 3.1(3)P
with internationally recognized standards and recommendations.

3.2. Geotechnical investigations

EN 1997-2 defines the following hierarchy of investigations:
• geotechnical investigations, which are ground investigations and other information
about the site
• ground investigations, which are field investigations, laboratory testing and desk studies
of geotechnical and geological information
• field investigations, which are direct investigations (drilling, sampling and trial pits) and
indirect investigations (in situ tests, such as the cone penetrometer test (CPT)).

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Clause 3.2.1(1)P As stated in clause 3.2.1(1)P, geotechnical investigations must provide sufficient data
concerning the ground and groundwater conditions at, and around, the site to enable a
proper description of the essential ground properties and a reliable assessment of the
characteristic values of the geotechnical parameters to be used in design calculations.
Hence, Eurocode 7 – Part 2 states that the objectives of ground investigations are to provide
a full description of all ground conditions relevant to the proposed works and to establish the
Clause 3.2.1(4) relevant geotechnical parameters. The character and extent of a geotechnical investigation
depend on the complexity of the ground conditions. When geotechnical categories are being
used, the ground conditions should be determined as early as possible as these are related
to, and may affect, the choice of geotechnical category. Three phases of geotechnical
investigation are considered in EN 1997-1:
• preliminary investigations
• design investigations
• control investigations.

Preliminary investigations
Clause 3.2.2(1)P Preliminary investigations are investigations carried out during the planning or feasibility
stage of a project with the objective to assess the general suitability of a site, compare
alternative sites, if relevant, plan the design and control investigations, and identify borrow
areas, if relevant. Preliminary investigations normally include:
• desk studies of geotechnical and geological information about the ground conditions,
including reports of previous investigations in the vicinity
• field reconnaissance (walk-over surveys)
Clause 3.2.3(1)P • consideration of construction experience in the vicinity.

Design investigations
Design investigations are the main geotechnical investigations carried out to obtain the
geotechnical data required for an adequate design of both the temporary and permanent
Clause 3.2.3(6)P works. Design investigations are also carried out to provide the information required for
construction and to identify any difficulties that may arise during construction. Design
investigations normally include in situ testing and sampling of the ground for laboratory
tests. Design investigations should be carried out at least through all the formations which
are considered likely to be relevant to the particular design. Some guidance on the spacing
and depth of investigations is provided in EN 1997-2.

Control investigations
Control investigations are inspections carried out during the construction phase of a project
Clause 4.2.2(1)P to check and compare the actual ground conditions encountered with those assumed in
the design. The clauses in Section 3 are only concerned with preliminary and design
Clause 4.2.3(1)P investigations; there are no clauses in this section concerned with control investigations. The
general requirements for inspections during construction are presented in Section 4, and
include the need to inspect the ground on a continuous basis and the need to record the
results of the inspections. In control investigations, additional tests may need to be carried
out to test that the ground conditions encountered, the quality of delivered construction
materials and the construction work correspond to those assumed in the design. The design
must then be assessed on the basis of the results of the inspections and the tests.

3.3. Evaluation of geotechnical parameters

3.3.1. General
Clause 2.4.3(1)P According to clause 2.4.3(1)P, the properties of soil and rock masses for use in design
calculations are quantified by geotechnical parameters, which are obtained from test results,

54

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 3. GEOTECHNICAL DATA

either directly or through correlation, theory or empiricism, and from other relevant data.
Clause 3.3 presents the general requirements for assessing geotechnical parameters from the Clause 3.3.1(1)P
most commonly used laboratory and field tests. Further requirements for evaluating
geotechnical parameters from particular tests are given in EN 1997-2. No specific standards
for carrying out the different tests are mentioned in EN 1997. Testing standards for the most
common soil tests are being prepared by CEN. Other tests may be used, provided that their
suitability has been demonstrated through comparable experience.

3.3.2. Characterization of soil and rock type

Prior to setting up a test programme to obtain the geotechnical parameters required for a
particular design, the stratigraphy of the site should be established. The character and basic
constituents of the soil or rock relevant for the design must be identified before the results of Clause 3.3.2(1)P
other tests are interpreted. The material must be examined, identified and described Clause 3.3.2(2)P
in accordance with a recognized nomenclature. Soils should be classified and described Clause 3.3.2(3)
according to an acknowledged geotechnical classification and description system. No
particular geotechnical classification and description system is mentioned in Eurocode 7,
but a new international standard, ISO 144688, Identification and Classification of Soil,
covers this.
EN 1997-2 gives the requirements for the use of routine classification tests, which
include soil class, water content, density, Atterberg limits, grain size distribution, undrained
shear strength and sensitivity. Guidance is offered on the suitability of these routine
classification tests for samples of different types of soil and for various degrees of sample
disturbance.

3.3.3. Procedure for evaluating geotechnical parameters

The procedure for evaluating the design values of geotechnical parameters from the results of
field and laboratory tests is a three-step procedure, and normally involves proceeding from the
measured values obtained from field and laboratory tests to the design values used in
calculations, via test results, geotechnical parameter values and characteristic values. The first
two steps, including the various aspects that need to be considered and the various factors, such
as calibration and correction factors, that need to be applied during these steps in order to
obtain the appropriate characteristic value from measured values are shown in Fig. 2.3 in
Chapter 2 of this guide, and are discussed below.
Step 1, going from measured values to geotechnical parameter values, is concerned with
evaluating and assessing the properties of soil and rock masses at the particular locations of the
field tests or laboratory samples, and does not involve taking account of the variability of
the soil properties or the design situation. The requirements of Step 1 are covered by clauses
2.4.3 and 3.3 of EN 1997-1 and by the clauses of EN 1997-2, and are the main concern of this Clause 2.4.3
chapter. Steps 2 and 3, going from the geotechnical parameter values to the characteristic Clause 3.3
value, and from the characteristic value to the design value, involve consideration of the design Clause 2.4.3
situation, and are covered by clauses 2.4.3, 2.4.5.2 and 2.4.6.2 of EN 1997-1, and are discussed Clause 2.4.5.2
in Chapter 2. Clause 2.4.6.2

Measured values and test results

A measured value is defined in EN 1997-2 as a value measured in a test. Often, corrections
need to be applied to the measured values in order to obtain the test results used to
determine the required geotechnical parameter values. These corrections are test related,
and are unrelated to the design situation for which the test results will be used. Examples of
test-related corrections are provided in EN 1997-2. One example of a test-related correction
is the correction to the stresses measured in triaxial tests to account for the effects of the
membrane. Another example is the correction factor applied to the measured blow count, N,
in a standard penetration test in sand to account for the energy delivered to the drive rods
and for the effect of the overburden pressure.

55

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Examples of test results include:

• the blow count, N, recorded in an standard penetration test
• the limit pressure, pLT, measured in a Ménard pressuremeter test
• the cone penetration resistance, qc, measured in a CPT
• the stresses, e.g. σ1¢ and σ3¢, s¢ and t, or p¢ and q, and deformations measured in a triaxial test
• the deformations measured in a triaxial test
• the pile bearing resistance, R, measured in a pile load test
• the Atterberg limits obtained from index tests.
Before evaluating the geotechnical parameter values from test results, and in order to
interpret them appropriately for the actual limit state being considered, it is necessary to
Clause 2.4.3(2)P select the test results which are relevant for the design. Depending on the limit state, these
may be the peak or constant-volume strength parameters.

Directly obtained and derived values

Clause 2.4.3(1)P As noted in Section 3.3.1 above, clause 2.4.3(1)P states that the values of geotechnical
parameters are obtained from test results, either directly or through correlation, theory or
empiricism, and from other relevant data. An example of a geotechnical parameter value
obtained directly from a test result without the use of correlation, theory or empiricism is the
pile bearing resistance obtained from a pile load test. In this situation, the geotechnical
parameter value is the measured pile resistance. Another example of a geotechnical
parameter used in design calculations that is obtained directly from test results, without the
use of correlation, theory or empiricism, is the weight density.
Clause 1.5.2.5 In most design situations, however, geotechnical parameters are obtained from test
Clause 2.4.5.2(1)P results using correlation, theory or empiricism. These values are defined in EN 1997-1 as
derived values, and form the basis for the selection of the characteristic values of geotechnical
parameters. The term ‘derived value’ is hardly used in EN 1997-1, but it has been found
useful in EN 1997-2 to describe the geotechnical parameter values obtained from the results
of the most common field and laboratory tests using the different correlations, theories, and
empirical relationships presented in EN 1997-2.
The use of different types of test can result in different derived values of the same
parameter being obtained at the same location. The example is given in EN 1997-2 of
different cu values being obtained by the following three different test methods:
• from CPT test results through correlations with qc
• from Ménard pressuremeter test results through correlations with pLM
• from laboratory triaxial test results evaluated using Mohr–Coulomb theory.
It should be noted that a derived value is the value of a geotechnical parameter, such as the
soil strength, resistance or stiffness, determined from the results of field or laboratory tests
on soil or rock at or from one particular point in the ground or from field test measurements
at one particular location on a site. It does not take account of the variability of the ground,
the scatter of the test results, the number of test results or the design situation, including the
volume of ground involved. Examples of derived parameters include:
• c¢ and ϕ¢ values obtained from triaxial test results using Mohr–Coulomb theory
• Em values obtained from measured triaxial stress-strain curves
• cu values derived from field vane test values (cfv) using an emperical correction factor
• cu values derived from index properties using empirical relationships
• ϕ¢ and Em values obtained from standard penetration test blow-counts using correlations
• ϕ¢ values obtained from CPT qc values using correlations
• R (bearing resistance) values obtained from measured pLM values and empirical
relationships.
When reporting derived geotechnical values, it is important to clarify how the derived
values have been obtained and to state the assumptions made. Also, when using correlations

56

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 3. GEOTECHNICAL DATA

to determine derived geotechnical parameter values from test results, it is important to

determine if the correlation is based on the mean value or on a conservative estimate of the
test values. Correlations should preferably relate to mean test results. When conservative
correlations are used, an unknown safety margin is introduced. A very cautious correlation
should be chosen when using standard tables relating geotechnical parameters to test results.

Assessment of geotechnical parameter values

Clause 2.4.3(3)P states that account should be taken of possible differences between Clause 2.4.3(3)P
the ground properties and geotechnical parameters obtained from test results and those
governing the behaviour of the geotechnical structure. These differences can be due to a
number of factors, such as stress level, time effects and construction activities, which are Clause 2.4.3(4)
listed in clause 2.4.3(4). The subclauses of clause 3.3 list the various factors and features Clause 3.3
that should either be considered or taken into account when assessing the values of the
geotechnical parameters obtained from particular field and laboratory tests. The factors to
be considered when assessing the other geotechnical information, such as the classification Clause 3.3.6(1)
and description of soil and rock, and the quality of rock and rock masses, are also provided.
Examples of the factors that need to be considered when assessing the shear strength of
soil include the following:
• the stress level imposed on the soil
• anisotropy of strength
• strain rate effects
• sample disturbance
• the level of confidence in the theory used to obtain the derived value.
Generally, the features that need to be considered or taken into account when assessing
the derived values of geotechnical parameter values are simply listed in clause 3.3 of Clause 3.3
EN 1997-1, without explicit guidance being given there as to how these features are to be
considered or taken into account.

Calibration and correction factors

According to clause 2.4.3(6)P, calibration factors must be applied where necessary to convert Clause 2.4.3(6)P
the parameter values obtained from test results into values representing the behaviour of the
soil or rock in the ground, for the actual limit state and to take account of correlations used to
obtain derived values from test results.
An example of a calibration factor is the correction factor which must be applied to the cfv Clause
values measured in field vane tests in order to derive the appropriate undrained shear 3.3.10.3(3)P
strength values of cu corresponding to those obtained from comparable experience and from
the back-analyses of slope failures. This factor depends on the liquid limit, plasticity index
and vertical effective stress, and examples of correction factors are given in EN 1997-2 for
normally consolidated and over-consolidated clays.
Another example of a correction factor is the factor of 1.1 applied to the ϕ¢ values
obtained from triaxial tests to convert them from axi-symmetric to plane strain conditions.

Other relevant data

It is noted in clause 2.4.3(1) that geotechnical parameter values must be obtained from test Clause 2.4.3(1)
results, either directly or through correlation, theory and empiricism, and from other
relevant data. The phrase ‘other relevant data’ is very important when establishing the values Clause 2.4.3(5)
of geotechnical parameters, and relates to the items listed in clause 2.4.3(5), which are the
other information and factors that should be considered, and include:
• published material for the test in appropriate ground conditions in order to interpret the
test results
• comparison of each derived parameter value with published data, local experience and
large-scale field trials, if available

57

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

• correlations between the results of more than one type of field test, whenever available,
the results of any large-scale field trials and measurements from neighbouring
constructions.

3.3.4. Characteristic values

Having obtained a set of geotechnical parameter values from test results in Step 1, by
applying the appropriate calibration and correction factors and taking account of the other
relevant data, the characteristic value for a particular design situation is then assessed in
Clause Step 2. The characteristic value of a geotechnical parameter is defined in clause 2.4.5.2(2)P
2.4.5.2(2)P as a cautious estimate of the value affecting the occurrence of the limit state.
Clause As indicated in Fig. 2.3, assessing the characteristic value of a geotechnical parameter
2.4.5.2(4)P involves taking account of the following factors:
• the variability of the ground
• the number of tests results
• the scatter of the test results
• the type of limit state
• the volume of the ground involved
• the nature of the structure.
The assessment of characteristic values is described in detail in Chapter 2.

3.4. Ground Investigation Report

All geotechnical designs, including small and relatively simple structures on straightforward
Clause 3.4.1(1)P ground conditions, should be based on the results of geotechnical investigations. Clause
3.4.1(1)P states that the information obtained from geotechnical investigations must be
compiled in a Ground Investigation Report, which forms part of the Geotechnical Design
Report. The person who prepares the Ground Investigation Report is often not the person
Clause 1.3(2) who prepares the Geotechnical Design Report. It is assumed that there will be adequate
continuity and communication between those involved in data collection, design and
construction. It is important, therefore, that the person who is responsible for the geotechnical
design, if not also involved in the ground investigation, has good communication with those
who carried out the investigation and obtained the geotechnical data for the design.
Clause 3.4.1(3) The information in a Ground Investigation Report normally consists of the following two
parts:
(1) a presentation of all the available geotechnical data and other information
(2) a geotechnical evaluation of the information, including the assumptions made in
interpreting the test results and the derived parameter values.
The first part of the Ground Investigation Report is the presentation of the geotechnical
Clause 3.4.2(1)P information from the ground investigation and corresponds to what is commonly referred to
as a ‘factual report’. This part must include:
• a factual account of all the field and laboratory investigations
• information about the methods used to carry out the field investigations and laboratory
testing.
Clause 3.4.2(2) A list of additional information that should also be included in this part of the Ground
Investigation Report, if relevant, is provided in clause 3.4.2(2). This list includes items such as:
• the names of all consultants and contractors
• the history of the site
• the geology of the site, including faulting
• local experience in the area.

58

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 3. GEOTECHNICAL DATA

The second part of the Ground Investigation Report is the evaluation of the geotechnical
information, and corresponds to what is commonly referred to as an ‘interpretative report’.
According to clause 3.4.3(1)P, this part must include the following items: Clause 3.4.3(1)P
• a review of the field and laboratory work, particularly any factors that need to be
considered when interpreting the results, for example limitations in the data, sampling,
sample transportation and storage procedures and any particularly adverse test results
• a review of the derived values of geotechnical parameters
• proposals for necessary further field and laboratory work and the purposes of such work.
A list is also provided in clause 3.4.3(2) of some additional items that should also be Clause 3.4.3(2)
included, if relevant, in this part of the Ground Investigation Report. This list includes items
such as:
• depth of the groundwater table and its seasonal fluctuations
• subsurface profile(s) showing the differentiation of the various formations
• detailed descriptions of all formations including their physical properties and their
deformation and strength characteristics.
In the case of simple design situations, for example Geotechnical Category 1 designs
involving small structures on ground that is known from comparable experience to be
straightforward, the two parts of the Ground Investigation Report together with the
Geotechnical Design Report may be presented on one page, as shown in Fig. 2.9.

59

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 4

Supervision of construction,
monitoring and maintenance

This chapter is concerned with the supervision of construction, monitoring and maintenance
of those geotechnical structures covered by Eurocode 7. The structure of the chapter
generally follows that of Section 4 in EN 1997-1.
4.1. Introduction Clause 4.1
4.2. Supervision Clause 4.2
4.3. Checking ground conditions Clause 4.3
4.4. Checking construction Clause 4.4
4.5. Monitoring Clause 4.5

4.1. Introduction
It is a Principle of EN 1997-1 that all geotechnical construction processes, including the Clause 4.1(1)P
workmanship applied, shall be supervised, that the performance of the structure shall
be monitored, both during and after construction, and that the finished structure shall be
adequately maintained.
Section 4 of EN 1997-1 has three main themes:
(1) Ensuring safety and quality through supervision, monitoring and maintenance. An Clause 4.1(8)P
important requirement is that the nature and quality of the supervision and monitoring
prescribed for the project shall be commensurate with the degree of precision assumed
in the design and in the selection of the engineering parameters and partial factors in the
design calculations. In other words, it may be necessary to prescribe an enhanced regime
of construction supervision and monitoring, where greater uncertainty surrounds the
reliability of the design parameters and the degree of accuracy of design calculations.
(2) Specifying what shall be done, and communicating this formally in contract documents Clause 2.8(1)P
and records such as the Geotechnical Design Report (GDR).
(3) Supervising, monitoring and maintaining in an orderly, planned manner, with the
keeping of records.
The section is primarily concerned with the designer’s responsibility for specifying the
requirements for construction supervision, monitoring and maintenance. The section is
concerned principally with the decisions of the designer and with information that others
should make available to him or her. Clearly, it cannot be the designer’s responsibility to
ensure that post-construction monitoring and maintenance are carried out on the structure;
however, it is the designer’s responsibility to prepare and communicate specifications for

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Box 4.1. Summary of items to be checked during construction

• Location and general layout of structure

• Ground conditions, by direct checking:
• during excavation, i.e.
– slopes and bottom of the excavation pit
– ground material when boring for piles and anchorages
• By indirect checks, i.e. during
– piling driving
– driving of sheet piles
• Groundwater conditions:
– flow and pore-water pressure regime
– effectiveness of groundwater control measures
• Structural settlements and movements, and stability of excavations
• Temporary support systems
• Effects of construction on nearby buildings and utilities
• Soil pressure on retaining structures
• Safety of workers

any such monitoring and maintenance, so that assumptions made in the design can be
confirmed during and subsequent to construction.
Section 4 gives no details of the requirements for workmanship, for which reference
should be made to the various ‘execution’ standards for special geotechnical structures; see
the list in Chapter 1 of this guide.
In many instances, it is left to the designer to specify what is ‘appropriate’, and, in most
respects, the whole section is simply a checklist reminding the designer of many of the items
which might be required or desirable in various circumstances. The amount of supervision,
monitoring and maintenance will depend on the nature of the project and the ground conditions.
As in other parts of Eurocode 7, this section does not specify contractual arrangements. It
does, however, specify what tasks must be undertaken and that an appropriate flow of
information must be maintained.
The section does not deal specifically with site safety matters, for which reference should
be made to national legislation.

Who is to do the supervision, monitoring and maintenance?

This is not stated, but it is assumed that adequate continuity and communication exist
between the designer and the constructor, and that adequate maintenance of the structure
will occur during its life. The GDR is intended to establish the requisite communication by
including a plan for the supervision and monitoring of listed items during construction,
Clause 2.8(6)P and items to be checked for maintenance in service. Extracts from the GDR concerning
any monitoring and maintenance requirements post-construction shall be provided to the
owner/client.
Annex J of EN 1997-1 provides a list of the more important items to be considered for
supervision during construction; a summary is given in Box 4.1.
The last item, safety of workers, is not specifically covered by the Eurocodes, being a
matter for national health and safety regulation. However, there is a requirement for the
person checking geotechnical construction activities to ensure the safety of workers against,
for example, the collapse of an excavation in which they are working.

4.2. Supervision
Clause 4.2 Supervision in EN 1997-1 means checking both the design and the construction. A plan is
required, to be included in the GDR, that will depend on the scale and complexity of the

62

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 4. SUPERVISION OF CONSTRUCTION, MONITORING AND MAINTENANCE

project, as represented by Geotechnical Categories (GCs). In the case of GC1, supervision

may be reduced to a visual inspection of the ground, superficial quality controls, and a
qualitative assessment of the performance of the structure during and immediately after
construction may be all that is required. Further monitoring and maintenance will probably
not be required at all. For GC2, measurements of the properties of the ground or the
movement of the structure may be necessary. Tests to check the movement under load and
the quality of piles, or tests of the density of fill behind a retaining wall, are examples of the
monitoring that might be required for GC2 projects.
Clause 4.2.1(1) states that ‘The construction shall be inspected on a continuous basis’.
‘Continuous’ means that inspection should not be so infrequent that important features of
the ground or of the installation of foundations, for example, may be missed.
That inspection records should systematically be kept and made available to the designer Clause 4.2.2(5)P
is an important requirement of EN 1997-1. The following features shall be recorded, as
appropriate:

• significant ground and groundwater features

• the sequence of works
• the quality of materials
• deviation from design
• as-built drawings
• results of measurements and their interpretation
• observations of the environmental conditions
• unforeseen events.

4.3. Checking ground conditions Clause 4.3

‘Control investigations’ concern checks of the ground and groundwater conditions. They
involve checking the design assumptions made and, if necessary, carrying out further
investigation of:

• the ground (soil or rock type and properties)

• the groundwater, including the level, pore-water pressures and chemistry, and the Clause 4.3.2
effects on the groundwater regime of such operations as dewatering, grouting and
tunnelling.
In the case of simple, GC1, designs, checking of ground and groundwater may be limited
to visual inspection and reliance on documented or other indirect evidence of site conditions.
For GC2 designs, additional sampling and laboratory testing or in situ testing and measuring Clause 4.3.1(5)
may be required as part of the control investigation. Indirect evidence of ground properties
from pile-driving records, for example, should be recorded and used in interpretation of
ground conditions.

4.4. Checking construction Clause 4.4(1)P

Construction operations shall be checked to ensure that they comply with the methods
assumed in the design and documented in the GDR. This Eurocode 7 requirement
may result in designers having to consider more carefully the likely construction methods,
and to state their design assumptions in the GDR (an example is the checking of fill
placement and compaction). However, this is not normally necessary for GC1 designs, Clause 4.4(4)
where construction sequences are frequently simple and straightforward and decided by the
contractor.
Deviations from design assumptions about ground and groundwater conditions and Clause 4.4(1)P
construction methods shall be reported without delay to the person responsible for the
design.

63

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

4.5. Monitoring
Clause 4.5(4) The measurements that may be required from a monitoring scheme include the following
features and their variation with time:
• deformations of the ground affected by the structure
• values of actions
• values of contact pressures between ground and structure
• pore-water pressures
• forces and displacements in structural elements.
Clause 4.5(8) For GC1 designs, a simple visual inspection may be sufficient, while for GC2 structures
Clause 4.5(10) monitoring may include measurements at selected points on the structure. For GC3
projects, EN 1997-1 recommends that analysis of the construction operations be
Clause 4.5(1)P added to measurements at selected points. The requirements for monitoring, with the
Clause 4.5(4) recommendation of formal keeping of records, need not be onerous for ‘routine’ geotechnical
activities (as described for GC1 and GC2). In the case of a simple foundation, they need only
be minimal, as shown in the sample Geotechnical Design Report for a strip footing in Fig. 2.9.

64

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 5

Fill, dewatering, ground

improvement and reinforcement

This chapter is concerned with design requirements related to fill, dewatering, ground
improvement and reinforcement. The chapter describes what could be termed ‘engineered
ground’ or ‘made ground’. The material is covered in Section 5 of EN 1997-1. The structure
of the chapter follows that of Section 5:
5.1. General Clause 5.1
5.2. Fundamental requirements Clause 5.2
5.3. Fill construction Clause 5.3
5.4. Dewatering Clause 5.4
5.5. Ground improvement and reinforcement Clause 5.5

5.1. General
The requirements for the design of fill, dewatering, ground improvement and (ground)
reinforcement according to EN 1997-1 are combined in one section, Section 5. This is due to
the fact that each of these processes is concerned with improving the properties of the
ground, while an earlier section of EN 1997-1, Section 3, is concerned with determining the
properties of existing ground. Section 5 provides the general requirements for the design Clause 5.1(1)P
of fill, dewatering, ground improvement and ground reinforcement giving few specific
requirements except for some aspects of fill. The section is effectively a checklist of items
which must be considered in the design.
The design procedures for geotechnical structures involving the use of fill, ground
improvement or reinforcement processes are covered in the sections of EN 1997-1 dealing
with the design of spread foundations, piles, anchorages, retaining structures, hydraulic
failure, overall stability and embankments. It should be noted that, in the context
of EN 1997-1, fill placed during execution is by definition part of the structure in that Clause 1.5.2.4
EN 1997-1 defines a structure as an organized combination of connected parts, including fill
placed during execution of the construction works, designed to carry loads and provide
adequate rigidity.

5.2. Fundamental requirements

Fill, dewatered, improved or reinforced ground all have to satisfy the same fundamental
requirements as existing ground, which are that they must be capable of sustaining the Clause 5.2(1)P
actions arising from their function and from their environment.

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

5.3. Fill construction

The design and construction of fill beneath foundations and floor slabs, as well as backfill to
Clause 5.3 excavations and retaining structures, is covered by the provisions in clause 5.3. The design
and construction of fill for general landfill, including hydraulic fill, landscape mounds,
spoil heaps and fill for embankments for dams (dykes) and transportation networks, are
also covered. This clause covers the selection, placement, compaction and checking of
fills. It provides a checklist of items, which should all be covered in specifications for
earthworks.
In the same way as for natural ground, design calculations involving fill materials require
assessment of the characteristic values of the material properties. At the time of design, the
fill to be used may not have been identified, though its properties will have been specified.
The assessment of the characteristic values of fill properties should follow the principles of
Clause 2.4.5.2 clause 2.4.5.2, requiring an assessment of a cautious estimate of the value affecting the
occurrence of the limit state.

5.4. Dewatering
Dewatering covers the abstraction of water both to improve the properties of ground and to
facilitate construction. In some situations, groundwater recharge may also be required as
part of a dewatering scheme to prevent drawdown causing unacceptable settlements of
neighbouring structures. The conditions to be satisfied and the aspects to be checked when
Clause 5.4 designing a dewatering scheme to EN 1997-1 are set out in Clause 5.4, which provides a
checklist of conditions to be considered by designers.
Clause 5.4(4) In the checklist of conditions to be considered, there is one condition that is particularly
important in the design of a dewatering scheme: it should be checked that the settlements
induced by the groundwater lowering do not lead to settlements of neighbouring structures,
which may cause damage or impair their serviceability. Especially dangerous are situations
where a permeable stratum overlying a compressible stratum is dewatered. Moreover, it
should be checked that the groundwater lowering does not have adverse effects on other
water extractions in the neighbourhood.
It should be noted that the extent to which the groundwater can be lowered in a specific
situation is highly dependent on the permeability of the ground and its variation over the
strata involved.
Clause 2.7 The design of dewatering schemes may very often benefit from the use of the observational
method.
The importance of considering the effects of groundwater is mentioned in many sections
Clause 2.4.2(9)P and clauses of EN 1997-1; for example, clause 2.4.2(9)P emphasizes the importance of
identifying for special consideration actions in which groundwater and free water forces
Clause 10.5(1)P predominate, while clause 10.5(1)P highlights the need for measures to control or block the
groundwater flow to prevent piping endangering the stability or serviceability of a structure.

5.5. Ground improvement and reinforcement

Clause 5.5 The design of ground improvement and reinforcement schemes is only covered in general
terms in EN 1997-1. In many cases such schemes would be classified as Geotechnical
Category 3 structures. As such, they would not be fully covered by the code requirements in
EN 1997-1, and their design should be carried out by a geotechnical specialist. It appears that
the basis for this is that ground improvement and reinforcement are not considered to
be conventional types of structures and foundations, which could be classified under
Geotechnical Category 2.

66

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 5. FILL, DEWATERING, GROUND IMPROVEMENT AND REINFORCEMENT

Some of the aspects which must be considered when designing a ground improvement or Clause 5.5(2)P
ground reinforcement scheme are listed. This list includes, for example, the thickness of the
in situ strata or fill material, the nature, size and position of the structure to be supported by
the ground, the importance of preventing damage to adjacent structures, etc. EN 1997-1 Clause 5.5(3)
requires that the effectiveness of ground improvement be checked against the acceptance
criteria by determining the changes in the appropriate ground properties or ground
conditions resulting from use of the improvement method. In order to do this, EN 1997-1 Clause 5.5(1)P
requires that a geotechnical investigation to determine the initial conditions be carried out
before any ground improvement is chosen or used.

67

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 6

Spread foundations

This chapter is concerned with checking the design of spread foundations. The material is
covered in Section 6 of EN 1997-1 and in the informative Annexes D (‘Sample analytical
method for bearing resistance calculation’), E (‘Sample semi-empirical method for bearing
resistance estimation’), F (‘Sample methods for settlement evaluation’), G (‘A sample method
for deriving presumed bearing resistance for spread foundations on rock’) and H (‘Limiting
values of structural deformation and foundation movement’).
Several sections and annexes of EN 1997-2 give additional information regarding
semi-empirical calculation models for bearing resistance and settlement evaluation using in
situ test results.
The structure of the chapter differs slightly from that of Section 6 of EN 1997-1, and
corresponds as follows:

6.1. Design methods Clause 6.4

6.2. Overall stability Clause 6.5.1
6.3. Direct method: ULS design Clause 6.5.2
6.4. Direct method: SLS design by settlement calculations Clause 6.6
6.5. Indirect method: simplified SLS method Clauses 6.4 and 2.4
6.6. Prescriptive method Clause 6.5.4.2
6.7. Structural design Clause 6.8

In common with other sections of EN 1997-1, Section 6 gives only the basic requirements
for the design of spread foundations. To assist the designer, this chapter describes typical
calculation methods for ultimate limit state (ULS) and serviceability limit state (SLS) design
checks, and includes the following examples:

• Example 6.1: a vertically loaded, square foundation pad on soft clay in drained and
undrained conditions. The ‘direct’ method is illustrated using the analytical calculation
model for bearing resistance of Appendix D of EN 1997-1.
• Example 6.2: the design of a square pad foundation, on sand and gravel, for a tower
subjected to a small, vertical, permanent action and a large, variable, horizontal action.
The ‘direct’ method, applying the analytical calculation model in Appendix D of
EN 1997-1, is used for a ULS design.
• Example 6.3: the design of a spread foundation using pressuremeter test results in
which both the ‘direct’ and ‘indirect’ methods are demonstrated for a semi-empirical
calculation model.

Many of the provisions of Section 6 also apply to gravity retaining structures; Example 9.1 Clause 9.7.3(1)P
in this guide, consisting of the ULS check against bearing capacity and sliding of the design of
the foundation of a gravity (stem) wall, illustrates the application.

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Section 6 may also be applicable to pile groups in compression (e.g. checking against
failure of a pile group by equivalent raft methods).

6.1. Design methods

Clause 6.2(1)P Some limit states are listed for consideration. The first four are clearly ULSs where an
‘infinite’ movement of the foundation occurs; the fifth on the list implies such large but
contained (differential) movements of the foundation that the supported structure reaches
a ULS, i.e. no longer fulfils the basic ULS requirements of its members or member
connections; the last three items concern SLSs.
Clause 6.4(1)P The attention of the designer is also drawn to several aspects which may play a role in the
Clause 2.5 choice of the depth of a spread foundation. Some of the listed items may be treated by
prescriptive measures. The checklist requires that future excavations should be considered.
This does not demand that the spread foundation be designed to withstand all conceivable
excavations; it merely requires that the designer considers possible future situations and
adopts a reasonable course of action. The same holds for other items in the checklist.
Clause 6.4(5)P One of three design methods has to be used to check the foundation design (see also Table
6.1 for a general overview), as follows:
(1) A direct method which involves two separate checks:
– firstly using a calculation model as close as possible to the ULS failure mechanism
– secondly using a settlement calculation to satisfy the SLS.
Clause 6.5.2.2 No restriction is placed on the calculation models that can be used for each of the two
Clause 6.5.3 steps. The ULS check can be performed using analytical calculation models for bearing
Clause 6.5.2.3 or sliding resistance, or using semi-empirical calculation models where the bearing
resistance is assessed directly as a derived value from in situ test results. The SLS check
must be performed by using settlement calculations (either an analytical calculation –
see Annex F, for example – or semi-empirical models for settlement calculation – see
EN 1997-2, Annex B2, C2 or D4, for example).
(2) An indirect method which is based on comparable experience (an essential prerequisite),
and which uses the results of field or laboratory measurements or other observations
and SLS loads. The comparable experience is related to SLSs, so that the use of this
method is intended automatically to satisfies the SLS requirement. The indirect method
also implicitly covers the ULS, at least for common structures with no exceptional
loading, provided the comparable experience is relevant. The indirect method is thus
a one-step method for checking both the SLS and the ULS. Calculations may be
performed using analytical or semi-empirical models. The choice of a Design Approach
(see Chapter 2) does not apply as the check is based on SLS conditions.
(3) A prescriptive method, which is usually based on comparable experience of the
observation of serviceability states.
Table 6.1 gives an overview of these three design methods as well as examples of
calculation models for spread foundations, with reference to relevant clauses and to sample
methods in informative annexes of EN 1997-1 and EN 1997-2.
It should be noted that simple structures, with no exceptional loading conditions, may be
designed using the indirect method or the prescriptive method, but that large or complex
structures, on difficult or compressible soil, should be designed using the two-step
directmethod. Furthermore, for structures subjected to significant horizontal loads, careful
attention should be paid to the eccentricity and inclination of the design loads.

6.2. Overall stability

Clause 6.5.1(1)P EN 1997-1 requires the design to be checked for the overall stability of the ground mass
according to the principles described in Section 11. The potentially unstable ground may

70

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 6. SPREAD FOUNDATIONS

Table 6.1. Overview of the design methods permitted by EN 1997-1 and corresponding
calculation models

Method and limit EN 1997-1

states Calculation model clause Sample method

Direct method
(a) ULS 6.5.2, 6.5.3
Analytical calculation model
– Bearing resistance 6.5.2.2 EN 1997-1, Annex D
– Sliding resistance 6.5.3
Semi-empirical calculation
model EN 1997-1, Annex E
– Pressuremeter 6.5.2.3 EN 1997-2, Annex C1
(b) SLS 6.6
Analytical settlement calculation EN 1997-1, Annex F
– Settlement calculation 6.6.2
Semi-empirical settlement EN 1997-2, Annexes
calculation B2, C2, D4
– Cone penetrometer,
pressuremeter, standard
penetration test
Indirect method 2.4.1(4), 2.4.8(4)
ULS and SLS Analytical calculation model EN 1997-1, Annex D
combined – Applying high ‘global’ factors
to ‘characteristic’ bearing
resistance
Semi-empirical calculation model EN 1997-2, Annex C1
– Applying high ‘global’ factors
to ‘characteristic’ bearing
resistance
Prescriptive method 6.5.2.4, 6.7
ULS and SLS Charts or tables for values of EN 1997-1, Annex G
combined presumed bearing resistance

D B

A C

Fig. 6.1. Overall stability has to be checked for failure surfaces A–B and C–D. The failure surface
C–D passes outside of the foundation but, if slope movement occurs, the vertical bearing resistance
of the foundation is substantially reduced

contain the foundation (Fig. 6.1, failure surface A–B) or the failure surface may pass close to
it (Fig. 6.1, failure surface C–D). Since failure along surface C–D may substantially affect the
bearing resistance of the foundation itself, the slope failure along C–D should also be
sufficiently improbable.
The qualification ‘sufficiently improbable’ is vague, and should be taken to mean that loss Clause 6.5.1(2)P
of overall stability should be as improbable as the loss of vertical bearing resistance of the

71

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

foundation. When Design Approaches 1 (DA-1) and 3 (DA-3) are used, this will usually be
ensured by using the same partial factors for the slope stability calculations and for the
calculation of the bearing resistance. When Design Approach 2 (DA-2) is used, the partial
resistance factors for overall stability and for bearing resistance are different (see Tables
A.14 and A.5, respectively).

6.3. Direct method: ULS design

6.3.1. Bearing resistance
Clause 6.5.2.1(1)P The fundamental ULS requirement is represented by the inequality
Vd £ Rd (6.1)
where Vd is the ULS design load normal to the foundation and Rd is the design bearing
resistance of the foundation against loads normal to it (Fig. 6.2). Rd may be calculated using
analytical or semi-empirical models. Vd includes the weight of the foundation and of any
backfill material (to be considered as a ‘structural action’) placed on top of it. Earth
pressures on structural elements above the foundation level are geotechnical actions and are
also included in Vd where relevant.
The basic inequality Vd £ Rd has to be checked for the recommended partial factors for
persistent and transient situations in Annex A (Table A.1 for partial factors on actions or the
effects of actions; Tables A.2 and A.5 for partial factors on soil parameters and resistances).
For accidental situations, the partial factors are usually put equal to 1.0. It should be noted
that the recommended values for the partial factors given in Annex A are established for
analytical methods in all three Design Approaches and for the semi-empirical method in
DA-2; thus semi-empirical methods in DA-1 and DA-3 may need the application of model
factors according to clause 2.4.7.1(6).
EN 1997-1 recommends that water pressures not caused by the foundation loads are
included explicitly in calculations. For drained conditions, it suggests that water pressures
are included as actions. A typical situation is illustrated in Fig. 6.3. It is consistent with the

FV, FHM

AH
W1 AV
Vd
W2
P

Rd

Fig. 6.2. Example of actions on the footing. Vd is the component, normal to the foundation, of the
design values of the following actions (the subscript d is omitted for clarity):
H, M and V, which are structural actions
A and P, which are earth pressures
W1 and W2,, which are weights of backfill
weight of the footing

72

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 6. SPREAD FOUNDATIONS

d1 g1

d2 F ¢1 U1 S F ¢1 U1

d3 W g2

U2 U2

Rd

Fig. 6.3. Actions on a footing with hydrostatic water pressure. (After Simpson and Driscoll, 1998.)
Ab = area of footing base
Ac = area of column cross-section
γ1 = total weight density of soil above the water table
γ2 = total weight density of soil below the water table
S = action from the superstructure
W = weight of footing
U1, U2 = forces due to water pressure
F1 = effect of the action of backfill on the foundation
= (γ1d1 + γ2d2)(Ab – Ac)
= F1¢+ U1
= (γ1d1 + (γ2 – γw)d2 + γwd2)(Ab – Ac)
where
F1¢ = effective effect of the action of the backfill on the foundation
U2 = γw(d2 + d3)Ab
Vd must be matched by design resistance Rd, which in this case is an effective force

definition of actions to treat all water pressures in drained conditions as actions, since they
are known at the start of the calculation. This implies that the resistance is calculated in
terms of effective stress. The question may arise: how to apply the partial factors to the
weight of a submerged or partially submerged structure? The force due to water pressure
acting on the underside of the foundation acts so as to reduce the value of Vd, and may then
be considered as ‘favourable’, while the (total) weight of the foundation is unfavourable.
Physically, however, it is the submerged weight (total weight minus upward force of water)
which has to be sustained by the soil, for which the resistance is then expressed in terms of Clause 2.4.2(9)P
effective forces; the same single partial factor may be applied to the sum of these actions.
This is illustrated in Fig. 6.3. The design values of the actions due to the weight of the
submerged footing and to the backfill become the design values of their effective weights.
Action factors of 1.0 (DA-1 Combination 2) and 1.35 (DA-1 Combination 1, and DA-2 and
DA-3) are applied to the effective weight of the submerged footing and backfill if they are
unfavourable. This is illustrated in Example 6.1. Action factors of 1.0 apply for all Design
Approaches if the effective weight of the foundation is favourable.
The calculations for undrained conditions are shown in Fig. 6.4 and in Example 6.1.
A method similar to that outlined in Fig. 6.4 applies in the semi-empirical calculation
model given in Annex E, which is expressed in terms of a total stress (independent of whether
the soil is in a drained or an undrained condition). Example 6.3 illustrates this.
For the design of structural members, water pressures may be unfavourable (e.g. for rafts
or closed caissons forming boxes), and the action factor for unfavourable structural actions
should be applied to the water pressures (see Section 6.7).

73

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

F 1¢ U1 S F 1¢ U1

Rd

Fig. 6.4. Footing with hydrostatic water pressure in granular soil over undrained clay. (After
Simpson and Driscoll, 1998.)
Arrows show hydrostatic water pressures (vertical components only). Those acting on the footing
are included in Vd whilst those alongside the footing augment Rd. In this (undrained) case, Rd is a total
force, acting on the base, which includes both the hydrostatic pore pressures and excess pore
pressures due to undrained shearing.
The action on the ground, Vd, which is to be supported by the undrained resistance Rd, is given by
Vd = S + W + F1 = S + W + (F1¢ + U1)
F1¢ is defined in Fig. 6.3.
A possible alternative approach is to include the hydrostatic water pressures in the clay as part of
Vd, as for the drained case; these act upwards on the base of the footing, reducing the value of Vd as
in Fig. 6.3. The excess pore pressures generated at the base of the footing due to shearing remain
part of Rd, however. This is consistent with the definition of actions in clause 2.4.2(2)P, since the
hydrostatic component is known, but the excess pore pressure is unknown at the start (and end) of
the calculation

Analytical calculation
Clause 6.5.2.2(1) The sample calculation model given in Annex D is widely recognized. Annex D gives an idea
of the degree of accuracy and conservatism that is appropriate. The recommended values for
the partial factors of Annex A (Tables A.1, A.2 and A.5) have been calibrated to the method in
Annex D.
Annex D is marked as ‘informative’. The designer may, therefore, use other methods that
may be specified in the National Annex.
The calculation model of Annex D applies to a single soil layer. Caution is advised for
Clauses 6.5.2.2(4) layered soils or ground containing discontinuities. Paragraph (5) may be over-conservative or
to 6.5.2.2(6) inappropriate in several situations. More refined calculation models may be employed, e.g.
for punching failure, for squeezing, or for undrained shear resistance increasing with depth.
The partial factors to be used depend on which Design Approach is adopted.
Basic input parameters for the analytical calculations are characteristic values of soil shear
strength (ck¢, ϕk¢ in drained conditions; cu, k in undrained conditions) and weight density (γk
and γk¢). Their characteristic value should be selected to account for the variability of the
ground relative to the size of the foundation and the stiffness of the structure:

• For a large, spread foundation, the characteristic value of the shear resistance parameters
will usually be a cautious estimate of the mean value under the foundation. When a
building is supported on several footings, the characteristic value should account for the
spatial variability of the shear strength parameters over the footprint of the building and
for the stiffness of the supported structure.

74

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 6. SPREAD FOUNDATIONS

• Local weak spots, leading to significantly lower bearing resistance, should be identified
when selecting the characteristic values. If no such weak spots exist, the characteristic
shear strength may be a cautious estimate of the mean values over the footprint of the
structure. If weak spots exist and the supported structure is not stiff enough to transfer
loads from footings at weak spots to footings at stronger spots, the characteristic value of
the shear strength parameters may be a cautious estimate of the mean or lower values
under each footing separately.

When selecting the characteristic value of ground shear strength parameters for the
design of a small footing, attention should be paid to failure surfaces which may develop
preferentially along points of weakness in the ground. Should this possibility be likely, then
the selection of the characteristic value should be governed by the lower values of the ground
strength parameters.
The characteristic value of the unit weight of the soil should be a cautious estimate of its
mean value.
The effective overburden pressure at the level of the foundation, q¢, is a cautious estimate
of its mean value in the vicinity of the footing. It should account for unfavourable water levels
and the adverse effect on bearing resistance of a reduction in overburden through any
removal of soil.

Some features to note when applying an analytical calculation model using DA-1 Clause 2.4.7.3.4.2
For persistent and transient situations, the design has to be checked for both Combinations 1
and 2:
Combination 1: A1 ‘+’ M1 ‘+’ R1

Combination 2: A2 ‘+’ M2 ‘+’ R1

The partial factors are applied at the source, i.e. on the actions (for recommended values,
see Table A.1) and on the material shear strength parameters c¢ and tan ϕ¢ or cu (for
recommended values, see Table A.2). For spread foundations, the factors on the resistance
are equal to 1.0 (see Table A.5).
Combination 2 usually determines the dimensions of the foundation, except in situations
where a horizontal variable load, the value of which is large compared with the permanent
vertical load, leads to a large overturning moment on the foundation. Such a load combination
may lead to a very eccentric design value of the resulting action. It is advisable, in a first step,
to calculate the size of the foundation for the partial factors of Combination 2, and then to
check, in a second step, that the resulting size fulfils the ULS requirements for the partial
factors of Combination 1.

Step 1 – sizing of foundation using Combination 2: A2 ‘+’ M2 ‘+’ R1. The design values of
the structural actions on the foundation are obtained by applying the factors in set A2
of Table A.3 (recommended values are γF = 1.0 on unfavourable, permanent loads and
γQ = 1.3 on unfavourable, variable loads; but these may be changed in the National Annex).
Design values of geotechnical actions (e.g. active pressure) are obtained by applying the
factors of set M2 of Table A.4 to the shear strength parameters, and the action factors of set
A2 of Table A.3 to unfavourable, permanent geotechnical actions (recommended value is
γF = 1.0) and to the unfavourable, variable geotechnical actions (recommended value:
γQ = 1.3).
The design values of the soil shear strength parameters are obtained by applying factors
larger than 1.0 to the characteristic values, as in set M2 of Table A.4 (recommended values
are γϕ = 1.25, γc¢ = 1.25 and γcu = 1.4; but these may be changed in the National Annex). The
partial factor on the resistance γRv is 1.0 according to set R1 of Table A.5.
A partial material factor of 1.0 is applied to the characteristic (effective) weight density of
the ground when calculating the overburden pressure as part of the bearing resistance.

75

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

The design values of the actions and of the shear strength parameters permit the
calculation of:

• The design values of the components of the resulting actions normal (Vd) and parallel
(Hd) to the foundation.
• The eccentricities eB, d and eL, d of the resulting actions and hence the effective
dimensions B¢ and L¢ of the foundation.
• The bearing resistance factors Nq(ϕd), Nγ(ϕd) and Nc(ϕd).
• The inclination factors iq, iγ and ic as functions of the design values of the shear strength
parameters and of Hd and of Vd.
• The shape factors sq, sγ and sc and the base inclination factors bq, bγ and bc.
• The design value of the bearing resistance, from Rd/A¢.

Usually, several iterations will be needed to obtain the optimal foundation size to fulfil the
ULS requirement.

Step 2 – checking the size of the foundation found in step 1 for the sets of partial factors of
Combination 1: A1 ‘+’ M1 ‘+’ R1. The design values of the actions on the foundation are
obtained by applying the factors according to Table A.3, set A1 (recommended values are:
γF = 1.35 on unfavourable, permanent, loads and γQ = 1.5 on unfavourable, variable loads;
but they may be changed in the National Annex). Design values of geotechnical actions (e.g.
active pressure) are obtained by applying these action factors to the characteristic values of
the geotechnical actions. The design values of the soil shear strength parameters are equal to
their characteristic values (with partial factors equal to 1.0, Table A.4, set M1). The partial
factor on the resistance γRv is 1.0 according to Table A.5.
It should be noted that a partial action factor equal to 1.0 is applied to the permanent
water pressures, and a partial material factor equal to 1.0 is applied to the characteristic
(effective) weight density of the ground, when calculating the (effective) overburden pressure.
Since the size of the foundation is usually determined by Combination 2, the calculation
for Combination 1 is reduced to a simple check that the size of the footing fulfils the
requirement of Combination 1.

Notes.

(1) For vertical or nearly vertical loads, it is often obvious that Combination 1 is not relevant
for determining the foundation dimensions. Calculations for step 2 are not then
necessary.
(2) Permanent actions may be favourable when, for example, they are acting in combination
with large variable actions. In such cases, it is necessary to perform two calculations
for Combination 1: one in which the permanent, vertical action is considered to be
favourable, and hence acquires an action factor of 1.0, and another one in which the
permanent vertical action is considered to be unfavourable, and hence acquires an
action factor of 1.35. The first case is denoted by ‘Vfavourable’ and the second by ‘Vunfavourable’
in the examples.

In Combination 1, the design value of the resistance is equal to its characteristic value
when the load is vertical and is somewhat lower when the load is inclined.

Clause 2.4.7.3.4.3 Some features when applying an analytical calculation model using DA-2
The design has to be checked for one set of partial factors using the combination
A1 ‘+’ M1 ‘+’ R2
There are two ways to introduce the partial factors in DA-2, either by applying them to the
actions (at the source) or by applying them to the effect of the actions. Both are discussed
below.

76

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 6. SPREAD FOUNDATIONS

When applying the partial factors to the actions, at the source, the design values of the
structural actions on the foundation are obtained by applying the action factors in Table A.3,
set A1 (recommended values are γF = 1.35 on unfavourable, permanent actions and
γQ = 1.5 on unfavourable, variable actions). Design values of geotechnical actions (e.g.
active pressure) are obtained by applying γF and γQ to the values of the geotechnical actions
calculated by applying a γM of 1.0 to the shear strength parameters (see Table A.4, set M1).
A partial material factor equal to 1.0 is applied to the characteristic (effective) weight
density of the ground when calculating the (effective) overburden pressure.
The design value of the bearing resistance is obtained by applying a partial resistance
factor γR larger than 1.0 (see Table A.5, set R2, where γR, v = 1.4) to the bearing resistance
calculated using design values of the soil shear strength parameters equal to their
characteristic values (i.e. bearing resistance calculated with γM = 1.0, see Table A.4, set M1).
The design value of the actions and the design value of the shear strength parameters
permit the calculation of:
• The design values of the components of the resulting action normal (Vd) and parallel
(Hd) to the foundation.
• The eccentricity eB, d and eL, d of the resulting action, and hence the effective dimensions
B¢ and L¢ of the foundation.
• The bearing resistance factors Nq(ϕd) = Nq(ϕk), Nγ(ϕd) = Nγ(ϕk) and Nc(ϕd) = Nc(ϕk).
• The inclination factors iq, iγ and ic as functions of the design values of the shear strength
parameters equal to their characteristic values and of Hd and Vd.
• The shape factors sq, sγ and sc and the base inclination factors bq, bγ and bc.
• The value of the bearing resistance R/A¢ is calculated using the above-mentioned factors
in a bearing resistance formula (e.g. see Annex D). The design value of the bearing
resistance Rd/A¢ is calculated by applying a partial resistance factor γR, v which is greater
than 1.0. (See Table A.4, set R2: recommended value γRv = 1.4.)
For the procedure in DA-2 where the partial factors are applied to the effect of all actions Clause 2.4.7.3.3(1)
(procedure denoted as DA-2*), the characteristic bearing resistance Rk/A¢ is calculated using
characteristic values of the actions and characteristic values of the shear strength parameters.
The inclined actions and the eccentricities eB, k and eL, k (and thus the effective width and
length B¢ and L¢) are also calculated using characteristic values of the actions and shear
strength parameters. As a result, the eccentricity will be smaller than if design values of
actions were used, with γF ≥ 1.0 applied at the source. As a consequence of this, when using
the procedure in DA-2 where the partial factors are applied to the effect of actions and the
resistance factor is applied to the bearing resistance calculated with characteristic values of
actions, special care must be taken regarding overturning of the foundation. When applying
this method, the eccentricities eB, k and eL, k are often limited to B/3, although this is not a
requirement of EN 1997-1.
For both of the above procedures, it must be remembered that permanent actions may be
favourable when, for example, they are acting in combination with large variable actions. In
such cases, it is necessary to perform two calculations: one in which the permanent vertical
action is considered to be favourable, and hence acquires a partial factor γF of 1.0, and
another one in which the permanent vertical action is considered to be unfavourable, and
hence acquires γF = 1.35.

Some features when applying an analytical calculation model using DA-3 Clause 2.4.7.3.4.4
The design has to be checked using the following sets of partial factors in combination:
(A1 or A2) ‘+’ M2 ‘+’ R3
The design values of the actions on the foundation are obtained as follows:
• Structural actions: by applying the factors of set A1 of Table A.3 (recommended values
are: γF = 1.35 on unfavourable, permanent actions and γQ = 1.5 on unfavourable,
variable actions).

77

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

• Geotechnical actions (e.g. active pressure): by applying the partial material factors of set
M2 of Table A.4 to the characteristic values of the shear strength parameters, and the
partial factors of set A2 of Table A.3 to the actions (recommended values are γF = 1.0 on
permanent geotechnical actions and γQ = 1.3 on unfavourable, variable geotechnical
actions).
The design values of the soil shear strength parameters are obtained by applying factors
larger than 1.0 to the characteristic values, as in set M2 of Table A.4 (recommended values
are γϕ = 1.25, γc¢ = 1.25 and γcu = 1.4; these values may be changed in the National Annex).
The partial factor on the resistance, γR, v, is 1.0, according to set R3 of Table A.5.
The design values of the actions and of the shear strength parameters as determined above
permit the calculation of:
• The design values of the components of the resulting action normal (Vd) and parallel
(Hd) to the foundation.
• The eccentricity eB, d and eL, d of the resulting action, and hence the effective dimensions
B¢ and L¢ of the foundation.
• The bearing resistance factors Nq(ϕd), Nγ(ϕd) and Nc(ϕd).
• The inclination factors iq, iγ and ic as functions of the design values of the shear strength
parameters and of Hd and Vd.
• The shape factors sq, sγ and sc and the base inclination factors bq, bγ and bc.
• The design value of the bearing resistance from Rd/A¢.
Usually, several iterations will be needed to obtain the optimal foundation size to fulfil the
requirement.
It should be noted that permanent structural actions may be favourable when, for
example, they are acting in combination with large variable actions. In such cases, it is
necessary to perform two calculations: one in which the permanent vertical action is
considered to be favourable, and hence acquires an action factor γF of 1.0, and another one
in which the permanent vertical action is considered to be unfavourable, and hence acquires
γF = 1.35.

Clause 6.5.2.3 Semi-empirical calculation

Semi-empirical models can be used to assess the bearing resistance of a spread foundation
from ground parameters (usually from the results of in situ tests). A sample method using
pressuremeter test results is given in Annex E. This annex gives no further details on how to
assess p le* and no values for the bearing factor k; the reader is referred to EN 1997-2, Annex
C1, or to the literature.
When using semi-empirical models for ULS design, the actions and the resistances
must be factored using the partial factors for which recommended values are given in
Annex A. The application of partial factors to resistance is most appropriate when using
semi-empirical models for ULS design. DA-2 can thus be applied straightforwardly
with semi-empirical models. In DA-1 and DA-3, values greater than 1.0 for the factor γR, v of
set R1 (DA-1) and set R3 (DA-3) (Table A.5) and values of γM equal to 1.0 could be applied for
semi-empirical models; the factor γR, v then acts merely as a model factor (see Example 6.3).

Clause 6.5.3 6.3.2. Sliding resistance

The fundamental ULS requirement is represented by the inequality (Fig. 6.5):
Hd £ Rd + Rp, d (6.2)
Clause 6.5.3(5) A limit state for the foundation may be reached even if the ground has not reached the
limits of its strength, that is, without the formation of a failure mechanism in the ground. It is
thus necessary to consider the displacement appropriate to the limit state considered. For
this reason, Rp, d may not necessarily be the limiting passive resistance of the ground. It
should be considered also that the maximum available sliding resistance, Rd, is likely to

78

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 6. SPREAD FOUNDATIONS

Fv

FH

AH
W1
AV
W2
Vd
Rp, d

Hd

Rd

Fig. 6.5. H is the resultant of the horizontal components of earth pressure A and the structural
action F (FH). V is the resulting force normal to the foundation. The resistance against sliding is the
sum of the frictional resistance under the footing, R, and the mobilized passive earth resistance, Rp

be mobilized with relatively little movement, though it could possibly reduce as large
movements take place. Hence it may not be possible to mobilize the maximum values
of Rd and Rp, d simultaneously. The selection of the characteristic values of the shear
strength parameters to assess Rd and Rp, d should take account of any incompatibility in the
movements in the limit state of sliding.
Consideration should be given to the effects of local excavation, erosion, shrinkage of clay, Clause 6.5.3(6)P
etc., acting to reduce or eliminate the passive resistance on relatively shallow foundations of Clause 6.5.3(7)P
retaining walls.
The application of equation (6.2) in the three Design Approaches is similar to the
procedure outlined in the section on bearing resistance. Rd can be written as follows: Clause 6.5.3(8)P
• Drained conditions:
Rd = Vd¢ tan δd when the ground properties are factored (equation (6.3a))
Rd = Vd¢ tan δk /γR, h when the ground resistances are factored and the actions
are factored at their source (equation (6.3b))
Rd = Vk¢ tan δk /γR, h when the ground resistances are factored and the effects
of actions are factored (equation (6.3a), see the note).
• Undrained conditions: Clause 6.5.4(11)P
Rd = Accu, d when the ground properties are factored (equation (6.4a))
Rd = Accu, k /γR, h when the ground resistances are factored (equation (6.4b)).
It should normally be assumed that the soil at the interface with a concrete structure will
be disturbed. Hence, critical state (constant volume) angles of shearing resistance are Clause 6.5.3(10)
usually relevant at the interface, even if higher values of the angle of shearing resistance are
used in the main body of the soil. Thus, the design value of the structure–ground interface
friction angle, δd, should be deduced from a cautious estimate of the critical state angle of
¢ d for cast-in-situ concrete and δd = 23 ϕ cv,
shearing resistance: δd = ϕ cv, ¢ d for smooth, precast
concrete foundations; any effective cohesion, c¢, should be neglected. This illustrates that,
for the same soil, two different characteristic values may be relevant for two different ULSs

79

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Light vertical load

Large horizontal load

(a)

Light vertical load

Large horizontal load

(b)

Fig. 6.6. (a) Light precast footing subjected to horizontal load. (b) Cast-in-situ footing fitting
intimately to the ground

of the same structure: vertical bearing resistance failure may be governed by peak values of
the angle of shearing resistance, while sliding at the soil–structure interface may be governed
by the critical state angle of shearing resistance.
Clause 6.5.3(11)P Equation (6.4) is mainly relevant for the sliding of bases on clay or weak rock, for which
undrained shear strength could be relevant, especially with rapid loading conditions. The
contact area, Ac, is restricted to the compressed area under design loads.
Because, in some circumstances, the vertical load may be insufficient to produce full
contact between the soil and the foundation, inequality (6.5) limits the design resistance to
0.4Vd in undrained conditions. Consider, for example, a light, precast base subjected to a
large horizontal force, as shown in Fig. 6.6a. In order to calculate the available horizontal
resistance, the contact area must be derived as a function of the vertical force and the vertical
bearing resistance, with allowance for load inclination. Mortensen (1983) has shown that
calculations of this type give a result which can be summarized as inequality (6.5). In some
cases of rapid loading, for footings which fit intimately onto the ground surface, suction may
Clause 6.5.3(13) ensure that no gap can form between the ground and the footing (Fig. 6.6b). In these cases,
inequality (6.5) can be disregarded.

6.3.3. Loads with large eccentricities

Clause 6.5.4(1)P EN 1997-1 does not place a specific limit on the degree of eccentricity of the load on the
foundation; however, special precautions are required when the resultant force lies outside
the middle two-thirds of the foundation. Although it is not stated, it may be assumed that the
resultant force considered is the ULS force. The commonly used ‘middle third model’ (on
the basis of SLS forces and material properties) is useful as it helps the designer to assess if
part of the foundation will lose contact with the ground during loading; it is, however, not a
requirement as EN 1997-1 puts no limits on the eccentricity of the load (see also Example 6.2).
When an eccentric load lies outside the middle-third of the foundation, it is likely that
elastic calculations of rocking will underestimate the movements, if these are required to
check if structural failure or second-order effects in the supported structure could occur, or
if serviceability limits are exceeded.
The precautions to be taken when large eccentricities could occur include:

80

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 6. SPREAD FOUNDATIONS

• careful review of the design value of the actions, as small deviations may have detrimental
effects on safety when the load is very eccentric
• taking account of the magnitude of the construction tolerances, as constructing a
foundation slightly too small may have detrimental effects on the safety when the load is
very eccentric.

6.3.4. Structural failure due to foundation movement Clause 6.5.5

A ULS may be reached in a supported structure by displacements of the ground, even if no
unlimited plastic mechanism has occurred in the ground. Examples of ground movements
that may cause a ULS in the structure, or severe damage, are:
• large settlements or horizontal displacements, e.g. in soft clays loaded beyond the
pre-consolidation pressure
• severe settlements caused by (artificial) groundwater lowering, desiccation due to tree
roots, etc.
• swelling of clays due to variations in water content (volume expansion of partly saturated
soil)
• collapsing soils
• settlements of loose fills due to vibrations, inundation, etc.
It is difficult to establish general rules for allowable foundation displacements, and
EN 1997-1 gives no advice on suitable values, except for some information in Annex H
indicating that a relative rotation of about 1/150 or more is likely to cause a ULS in the
supported structure.
For conventional problems (i.e. conventional structures and loadings in ground conditions
for which experience exists), the ULS partial factors given in Annex A may be regarded as
providing sufficiently low mobilization of ground strength to avoid ULSs in the structure due
to ground movements, at least for drained soil with angles of shearing resistance not less than
about 25–27°, and of reasonable stiffness and density.
For clays which are loaded rapidly enough to remain in an undrained condition, the partial
factor on cu in DA-1, for instance, is small, and is usually insufficient to prevent large
settlements due to approaching plastic failure. For soils with low angles of shearing resistance,
such as soft clays and organic soils, settlement analyses are required to check against ULSs in
the supported structure.
Presumed bearing pressures based on comparable experience should be low enough to
avoid settlement problems leading to failure in the supported structure.
When ULSs in the supported structure subjected to (differential) settlements are checked,
design values of (differential) settlements are to be imposed.
Figure 6.7 illustrates the situation for a continuous beam on three supports for which it is
required to check the bending moment and shear force at the middle support induced by the
imposed settlement of the end support.
Design values of the settlement under the relevant design value of the load may, ideally, be
assessed from a ‘characteristic’ load–settlement curve of the foundation; a ‘characteristic’
load–settlement curve is a cautious estimate of the load–settlement curve. This can be
obtained by test (but this is not common) or by calculation. A load–settlement curve up to
failure usually has the shape of curve 1 in Fig. 6.8a. Above a certain value of the load, the
curve becomes significantly non-linear, and the effect of the action (i.e. the settlement)
increases more than the action (i.e. the load). This downward curvature is due to local
yielding of the soil beneath the foundation. For such a shape, according to EN 1990, clause
6.3.2(4), ‘design’ values of settlements are read from the characteristic load–settlement
curve at Vd, the design value of the ULS load to be checked.
The characteristic load–settlement curve may be assessed using Terzaghi’s logarithmic
settlement formula (Fig. 6.8b) using characteristic values of deformation parameters. This
formula assumes that no significant yield occurs below the foundation. Using this method,
the calculated settlement increases less than the load. For such a shape, according

81

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Design value of
settlement support

Md
Bending moment induced by
settlement of end support

Fig.6.7. Example of check against structural failure in a continuous beam on three supports by
exceeding the design value of the bending resistance at the middle support due to large settlement
of end support

to EN 1990 (clause 6.3.2(4)), ULS ‘design’ values of settlements are deduced from the
characteristic load–settlement curve by applying the ULS action factor to the ‘characteristic’
settlement obtained for Vk, the characteristic value of the action.
The characteristic load–settlement curve may be assessed using adjusted elasticity formulae
(Fig. 6.8c) and characteristic values of deformation parameters. Adjusted elasticity formulae
assume that no significant yield occurs below the foundation. Using such formulae, the
calculated settlement increases linearly with the load. For a linear relationship, the ‘design’
value of settlement, sd, obtained at Vd, the ULS design value of the load, and sd obtained by
multiplying the settlement sk at Vk, the ULS characteristic value of the load, by the action
factor, are the same.
It should be emphasized that Terzaghi’s settlement formula and adjusted elasticity methods
are only reliable when the value of the load is small compared with the failure load. The
resulting settlements cannot be extrapolated to a complete load–settlement curve. Their use
to check for ULSs in the structure should be restricted to situations where plastic behaviour
in the soil does not affect the load–settlement curve significantly, or where a correction may
be made for plastic behaviour.
When such ULS design values of settlements are entered into the structural calculation,
the resulting internal moments and shear forces are ULS design values (i.e. Md, Sd) to which
the requirements of the relevant structural Eurocode apply.

6.4. Direct method: SLS design by settlement calculations

Clause 6.4(5)P When the direct method is used, settlement calculations have to be performed and their
Clause 2.4.9 results compared with limiting values. Limiting values should be established for the foundation
movements, depending on several factors, of which a list is given. In the absence of specified
limiting values of deformations of the supported structure, the values of structural deformations
and foundation movements given in Annex H (informative) may be used. Annex H is based on
the state-of-the-art report by Burland et al. (1977).
Not only should deformations due to the load on the foundation be considered, but also
any other source of settlement such as self-weight compaction, crushing of sand, effects of
vibrations, flooding, water table lowering or rise, and collapse of soil structure.
Clause 6.6.1 General requirements and recommendations are given for the assessment of SLSs and
Clause 6.6.2 limiting values of foundation displacements. Annex F gives sample (analytical) methods for
the evaluation of settlements. Although not quoted in EN 1997-1, EN 1997-2 gives four

82

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 6. SPREAD FOUNDATIONS

Vk Vd Loads

Sk: settlement for

characteristic value of action

Sd: settlement for design

value of action, to be used
for ULS check of the structure

Settlement Simplified linear spring

for ULS design
Simplified linear spring Curve 1
for SLS design

Fig. 6.8. (a) Evaluation of the design value of the imposed settlement sd when the effect of the load
(the settlements) increases more than the load: the design value of the settlement is read from the
characteristic load settlement curve for the design value of the action Vd

Vk Loads

Simplified linear spring

for ULS design

Sk: settlement for

characteristic value of action Load–settlement curve obtained from
Terzaghi's logarithmic settlement formula
Sd: settlement for design
value of action, to be used
for ULS check of the structure
Settlement

Fig. 6.8. (b) Evaluation of the design value of the imposed settlement sd when the effect of the load
(the settlement) increases less than the load: the design value of the settlement is obtained by
multiplying the settlement for characteristic load, read from the characteristic load–settlement
curve, by the action factor

Vk Vd Loads

Sk: settlement for

characteristic value of action

Sd: settlement for design Load–settlement curve obtained

value of action to be used using adjusted elasticity methods
for ULS check of the structure
Settlement

Fig. 6.8. (c) Evaluation of the design value of the imposed settlement sd when the effect of the load
(the settlements) increases linearly with the load

83

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

sample methods for calculating settlements from in situ measurements using semi-empirical
calculation models (EN 1997-2, Annexes B2, C2 and D4).
Clause 2.4.8(2) Usually, the partial factors for checking SLSs by settlement calculations are equal to 1.0,
and are applied to the representative values of the actions in the SLS (see also EN 1990 for
load combinations), as well as to the characteristic values of the deformation parameters of
the ground. The characteristic values of the deformation parameters should be selected
according to the principles of clause 2.4.5, and thus are cautious estimates. Items such
Clauses 6.6.2(1) as stress and strain level, variability, and stiffness of the structure should be carefully
to 6.6.2(15) considered. EN 1997-1 offers some generally accepted rules of good practice for settlement
calculation. The designer should concentrate on variability and the range of values for the
soil volume affected rather than on the ‘mean’ values.
Clause 6.6.1(8)P When calculating the settlement of a footing, interaction with neighbouring footings has
to be considered.

6.5. Indirect method: simplified SLS method

6.5.1. General
Clause 6.4(5)P Indirect methods use comparable experience and the results of field or laboratory
measurements or observations, chosen in relation to SLS loads, to satisfy the requirement of
all relevant limit states. The indirect method, therefore, satisfies the SLS and, implicitly, also
the ULS, at least for common structures for which comparable experience exists and for
which no exceptional loading or ground conditions apply. The indirect method differs from
the direct method in that only one analysis is performed to check both ULSs and SLSs. The
indirect method is well-suited to routine Geotechnical Category 2 problems. As a minimum,
Geotechnical Category 3 problems should be treated using the direct method.
Besides the requirement of comparable experience and the relation to SLS loads,
Clause 2.4.1(4)P EN 1997-1 gives no clear guidance on how to apply the indirect method. Based on the stated
general principle, it can be understood that the indirect method may consist of:
• either performing an SLS settlement calculation or
• limiting the mobilization of the ground resistance calculated using the models (analytical
or semi-empirical) for the bearing resistance, to avoid exceeding the SLS.
As there is rather limited experience with the first-mentioned procedure, this will not be
treated further in this guide.

6.5.2. Indirect method based on limiting the mobilization of bearing

resistance
At least for simple ground and loading conditions, experience of classical design indicates
that foundations behave satisfactorily for serviceability, and therefore automatically for
persistent and transient ULS requirements, when the foundation dimensions have been
determined with a sufficiently high value of the overall safety factor on bearing resistance
(e.g. safety factor greater than 2.5–3). This experience has been gained with analytical as well
as with semi-empirical calculation models.
Clause 2.4.8(4) For the indirect method, EN 1997-1 states that ‘it may be verified that a sufficiently
low fraction of the ground strength is mobilised to keep deformations within the required
serviceability limits’. This is a statement of traditional practice. Limiting the mobilization of
ground resistance can be achieved by:
• Comparing the serviceability loads to the bearing resistance divided by a sufficiently
large ‘global’ mobilization factor. Values of this ‘global’ factor may be taken from
traditional practice.
• Comparing the serviceability loads to the bearing resistance calculated by applying a
sufficiently large ‘mobilization factor’ directly to the shear strength parameters. There is

84

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 6. SPREAD FOUNDATIONS

less experience with values of these ‘partial mobilization’ factors, but it may be expected
that such experience will quickly grow in the future.
Both methods can be applied when analytical formulae (e.g. in Annex D) are used; when
semi-empirical rules (e.g. in Annex E) are used, only a ‘global’ mobilization factor can be
applied.
EN 1997 gives no indication on which combination of actions for SLSs as defined
in EN 1990 (clause 6.5.3) is to be considered when applying the indirect method. As
irreversible limit states are considered in the indirect method, the characteristic combination
is appropriate.
In this indirect method, the SLS settlements are expected not to be exceeded. This
‘simplified method’ for checking SLS is thus a way to apply the indirect method in which
SLSs and ULSs are checked together.
Some indications are given of what could be considered to be ‘a sufficiently low mobilized
resistance’ in (medium-to-stiff) clays when using the simplified method. Similar figures are
often applied to medium-dense and dense sand, but are, strangely, not given in EN 1997-1.
The use of the simplified method for SLS checking is subject to the following conditions: Clause 2.4.8(4)
(1) Well-established and documented successful experience must exist.
(2) There is no explicit settlement limit specified for checking SLSs and ULSs in the
supported structure due to foundation movements.
(3) Exceptional loading conditions do not prevail, such as for highly inclined or eccentric
loads, highly variable or cyclic loads or climatic loads (such as snow and wind). Where
such exceptional loading conditions prevail, extreme care is required, and the authors do
not recommend the use of semi-empirical methods for lightweight structures.
(4) The method is not applicable for soft clays and for highly organic soils, for which Clause 6.6.1(3)P
settlement calculations are always required.
The indirect method consists of a very simple procedure that corresponds to traditional
practice, with no calculation of settlements being required. In many design situations, the
strength parameters of the ground (tan ϕ¢ and c¢ or cu) are known with much greater
confidence than the ground deformation parameters, and consequently the simplified
method may be more appropriate than settlement calculations.

6.6. Prescriptive method

Prescriptive methods should be applied in the context defined in clause 2.5. A sample Clause 6.5.2.4
method for deriving the presumed bearing resistance for spread foundations on rock is given
in Annex G.

6.7. Structural design Clause 6.8(1)P

The structural design of the foundation involves checks against ULSs occurring in the
foundation element. A check against SLSs (e.g. crack width) may also be relevant.
The values of the internal bending moments and shear forces in the foundation element,
for ULSs and SLSs, are obtained by integrating the bearing pressures under the foundation,
the size of which has been determined to resist geotechnical failure (or the requirements of
serviceability should these lead to a larger foundation size).
For ULS structural design, the bearing pressures under the foundation are deduced from
the ULS design values of the actions, i.e. values including the partial action factors γG and γQ
and the partial material factors γM. It is important to treat all known values of water
pressures as actions as their effect is often unfavourable for the bending moments and
internal forces (e.g. upward water pressure under a raft); the water pressures are then
multiplied by the corresponding action factor (recommended values for the three Design
Approaches are given in Table A.3).

85

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

It should be noted that DA-1 Combination 2 is not relevant for the structural design when
the ground strength plays no role in the assessment of the bearing pressure under the
foundation (this is always the case for conventional methods for the assessment of the
bearing pressures). In DA-1, the structural design of spread foundations is then performed
with the set of action factors of Combination 1, although usually the size of the foundation
has been determined using Combination 2.
Clauses 6.8(2) EN 1997-1 differentiates between stiff and flexible foundations, without giving any guidance
to 6.8(6) on how to do this. Similarly, no guidance is given on the domains of applicability or the
relative merits and limitations of subgrade reaction (spring) and continuum models. The
Clause 2.4.1(13) choice is left to the designer’s judgement and experience and to guidance in the literature.
Clause 6.8(6) Compatibility of strains at a limit state should be considered. Detailed interaction analysis
may be needed.
The distribution of bearing pressures to be used for the SLS check of the foundation (e.g.
crack width) is deduced from the relevant SLS combinations of actions according to EN 1990
and by accounting for the deformations of the ground and the foundation.

Example 6.1: square pad foundation on soft clay

Introduction
This is an example of the ULS design of a footing, using the direct method with the sample
analytical procedure of Annex D. Undrained and drained conditions are considered. ULS
calculations, for persistent and transient situations, are performed for the three Design
Approaches. ULS accidental situations should also be checked, where relevant, but, for
simplicity, the calculations are not presented in this example. When using the direct
method, the SLS should be checked using settlement calculations, which are also not
presented in this example.

Description of the problem

A 0.5 m thick, square pad foundation, bearing on soft clay 1 m below ground level,
carries a vertical, centric, permanent load of 270 kN and a variable load of 70 kN. The
characteristic values of soil properties and other relevant data are shown in Fig. 6.9. The
water table is at the ground surface. For simplicity, the cross-section of the column on the
foundation is neglected.
As the load is applied at the centre of the pad and no moment acts on it, the effective
width and the effective length of the foundation are equal to their nominal values, i.e.
B ¢ = B and L¢ = L; thus A¢ = A.

Pk = 270 kN Qk = 70 kN

h2 = 0.5 m

gk = 25 kN/m3
g k¢ = 15 kN/m 3
h1 = 0.5 m

B×B

Ground properties: gk = 18 kN/m3

gwater = 10 kN/m3
gk¢ = 8 kN/m3
jk¢ = 20˚
ck¢ = 5 kPa
cu, k = 30 kPa

Fig. 6.9. Problem geometry, characteristic values of actions and ground properties

86

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 6. SPREAD FOUNDATIONS

ULS design using analytical calculations

The minimum footing size is assessed for each Design Approach.

(a) Undrained conditions

The analysis is performed in total stresses.

Design Approach 1 Combination 2

The requirement Vd £ Rd is checked for a 1.7 m × 1.7 m pad.
The design value of the actions (including the weight of the foundation and the backfill
on it (Gpad, k = 1.72 × (0.5 × 18 + 0.5 × 25) = 62 kN) is obtained using the action factors
of set A2 in Table A.3:
Vd = γG(Pk + Gpad, k ) + γQQk
Vd = 1.0 × (270 + 62) + 1.3 × 70 = 423 kN
The design value of the vertical bearing resistance is calculated using equation (D.1) of
Annex D, and applying the partial factors to undrained strength of set M2 in Table A.4; the
partial resistance factor γR, v is equal to 1.0 according to set R1 in Table A.5.
Rd /A¢ = (π + 2) cu, d bc sc ic + qd
Thus
cu, d = cu, k/γcu = 30/1.4 = 21.4 kPa
sc = 1.2 (square shape: B¢/L¢ = 1)
bc = 1 (horizontal base and soil surface); ic = 1 (vertical loads)
qd = (γ/γγ)(h1 + h2) = (18/1.0) × (0.5 + 0.5) = 18 kPa
and
Rd /A¢ = (3.14 + 2) × 21.4 × 1 × 1.2 × 1 + 18 × 1.0 = 150 kPa
The design value of the vertical bearing resistance of the foundation (1.70 m × 1.70 m)
is then
Rd = 150 × 1.7 × 1.7 = 433 kN
The requirement Vd £ Rd is fulfilled as 423 kN < 433 kN.

Design Approach 1 Combination 1

The requirement Vd £ Rd is checked for the 1.7 m × 1.7 m pad sized in Combination 2.
The design value of the actions (including the weight of the foundation and the backfill
on it) is obtained using the action factors of set A1 in Table A.3:
Vd = 1.35 × (270 + 62) + 1.5 × 70 = 553 kN
The design value of the vertical bearing resistance is calculated using equation (D.1) of
Annex D, and applying the partial factors on undrained strength of set M1 of Table A.4; the
partial resistance factor γR, v is equal to 1.0 according to set R1 in Table A.5. Thus
cu, d = 30/1.0 = 30 kPa
sc = 1.2 (square shape: B¢/L¢ = 1)
bc = 1 (horizontal base and soil surface); ic = 1 (vertical loads)
qd = (γ/γγ)(h1 + h2) = (18/1.0) × (0.5 + 0.5) = 18 kPa
and

87

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Rd /A¢ = (3.14 + 2) × 30 × 1.2 × 1 × 1 + 18 × 1.0 = 203 kPa

The design value of the vertical bearing resistance of the foundation (1.70 m × 1.70 m)
is then
Rd = 203 × 1.7 × 1.7 = 586 kN
The requirement Vd £ Rd is fulfilled as 553 kN < 586 kN

The GEO ULS design requirements are fulfilled for both Combinations 1 and 2. The
design for DA-1 is governed by Combination 2 as the ratio Rd /Vd is smaller for
Combination 2 than for Combination 1.
The equivalent deterministic overall factor of safety is
OFS = Rk /(Pk + Qk + Gpad ) = 586/(270 + 70 + 62) = 586/402 = 1.46
This value is very low compared with the values used in traditional design.

Design Approach 2
The requirement Vd £ Rd is checked for a 2.0 m × 2.0 m pad.
The design value of the actions (including the weight of the foundation and the backfill
on it) is obtained using the action factors of set A1 in Table A.3:
Vd = 1.35 × (270 + 86) + 1.5 × 70 = 585 kN
The design value of the vertical bearing resistance is calculated using equation (D.1) of
Annex D, in which characteristic values of soil properties are introduced (the material
factor γc, u is equal to 1.0 in Table A.4), and applying the partial factor γR, v of set R2 to the
calculated (characteristic) undrained resistance (γR, v = 1.4 in Table A.5). Thus
cu, k = 30 kPa
sc = 1.2 (square shape: B¢/L¢ = 1)
bc = 1 (horizontal base and soil surface); ic = 1 (vertical loads)
qk = 1.0 × 18 = 18 kPa
and
Rk /A¢ = (3.14 + 2) × 30 × 1.2 × 1 × 1 + 18 × 1.0 = 203 kPa
As Rd = Rk /γR, v = Rk/1.4, the design resistance of the foundation (A¢ = 2.0 m × 2.0 m)
becomes
Rd = 203 × 2 × 2/1.4 = 580 kN
The GEO requirement Vd £ Rd may be considered as being fulfilled as
585 kN @ 580 kN.
The equivalent deterministic overal lfactor of safety is OFS = 812/426 = 1.91.

Design Approach 3
The requirement Vd £ Rd is checked for a 2.0 m × 2.0 m pad.
The actions are ‘structural actions’. Their design values (including the weight of the
foundation and the backfill on it) are obtained using the action factors of set A1 in Table
A.3, because all actions are structural actions:
Vd = 1.35 × (270 + 86) + 1.5 × 70 = 585 kN
The design value of the vertical bearing resistance is calculated using equation (D.1) of
Annex D, and applying the partial factors to undrained strength of set M2 of Table A.4; the
partial resistance factor γR, v is equal to 1.0 according to set R3 in Table A.5. Thus

88

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 6. SPREAD FOUNDATIONS

Table 6.2. Results for undrained conditions

DA-1 DA-2 DA-3

Size (m × m) 1.70 × 1.70 2.00 × 2.00 2.00 × 2.00

Equivalent overall factor of safety 1.46 1.91 1.91

cu, d = cu, k/γcu = 30/1.4 = 21.4 kPa

sc = 1.2 (square shape: B¢/L¢ = 1)

bc = 1 (horizontal base and soil surface); ic = 1 (vertical loads)

qd = (γ/γγ)(h1 + h2) = (18/1.0) × (0.5 + 0.5) = 18 kPa

and
Rd /A¢ = (3.14 + 2) × 21.4 × 1.2 × 1 × 1 + 1.0 × 18 = 150 kPa
The design value of the vertical bearing resistance of the foundation (2.0 m × 2.0 m) is
then
Rd = 150 × 2.0 × 2.0 = 600 kN
The requirement Vd £ Rd is fulfilled as 586 kN < 600 kN.
The equivalent deterministic overall factor of safety is OFS = 812/426 = 1.91

Comparison of results for undrained conditions

Table 6.2 summarizes the calculation results for the three Design Approaches in undrained
conditions.
The equivalent overall factor of safety found in DA-1 may seem small compared with
values commonly found in traditional practice. But the final size of the footing depends
also on the SLS check, and it can be argued that, in traditional practice, the SLS
requirement usually governs the size of the footing. Thus, the result can be compared with Clause 2.4.8(4)
traditional practice only when the SLS has been checked. The SLS can be checked by Clause 6.6.2(16)
calculating the settlement or by the simplified method.

(b) Drained conditions

The ULS requirement is written in terms of effective stress. As all the water pressures are
from the same single source, the same factor is applied to them, and they cancel out.
The actions due to the weight of the submerged footing and to the backfill are their
effective weights. Thus, the action factor applies to the effective (buoyant) weight of the
submerged soil and of the footing. The design value of the action on the ground is then
given by
Vd = γG, unfavVk + γQ, unfavQk + (γG, unfav h2γk¢ + γG, unfav γ ¢concrete h1 ) A

The values of the partial factors γG, unfav and γQ, unfav depend on the Design Approach
adopted. The cross-section of the column has been neglected in this equation.
It should be noted that, using this equation, the action factor for unfavourable actions is
applied to the effective weight of the footing and of the backfill; thus, the action factor for
unfavourable actions is also applied to the favourable water pressure. This is questionable
in some special cases where the water pressure plays an important role in the equilibrium
and a separate factoring of the (total) weight and water forces should be considered.

89

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Design Approach 1 Combination 2

The requirement Vd £ Rd is checked for a 1.85 m × 1.85 m pad.
Again, we start with Combination 2 as, for vertical (and slightly inclined loads), it
always governs the sizing of the foundation.
The design value of the actions (including the effective weight of the foundation and of
the backfill on it) is obtained using the action factors of set A2 in Table A.3:
Vd = γG, unfav(Pk + G¢pad, k ) + γQQk
Vd = 1.0 × (270+39) + 1.3 × 70 = 400 kN
where
G¢pad, k = 1.852 × [(25–10) × 0.5 + (18–10) × 0.5] = 39 kN
The design value of the vertical bearing resistance is calculated using equation (D.2) of
Annex D and applying the partial factors of set M2 in Table A.4 to drained strength
parameters c¢ and tan ϕ¢; the partial resistance factor γR, v is equal to 1.0 according to set
R1 in Table A.5.
Rd /A¢ = q¢Nq, d bq iq sq, d + 0.5γ ¢B¢Nγ, d bγ iγ sγ + c¢Nc, d bc ic sc, d
in which
ϕd¢ = 16.23° (tan ϕd¢ = tan ϕk¢ /1.25)
cd¢ = 5/1.25 = 4 kPa
The design values of bearing capacity factors and shape factors are (see the equations in
Annex D, inserting ϕ = ϕd¢)
Nq, d = 4.43 sq, d = 1.28 iq = 1.0 bq = 1.0
Nγ, d = 2.00 sγ = 0.7 iγ = 1.0 bγ = 1.0
Nc, d = 11.79 sc, d = 1.36 ic = 1.0 bc = 1.0
The effective overburden pressure is q¢ = γ ¢D = 8 × 1.0 = 8 kPa.
Rd /A¢ = 8 × 4.43 × 1.28 + 0.5 × 1.85 × 2.0 × 8 × 0.7 + 4 × 11.79 × 1.36 = 120 kPa
The design value of the vertical bearing resistance of the foundation (1.85 m × 1.85 m)
is then
Rd = 120 × 1.85 × 1.85 = 410 kN
The requirement Vd £ Rd is fulfilled as 400 kN < 410 kN.

Design Approach 1 Combination 1

The design values of the actions (including the effective weight of the foundation pad and
of the backfill on it) is obtained using the action factors of set A1 in Table A.3:
Vd = 1.35 × (270 + 39) + 1.5 × 70 = 522 kN
The design value of the vertical bearing resistance is calculated using equation (D.2) of
Annex D, and applying the partial factors of set M1 (equal to 1.0) in Table A.4 to drained
strength parameters c¢ and tan ϕ¢; the partial resistance factor is equal to 1.0, according to
set R1 in Table A.5.
The design values of bearing capacity factors and shape factors are obtained using
ϕ = ϕd¢ = ϕ k¢; c¢ = cd¢ = ck¢ and the inclination factors i and b are equal to 1.0:
Nq, d = 6.40 sq, d = 1.34 iq = 1.0 bq = 1.0
Nγ, d = 3.93 sγ = 0.7 iγ = 1.0 bγ = 1.0
Nc, d = 14.83 sc, d = 1.40 ic = 1.0 bc = 1.0

90

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 6. SPREAD FOUNDATIONS

The effective overburden pressure is qd = (γ/γγ)(h1 + h2) = (8/1.0) × (0.5 + 0.5) = 8 kPa.
Rd /A¢ = 8 × 6.40 × 1.34 + 0.5 × 1.85 × 3.93 × 8 × 0.7 + 5 × 14.83 × 1.40 = 193 kPa
The design value of the vertical bearing resistance of the foundation (1.85 m × 1.85 m)
is then
Rd = 193 × 1.85 × 1.85 = 660 kN
The requirement Vd £ Rd is fulfilled as 522 kN < 660 kN.

The GEO ULS design requirements are fulfilled for both Combinations 1 and 2. The
design for DA-1 is governed by Combination 2.
Note that the design value of the bearing resistance for Combination 1 is equal to the
characteristic value of the bearing resistance.
The equivalent deterministic overall factor of safety is
OFS = Rk/(Pk + Qk – Wk + Gpad) = 193 × 1.852/(270 + 39 + 70) = 660/379 = 1.74

Design Approach 2
The requirement Vd £ Rd is checked for a 1.95 m × 1.95 m pad.
The design value of the actions (including the effective weight of the foundation and of
the backfill on it) is obtained using the action factors of set A1 in Table A.3:
Vd = 1.35 × (270 + 44) + 1.5 × 70 = 530 kN
The design value of the vertical bearing resistance is calculated using equation (D.2) of
Annex D, in which the partial factors of set M1 in Table A.4 are applied to characteristic
values of soil shear strength parameters c¢ and tan ϕ¢ and the partial resistance factor of
set R2 of Table A.5 (γR, v = 1.4) is applied to the calculated (characteristic) resistance. The
design value of the bearing capacity factors, shape and inclination factors are equal to
their ‘characteristic values’:
Nq, k = 6.40 sq, k = 1.34 iq = 1.0 bq = 1.0
Nγ, k = 3.93 sγ = 0.70 iγ = 1.0 bγ = 1.0
Nc, k = 14.83 sc, k = 1.40 ic = 1.0 bc = 1.0
Rk/A¢ = 8 × 6.40 × 1.34 + 0.5 × 1.95 × 3.93 × 8 × 0.70 + 5 × 14.83 × 1.40 = 194 kPa
Applying the partial resistance factor γRv = 1.4 to the characteristic resistance yields
Rd /A¢ = 194/1.4 = 139 kPa
Rd = 139 × 1.95 × 1.95 = 529 kN
The GEO requirement Vd £ Rd is fulfilled as 530 kN < 529 kN.
The equivalent deterministic overall factor of safety is
OFS = Rk/Vk = Rk/(Pk + Qk – Wk + Gpad ) = 738/(270 + 44 + 70) = 738/384 = 1.92

Design Approach 3
The requirement Vd £ Rd is checked for a 2.15 m × 2.15 m pad.
The actions are ‘structural actions’. Their design value (including the effective weight
of the foundation pad and of the backfill on it) is obtained using the action factors of set
A1 in Table A.3:
Vd = 1.35 × (270 + 53) + 1.5 × 70 = 541 kN
The design value of the vertical bearing resistance is calculated using equation (D.2) of
Annex D, and applying the partial factors of set M2 in Table A.4 to drained strength

91

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Table 6.3. Results for drained conditions

DA-1 DA-2 DA-3

Size (m × m) 1.85 × 1.85 1.95 × 1.95 2.15 × 2.15

Equivalent overall factor of safety 1.74 1.92 2.30

parameters c¢ and tan ϕ¢; the partial resistance factor γRv is equal to 1.0, according to set
R3 in Table A.5:
Nq, d = 4.43 sq, d = 1.28 iq = 1.0 bq = 1.0
Nγ, d = 2.00 sγ = 0.7 iγ = 1.0 bγ = 1.0
Nc, d = 11.79 sc, d = 1.36 ic = 1.0 bc = 1.0
The effective overburden pressure is q¢ = γ¢D = 8 × 1.0 = 8 kPa.
Rd /A¢ = 8 × 4.43 × 1.28 + 0.5 × 2.15 × 2.0 × 8 × 0.7 + 4 × 11.79 × 1.36
Rd /A¢ = 121.5 kPa
Rd = 2.15 × 2.15 × 121.5 = 562 kN
The GEO requirement Vd £ Rd is fulfilled as 541 kN < 562 kN.
The equivalent overall factor of safety is
OFS = Rk /Vk = Rk/(Pk + Qk – Wk + Gpad ) = 906/(270 + 70 + 53) = 906/393 = 2.30

Comparison of the results for drained conditions

Table 6.3 presents a comparison of the results.

Design for SLS

Clause 6.6.2(16) As the direct method is used, settlement calculations are required to check the SLS.
Furthermore, as the ratio of the bearing capacity of the ground at its initial undrained
strength to the applied serviceability load is less than 2.5–3 (in all the Design Approaches),
EN 1997-1 requires that assessments of settlements should be undertaken. The settlement
calculations should be performed using SLS combinations of the actions, with action
factors equal to 1.0 (see EN 1990) and characteristic values of deformation parameters,
for the undrained as well as for the drained loading conditions.
The calculated settlements should be compared with the limiting values established for
the structure being considered. The settlement calculations are necessary before the size
of the footing can be finalized, but they are not presented here.

ULS structural design

The structural design is performed for the foundation, the dimensions of which fulfil the
GEO ULS requirements and the SLS requirements. As the foundation isClause very 6.8(2)
stiff
compared with the ground, a uniform linear distribution of the bearing pressures can be
assumed.
The design value of the bending moments is
Md = [σd × (B/2)2]/2
Md = [(Vd /B2)(B/2)2]/2
Md =Vd /8
where σd is the ULS design value of the bearing pressure due to column load.
Table 6.4 summarizes the calculation results for the three Design Approaches.
The required reinforcement is determined using EN 1992.

92

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 6. SPREAD FOUNDATIONS

Table 6.4. ULS structural design results for the three Design Approaches

Pad dimension, Md
Design Approach Vd (kN) B (m) (kN m/m)

DA-1
Combination 1 470 1.85 59
Combination 2 361 1.85 Not relevant
DA-2 470 1.95 59
DA-3 470 2.15 59

Example 6.2: ULS design of spread foundation for a tower

Introduction
This is an example of footing design by the direct method and using the sample ULS
analysis given in Annex D. The ULS bearing capacity calculations are performed
for persistent and transient design situations using the three Design Approaches. The
checking of the serviceability limit state should be done by calculating the settlement and
the rocking of the footing; these calculations are, however, not presented in this example.

Description of the problem

The footing shown in Fig. 6.10 carries a tall, lightweight structure which is subjected to
significant variable, horizontal loading, for example a windmill or chimney. A moment
results from the horizontal load which is applied to the structure at a point 10 m above the
top of the square pad footing. The footing is 2 m deep, and rests on a dry, medium-dense
sand and gravel layer, the strength parameters of which are ϕk¢ = 35° and ck¢ = 0 kPa. The
critical state (constant volume) value of the angle of shearing resistance is ϕ cv,¢ k = 30°.
The characteristic values of the permanent and variable actions and the points where they
act on the structure are shown in the figure. A low characteristic value for the concrete
weight density is selected, as the permanent weight would appear to be favourable.

Qhk = 300 kN

h = 10 m
Gvk = 600 kN

2m gconcrete = 24.5 kN/m3

Ground properties: g ¢ = 20 kN/m3

j k¢ = 35˚
c k¢ = 0 kPa
¢ k = 30˚
j cv,

Fig. 6.10. Geometry, actions and ground properties for the example of a tall, lightweight
structure

93

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Table 6.5. Characteristic and design values of actions for the different Design Approaches
(weight of the footing not included). The recommended values of partial factors in Table A.3 are
used

DA-1
DA-1 Combination 1, DA-2, DA-3: Combination 2:
set A1 set A2

Vunfavourable Vfavourable Vunfavourable

Characteristic
value of action Partial Design Partial Design Partial Design
Action (kN) Symbol factor load factor load factor load

Permanent
Vertical Gv, k = 600 γG 1.35 810 1.0 600 1.0 600
Variable
Vertical Qv, k = 0 γQ 1.5 0 0 0 1.3 0
Horizontal Qh, k = 300 γQ 1.5 450 1.5 450 1.3 390

The partial factors and design values of the actions are indicated in Table 6.5 for DA-1,
DA-2 and DA-3. As it is not known in advance if the vertical action Gv, k and the weight of
the footing are favourable or unfavourable to the bearing resistance, two values for γG in
set A1 will be considered in DA-1 Combination 1, DA-2 and DA-3. The design situation
where γG = 1.35 is applied to the permanent vertical actions will be denoted ‘Vunfavourable’.

ULS design
Design Approach 1
Vertical bearing resistance
A footing having a size of B × L = 5.6 m × 5.6 m (the characteristic value of its weight
Gpad, k is 1537 kN) is checked for both Combinations 1 and 2. The design values of actions,
ground parameters, bearing capacity factors, shape factors and inclination factors are
summarized in the Table 6.6. As the foundation base and soil surface are horizontal, the
inclination factors b are all equal to 1.0. For Combination 1, it is necessary to consider
the permanent vertical actions as both ‘favourable’ and ‘unfavourable’. This is not
necessary for Combination 2 as the permanent actions are multiplied by a partial factor
γG = 1.0.
The recommended values of the partial factors are taken from Table A.4 (Combination
2 – set M2, γϕ = 1.25; Combination 1 – set M1, γϕ = 1.0) and Table A.5 (set R1, γRv = 1.0).
For all cases analysed in this example the basic requirement Vd £ Rd is fulfilled.
Combination 1, with the permanent vertical actions taken as ‘favourable’, is found to be
the critical case, with a required footing dimensions of 5.6 m × 5.6 m. The required
minimum dimension for Combination 2 is 5.4 m × 5.4 m. Hence, Combination 1 is
marginally critical for sizing of the pad footing (i.e. critical for geotechnical stability).

Sliding resistance
Clause 6.5.3(10) The angle of shearing resistance at constant volume, ϕ¢cv , k, is used when checking sliding
resistance. The passive resistance in front of the foundation is neglected in this example.
The 5.6 m square footing is checked against sliding resistance for both Combinations 2
and 1.

DA-1 Combination 2. The recommended value γϕ = 1.25 indicated in set M2 of Table A.4
is applied to ϕ¢cv, k. For the 5.6 m square footing, the design value of the sliding resistance
is given by

94

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 6. SPREAD FOUNDATIONS

Table 6.6. Calculations for DA-1 Combination 2 and Combination 1, ‘ Vfavourable’ and ‘Vunfavourable’

DA-1 Combination 1

Design value of DA-1 Combination 2 Vfavourable Vunfavourable

Footing size B × L (m2) 5.6 × 5.6 5.6 × 5.6 5.6 × 5.6

Gpad, d (kN) 1537 1537 2075
Gd (kN) 600 600 810
Vd (kN) 2137 2137 2885
Hd (kN) 390 450 450
Md = Hd × 12m (kN m) 4680 5400 5400
i = Hd/Vd 0.182 0.210 0.156
ed = Md/Vd (m) 2.19 2.53 1.87
B¢ = 2 × (B/2 – ed) (m) 1.22 0.54 1.85
A¢ = B¢L (m2) 6.83 3.02 10.36
B¢/L 0.22 0.10 0.33
ϕd (°) 29.3 35 35
cd¢ (kPa) 0 0 0
γ ¢ (kN/m3) 20 20 20
q¢ (kPa) 40 40 40
Nq, d 17.00 33.30 33.30
Nγ, d 17.96 45.23 45.23
Nc, d Not relevant Not relevant Not relevant
sq, d 1.108 1.057 1.189
sγ, d 0.93 0.97 0.90
sc, d Not relevant Not relevant Not relevant
mB 1.82 1.91 1.75
iq, d 0.69 0.52 0.74
iγ, d 0.57 0.50 0.63
ic, d Not relevant Not relevant Not relevant
Rd /A¢ (kPa) 520 + 117 = 637 725 + 122 = 847 1176 + 472 = 1648
Rd (kN) 4330 2557 1790
Rd/Vd 2.03 1.19 5.92
Vd /A¢ (kPa) 313 707 278

RH, d = [γGVk tan(ϕcv, k)/γϕ]/γR, h = (1.0 × 2137 × tan 30°/1.25)/1.0 = 996 kN

Hd = 1.3 × 300 = 390 kN
The design is satisfactory since Hd < Rd, H.

DA-1 Combination 1. The case ‘Vfavourable’ clearly governs the design. The recommended
value γϕ = 1.0 indicated in set M1 in Table A.4 is applied to ϕcv, k. For the 5.6 m square
footing, the design value of the sliding resistance is given by
RH, d = (Vd tan ϕcvd)/γR, h = (2137 × tan 30°)/1.0 = 1136 kN
Hd = 1.5 × 300 = 450 kN
The design is satisfactory since Hd < Rd, H.

The GEO requirement for sliding is fulfilled for both Combinations 1 and 2.

Design Approach 2
Bearing resistance
Calculations using DA-2 can be performed in two ways:

95

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

(1) DA-2 – by factoring the actions at their source and calculating the bearing resistance
using factored values of actions; thus, design values of eccentricity and load
inclination depend on design values of the actions.
(2) DA-2* – by factoring the effects of actions and calculating the bearing resistance
using characteristic values of actions; thus, design values of eccentricity and load
inclination depend on the unfactored (characteristic) values of the actions.
In the first calculation, it is necessary to consider two cases: one with the permanent
vertical actions as ‘favourable’ and the other with the actions as ‘unfavourable’. In the
second calculation, only the case where the vertical actions are unfavourable has to be
considered, as the other case is not relevant.
Square footings having dimensions of 5.65 m × 5.65 m (characteristic value of its
weight Gpad, k = 1564 kN) and 4.8 m × 4.8 m (1129 kN) are checked in calculations 1 and
2, respectively. The design value of the bearing resistance is obtained for both calculations
by applying the partial factor γRv to R/A¢. The recommended resistance factor value of 1.4
is used, as indicated in set R2 of Table A.5. Values used in the calculations are shown in
Table 6.7.

Discussion. When performing the second calculation (i.e. by calculating the eccentricity
of the actions, the inclination factors, and hence Rk, using characteristic values of
Clause 6.5.4(1)P actions), the resulting force lies outside the middle two-thirds of the foundation
(e = 2.08 m > B/3 = 1.6). The foundation will lose contact with the ground over more
than half its width under the service loads. It is common practice (although not required
by EN 1997-1) to put some limit on the eccentricity under characteristic values of actions.
For instance, should it be required that less than half the foundation loses contact with the
soil, the foundation should be at least 6.24 m wide.

Sliding resistance
The angle of shearing resistance at constant volume, ϕ¢cv, is used when checking sliding
resistance. The passive resistance in front of the foundation is neglected in this example.
In the first calculation, where the actions are factored at their source, the case ‘Vfavourable’
clearly governs the design. The recommended value is used for the partial resistance
factor γR, h in set R2 of Table A.5. The design value of the sliding resistance is given by
RH, d = (Vd tan ϕcv, k)/γR, h = (1.0 × (1564 + 600) × tan 30°/1.1 = 1136 kN
Hd = 1.5 × 300 = 450 kN
The design is satisfactory since Hd < RH, d.
In the second calculation, where the effects of actions are factored,
RH, d = (Vk tan ϕcv, k)/γR, h = (1129 + 600) × tan 30°/1.1 = 907 kN
Hd = γQHk = 1.5 × 300 = 450 kN
Again, the design is satisfactory since Hd < RH, d.

Design Approach 3
Vertical bearing resistance
A square footing of dimensions 5.7 m × 5.7 m (characteristic value of its weight
Gpad, k = 1592 kN) is checked for DA-3. The design values of the actions, the ground
parameters, the bearing capacity factors, the shape factors and inclination factors are
summarized in Table 6.8. It is necessary to consider the permanent vertical actions both as
‘Vfavourable’ and as ‘Vunfavourable’.
The recommended values of the partial factors for material properties and resistances
are taken from Tables A.4 (set M2, γϕ = 1.25) and A.5 (set R3, γRv = 1.0).

96

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 6. SPREAD FOUNDATIONS

Table 6.7. Calculations for DA-2, Rd being calculated using design values of actions (columns 1
and 2) and Rd being calculated using characteristic values of actions (column 3)

Calculation 2
(factoring effects of
actions: R calculated
Calculation 1 (factoring actions at their using characteristic
source: R calculated using design values of values of actions:
actions) DA-2*)

Design value of Vfavourable Vunfavourable

Footing size BL (m2) 5.65 × 5.65 5.65 × 5.65 4.80 × 4.80

Gpad, d (kN) 1564 2112 Gpad, k = 1129
Gd (kN) 600 810 Gk = 600
Vd = Gpad, d + Gd (kN) 2164 2922 Vk = 1729, Vd = 2334
Hd (kN) 450 450 Hk = 300, Hd = 450
Md = Hd × 12m (kN m) 5400 5400 Mk = 3600
I = Hd /Vd (–) 0.208 0.154 Hk/Vk = 0.173
ed = Md /Vd (m) 2.50 1.85 ek = Mk/Vk = 2.08
B¢ = 2 × (B/2 – ed) (m) 0.65 1.95 0.64
A¢ = B¢L (m2) 3.67 11.04 3.05
B¢/L (–) 0.12 0.35 0.13
ϕ¢d (°) 35 35 35
c¢d (kPa) 0 0 0
γ ¢ (kN/m3) 20 20 20
q¢ (kPa) 40 40 40
Nq, d (–) 33.30 33.30 33.30
Nγ, d (–) 45.23 45.23 45.23
Nc, d (–) Not relevant Not relevant Not relevant
sq, d (–) 1.069 1.189 sq, k = 1.08
sγ, d (–) 0.96 0.90 sγ, k = 0.96
sc, d (–) Not relevant Not relevant Not relevant
mB (–) 1.89 1.75 1.88
iq, d (–) 0.64 0.743 iq, k = 0.70
iγ, d (–) 0.51 0.627 iγ, k = 0.58
ic, d (–) Not relevant Not relevant Not relevant
Rd /A¢ (kPa) (911 + 144)/1.4 (1176 + 472)/1.4 (1006 + 161)/1.4
= 753 = 1178 = 833
Rd (kN) 2763 13005 2540
Rd /Vd (–) 1.27 4.45 1.09
Vd/A¢ (kPa) 707 265 760

Sliding resistance
The angle of shearing resistance at constant volume, ϕ¢cv, is used when checking sliding
resistance. The recommended strength factor value γϕ= 1.25 in set M2 in Table A.4
is applied to ϕ¢cv, k. The passive resistance in front of the foundation is neglected in
this example. The case ‘Vfavourable’ clearly governs the design. For the square footing of
dimensions 5.7 m × 5.7 m, the design value of the sliding resistance is given by
RH, d = (γG, favVk tan ϕcv, k/γϕ)/γR,h = (1.0 × 2192 × tan 30°/1.25)/1.0 = 1022 kN
Hd = γQHk = 1.5 × 300 = 450 kN
The design is satisfactory since Hd < RH, d.

97

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Table 6.8. Calculations for DA-3

Design value of Vfavourable Vunfavourable

Footing size BL (m2) 5.7 × 5.7 5.7 × 5.7

Gpad, d (kN) 1592 2149
Gd (kN) 600 810
Vd (kN) 2192 2959
Hd (kN) 450 450
Md = Hd × 12m (kN m) 5400 5400
i = Hd/Vd (–) 0.205 0.152
ed = Md/Vd (m) 2.46 1.83
B¢ = 2(B/2 – ed) (m) 0.77 2.05
A¢ = B¢L (m2) 4.41 11.68
B¢/L (–) 0.14 0.36
ϕ¢d (°) 29.3 29.3
c¢d (kPa) 0 0
γ ¢ (kN/m3) 20 20
q¢ (kPa) 40 40
Nq, d (–) 17.00 17.00
Nγ, d (–) 17.96 17.96
Nc, d (–) Not relevant Not relevant
sq, d (–) 1.07 1.18
sγ, d (–) 0.96 0.89
sc, d (–) Not relevant Not relevant
mB (–) 1.88 1.74
iq, d (–) 0.65 0.75
iγ, d (–) 0.52 0.64
ic, d (–) Not relevant Not relevant
Rd /A¢ (kPa) 470 + 68 = 539 602 + 210 = 812
Rd (kN) 2325 9483
Rd /Vd (–) 1.06 3.20
Vd /A¢ (kPa) 500 253

Eccentricity of the load

For all three Design Approaches it appears that, for the large eccentricity that acts in this
example, a small increase in applied horizontal force to the structure (and hence an
Clause 6.5.4 increase in moment) would lead to overturning of the foundation. It is vital to dimension
the footing to cater for this and any other unforeseen situation. In all three Design
Approaches, stability is strongly adversely affected by very small decreases (of a few
centimetres) of the size of the footing. EN 1997-1 therefore requires that, where the
design load passes outside of the middle two-thirds of the footing, an allowance must be
made for construction tolerances. This will typically require that dimensions B¢ and L be
increased by 0.1 m.

ULS due to soil deformations

As the design action is highly eccentric and the structure is tall, the second-order effects of
rotation of the foundation should be checked; as a consequence of rocking, the vertical
load also becomes eccentric. The eccentricity under design values of the vertical load
has to be combined with the eccentricity due to the horizontal load. If the combined
eccentricity becomes significant, the ULS check of bearing resistance must be repeated
for the increased eccentricity.

98

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 6. SPREAD FOUNDATIONS

Design for the SLS

Owing to the large eccentricity of the load, it is strongly recommended that deformation
be calculated. Rocking of the foundation and subsequent horizontal displacement of the
top of the structure particularly should be examined and compared with allowable values
in serviceability conditions. Calculations of deformation should be performed using
characteristic values of the soil deformation parameters and of the actions.

ULS structural design

The calculations for the structural sections of the foundation require an assumption of the
distribution of the ULS bearing pressures, taking due account of the area over which the
foundation loses contact with the soil as a consequence of the large eccentricity of the
action.
For DA-1, Combination 1 will apply.

Example 6.3: design based on the indirect method using pressuremeter test
results
Introduction
In this example, a footing is designed using the indirect method, and the sample semi- Clause 6.4(5)P
empirical model of bearing resistance estimation based on pressuremeter tests results and
given in Annex E. This annex should be complemented by values of the pressuremeter
bearing resistance factors kp for spread foundations given in EN 1997-2, Annex C1.
The design of spread foundations using semi-empirical rules embodies much previous
experience; for this example, using the results of pressuremeter tests, the previous
experience is borrowed from French practice (Frank, 1999).
In the indirect method, the concept of ‘Design Approaches’ does not exist, as the check
is based on the SLS loads.

Description of the problem

Square pads (of approximate dimensions 2 m × 2 m, depth 1 m, thickness 0.5 m, covered
by 0.5 m of backfill, i.e. Gk = 84 kN) bear on medium-dense sand and support a precast,
flexible industrial building. The water table is deep below the foundation level. The
weight density, γ, of the soil is equal to 18 kN/m3. The loads are vertical and act through
the pad centres. The characteristic values of the permanent and variable loads are
Gk = 1600 kN and Qk = 300 kN.
Four pressuremeter (PMT) tests are performed with equal spread over the construction
site. From Annex C1 of EN 1997-2, the value of the bearing resistance is given as (see also
Annex E in EN 1997-1)
R/A = σv0 + kp p*le
where:

• p*le is the equivalent (characteristic) net limit pressure at the base of the foundation,
calculated by a smoothing procedure (e.g. see Frank, 1999) of a cautious ‘characteristic’
PMT profile deduced from the four tests. In this example, p*le = 1.1 MPa.
• kp is the bearing resistance factor. In this example the sand is ‘type B’, and kp = 1.2.
• σv0 is the total initial vertical stress acting at foundation level: σv0 = Dγ.

Design using the indirect method

The indirect method uses comparable experience, and covers SLS and ULS in one Clause 6.4(5)P
calculation. This example uses the semi-empirical rule given in Annex E for the
assessment of the bearing resistance from pressuremeter test results. The comparable
experience chosen in relation to the SLS loads is that the design satisfies both ULS and

99

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

SLS requirements if the ratio of available bearing resistance to the serviceability load
combination is greater than 3.0. The indirect method becomes identical to the simplified
method for checking the serviceability conditions.
The characteristic value of the equivalent limit pressure is 1.1 MPa. The characteristic
value of the bearing resistance is
Rk /A = 18 + 1.2 × 1100 = 1338 kPa
The most unfavourable load applied in serviceability conditions is assumed in this
example to be the characteristic combination (see EN 1990, clause 6.5.3, and Table A1.4):
Vk = 1600 + 84 + 300 = 1984 kN
Adopting the ratio Rk /Vk = 3.0 for the indirect method leads to the required size of the
footing:
A > 1984/(1338/3.0) = 4.45 m2
A square footing of 2.10 × 2.10 m fulfils the requirement.
It should be pointed out that when applying the indirect method, no value of the
deformation is calculated, and the real safety margin in the ULS remains unknown.

100

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 7

Pile foundations

This chapter is concerned with the design of pile foundations. The material described in this
chapter is covered in Section 7 of EN 1997-1, with the values of the partial factors taken from
Annex A. The structure of the chapter follows that of Section 7:
7.1. General Clause 7.1
7.2. Limit states Clause 7.2
7.3. Actions and design situations Clause 7.3
7.4. Design methods and design considerations Clause 7.4
7.5. Pile load tests Clause 7.5
7.6. Axially loaded piles Clause 7.6
7.7. Transversely loaded piles Clause 7.7
7.8. Structural design of piles Clause 7.8
7.9. Supervision of construction Clause 7.9
Section 7 is one of the Sections of EN 1997-1 providing the most comprehensive guidance
for actually performing a design.
The core of Section 7 is devoted to the behaviour of pile foundations under axial (vertical) Clause 7.6
loads. The importance of static load tests is clearly recognized as the basis of pile design Clause 7.4.1(1)P
methods. An innovative concept introduced in this section, with regard to traditional pile
design, is the use of correlation factors ξ for deriving the characteristic compressive and
tensile resistances of piles either from static pile load tests or from ground test results (for Clause 7.6.2.2(8)P
compressive resistance: clauses 7.6.2.2(8)P and 7.6.2.3(5)P, respectively, are relevant). In Clause 7.6.2.3(5)P
both cases, the correlation factor ξ depends mainly on the number of tests performed,
whether pile load tests or profiles of ground tests.

7.1. General
Section 7 of EN 1997-1 applies to all piles, regardless of the installation method (driving, Clause 7.1(1)P
jacking, screwing or boring with or without grouting) and their expected behaviour (whether
end-bearing or friction).
Piles are meant to be used to support loads arising from a structure. Though micropiles
are not mentioned explicitly, it is understood that the provisions of Section 7 also apply to
them, where relevant. It is the same for ‘barrettes’ and panels of diaphragm wall or sheet
piles in a wall, when they are located beneath a structure and carry loads at their head
(whether vertical or horizontal). When piles are only used for reducing the settlement of a Clause 7.1(2)
spread or raft foundation (such piles are often referred to as ‘creep’ piles), it is stated that the
provisions do not apply directly. This is because the overall factor of safety with regard to
resistance failure for such piles can be (much) smaller than for standard piles, as the bearing
function is provided by the spread or raft foundation.

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Clause 7.1(3)P It is assumed that the piles are installed in conformity with the appropriate execution
standards. Such standards have already been established by CEN Technical Committee 288
(CEN/TC 288), or will be established in the near future (see Table 1.1).

7.2. Limit states

Clause 7.2(1)P When designing a pile foundation, a number of limit states need to be considered. Clause
7.2(1)P provides the list of the 11 most common limit states that need to be considered in the
case of pile foundations. This list does not imply that calculations or other explicit checks
should be carried out for all of them. The first seven limit states listed are the usual failure
modes, either geotechnical or structural failures, and hence are ultimate limit states (ULSs).
Of the remaining four, excessive movements (settlement, heave and lateral movements) can
lead to a serviceability limit state (SLS) or to a ULS in the supported structure, while
unacceptable vibrations are an SLS (see Chapter 3 in the Designer’s Guide to EN 1990, or
Chapter 2 in this guide, for a general definition of limit states).

7.3. Actions and design situations

Clause 7.3.1 When designing a pile foundation, the actions listed under clause 2.4.2(4) should be
Clause 2.4.2(4) considered. The actions to be taken into account depend on the type of pile foundation and
the particular design situation.
For pile foundations, permanent and variable loads from supported structures, surcharges
and traffic loads, as well as ground movements, are the most frequent actions to be
considered.
Pile foundations are usually studied separately for axial (normally vertical) and for
transverse (normally horizontal) loadings.

Actions due to ground movements

The actions caused by ground movements are very specific to pile foundations, and the soil–
structure interaction resulting from these ground movements, which is usually hidden in the
Clause 7.3.2.1(1)P design calculations, complicates the analyses. Typical ground movements to be considered
are:
Clause 7.3.2.2 • vertical settlements causing downdrag, i.e. negative shaft friction on the piles
Clause 7.3.2.3 • upwards soil displacements causing heave
Clause 7.3.2.4 • horizontal soil displacements causing transverse loadings.
Clause 7.3.2.1(2) Most design methods for piles located in ground subject to displacements are based on a
value for the action which is fixed by other considerations. In order to be on the safe side, the
values of the stiffness or the resistance of the ground should be chosen as upper values.
Indeed, upper values will lead in most cases to a safe estimate of the adverse effects of the
ground displacements on the piles.
Clause 7.3.2.1(3)P Eurocode 7 requires that one of the following two methods is used:
• Either the ground displacement is taken as the action in the design calculations and an
interaction analysis is carried out; this is the case when methods based on load transfer
functions, such as t–z and p–y curves, are being used (z and y being the relative
displacements of the pile with respect to the ground). The absolute ground displacement
is inserted, at the beginning of the calculation, as a correction to the displacement z or y
(Alonso et al., 1984).
• Or the load on the pile caused by the ground movement is taken as the action. This value
must be an upper bound to the characteristic value.
Note that the partial material and resistance factors given in Annex A are intended to
provide low (conservative) design values. They are not meant to yield high (unconservative)

102

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 7. PILE FOUNDATIONS

design values, such as the ones needed in the case of loads acting on piles caused by ground
movements (e.g. downdrag). This issue is not addressed in EN 1997-1, but for this design
situation the authors of this guide recommend that the partial material and resistance
factors, e.g. γϕ , are treated as partial action factors, and the downdrag load (a geotechnical
action) is multiplied rather than divided by of the values of these factors in Annex A (see
Example 7.4).

Downdrag (negative friction)

The usual conservative way of designing piles subjected to downdrag is to estimate the Clauses 7.3.2.2(1)P
maximum (long-term) possible value of the downdrag load. An example of a pile foundation to 7.3.2.2(3)
designed using the three Design Approaches of Eurocode 7 and taking into account the
maximum downdrag load is given below in Example 7.4.
Use of the maximum possible downdrag load can lead to very conservative or even
unrealistic designs, particularly when the settlements of the ground are small and/or the Clauses 7.3.2.2(4)
compressible layer is very thick. In these situations, a careful soil–pile interaction analysis to 7.3.2.2(6)
should be carried out. The aim of such an analysis is to estimate, in particular, the ‘neutral
point’ in the compressible layer, which is the depth where the downward displacement of the
pile equals the downward movement of the ground around the pile, and hence the force
between the soil and the pile shaft ceases being a downward action on the pile and starts to
become an upward shaft resistance. As mentioned above, a design procedure using ‘t–z
curves’, where the profile of ground displacement is an input, is well adapted to the
calculations mentioned under clause 7.3.2.2(6). Clause 7.3.2.2(6)

Heave
For heave, EN 1997-1 requires that the upward movement is treated as an action and that an Clause 7.3.2.3(1)P
interaction analysis is carried out. This is because the movements are usually small enough to
allow treatment in this way.

Transverse loading
Many situations are encountered where piles are subjected to transverse loadings caused by
ground movements. The design situations which need to be considered in the case of Clause 7.3.2.4(2)
transverse loading are listed under clause 7.3.2.4(2). A particular situation where this type of
loading is very common is bridge abutments on piles.
Eurocode 7 recommends a soil–pile interaction analysis in the case of transverse loading. Clause 7.3.2.4(3)
This can be carried out using the theory of beams on linear or non-linear supports and a
horizontal modulus of reaction or generalized p–y curves (see also clause 7.7). Clause 7.7

7.4. Design methods and design considerations

As mentioned above, the methods accepted by Eurocode 7 for the design of piles must nearly Clause 7.4.1(1)P
all be based, directly or indirectly, on the results of static pile load tests. This is the case, of
course, when the results of static pile load tests on a specific site form the direct basis for the Clauses 7.6.2.2
design. It is also the case when calculation methods are used, as their validity should be based Clause 7.6.3.2
on the results of static pile load tests ‘in comparable situations’, or when dynamic tests Clauses 7.6.2.4
are used, as their validity should also have been demonstrated by static tests ‘in to 7.6.2.6
comparable situations’. One notable exception is when a design method based on the
observed performance of comparable pile foundations is used, provided it is ‘supported by
the results of ground investigation and ground testing’.
The design method based on the results of pile load tests is applicable to the design of piles
under either axial loads and/or transverse loads. Pile load tests are the subject of the entire Clause 7.5
clause 7.5.
Clause 7.4.2 provides rather long lists of items to be considered or deserving attention in Clause 7.4.2
the design of piles, including the choice of pile type. While it should be recognized that some

103

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

of the listed requirements are not easy to fulfil explicitly, some others are normally addressed
Clause 7.4.2(2)P in common practice. When using pile load test results, it is particularly important to
take account of the nature of the loading and the potential changes in the ground and
groundwater conditions.

7.5. Pile load tests

Clause 7.5 The requirements for pile load tests in Clause 7.5 apply to both static and dynamic tests,
unless otherwise stated. While some provisions clearly apply only to compression or tension
piles (axial loadings), others apply to transverse as well as axial loadings.
Pile load tests are especially required when there is great uncertainty about the behaviour
of piles, either because of the installation method, the ground conditions and the anticipated
loadings or because unexpected behaviour was noted during installation. The general way to
Clause 7.5.1 use the pile load test results is explained in clause 7.5.1.
Clause 7.5.1(3) In the case of complex loadings, which are not easy to reproduce in pile load tests,
Eurocode 7 states the use of load tests should be replaced by design calculations with very
cautious design values for the ground parameters. This is the situation in the case of cyclic
loads, for instance.
Clause 7.5.1(4)P The number and location of pile load tests on trial piles must be chosen according to the
Clause 7.5.2.2(1)P ground conditions. In particular, the most unfavourable location must be tested.
Clause 7.5.1(4)P The representivity of pile load tests is very important, as the results will directly influence
Clause 7.5.1(5)P the design of the pile.

Static load tests

Clause 7.4.1(3) Static load tests can be performed on trial piles or on working piles which will form part of
the foundation.
Clause 7.5.2.1(1)P With regard to the exact test procedure, clause 7.5.2.1(1)P only refers, in a note, to the
suggested method published in the ASTM Geotechnical Testing Journal (Smoltczyk, 1985),
which is essentially the method normally followed in current European practice. Recently,
the European Technical Committee on piles (ETC3) of the International Society for Soil
Mechanics and Geotechnical Engineering (ISSMGE) has produced a recommendation
document for the execution and interpretation of axial static pile load tests which is fully
consistent with the definitions and philosophy of Eurocode 7 (De Cock et al., 2003). This
document will form the basis for the European standard being prepared by CEN/TC 341.
Clause 7.5.2.1(1)P From the measurements during static load tests, conclusions must be able to be drawn
Clause 7.6.2.2 about the deformation, creep and rebound of a pile foundation. These aspects are useful, in
Clause 7.6.3.2 particular, to check the SLS requirements. ULS calculations are often based on measured
ultimate failure loads, at least for axial loading. Suprisingly, Eurocode 7 states that the
designer should be able to draw conclusions about the ultimate failure load from only the
loading of trial piles. However, it should be understood that it is not always necessary to
bring trial piles to failure: the common practice of deriving the ultimate failure load by
Clause 7.5.2.3(2)P extrapolating the load–displacement graph can be used. In clause 7.5.2.3(2)P, Eurocode 7
states that working piles must be loaded to at least the design load for the foundation.
Although it is not mentioned, it should be understood that this design load is the one
corresponding to persistent situations.
Clause 7.5.2.1(4) When designing pile foundations for tensile loading, static load tests should always be
performed to failure. The reasons for this are because it is easier to reach failure in tensile
tests and because, due to the ‘brittle’ behaviour of piles in tension, it is dangerous to make
an optimistic estimate of the ground tensile resistance by extrapolating from the load–
displacement graph.
Clause 7.5.2.2(2)P For trial piles, Eurocode 7 requires a thorough ground investigation in order to be sure of
Clause 7.5.2.2(3)P the nature of the ground around the piles and beneath their bases, and full documentation of
Clause 7.9 the installation method, as for the permanent piles of a foundation.

104

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 7. PILE FOUNDATIONS

Dynamic load tests

The use of the results of dynamic load tests for assessing the compressive resistance of piles is Clauses 7.5.3(1)
the subject of clause 7.5.3(1) and also clauses 7.6.2.4, 7.6.2.5 and 7.6.2.6, which are concerned Clauses 7.6.2.4
with dynamic impact tests, pile-driving formulae and wave equation analyses, respectively. to 7.6.2.6
Eurocode 7 sets the following very stringent criteria restricting the use of dynamic load tests
for assessing the compressive resistance of piles to when:
• an adequate site investigation has been carried out
• the method has been calibrated against static load tests on the same type of pile, of
similar length and cross-section, and in comparable soil conditions.
The sentence on the stringent criteria restricting the use of dynamic load tests is repeated Clauses 7.6.2.4
in each of the three clauses concerned with the determination of the compressive resistance to 7.6.2.6
from dynamic impact tests, pile-driving formulae and wave equation analysis.
If more than one type of dynamic test is performed, cross-checking of their results is Clause 7.5.3(2)P
mandatory.

Load test report

Pile load tests reports are of prime importance, not only to ensure that there is no
misunderstanding about the precise meaning of the test results before the actual design is
finalized but also in order to extract efficiently all possible information from the tests. Thus,
an extensive list of items to be reported from pile load tests is given in clause 7.5.4(1)P. Clause 7.5.4(1)P

7.6. Axially loaded piles

7.6.1. General
As already stated, clause 7.6 is the core of the section of EN 1997-1 on pile foundations. After Clause 7.6
some general considerations, it deals successively with the determination of the compressive Clauses 7.6.1
ground resistance and the tensile ground resistance for ULS design calculations and the to 7.6.4
vertical displacements for checking the SLS of supported structures. The structural design of
the piles (including possible buckling) is the subject of clause 7.8. Clause 7.8
The term ‘ground resistance’, used in Section 7, means the bearing or tensile capacity of a
pile. The word ‘resistance’ is used here to indicate that it is the maximum reaction of the
ground (maximum shaft resistance and, for a pile in compression, the maximum base
resistance), as opposed to being the actions applied to the ground (through a pile).
The limit states to be checked explicitly are given in clause 7.6.1.1(1)P. Some of the items Clause 7.6.1.1(1)P
are also listed in clause 7.2(1)P. The first three items on the list are ULSs: the bearing Clause 7.2(1)P
resistance failure of a single pile, the bearing resistance failure of a whole pile foundation,
and the collapse or severe damage to a supported structure caused by excessive pile
foundation displacements. The last item is the SLS of the supported structure caused by
displacements of the foundation that exceed the agreed limiting value.
The ULS bearing resistance failure of a single pile and of a whole pile foundation
corresponds to the traditional bearing capacity design, even if Eurocode 7 introduces some
innovative features for calculating the characteristic ground resistance and if the calculations
are carried out with partial factors modifying both the applied loads and the ground
resistance.
For pile foundations, the calculation of displacements is not common practice, especially Clause 7.6.1.1(4)P
for a ULS in the supported structure. For piles which undergo significant settlements,
Eurocode 7 requires that a cautious range of settlements is estimated. It should be noted that
no further guidance is given in Section 7 concerning the method for checking displacements
corresponding to a ULS in the supported structure. The guidelines given in Chapter 6 of this
guide for spread foundations are considered to be relevant to pile foundations as well.
Compressive or tensile resistance failure is defined as the state in which the pile foundation Clause 7.6.1.1(2)
displaces significantly downwards or upwards with negligible increase or decrease in

105

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Table 7.1. Sets of partial factors for pile design in persistent and transient situations

Ground
Design Approach Structural action Geotechnical actiona resistance

DA-1
Combination 1 Set A1 Sets (M1) + A1 Sets (M1) + (R1)
Combination 2 Set (A2) Sets M2 + (A2) Sets (M1) + R4
DA-2 Set A1 Sets (M1) + A1 Sets (M1) + R2
DA-3 Set A1 Sets M2 + (A2) Sets M2 + (R3)

(A2), (M1), (R1) and (R3): sets A2 (permanent actions), M1, R1 (driven piles in compression) and R3 (all
piles in compression) have values of partial factors equal to 1.0.
a
Unfavourable action, e.g. negative friction or transverse loading

resistance. During load tests of piles in compression it is often difficult to reach this state, or to
Clause 7.6.1.1(3) derive it from the load–settlement plot; in these cases, Eurocode 7 recommends that a
pile head settlement of 10% of the pile diameter is used as the ‘failure’ criterion. This
recommendation is important, since the ground resistance calculation methods are based on
failure loads measured during static load tests, which can vary significantly depending on the
failure criterion used.
Clause 7.6.1.2(1)P As in the design of spread foundations, the overall stability of pile foundations must be
considered. The provisions of Section 11 apply in the design of pile foundations as in other
Clause 7.6.1.2(2) geotechnical designs (see also Chapter 6 of this guide). In the case of pile foundations,
failure surfaces passing below the piles and also failure surfaces intersecting the piles should
both be considered.

7.6.2. Compressive ground resistance (ULS)

General
Clause 7.6.2.1(1)P Compressive ground resistance for all ULSs is checked using the following basic inequality:
Fc, d £ Rc, d (7.1)
where Fc, d is the design axial compression load on a single pile or group of piles at the ULS,
and Rc, d is the design value of the compressive ground resistance of a single pile or group of
piles at the ULS.
The ULS loads Fc, d are determined from the combinations of actions which follow the
general format of the partial factor method (see the Designer’s Guide to EN 1990), i.e.
the combinations for persistent or transient design situations (also called ‘fundamental’
combinations in EN 1990), the combinations for accidental situations and the combinations
for seismic situations.
Clauses 7.6.2.2 The ultimate compressive resistance Rc, d in inequality (7.1) can be determined from static
to 7.6.2.6 load tests, from ground test results or on the basis of dynamic load tests.
Clause 2.4.7.3.4 The sets of partial factors of Annex A, applicable to pile design in persistent and transient
situations for use with the three Design Approaches, are given in Table 7.1.
In the design of piles, Design Approaches 1 (DA-1) and 2 (DA-2) include resistance
factoring, i.e. when calculating the ground resistance the partial factors for the ground
parameters are equal to 1.0 (set M1 in Table A.4) while those for resistance are greater than
1.0 (sets R4 or R1 – except for driven piles – for DA-1, and set R2 for DA-2 in Tables A. 6 to
A.8). In contrast, Design Approach 3 (DA-3) involves material factoring, i.e. partial factors
greater than 1.0 are applied to ground parameters when calculating the ground resistance
(set M2 in Table A.4) while those for resistance are equal to 1.0 (set R3 for all piles in
compression, in Tables A.6 to A.8.
Clause 2.4.7.1(3) For accidental and seismic design situations, the combinations of actions are formed with
partial factors for actions normally equal to 1.0. Eurocode 7 does not recommend values for

106

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 7. PILE FOUNDATIONS

the partial factors on the resistances in the case of accidental situations. It is usual practice to
adopt values equal to 1.0, but they can depend on the particular design situation. For the
values of the partial factors on pile resistances in the case of seismic design situations, see
Eurocode 8 – Part 5.
It should be noted that inequality (7.1) is valid for the whole foundation, and does not
necessarily have to be checked for each individual pile independently. Equations (7.2) to
(7.11) for Rc, d later on in Section 7 are intended to be used for the design of individual piles.
It is also important to note that Eurocode 7 draws attention to the fact that a ULS is only Clause 7.6.2.1(6)
reached when a significant number of piles fail simultaneously. In fact, as long as a pile
foundation does not become a mechanism (in the static sense), a ULS is not reached. A
mechanism can be created by the resistance of several piles being exceeded, as well as by Clause 7.6.2.1(5)P
failure of the structure connecting the piles.
The analysis of a ULS failure involving a whole pile foundation is not frequently applied.
It requires a non-linear analysis of the load redistribution among the piles. It may be
assumed, however, that similar load redistribution effects may be accounted for in a
simplified way, by selecting different characteristic pile resistances depending on the
stiffness of the structure. Account of the structural stiffness may be introduced in a
simplified way through the factor 1.1 of clauses 7.6.2.2(9) and 7.6.2.3(7). This factor should Clause 7.6.2.2(9)
not be used when the non-elastic load redistribution is explicitly taken into account. Clause 7.6.2.3(7)
Eurocode 7 draws attention to two other aspects for groups of piles:

• Special care is needed for edge piles, as they are usually more heavily loaded than the Clause 7.6.2.1(8)
other piles in the case of eccentrically loaded stiff foundations or inclined loads and have
a stiffer behaviour than centre piles, and hence attract more load.
• A group of piles loaded in compression may also fail as a block, consisting of the piles Clause 7.6.2.1(3)P
and the ground contained between the piles. The code allows such a block to be treated Clause 7.6.2.1(4)
as a single pile of large diameter. The wording of clause 7.6.2 might be misleading
because the methods and equations given in this clause do not apply to such a case; the
single-pile block is usually analysed by considering the vertical bearing resistance of an
equivalent raft foundation, e.g. according to the calculation methods of Section 6.
Clearly, a check against punching failure of such an equivalent raft should be carried Clause 7.6.2.1(11)
out, if relevant, as clause 7.6.2.1(11) also applies.
Clauses 7.6.2.1(9)P to 7.6.2.1(13) draw attention to other factors to be considered when Clauses 7.6.2.1(9)P
assessing the compressive resistance of single piles: to 7.6.2.1(13)
• the possible presence of a weak stratum below the pile base
• the geometry of the base compared with that of the shaft
• the possibility of a plug ‘effect’ for open-ended driven piles.

Ultimate compressive resistance from static load tests

The procedure for determining the compressive resistance of a pile from static load tests is
based on analysing the compressive resistance, Rc, m, values measured in static load tests on Clause 7.6.2.2(2)P
one or several trial piles. The trial piles must be of the same type as the piles of the
foundation, and must be founded in the same stratum.
An important requirement stated in Eurocode 7 is that the interpretation of the results of Clause 7.6.2.2(7)P
the pile load tests must take into account the variability of the ground over the site and the Clause 7.6.2.2(11)P
variability due to deviation from the normal method of pile installation. In other words,
there must be a careful examination of the results of the ground investigation and of the pile
load tests results. The results of the pile load tests might lead, for example, to different
‘homogeneous’ parts of the site being identified, each with its own particular characteristic
pile compressive resistance.
The steps of the procedure for determining the design compressive resistance of a pile
foundation from static load tests are as follows:

107

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Clause 7.6.2.2(8)P (1) From the measured compressive resistances Rc, m determine the characteristic value Rc, k
from the following equation:
Rc, k = Min{(Rc, m)mean/ξ1; (Rc, m)min/ξ2} (7.2)
where ξ1 and ξ2 are correlation factors related to the number n of piles tested, and are
Clause 7.6.2.2(7)P applied to the mean (Rc, m)mean and to the lowest (Rc, m)min of Rc, m, respectively. The
recommended values for these correlation factors, given in Annex A, are intended
primarily to cover the variability of the ground conditions over the site. However, they
may also cover some variability due to the effects of pile installation.
Clause 7.6.2.2(9) (2) If the structure connecting the piles is stiff and strong enough to transfer loads from
weaker piles to stronger piles, the values for ξ1 and ξ2 may be divided by 1.1, provided the
value of ξ1, which is applied to the mean pile resistance, is not lower than 1.0.
Clause 7.6.2.2(7)P (3) As mentioned above, when several load tests have been performed, the first estimate
Clause 7.6.2.2(10)P of an overall Rc, k value, based on all the load test results, might lead to different
‘homogeneous’ parts of the site being identified. In this situation, the procedure is
started again from step 1 for each ‘homogeneous’ part of the site. This can result in less
conservative Rc, k values than would be obtained if, as in the first estimate, the soil
conditions were assumed to be the same over the whole site and the minimum value of
Rc, m governs the Rc, k value over the whole site.
Clause 7.6.2.2(12) (4) If necessary, Rc, k can be split into the characteristic shaft resistance Rs, k and the
Clause 7.6.2.2(13) characteristic base resistance Rb, k. This is possible particularly when measurements to
determine the load in the pile have been made along the shaft of the tested piles
or when calculations based on the results of ground tests or dynamic tests have been
carried out.
Clause 7.6.2.2(14)P (5) The design pile compressive resistance, Rc, d is obtained by applying the partial factor γt
to the total characteristic resistance or the partial factors γs and γb to the characteristic
shaft resistance and characteristic base resistance, respectively, in accordance with the
following equations:
Rc, d = Rc, k /γt (7.4)
or
Rc, d = Rb, k /γb + Rs, k /γs (7.5)
Rc, d for persistent and transient situations may be obtained from the results of pile load
tests using DA-1 and DA-2 and the recommended values for the partial factors γt or γs
and γb given in Tables A.6, A.7 and A.8. DA-3 is not applicable in the case of pile load
tests because the procedure using the results of load tests involves applying partial
factors directly to the characteristic resistances Rc, k or Rs, k and Rb, k, derived from load
tests, while DA-3 involves applying partial factors to the characteristic values of the
ground strength parameters. For accidental situations, unless stated otherwise in the
National Annex, it can be assumed that γt = γs = γb = 1.0.
Clause 7.6.2.1(3)P In the case of pile groups, the compressive resistance of the whole foundation is
Clausse 7.6.2.1(4) determined either by summing the compressive resistances of the individual piles or by
assuming a block failure, whichever yields the lower value.

Some background information on the recommended values for ξ1 and ξ2 given in Annex A
can be found in Bauduin (2001). These values are based on a reference value of about 10%
for the coefficient of variation of the pile compressive resistance: for coefficient of variation
values less than 10%, the mean of the measured resistances should govern the design,
whereas for coefficient of variation values greater than 10%, the lowest measured resistance
should govern.
Example 7.1 below illustrates the calculation of the design compressive resistance of a
single pile from the results of static load tests following the procedure described above.

108

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 7. PILE FOUNDATIONS

Ultimate compressive resistance from ground tests results

General
As already mentioned, methods for predicting the compressive resistance of piles from Clause 7.6.2.3(1)P
ground test results can be used, provided that they have been established from pile load tests
and from comparable experience.
To account for the additional uncertainty of the calculation method, Eurocode 7 allows a Clause 7.6.2.3(2)
model factor to be introduced in order to ensure that the prediction is safe enough. This
model factor is, in fact, a calibration factor for the method. However, no further information
is given about it in the code, but it is anticipated that it will play an important role when the
existing prediction methods are adapted to the new design procedures, partial factors and
coefficients being introduced by the Eurocodes. The value of the model factor should be
obtained by comparing static load test results and corresponding predictions. It can be
selected with the aim to provide a given reliability to the predictions: for instance, one may
wish that, if static load tests were performed, 95% (or some other reliability level) of the
measured compressive resistances would be greater than the calculated predictions. The
value of the model factor can be obtained by statistical analyses of the data bank of load tests
results (e.g. see Bauduin, 2002b; Frank and Kovarik, 2005).

Design procedures with Design Approaches 1 and 2

As in the design method based on the results of static load tests, the variability of the ground,
and hence of the pile resistance over the site considered, should also be taken into account
when assessing the characteristic value of the compressive resistance. The following two
procedures for accounting for the variability of the ground are introduced in Eurocode 7:
• The first procedure, called the ‘model pile’ procedure in this guide, is where the values Clause 7.6.2.3(5)P
of the ground test results at each individual tested profile are used to calculate
the compressive resistance of a model, e.g. fictitious, pile at the same location. The
procedure is, in fact, similar to that used with the results of static load tests, e.g. it
involves applying ξ factors to the calculated resistance to account for the variability of
the pile resistance and obtain the characteristic compressive resistance.
• The other procedure, referred to as the ‘alternative’ procedure, is where the ground test Clause 7.6.2.3(8)
results (shear strength, cone resistance, etc.) of all tested locations are first brought
together before evaluating the characteristic values of base resistance and shaft resistance
in the various strata based on a cautious assessment of the test results and without
applying the ξ factors.

‘Model pile’ procedure

The ‘model pile’ procedure for determining the design compressive resistance Rc, d from the Clause 7.6.2.3(5)P
results of tests on the ground (these can be either laboratory or in situ tests) is based on the
results of one or more profiles of such tests. As in the case of pile load tests, Eurocode 7 Clause 7.6.2.3(6)P
requires that, when determining Rc, d using this semi-empirical method, the interpretation of
the results of the ground tests and the calculated resistances take into account the variability
of the ground over the site. Careful examination of the calculated pile resistances might lead
to different parts of the site being identified, each with different resistances. A clear example
of this, involving the use of cone penetrometer test (CPT) results, is given by Bauduin (2002b).
The steps in the ‘model pile’ procedure for determining the design compressive resistance
of a pile foundation from ground tests results are as follows:
(1) Determine the calculated compressive resistance Rc, cal by applying the prediction method
to the results of each profile of tests independently in accordance with the following
equation:
Rc, cal = Rb, cal + Rs, cal (D7.1)
where Rb, cal and Rs, cal are the calculated base resistance and calculated shaft resistance,
respectively, determined for the profile of ground tests results under consideration.

109

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

They include the model (calibration) factor. The result is the predicted resistance of a
pile if it were exactly at the location of the profile of ground tests. Hence this pile can be
called a model pile.
Clause 7.6.2.3(5)P (2) The characteristic values Rc, k, Rb, k and Rs, k are determined from the following equation:
Rc, k = (Rb, k+ Rs, k)
Rc, k = (Rb, cal + Rs, cal)/ξ = Rc, cal /ξ = Min{(Rc, cal)mean/ξ3; (Rc, cal)min /ξ4} (7.8)
where ξ3 and ξ4 are correlation factors that depend on the number of profiles of tests, n,
and are applied respectively
– to the mean values
(Rc, cal)mean = (Rb, cal + Rs, cal)mean = (Rb, cal)mean+ (Rs, cal)mean
– and to the lowest values
(Rc, cal)min = (Rb, cal + Rs, cal)min
The recommended values for ξ1 and ξ2 given in Annex A are intended to cover the
variability in the ground and the calculated resistances over the site (see above).
Clause 7.6.2.3(7) (3) If the structure connecting the piles is stiff and strong enough to transfer loads from
weaker piles to stronger piles, the values for ξ3 and ξ4 may be divided by 1.1, provided the
value of ξ3, which is applied to the mean value, is not lower than 1.0.
Clause 7.6.2.3(6)P (4) If several profiles of ground tests are available, the first estimate of an overall Rc, k value,
based on all the profiles of ground tests, might lead to different ‘homogeneous’ parts of
the site being identified. In this situation, the procedure is started again from step 2 for
each ‘homogeneous’ part of the site. This can result in less conservative Rc, k values than
would be obtained if, as in the first estimate, the soil conditions were assumed to be the
same over the whole site and the minimum value of Rc, cal governs the Rc, k value over the
whole site.
Clause 7.6.2.3(3)P (5) The design compressive resistance is obtained by applying the partial factors γs and γb to
Clause 7.6.2.3(4)P the characteristic shaft resistance and to the characteristic base resistance, respectively,
in accordance with the following equations:
Rc, d = Rb, d + Rs, d (7.6)
Rc, d = Rb, k /γb + Rs, k /γs (7.7)
This model pile design procedure may be used with DA-1 and DA-2 and the
recommended values for the partial factors γs and γb given in Annex A for persistent or
transient situations. The model pile procedure is not applicable to DA-3 because, as
explained in the case of compressive resistance obtained from pile load tests, the
procedure involves applying partial factors to the resistances while DA-3 involves
applying partial factors to the characteristic values of the ground strength parameters.
For accidental situations, unless otherwise stated in the National Annexes, it can be
assumed that γs = γb = 1.0.
Clause 7.6.2.1(3)P (6) In the case of groups of piles, the bearing capacity of the whole foundation is determined
Clause 7.6.2.1(4) either by summing up the individual compressive resistances or by assuming a block
failure, whichever provides the lower value.
Examples 7.2 and 7.3 demonstrate the calculation of pile compressive resistance from the
results of ground tests following the procedure outlined above. Example 7.2 uses the
results of pressuremeter tests (PMTs) while Example 7.3 uses undrained shear strength (cu)
measurements.

‘Alternative’ procedure
Clause 7.6.2.3(8) As an alternative to the ‘model pile’ procedure, Eurocode 7 allows the characteristic
resistance values Rb, k and Rs, k to be determined directly from the values of ground
parameters:

110

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 7. PILE FOUNDATIONS

Rb, k = Ab qb, k
Rs, k = Â As, i qs, k, i (7.9)
i

where qb, k and qs, k, i are characteristic values of base resistance and shaft resistance per unit
area in the various strata, obtained using an appropriate method and the values of the
ground parameters.
The ‘alternative’ procedure then follows steps 5 and 6 of the ‘model pile’ procedure. This
alternative procedure is intended to allow use of the traditional methods for calculating the
compressive resistance from ground test results.
The values of qb, k and qs, k, i (obtained from tabulated or chart values or from correlations)
should account for the variability of the ground parameters, the volume of soil involved in
the failure mechanism considered, the spatial variability of the pile resistance, the variability
due to the effect of pile installation, and the stiffness of the structure. This is because the ξ
factors are not used explicitly in this alternative method.
The ‘alternative’ procedure treats the soil variability differently from the method based on
the results of static load tests or from the ‘model pile’ procedure based on ground test results.
As the partial factors γb, γs and γt in Tables A.6, A.7 and A.8 are intended to be used in
conjunction with the ξ factors, they may not be appropriate for use with the ‘alternative’ Clause 7.6.2.3(8)
procedure. A model factor may have to be introduced or directly included when determining
the values of qb, k and qs, k.
Examples 7.2 and 7.3 demonstrate the use of the ‘alternative’ procedure.

Design Approach 3
As noted above, since DA-3 uses partial factors applied to the ground parameters, neither
the ‘model pile’ procedure nor the ‘alternative’ procedure can be used with DA-3 because
they both involve factors applied to the resistances. DA-3 is usually intended to be used
with calculation models involving the design values of ground parameters obtained from
the results of laboratory tests. When using DA-3, clause 7.6.2.3(9)P explains that the Clause 7.6.2.3(9)P
characteristic values of the ground parameters Xk are first determined in accordance with Clause 2.4.5
clause 2.4.5, and the design values Xd are then obtained by applying the material factors γM Clause 2.4.6.2(1)P
(column M2 of Table A.4) in accordance with equation (2.2):
Xd = Xk /γM (2.2)
For pile design using DA-3, the design value of the compressive resistance Rc, d is derived
directly by inputting the design values of the ground strength parameters Xd in the
calculation model:
Rc, d = Rb, d + Rs, d = Rb, cal(Xd) + Rs, cal(Xd) (D7.2)
Example 7.3 shows the various possibilities for using DA-3 with the results of undrained
shear strength (cu) measurements.
Note that, where relevant in the above procedures, the values of the parameters to be used
to assess the shaft resistance should be cautious estimates of their mean values because the
length of the pile shaft in a given stratum is usually ‘large’. For the base resistance, as the
volume involved in the failure mechanism around the base is usually ‘small’, the values of the
parameters to be used should be cautious estimates of their local values around the base
level.

Ultimate compressive resistance from dynamic tests results

The general requirements in Eurocode 7 for using dynamic tests to determine the compressive
resistance of piles are set out in clause 7.5.3. The use of different types of dynamic Clause 7.5.3
test results, as well as the use of redriving results, to assess the compressive resistance
of individual piles in compression, are mentioned in clause 7.6.2. The dynamic tests results
are:

111

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Clause 7.6.2.4 • Dynamic impact (hammer blow) tests. During this type of test, it is assumed that the
Clause 7.6.2.4(3)P strain is high enough to reach the (dynamic) ultimate compressive resistance. The
Clause 7.6.2.4(2) dynamic resistance should be measured. The test can also include a signal-matching
procedure leading to an approximate estimate of the shaft and base resistances and of
the load–settlement curve.
Clause 7.6.2.5 • Pile-driving formulae. The use of pile-driving formulae is the traditional way of taking
advantage of the penetration records during the driving of piles to determine the
bearing resistance.
Clause 7.6.2.6 • Wave equation analyses. Wave equation analyses may be used to assess the design
compressive resistance of piles. However, this mathematical model (also called the
mass–spring model) is usually employed to study the driving conditions (hammer
performance, stresses in the pile).
When determining the compressive resistance of piles from dynamic tests, it is often
recommended that the driving records after redriving (by a few centimetres and some time
after the end of the initial pile driving) are used.
The procedure for determining the compressive resistance of a pile from dynamic test
Clause 7.6.2.4(4)P results is the same for the three types of test, although some specific requirements need to be
met for each type of test. The characteristic value is determined from the following equation:
Rc, k = Min{(Rc, m)mean /ξ5; (Rc, m)min /ξ6} (7.11)
where Rc, m are the static compressive resistances derived from the dynamic measurements,
and ξ5 and ξ6 are correlation factors related to the number of piles tested, n, and are applied
to the mean (Rc, m)mean and the lowest (Rc, m)min. Recommended values for ξ5 and ξ6 are given
in Annex A. Note that the ξ values are not the same for the three types of dynamic
measurements.
The design compressive resistance is then obtained from
Rc, d = Rc, k /γt (7.10)
where the partial factor on the total resistance γt is the same as for the other methods for
determining the compressive resistance of pile foundations. For DA-1 and DA-2, the values
of γt are given in Annex A. For accidental situations it can be assumed that γt = 1.0.
Clause 7.6.2.5(3)P Pile-driving formulae should only be used with the above procedure to assess the design
compressive resistance in the case of end-bearing piles in cohesionless soils. For other types
of pile and/or other types of soil the uncertainty is larger, and other methods or higher partial
factors should be used.

7.6.3. Ground tensile resistance

General requirements
Clause 7.6.3.1(1)P Many of the Eurocode 7 requirements for piles in compression also apply to piles in tension.
Clause 7.6.3.1(3)P For piles in tension, two ULSs must be checked:
• pull-out failure (tensile resistance) of individual piles
• uplift failure of the block of ground containing the piles.

As in the case of the compressive resistance failure of piles, the pull-out resistance of
Clause 7.6.3.1(4)P individual piles should be checked using the GEO set of partial factors and relevant
inequality. However, for groups of piles in tension, possible uplift failure of the block of
ground containing the piles should be checked using the UPL set of partial factors and
relevant inequality.
Clause 7.6.3.1(8)P Clause 7.6.3.1(8)P draws the designer’s attention to the situation where tensile piles
interact in a group. Tension on one pile reduces the effective vertical stress close to the
surrounding piles, and hence reduces their shaft resistance. This applies to the resistance of
the individual piles in the group and consequently to the resistance of the whole foundation.

112

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 7. PILE FOUNDATIONS

Tensile resistance of individual piles

The methods for determining the ground resistance of piles in tension, either from static pile Clause 7.6.3.1
load tests or from ground test results, are very similar to those for piles in compression. The
main differences are:
• The base resistance is always neglected.
• The correlation factors ξ should not be divided by 1.1, even in the case of stiff structures.
This is due to the fact that ground failure of tensile piles is usually brittle (loss of strength
after having reached the peak resistance). Hence, it is unsafe to assume load transfer
from failed to not yet failed piles, and therefore this is not allowed.
• The value of the partial factor on the total resistance (i.e. shaft resistance) in tension,
γs, t, is larger than the value of γs in compression (see the tables in Annex A). It is common
practice to be somewhat more cautious when attributing safety to tensile piles than to
compressive piles.
Given these differences, the general design requirements for piles in tension in clause Clause 7.6.3
7.6.3 are the same as those in clause 7.6.2 for piles in compression. Also, the correlation Clause 7.6.2
factors ξ1 and ξ2 (for measured tensile resistances) and ξ3 and ξ4 (for calculated tensile
resistances) recommended by EN 1997-1-1 have the same values (Tables A.9 and A.10,
respectively). When static load tests are used for designing tensile pile foundations, these
should, as already noted in Section 7.5 of this guide, be carried to failure. For the design of Clause 7.6.3.3(4)P
tensile piles using ground tests results, the model pile procedure may be used, and EN 1997-1-1 Clause 7.6.3.3(6)
also allows the alternative procedure to be used. When using DA-3, partial factors are Clause 7.6.3.3(7)P
applied to the characteristic ground parameters to calculate the design tensile resistance.
Clause 7.6.3.1(8)P draws the attention of the designer to the case where tensile piles Clause 7.6.3.1(8)P
interact in a group: tension on one pile reduces the effective vertical stress close to the
surrounding piles, and hence their shaft resistance. This applies to the resistance of the
individual piles in the group, and consequently to the resistance of the whole foundation.
The partial factors of Tables A.6, A.7 and A.10 are, in principle, valid for quasi-static loads.
Another important requirement of EN 1997-1 is that cyclic loadings or reversals of load, Clause 7.6.3.1(9)P
which might be very detrimental to tensile resistance of a pile, must be considered.
Unfortunately, no precise provision as to how to deal with these loading conditions is given
in the code. Unless special calculation rules are applied, the effects of such loadings may
either be taken into account by applying a model factor to the calculated results so that they
err on the safe side or else, as is common practice, increased partial factors are simply
applied to the actions for cyclic or reversed loading conditions.
It should be noted that when a large variable tensile load has to be combined with a
permanent compressive load of similar magnitude, the design load on the pile can become a
tensile load, while the resulting characteristic load remains compressive. The foundation
should be designed for this tensile design value of the action.

Block failure of a group of piles

Block failure of a group of piles occurs when all the piles and the soil included between them
are lifted up by a tensile force. The tensile force causing this may come from a structure
above the water level and the pile group, or from upwards water pressure acting on the pile
group and that part of the structure below water level, as shown in Figs 7.1a and 7.1b of
EN 1997-1-1.
It should be noted that, both for isolated piles and for group of piles, EN 1997-1-1 Clause 7.6.3.1(5)
recommends that the pull-out mechanism involving a cone of ground surrounding the piles is
checked, especially when an enlarged base or rock socket is used.
Eurocode 7 requires that the design against uplift failure of the block is treated as an uplift
(UPL) ULS calculation with the appropriate partial factors (see Tables A.15 and A.16 in
Annex A, as well as Chapter 2 in this guide). Example 7.5 demonstrates the design of a group
of piles from an upwards water pressure against uplift in accordance with the requirements Clause 7.6.3.1
of clause 7.6.3.1.

113

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Clause 2.4.7.1(1)P It may also be appropriate to check the GEO ULS of groups of piles subjected to a tensile
force, in particular when the tensile force comes from the structure.

7.6.4. Vertical displacements of pile foundations

The design of foundations should be such that the movements are limited and do not lead to
the occurrence of limit states (serviceability or ultimate) in the supported structure. This
Clause 2.4.8 general principle is clearly recognized in EN 1997-1-1, in particular in clauses 2.4.8 and 2.4.9.
Clause 2.4.9 The values of the allowable movements should be selected during the design. Some
Clause 2.4.9(3)P indicative limiting values of structural deformation are given in Annex H (informative).
Clause 7.6.4.1(1)P According to EN 1997-1, it is necessary to assess the settlements of axially loaded pile
foundations, such as those of shallow foundations, in order to ensure that the occurrence of
an SLS in the structure is sufficiently unlikely. Nevertheless, it is understood that most of the
Clause 7.6.4.1(2) time this assessment can only provide an estimate of the settlements because of all the
Clause 2.4.8(4) uncertainties in the models for pile settlements. EN 1997-1 allows settlement calculations to
be replaced, in many circumstances, by a simplified design approach involving bearing
resistance calculations with partial factors high enough such that a sufficiently low fraction of
the ground strength is mobilized and no unacceptable deformation occurs in the supported
structure, at least when common piles are used in known ground conditions and successful
previous experience is available. This simplified design approach is common practice in
traditional design (see also Chapter 6 of this guide).
Methods for calculating the displacement of pile foundations include the linear elastic
approach of Poulos and Davis (1980), as well as elasto-plastic finite-element calculations and
t–z load transfer functions (curves of mobilization of shaft resistance, e.g. see Baguelin et al.,
1982).

7.7. Transversely loaded piles

Clause 7.4 The design of transversely loaded piles must be consistent with the general requirements for
Clause 7.5 the design and load testing of piles given in clauses 7.4 and 7.5. In addition, the following
requirements apply:
Clause 7.7.1(3) • the transverse resistance should be checked for situations involving rotational or
translational failure in the ground in the case of short rigid piles, and for bending failure
of the pile together with yield of the soil near the surface in the case of long slender piles
Clause 7.7.1(4)P • the group effect must be taken into account
Clause 7.7.2 • the transverse load resistance should be determined by means of static load tests or on
Clause 7.7.3 the basis of ground test results and pile strength parameters
• the strength of the pile to resist structural failure should be checked
• the transverse displacement should be assessed.

Clause 7.7.2(3)P Other requirements specifically for transversely loaded piles, which also should be
Clause 7.7.2(4) accounted for, concern the variability of the ground near the surface as well as the head fixity
Clause 7.7.3(4)P conditions at the connection between the piles and the structure.
Clause 7.7.3(3) The theory of a beam supported by elastic springs, characterized by the horizontal modulus
of subgrade reaction, is explicitly mentioned in EN 1997-1 as an acceptable calculation
method for the design of a long slender piles subjected to a transverse load at the top.

7.8. Structural design of piles

The structural design of piles is an important aspect, which forms an integral part of
the design of pile foundations. However, most structural designs involve applying code
Clause 7.8(1)P requirements relating to the behaviour of concrete, steel or wood. EN 1997-1 requires that
Clause 7.8(2)P piles are checked against structural failure in all relevant design situations, in accordance

114

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 7. PILE FOUNDATIONS

with the requirements of the relevant material Eurocodes, in particular EN 1992, EN 1993
and EN 1995.
Slender piles passing through water or through a thick layer of very weak soil must be Clause 7.8(4)P
checked against buckling. This check against buckling can be omitted if the undrained shear Clause 7.8(5)
strength cu is greater than 10 kPa.
For persistent or transient situations, the design values of actions for the ULS structural
design of piles are determined using DA-1, DA-2 or DA-3, depending on the national
choice. When DA-1 is used, Combination 1 usually governs the structural design of piles.

7.9. Supervision of construction

The way piles are installed not only influences the design but also plays a fundamental role in
ensuring an adequate behaviour of the supported structure. EN 1997-1 requires that a Clause 7.9(1)P
detailed plan of the pile installation is prepared as a basis for the piling works and that the Clause 7.9(2)
installation of all the piles is monitored and recorded. If there is doubt about the quality of Clause 7.9(3)P
installed piles, relevant measures must be taken, including full-scale tests, the installation of Clause 7.9(6)P
new piles, redriving, etc.
Many aspects of pile design are often linked to the pile installation method. Hence Clause 7.9(4)
EN 1997-1 repeats some requirements included in the relevant execution standards issued
by CEN/TC 288.
When relevant, the integrity of piles must be checked. Dynamic low-strain tests are not Clause 7.9(7)P
always suitable for this purpose, and hence other tests, such as sonic and vibration tests, or Clause 7.9(8)
coring, may be needed.

Example 7.1: design of a pile in compression from static load test results
The deep foundations of a major bridge in Europe have been designed by interpreting the
results of full-scale pile tests under both axial and transverse loadings (see Wastiaux et al.,
1998). The present example, inspired by this project, deals only with the axial compressive
resistance of the piles, and aims to show how the results of static compression load tests
should be interpreted following Eurocode 7.
The foundations on open-ended driven piles have to be designed in order to carry
vertical compressive loads. In the following design calculations, only ULSs in persistent
and transient situations and accidental situations are considered. The permanent vertical
compressive load is 31 MN, and the vertical accidental load is 16 MN. The purpose of the
design is to determine the number of driven piles of length 55.5 m required to carry this
load.
It is assumed that there is no need to take into account any group effect for the pile
foundation. The full design of the foundation would also require the SLSs to be checked,
e.g. a check that the settlement of the foundation is acceptable.
A typical soil profile consists of 20–30 m of very soft clay (mud), muddy sands, sands
and clays, and, finally, sands and gravels at the level of the expected location of the base of
the piles.
The results of four static load tests on driven piles with different lengths of embedment
are available (Fig. 7.1). In what follows, the resistance Rm of a pile of length L = 55.5 m is
deduced from the resistances measured in the static tests, taking into account the
differences in pile length (and after subtracting the positive shaft resistance in the layers
which will be submitted to downdrag). Hence, the estimated values of the resistances for a
pile of length L = 55.5 m, are the following:
• from pile test P8: Rm = 14.0 MN
• from pile test P31: Rm = 14.4 MN
• from pile test P79: Rm = 12.1 MN
• from pile test P79b: Rm = 13.9 MN.

115

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Force (kN)
0 2500 5000 7500 10 000 12 500 15 000 17 500 20 000
0

10 P8 P311
P31 P79bis
20
P79
30

40
Displacement (mm)

50

60

70

80

90

100

110

120

130

Fig. 7.1. Results of static load tests on driven piles of diameter B = 1220 mm and of different
lengths of embedment (Wastiaux et al., 1998)

Clause 7.6.2.2(8)P According to EN 1997-1, the characteristic value of the axial pile resistance, Rk, is
determined from the corresponding mean and minimum measured values (Rm, mean,
Rm, min) using the following formula:
Rk = Min{Rm, mean /ξ1; Rm, min /ξ2} (7.2)
The correlation factors (ξ1, ξ2) depend on the number n of static pile load tests. For n = 4,
Table A.9 recommends ξ1 = 1.10 and ξ2 = 1.00. It is assumed that redistribution between
weak and strong piles cannot be accounted for because the expected number of piles is too
Clause 7.6.2.2(9) small. Thus, the reduction factor 1.1 on ξ1 and ξ2 proposed in clause 7.6.2.2(9) is not
applied.
Rm, mean = 13.6 MN
Rm, min = 12.1 MN
Thus,
Rk = Min{13.6/1.10; 12.1/1.00} = Min(12.4, 12.1) = 12.1 MN
which shows that the minimum measured pile resistance governs the design.
For a single pier of the bridge, the characteristic values of the vertical loads, taking into
account the downdrag, are assumed to be:

• for persistent and transient situations: permanent loads Gk = 31 MN (the variable

loads are considered to be negligible for the critical piers)
• for accidental situations: permanent and accidental loads Gk + Ak = 47 MN (the
accidental load is 16 kN).

Clause 7.6.2.2(14)P For persistent and transient situations, DA-1 (Combinations 1 and 2) and DA-2 are
considered. The load tests yield directly the total (or shaft and base) characteristic
resistances, and the design values are obtained by applying the resistance factors γt (or γs

116

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 7. PILE FOUNDATIONS

and γb). DA-3 is not intended to be used with pile load test results, as it is a ‘material’
factor approach involving the use of ground strength parameters.

Design Approach 1
DA-1 Combination 1 involves applying the partial factors from sets A1 and R1 of Annex A. Clause
This leads to the following design actions and design resistances: 2.4.7.3.4.2(2)P
Fc, d = 1.35Gk = 1.35 × 31 = 41.85 MN
Rc, d = Rc, k /γt = 12.1/1.0 = 12.1 MN
Thus, the number of piles needed is 41.85/12.1 = 4.
DA-1 Combination 2 involves applying the partial factors of sets A2 and R4 of Annex A. Clause
This leads to the following design actions and design resistances: 2.4.7.3.4.2(2)P
Fc, d = 1.0Gk = 1.0 × 31 = 31 MN
Rc, d = Rc, k /γt = 12.1/1.3 = 9.3 MN
Thus, the number of piles needed is 31/9.3 = 4.
The requirement for DA-1 is that the more conservative result of the two combinations
should be used. The number of piles being the same, there is no difference in this example
between the two methods for the design number of piles.
It should be noted that, in this Design Approach, Combination 1 provides an overall
factor of safety (OFS) on the characteristic resistance of at least
γFγt = 1.35 × 1.0 = 1.35
whereas Combination 2 provides an OFS of at least
γFγt = 1.0 × 1.3 = 1.3
In relation to the measured mean pile resistance, i.e. including ξ = 1.10, these figures
become OFS = 1.49 and 1.43, respectively.

Design Approach 2
For DA-2, sets A1 and R2 of Annex A must be applied. Thus the design actions and design Clause
resistances are 2.4.7.3.4.3(1)P
Fc, d = 1.35Gk = 41.85 MN
Rc, d = Rc, k /γt = 12.1/1.1 = 11.0 MN
Thus, the number of piles needed is 41.85/11.0 = 4.
Note that DA-2 provides an OFS on the characteristic resistance of at least
γFγt = 1.35 × 1.1 = 1.5
and an OFS of at least 1.65 on the measured mean pile resistance.

Accidental situation
For an accidental situation, all the partial factors on loads should be taken equal to 1.0. It Clause 2.4.7.1(3)
is assumed that a resistance factor equal to 1.0 is also relevant for the circumstances in this
particular accidental situation. Thus, the design actions and design resistances are
Fc, d = 1.0Gk + 1.0Ak = 47 MN
Rc, d = Rc, k/γt = 18.1/1.0 = 12.1 MN
The number of piles needed is thus 47/12.1 = 4.
In conclusion, for the given loads, the ULS designs of the piles for the bridge pier for
persistent and accidental situations, both require four piles of length L = 55.5 m.

117

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Example 7.2: design of a pile in compression from in situ test results

A foundation on driven piles in a slightly overconsolidated clay has to be designed to carry
permanent and variable vertical compressive loads. The characteristic values of the loads
are permanent load Gk = 3900 kN and variable load Qk = 800 kN.
Clause 7.6.2.3(6)P The results of three PMT profiles are available (P1, P2 and P3), and the area can be
considered as ‘homogeneous’ (there is only a random variation in the ground test results).
The diameter of the piles is B = 400 mm, and the expected embedment is L = 13 m.
In the following design calculation, only ULSs for persistent and transient situations
are considered. It is also assumed that there is no need to take into account any group
effect in the design of the pile foundation. The full design of the foundation would also
require checking of:

• accidental situations, if relevant

• SLSs, e.g. checking that the settlement of the foundation is acceptable.

Only DA-1 (Combinations 1 and 2) and DA-2 are considered. DA-3 is not intended, in
principle, to be used with in situ test results. DA-3 uses material factoring, and is only
meant to be used with laboratory test results and ground strength parameters.
Note that the sets of calculations for DA-1 and DA-2 would be exactly the same if CPT
results were to be used with the ‘model pile’ procedure.
Clause 7.6.2.3(2) The PMT calculation rules given in Annex D.3 of EN 1997-2 are used, and a model
(calibration) factor γRD equal to 1.05 is applied. This value has been selected for use with
the recommended values of the partial factors given in Annex A for DA-2, in order to yield
the same value of the overall factor of safety as in the present French code of practice with
PMT calculation rules (see Frank, 1999). For the sake of comparison, γRD = 1.05 is also
used below for DA-1.
From the three profiles of limit pressures pl, the mean net limit pressure values
determined at the base, pl, base, and the net limit pressure values determined along the
shaft, pl, shaft, are given in Table 7.2.
Clause 7.6.2.3(3)P The calculated design ultimate compressive resistance of the foundation is obtained
from
Rc, d = Rb, d + Rs, d (7.6)
where Rb, d is the design calculated pile base resistance obtained from Rs, d = Rs, k /γs, and
Rs, d is the design calculated shaft resistance obtained from Rb, d = Rb, k /γb.

‘Model pile’ procedure

Clause 7.6.2.3(5)P For DA-1 and DA-2, according to EN 1997-1, the characteristic value of the compressive
resistance Rc, k of a pile is determined from the corresponding mean and minimum
calculated values (Rcal, mean, Rcal, min) using the following formula:
Rc, k = Min{Rcal, mean /ξ3; Rcal, min /ξ4} (7.8)
The correlation factors (ξ3, ξ4) depend on the number n of ground profiles investigated,
and Rcal, mean and Rcal, min are obtained by calculating the pile compressive resistance for
each profile (‘model pile’ procedure).
In the following, pressuremeter test results are used. All the calculation steps would be
exactly the same if CPT results were to be used with the ‘model pile’ procedure, instead of
pressuremeter test results.
For the pressuremeter, the following design criteria are obtained (see EN 1997-2):

• the soil falls into the ‘Clay A’ category

• the bearing (base) resistance factor is k = 1.4
• for the shaft resistance, design curve 1 is used.

118

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 7. PILE FOUNDATIONS

Table 7.2. Net limit pressures determined on the pile

Pressuremeter test profiles

P1 P2 P3 Mean (of all values)

pl, base (MPa) 0.98 0.78 0.81 0.86

pl, shaft (MPa) 0.77 0.73 0.78 0.76

Table 7.3. Values of predicted resistances by the PMT method and calculated pile resistances

PMT profile Rb (kN) Rs (kN) Rb+ Rs (kN) Rc, cal = (Rb+ Rs)/γRD (kN)

1 172 498 670 638

2 137 482 619 590
3 143 503 646 615
Minimum 590
Mean 614

Table 7.3 gives the predicted values of the base and shaft resistances based on the three
PMT profiles, as well as the calculated compressive resistance Rc, cal, taking into account
the model factor γRD = 1.05.
In the present example, from Table A.10, n = 3, thus ξ3 = 1.33 and ξ4 = 1.23. If it is
assumed that the structure is stiff and strong enough to transfer loads from weak to strong Clause 7.6.2.3(7)
piles, these values may be divided by 1.1. Thus, the values used are ξ3 = 1.33/1.1 = 1.21
and ξ4 = 1.23/1.1 = 1.12, and hence the characteristic pile resistance is
Rk = Min{614/1.21; 590/1.12} = Min(507; 527) = 507 kN
which shows that the mean value governs (this is for ground where the standard deviation
of the resistance is less than 10%, and hence is consistent with considering the area as
‘homogeneous’).
In this example, as the mean resistance governs the design, applying ξ3 = 1.21 to the
mean calculated shaft and base resistances leads to their characteristic values:
Rb, k = (172/1.05 + 137/1.05 + 143/1.05)/(3 × 1.21) = 143/1.21 = 118 kN
Rs, k = (498/1.05 + 482/1.05 + 503/1.05)/(3 × 1.21) = 471/1.21 = 389 kN
Rc, k = 118 + 389 = 507 kN

Design Approach 1
For DA-1, it is usually logical to apply first the factors corresponding to Combination 2, as
this combination usually determines the geotechnical sizing, and then to check that the
sizing is adequate with Combination 1, which controls the design with regard to structural
safety.
Combination 2 involves applying the partial factors of sets A2 and R4 of Annex A. This Clause
leads to the following design actions and design resistances: 2.4.7.3.4.2(2)P
Fc, d = 1.0Gk + 1.3Qk = 1.0 × 3900 + 1.3 × 800 = 4940 kN
Rc, d = Rb, k/γb + Rs, k/γs = 118/1.3 + 389/1.3 = 390 kN

119

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Thus, the number of piles needed is 4940/390 = 12.7, which is rounded up to 13 piles. The
equivalent overall factor of safety, OFS, can be considered as the total predicted mean
resistance of the foundation (Rb + Rs), i.e.
13 × (670 + 619 + 646)/3 = 13 × 645 kN
from Table 7.3, divided by the total applied characteristic load, i.e. 3800 kN + 800 kN,
which gives
OFS = 13 × 645/4700 = 1.78
Clause Combination 1 involves applying the partial factors of sets A1 and R1 of Annex A. This
2.4.7.3.4.2(2)P leads to the following design actions and design resistances:
Fc, d = 1.35Gk + 1.5Qk = 1.35 × 3900 + 1.5 × 800 = 6465 kN

Rc, d = Rb, k /γb + Rs, k /γs = 118/1.0 + 389/1.0 = 507 kN

Thus, the number of piles needed is 6465/507 = 12.8, which also rounds up to 13 piles,
and, as for Combination 2, the OFS = 1.78.
The requirement for DA-1 is that the more conservative result of the two combinations
should be used. In this example, Combination 1 is marginally more conservative than
Combination 2, but as the number of piles is the same for both combinations, there is no
difference for the final foundation design using the two combinations.

Design Approach 2
Clause For DA-2, sets A1 and R2 of Annex A are applied. Thus, the design actions and design
2.4.7.3.4.3(1)P resistances are
Fc, d = 1.35Gk + 1.5Qk = 1.35 × 3900 + 1.5 × 800 = 6465 kN

Rc, d = Rb, k /γb + Rs, k /γs = 118/1.1 + 389/1.1 = 461 kN

Thus, the number of piles needed is 6465/461 = 14. The equivalent overall factor of safety
can be considered as the total calculated mean resistance of the foundation, i.e. 14 × 645,
divided by the total applied characteristic load, i.e. 3800 kN + 800 kN, which gives
OFS = 14 × 645/4700 = 1.92

‘Alternative’ procedure
Clause 7.6.2.3(8) For DA-1 (Combinations 1 and 2) and DA-2, clause 7.6.2.3(8) allows Rb, k and Rs, k to be
determined directly from the ground parameter values. It is assumed that the same
calculation rules are used as for the ‘model pile’ procedure. They can be applied in two
ways: with either the characteristic or mean values of the ground parameters.
Using the mean pl values along the shaft and at the base of the piles to predict the shaft
and base resistances given in Table 7.2, which corresponds to more to traditional practice,
one finds
Rmean = (Rs, k + Rb, k)/γRD = (495 + 151)/1.05 = 615 kN
This value is 615/507 = 1.21 times larger than the characteristic value for the ‘model pile’
procedure (it corresponds to the value of ξ3). This shows that if the same values for
the partial factors on loads and base and shaft resistances are to be used in DA-1
(Combinations 1 and 2) and DA-2 when the alternative procedure is applied, the values of
Clause 7.6.2.3(8) the partial factors in Annex A may need to be corrected by a model factor greater than 1.0
(see the note in Clause 7.6.2.3(8)). The results of the alternative procedure, as compared
with the ‘model pile’ procedure, will then be directly linked to the choice of this model
factor.

120

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 7. PILE FOUNDATIONS

Table 7.4 summarizes the values of Rk and Rd obtained in the various Design
Approaches and procedures.

Table 7.4. Values of Rk (or Rmean) and Rd (in kN)

Model pile procedure Alternative procedure

DA-1

Combination 2 Combination 1 DA-2 DA-1, DA-2

Rk (kN) 507 507 507 Rmean: 615

a
Rd (kN) 390 507 461
a
Depends on the Design Approach used and the value of the supplementary model factor introduced

Example 7.3: design of a pile in compression from laboratory test results

This example presents the ULS design of a single bored pile in clay (pile diameter
B = 0.80 m using the undrained shear strength cu to avoid bearing resistance failure for
a persistent or transient design situation. The characteristic values of the loads are
permanent load Gk = 600 kN and one single variable load Qk = 300 kN. The design aims
to check the required length (L) of the pile, which is expected to be 18.5 m.
The ground was investigated by boreholes and tests in three profiles (n = 3: BH1, BH2
and BH3). Figure 7.2 shows the three profiles of undrained shear strength cu measured on
samples recovered from the boreholes.

Undrained shear strength (kPa)

0 20 40 60 80
0
BH1
BH2
BH3

8
Depth (m)

12

16

20

Fig. 7.2. Undrained shear strength cu profiles for boreholes BH1, BH2 and BH3

121

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Table 7.5. Determination of characteristic values of cu, shaft and cu, base

Mean of all Characteristic

Mean cu values values value of all values

BH1 BH2 BH3

cu, shaft (kPa) 52 46 51 50 47

cu, base (kPa) 33 30 42 35 32

The corresponding mean values of the undrained shear of the clay cu along the shaft of
the pile (cu, shaft) and in the vicinity of pile base at a depth of 18.5 m (cu, base) are given in
Table 7.5. The characteristic values have been determined using the method given in
Chapter 2 of this guide.
The axial bearing resistance of the pile is calculated by the formula
R = Rs + Rb
where
Rs = πBLαcu
is the shaft resistance (α = 0.75 in this case) and
Rb = (πB2/4)9cu
is the pile base resistance, and it is assumed that there is no need for a model factor (i.e.
the model factor is taken equal to 1.0). The shaft resistance is governed by the mean
values of the undrained shear resistance over the length of the shaft, while the base
resistance is governed by the local mean values close to the pile base.
The calculations are performed according to the ‘model pile’ and ‘alternative’
procedures for DA-1 and DA-2, as well as for DA-3. The structure is considered not to
have sufficient strength and stiffness to allow redistribution of loads.

‘Model pile’ procedure

Clause 7.6.2.3(5)P For DA-1 and DA-2, according to EN 1997-1, the characteristic axial resistance, Rk, of a
pile is calculated from the corresponding mean and minimum values (Rmean, Rmin) by the
formula (called the ‘model pile’ procedure)
Rk = Min{Rmean/ξ3; Rmin/ξ4}
The correlation factors (ξ3, ξ4) depend on the number n of ground profiles investigated;
for n = 3, ξ3 = 1.33 and ξ4 = 1.23 (see Table A.10).
Rmean and Rmin are obtained by calculating the compressive resistance, Rc, for each
borehole using the mean cu values (cu in kilopascals and resistances in kilonewtons):
Rc = Rs + Rb = πBLαcu, shaft + (πB2/4)9cu, base
Rc = 34.9cu, shaft + 4.5cu, base
Thus,
R(BH1) = 1815 + 148 = 1963 kN
R(BH2) = 1605 + 135 = 1740 kN
R(BH3) = 1780 + 189 = 1969 kN
Hence
Rmean = (1963 + 1740 + 1969)/3 = 1891 kN
Rmin = 1740 kN

122

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 7. PILE FOUNDATIONS

and
Rk = Min{1891/1.33; 1740/1.23}
Rk = Min{1422 kN; 1415 kN} = 1415 kN
The minimum value governs the resistance, indicating a variability greater than 10% of
the undrained shear strength over the site and hence of the resulting pile resistance.
Hence, using the minimum mean cu values obtained from the BH2 profile, one can now
deduce the characteristic base and shaft resistances:
Rs, k = 1605/1.23 = 1305 kN
Rb, k = 135/1.23 = 110 kN
A careful examination of the ground investigation results does not enable different
‘homogeneous’ areas to be distinguished. These characteristic values are thus adopted for
the entire site.

Design Approach 1
For DA-1, it is normally logical to apply first the factors corresponding to Combination 2,
as this combination usually determines the geotechnical sizing, and then to check if the
sizing is adequate with Combination 1, which is the relevant combination for determining
the structural sizing.
Combination 2 consists of applying the partial factors of sets A2 and R4 of Annex A. Clause
This results in the following design actions and resistances: 2.4.7.3.4.2(2)P
Fc, d = 1.0Gk + 1.3Qk = 1.0 × 600 + 1.3 × 300 = 990 kN
Rc, d = Rb, k /γb + Rs, k /γs = 1305/1.3 + 110/1.6 = 1073 kN
If the ‘degree of optimization’ is defined as DO = (Rc, d/Fc, d) – 1 expressed as a percentage,
then in the present case DO = 8%.
Combination 1 consists of applying the partial factors of sets A1 and R1 of Annex A. Clause
This results in the following design actions and resistances: 2.4.7.3.4.2(2)P
Fc, d = 1.35Gk + 1.5Qk = 1.35 × 600 + 1.5 × 300 = 1260 kN
Rc, d = Rb, k /γb + Rs, k /γs = 1305/1.0 + 110/1.25 = 1393 kN
Hence, DO = 11%.
The requirement for DA-1 is that the more conservative result of the two combinations
should be used. In the present case, there is sufficient length (18.5 m) to fulfil both
conditions. The conclusion is that the piles can be slightly shortened.

Design Approach 2
For DA-2, sets A1 and R2 of Annex A must be applied. Thus Clause
2.4.7.3.4.3(1)P
Fc, d = 1.35Gk + 1.5Qk = 1.35 × 600 + 1.5 × 300 = 1260 kN
Rc, d = Rb, k /γb + Rs, k /γs = 1305/1.1 + 110/1.1 = 1286 kN
DO = 2%, which shows that the length of the pile (L = 18.5 m) is optimum with regard
to DA-2.

Alternative procedure
Design Approaches 1 and 2
For DA-1 (Combinations 1 and 2) and DA-2, clause 7.6.2.3(8) allows Rb, k and Rs, k to be Clause 7.6.2.3(8)
determined directly from the ground test results using the alternative procedure. It is
assumed that the same calculation rules are used as for the ‘model pile’ procedure. In the

123

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

case where ground strength parameters are used as inputs, it seems in the logic of
Eurocode 7 that the alternative method should be used with their characteristic values.
Taking the characteristic values of cu from Table 7.5 results in the characteristic resistance
being calculated as
Rk = Rs, k + Rb, k = 34.9cu, shaft, k + 4.5cu, base, k = 1640 + 144 = 1784 kN
This value is 26% greater than the value of 1415 kN obtained by applying the ξ correlation
factors as in the case of the ‘model pile’ procedure, and, thus, the following ‘degrees of
overestimation’ are obtained:
• for DA-1 (Combination 2)
DO = 1.26 × 1.08 – 1 = 36%
• for DA-2
DO = 1.26 × 1.02 – 1 = 29%
This arises directly from the fact that the characteristic values of cu are very near the ones
governing in the ‘model pile’ procedure (i.e. the minimum ones given by BH2) and that no
correlation factor ξ is applied here.

Design Approach 3
For DA-3, sets A1 and M2 (material factors) of Annex A must be applied (with resistance
factors γR = 1.0), thus
Fc, d = 1.35Gk + 1.5Qk = 1.35 × 600 + 1.5 × 300 = 1260 kN
The design resistance is calculated directly from the design values of the undrained
shear strength, which are derived from their characteristic values:
• for the pile shaft
cu, shaft, d = cu, shaft, k /γcu = 47/1.40 = 33.6 kPa
• for the pile base
cu, base, d = cu, base, k /γcu = 32/1.40 = 22.9 kPa
and
Rc, d = 34.9cu, shaft, d + 4.5cu, base, d = 1640 + 144 = 1276 kN
DO = 1%, which shows that the length of the pile (D = 18.5 m) in this particular
example is optimal with regard to DA-3.

Conclusion
If the alternative method is excluded, the three Eurocode Design Approaches give similar
results. DA-1 is the least conservative, DA-3 is the most conservative, while DA-2 gives an
intermediate value (very near to that of DA-1). In the present example, the maximum
deviation between the three Design Approaches is very small in view of the usual
uncertainties in the calculation models and in the estimation of the soil material
properties. It is pointed out that the above conclusion is not universal; other combinations
of loads and material properties may result in different levels of conservatism for each
approach.
The design of a pile may also have to include ULS calculations to check ULS accidental
and seismic situations as well as calculations to check SLS situations, i.e. to check that the
settlement of the pile is acceptable under the common working load combinations. The
SLS design calculations may demonstrate that a further increase of the pile length is
required if the settlement tolerances are very small.

124

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 7. PILE FOUNDATIONS

Surcharge placed

FD
Soft clay
5 m = LD
FC

RS
Stiff clay
L R?

Rb = 0

0.3 m

Fig. 7.3. Pile subjected to downdrag (Simpson and Driscoll, 1998)

Example 7.4: design of a pile subject to downdrag

This example involves a bored pile of diameter B = 300 mm, embedded in a stiff clay
overlain by 5 m of soft clay as shown in Fig. 7.3.
A surcharge is placed at ground level, causing settlement of the soft clay and downdrag
(negative skin friction) on the pile. The pile is also subjected to a permanent vertical load
with a characteristic value Gk = 300 kN.
In the following, the required embedment LR of the pile in the stiff clay is calculated for
ULSs in persistent situations, using the three Design Approaches (DA-1, DA-2 and
DA-3).
The checks of SLSs would require prediction of the settlements under the characteristic
values of the actions together with the characteristic values of the ground deformation
parameters.
According to clause 7.3.2.1(3)P, the designer is free to choose either the ground Clause 7.3.2.1(3)P
settlement or the downdrag (negative friction) load as the action. Taking the ground
settlement as the action is generally carried out using a full ground–pile interaction
analysis. It can be performed using the well-known ‘t–z curve’ approach, where ‘t’ is the
mobilized shaft friction and ‘z’ is the corresponding relative pile–ground displacement
∆s = s – sg (where s is the settlement of the pile and sg is the ‘free’ ground settlement).
The ground settlement sg can be taken as a function of depth, and the analysis enables, in
particular, the neutral point where ∆s = 0 to be determined. Below the neutral point the
pile settles more than the ground, so that the ground provides resistance to rather than
loads the pile, i.e. the skin friction becomes positive. This analysis is more complicated Clause 7.3.2.2(1)P
than the approach which consists of estimating a maximum (limiting) downdrag load
along the entire length of the pile in the settling layer. The interaction analysis may be
used to demonstrate that the actual downdrag load is less than the limiting value (either
because the downdrag is not fully mobilized and/or because the neutral point is located
somewhat above the bottom of the layer).
In this example, the main goal is to show how the three Design Approaches are applied Clause 2.4.7.3.4
and, for simplicity, how the maximum downdrag load approach is used.

125

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Table 7.6. Recommended values of partial factors to apply when designing for downdrag
(negative skin friction) treated as an action (effective stress analysis)

Downdrag (negative skin friction)

Permanent Compressive
structural action, Shear strength resistance (positive
Design Approach γG parameter, γϕ Load, γG friction), γ s (or γϕ )

DA-1
Combination 1 Set A1: 1.35 Set M1: 1.0 Set A1: 1.35 Set R1: 1.0
Combination 2 Set A2: 1.00 Set M2: 1.25a Set A2: 1.00 Set R4: 1.3
DA-2 Set A1: 1.35 Set M1: 1.0 Set A1: 1.35 Set R2: 1.1
DA-3 Set A1: 1.35 Set M2: 1.25a Set A2: 1.00 Set M2: 1.25
a
This guide recommends that this M2 value is applied as a partial action factor, not a material factor,
because the downdrag force is usually estimated without recourse to the angle of shearing resistance
of the ground

The following assumptions are made:

Clause 7.3.2.2(1)P • The settlement of the soft clay is sufficient to mobilize the maximum downdrag
(limiting negative skin friction) along the entire length of the pile in the soft clay layer,
LD = 5 m.
• The characteristic value of the long-term downdrag per unit area (calculated in terms
Clause 2.4.5.2(8) of effective stress), on average over the depth LD, is q¢Dk = 20 kPa; this is an upper
value, as downdrag is detrimental to the pile behaviour.
• The shaft resistance in the stiff clay is upwards (positive skin friction); i.e. it resists the
downdrag on the pile along the whole length LR in this layer.
• The characteristic value of the long-term shaft resistance per unit area in the stiff clay
(calculated in terms of effective stress), on average over the depth LR, is q¢sk = 50 kPa.
• The ground resistance at the base of the pile is assumed to be negligible (Rb, k = 0).
Clause 2.4.7.3.4 From clause 2.4.7.3.4 and the corresponding tables in Annex A, the recommended
values of the partial factors for the three Design Approaches are given in Table 7.6.

Characteristic and design values of loads

The characteristic loads and resistance are:

• Permanent load:
Gk = 300 kN
• Total downdrag load:
FD, k = πBLD q¢Dk = π × 0.3 × 5 × 20 = 94.2 kN
• Shaft resistance in the stiff clay layer:

Rs, k = πBLR q¢sk = 47.1LR (in kN, with LR in m)

The design action (total vertical compression load) is

Fc, d = Gd + FD, d = γGGk + FD, d
and the design resistance is
Rs, d = Rs, k /γs
The condition to be fulfilled for all ULSs is Fc, d £ Rs, d.

126

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 7. PILE FOUNDATIONS

The way the design value of the downdrag load FD, d is obtained from the characteristic
value FD, k depends on the Design Approach being used.

Design Approach 1
For geotechnical design to DA-1, Combination 2 is normally considered first, as it is often
the prevailing approach, and the result is then checked against Combination 1.
For Combination 2, set A2 in Table A.3, set M1 (on Rs, k) and M2 (on FD, k) in Table A.4 Clause
and set R4 in Table A.7 are used: 2.4.7.3.4.2(2)P
• the design value of the downdrag load is taken as
FD, d = γϕ FD, k
where γϕ = 1.25 is selected from Table 7.6. Thus,
FD, d = 1.25 × 94.2 = 117.8 kN
• the total design action is
Fc, d = γG Gk + FD, d = 1.0 × 300 + 117.8 = 417.8 kN
• the design resistance is
Rs, d = Rs, k /γs = 47.1LR/1.3 = 36.2 LR (in kN, with LR in m)

The condition Fc, d £ Rs, d leads to LR ≥ 417.8/36.2 = 11.54 m.

Clause
For Combination 1, sets A1, M1 and R1 are used:
2.4.7.3.4.2(2)P
• the design value of the downdrag load, considered as an unfavourable action, is
FD, d = γG FD, k = 1.35 × 94.2 = 127.2 kN
• the total design action is
Fc, d = γG Gk + FD, d = 1.35 × 300 + 127.2 = 532.2 kN
• with LR = 11.54 m (from Combination 1), the design resistance is
Rs, d = Rs, k /γs = 47.1LR/1.0 = 47.1 × 11.54 = 543.5 kN

Thus, the condition Fc, d £ Rs, d is fulfilled and LR = 11.54 m is the final result of the
design according to DA-1.
Note that for the structural design of the pile, the larger design value of the action
(total vertical compression load) should be considered and is given by Combination 1:
Fc, d = 532.2 kN.

Design Approach 2
Set A1, set M1 and set R2 are used: Clause
2.4.7.3.4.3(1)P
• the design value of the downdrag load, considered as an unfavourable action, is
FD, d = γG FD, k = 1.35 × 94.2 = 127.2 kN
• the total design action is
Fc, d = γG Gk + FD, d = 1.35 × 300 + 127.2 = 532.2 kN
• the design resistance is
Rs, d = Rs, k /γs = 47.1LR/1.1 = 42.8LR (in kN, with DR in m)
The condition Fc, d £ Rs, d leads to LR ≥ 532.2/42.8 = 12.43 m, and for the structural
design of the pile Fc, d = 532.2 kN should be used.

127

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Design Approach 3
Clause Sets A1 (on Gk) or A2 (on FD, k), M2 (on FD, k and Rs, k) and R3 are used:
2.4.7.3.4.4(1)P
• the design value of the downdrag load is obtained from FD, d = γϕ FD, k, where
γϕ = 1.25 is selected from Table 7.6. Thus,
FD, d = 1.25 × 94.2 = 117.8 kN
• the total design action is
Fc, d = γG Gk + FD, d = 1.35 × 300 + 117.8 = 522.8 kN
• the design resistance in the stiff clay is taken as
Rs, d = Rs, k(tan ϕ¢d) = Rs, k /γϕ
i.e. by applying the material partial factor γϕ directly to Rs, k (or q¢s, k) because Rs (or q¢s)
is rather insensitive to the variations of tan ϕ¢k. Thus,
Rs, d = 47.1LR/1.25 = 37.7LR (in kN, with LR in m)
The condition Fc, d £ Rs, d leads to LR ≥ 522.8/37.7 = 13.87 m, and for the structural
design of the pile Fc, d = 522.8 kN should be used.

Conclusion
DA-3 requires the longest pile length of the three Design Approaches: LR = 13.87 m,
compared with LR = 11.54 m for DA-1 and LR = 12.43 m for DA-2. This is clearly due to
the fact that for DA-3 the values of the three partial factors are equal to 1.25 or 1.35. It can
also be argued that the application of the correlation factor ξ to the estimated values of
Clause 7.6.2.2(8)P shaft friction qs in DA-1 and DA-2 (see clauses 7.6.2.2(8)P and 7.6.2.3(5)P) would have led
Clause 7.6.2.3(5)P to lower values for q¢s, k than in DA-3 (for which they are not used).

Example 7.5: uplift of piled structures

Introduction and description of the problem
In this example of a buried structure subjected to uplift due to water pressure, only
the UPL ultimate limit state is checked. It is found that tensile piles are needed to
Clause 7.6.3.1(3)P ensure stability against failure due to uplift. The two failure mechanisms given in clause
7.6.3.1(3)P which need to be considered in this situation are pull-out of the piles from the
ground mass, and uplift of the block of ground containing the piles. The stability of the
structure against both of these failure mechanisms is assessed in this example using only
the UPL partial factors. However, as stability against failure due to pull-out of the piles
involves the strength of the ground providing resistance, this mechanism may also be
assessed using the GEO partial factors.
The raft of a buried structure is located H = 20 m below the groundwater level, as
shown in Fig. 7.4. The water pressures are hydrostatic. The structure is thus subjected to
an upward water pressure with a characteristic value of γwater H = 10 kN/m3 × 20 m =
200 kN/m2. The structure extends above the groundwater level, so there is no downward
water pressure acting on the structure.
The characteristic value of the permanent weight of the structure (calculated with low
characteristic values of the unit weights of the materials , e.g. 23.5 kN/m3 for concrete) is
gk = 100 kN per square metre of the raft plan area.
The structure is founded on tensile piles: initially, it is assumed that there will be one
pile every 5 m2. The pile diameter is B = 0.5 m. The characteristic value of the unit
Clause 7.6.3.3(4)P shaft tensile resistance of the piles is qs, k = 70 kPa. This characteristic value has been
Clause 7.6.3.3(6) established according to one of the methods of Section 7 (using the ‘model pile’ procedure
Clause 7.6.3.3(2) including the use of the ξ values or using the ‘alternative’ procedure; the model factor has
been taken equal to 1.0). The characteristic resistance has to account for the adverse

128

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 7. PILE FOUNDATIONS

0 swater
200 kPa

H = 20 m
Permanent load, gk = 100 kN/m2

swater = 200 kPa

Characteristic tensile
resistance, qs, k = 70 kPa
L=?

Hydrostatic
water pressure

Depth

Fig. 7.4. Description of the problem and data for Example 7.5

effect of interaction between the tensile piles when they are close together. The weight
density of the soil containing the piles is 20 kN/m3.
The length L of the piles and their optimum spacing are to be calculated.
For the sake of simplicity, the construction is considered as being infinitely long;
side-effects (e.g. friction along the walls of the buried structure and the sides of the block
of soil containing the piles) are not considered in the example presented here.

Check against uplift failure

The design requirements to prevent uplift failure are checked according to the following Clause 2.4.7.4
inequality:
Vdst, d £ Gstb, d + Rd (2.8)
The recommended values of the partial factors for this UPL ULS in Table A.15 are
γG, dstb = 1.0 on the permanent destabilizing action and γG, stb = 0.9 on the permanent
stabilizing action. Hence:

• the design value of destabilizing uplift force (per m2) is

Vdst, d = γG, dstγwater H = 1.0 × 10 × 20 = 200 kN
• the design value of stabilizing weight (per m2) is

Gstb, d = γG, stb gk = 0.9 × 100 = 90 kN

As Vdst, d = 200 kN > Gstb, d = 90 kN, the design requirement to prevent uplift failure is
Rd ≥ 110 kN (per m2), to be provided by tensile piles.

Design value of pile resistance and pile length: single-pile tensile failure
If the spacing of the piles is one pile every 5 m2, as shown in Fig. 7.5, the design values of
the destabilizing and stabilizing forces over each 5 m2 area are Vdst, d = 5.0 × 200 =
1000 kN and Gstb, d = 5.0 × 100 = 450 kN, respectively, so that tensile load to be carried
by each pile is
Rd ≥ Vdst, d – Gstb, d = 1000 – 450 = 550 kN
The characteristic tensile resistance of a single pile is given by

129

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Gstb, d = 0.9 × 100 × 5 = 450 kN: stabilizing action

Area = 5 m2 Vdst, d = 1.0 × 10 × 20 × 5 = 1000 kN: destabilizing action

Rd = Rk /g = LB pqsk / 1.4 = L × 0.5 × p × 70/1.4

Fig. 7.5. Equilibrium equation for a single pile every 5 m2 under the UPL condition

Rk = πBqs, k L = 3.14 × 0.5 × 70 × L = 110L (in kN, with L in m)

The design tensile resistance of the single pile Rd is determined by applying the UPL
partial factor γt, s = 1.40 to the characteristic tensile resistance of the pile Rk (see Table
A.16):
Rd = Rk /γt, s = (110 × L)/1.4
Its length L is readily found from inequality (2.8):
550 £ 110 × L/1.4
Thus, L ≥ 7.0 m; i.e. the minimum required length for a single pile to fulfil the UPL
requirement is 7.0 m.
Note: the self-weight of the piles has been neglected in the previous calculation. If the
submerged weight of the pile is included, one finds L = 6.85 m.

Block failure of the structure and the piles and soil included between the piles
The length of the piles and their spacing should be checked to ensure that the ULS of
Clause 7.6.3.1(4)P block failure, i.e. uplift failure of the structure and the block of soil containing the
Clause 2.4.7.4(1)P piles, is not exceeded. Clause 7.6.3.1(4)P, together with clause 2.4.7.4(1)P, imply that
the inequality for block failure should be written in terms of total stresses. In the
following, both total stress and the effective stress analyses are carried out for the sake of
comparison.

(a) Total stress analysis

• In a total stress analysis, the downward force per square metre is the weight of the
structure (100 kN) plus the total weight per square metre of the block of soil
containing the piles:
gk + γsoil L
where L is the length of the piles in metres, and γsoil is the weight density of the soil in
kilonewtons per cubic metre. The greater weight of the piles compared with the
weight of the block of soil has been ignored in this calculation as this is conservative
and has little effect on the result.
• The upward force per square metre on the plane through the pile bases is the water
pressure:

130

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 7. PILE FOUNDATIONS

γwater(H + L)
The partial factors γG, stb and γG, dst are applied to the stabilizing actions and to the Clause A.4(1)P
destabilizing actions on the block of soil and structure. The design inequality for the UPL
ULS becomes
γG, dstγwater(H + L) £ γG, stb(gk + γsoilL)
Inserting the recommended values in Table A.15 for the partial factors, γG, stb = 0.9 and
γG, dst = 1.0, and rearranging yields
L ≥ (1.0 × 10 × 20 – 0.9 × 100)/(0.9 × 20 – 1.0 × 10) = 13.75 m

(b) Effective stress analysis

• Using an effective stress analysis, the downward force per unit area is the weight of
the structure per square metre plus the submerged (effective) weight per square
metre of the block of soil containing the piles:
gk + γ ¢soil L
• The upward force per square metre is the water pressure acting on the raft:

γwater H

Applying the partial factors γG, stb and γG, dst, the design inequality becomes
γG, dstγwater H £ γG, stb(gk + γ ¢soil L)
Inserting the recommended values for the partial factors, γG, stb = 0.9 and γG, dst = 1.0,
yields
L ≥ (1.0 × 10 × 20 – 0.9 × 100)/(0.9 × 10) = 12.2 m

Comparison between total stress and effective stress analyses

The total stress analysis yields a more conservative result than the effective stress analysis
(L ≥ 13.75 m compared with L ≥ 12.2 m). Note that the total stress analysis would have
yielded the same result as the effective stress analysis, if the water pressures on both
sides of the inequality had been treated in the same manner, i.e. applying the same
(unfavourable) partial factor γG, dst to them on both sides of the inequality (according to
the ‘single-source’ principle).
In the following section, the more conservative result (the result of the total stress
analysis), is adopted: L ≥ 13.75 m. This is because:

• the recommended values of Table A.15 are meant to be used with the total stress Clause 7.6.3.1(4)P
approach, as indicated in clause 7.6.3.1(4)P
• the equivalent overall factor of safety obtained by using the recommended partial Clause 2.4.7.4(2)P
factor values from Table A.15 for γG, stb and γG, dst is already very low (γG, dst /γG, stb = 1.1).
As this pile length is longer than L = 7 m, as previously obtained, it is found that, in this
example, the length of the piles is determined by the uplift failure criterion of the block of
soil containing the piles and not by the sum of the tensile resistances of the single piles.

Optimization of the pile spacing

Since the pile length required to ensure that the ULS of uplift due to tensile resistance
failure of the piles is not exceeded, with one pile every 5 m2, is 7 m, while the length of the
piles required to ensure safety against uplift of the block of soil containing the structure
and the piles is 13.75 m, the optimum spacing of the piles so that these two failure
mechanisms are just not exceeded can now be determined.

131

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

As the piles have a design length of 13.75 m and a characteristic tensile resistance per
unit length of 110 kN/m, their design tensile resistance becomes
Rd = 13.75 × 110/1.4 = 1080 kN
Since the design load to be supported per square metre is 110 kN, one pile is needed
every 1080/110 = 9.8 m2. This is the optimum pile spacing for both pile resistance failure
and block failure to be just not exceeded.

Design value of the tensile force in a pile for ULS structural pile design
The structural design of the piles should be performed with the most severe UPL and STR
action factors. Clearly, STR factors are more severe than UPL factors, so that only the
STR check has to be performed with the STR action factors.
The permanent weight of the structure is a favourable action, so its design value is
calculated using the action factor for permanent favourable loads in STR and GEO ULSs
(Table A.3):
Gd = 1.0 × 100 × 9.8 = 980 kN
The upward force due to the water pressure of 200 kN/m2 is an unfavourable action, so
its design value acting on the pile is calculated using the action factor for permanent
unfavourable loads in STR and GEO ULSs (Table A.3):
Wd = 1.35 × 200 × 9.8 = 2646 kN
Hence, the design action for structural design of the pile, denoted here as negative as it
is a tensile force, is
Fd = Gd – Wd = 980 – 2646 = –1666 kN

Discussion
(1) In the example above, for convenience, the resisting block of soil has a simplified
shape. More refined shapes may be used, accounting for the conical shape of the
stabilizing soil mass at the base of the piles and around the edge piles.
(2) The equivalent overall factor of safety for the single pile under UPL conditions
is about 1.55ξ (i.e. (1.4/0.9)ξ), ξ being the correlation factor used to obtain the
characteristic value of the shaft resistance for the piles. This is a rather low value,
which can be increased by a model factor. The characteristic value of the skin friction
should account for the adverse effect of interacting tensile piles.
(3) The recommended value for the equivalent overall factor of safety for block failure
under UPL conditions is about 1.1 (γG, dst /γG, stb). This is a rather low value, which may
be increased by a model factor, if necessary.

132

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 8

Anchorages

This chapter is concerned with the geotechnical design of temporary and permanent
anchorages as presented in Section 8 of EN 1997-1. The structure of the chapter largely
follows that of Section 8:
8.1. General Clause 8.1
8.2. Ultimate limit state design Clause 8.5
8.3. Structural design of anchorages Clause 8.5.4
8.4. Load testing of ground anchorages Clauses 8.7 and 8.8
In common with other sections of EN 1997-1, Section 8 gives only the basic requirements
for the design of anchorages without describing or specifying methods. To assist the
designer, this chapter describes how to determine the design value of the anchorage load
from calculations of structures tied with anchorages and explains how to assess their design
resistance. An example, using the anchorage reaction calculated in Example 9.2, illustrates
the process, from the calculated anchorage reactions obtained from the design of the
retaining wall to the specification of the anchorage resistances to be proved by the suitability
and acceptability tests.

8.1. General
Anchorages are characterized by a free length and a system to transmit tensile force Clause 8.1.1(2)P
to the resisting ground. Anchorages may be prestressed (e.g. grouted anchorages) or
non-prestressed (e.g. deadman anchorages). Systems with bonding to the ground over the
whole length of the anchorage, such as nails or piles, are not covered in Section 8 of
EN 1997-1.
Section 8 is related to Sections 9 (‘Retaining structures’), 10 (‘Hydraulic failure’), 11 (‘Overall
stability’) and 12 (‘Embankments’), since the design value of the action to be sustained by the
anchorage in ultimate limit states (ULSs) and serviceability limit states (SLSs) has to be
determined from the requirements of one or more of these sections. Uplift failure of a group
of anchorages is checked using the same principles as for pile groups subjected to UPL
conditions, and is described in Chapter 7 of this guide.
Section 8 refers to EN 1537: 1999, Ground Anchors, for definitions and guidance on the Clause 8.1.2
execution and testing of grouted anchorages. It should be noted that there is not always Clause 8.4
perfect consistency between EN 1537 and EN 1997-1. The requirements for geotechnical Clause 8.5.4
design in EN 1997-1 supersede those in EN 1537. Section 8.4 of this guide gives some Clause 8.7
consideration to anchorage load testing.
The tendon inclination should be such that self-prestressing is provided by deformations Clause 8.4(9)P
due to a potential failure mechanism. This requirement will usually be met with flat
anchorages that cross the failure surface at an angle of less than 90°.

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

8.2. Ultimate limit state design

8.2.1. Design of the anchorage
The basic requirement of the ULS design of anchorages is:
Clause 8.5.1(1)P Pd £ Ra, d (8.1)
where Ra, d is the design value of the anchorage pull-out resistance and Pd is the design value
of the anchorage load (i.e. the design value of the load to be sustained by the anchorage).
The anchorage load Pd is obtained as the ‘reaction’ in the design of the structure (e.g. a
retaining wall where the anchorage resists earth pressures, a slope where anchorages resist
sliding of the soil mass, or a buried structure where anchorages resist uplift pressures). For
prestressed anchorages, Pd is composed of the prestress force P0 (at lock-off) and the
variations in the force during construction (further excavations) and the life of the structure
(loadings). The anchorage prestress (lock-off) force P0 is often a favourable action in the
design of geotechnical structures (e.g. a retaining wall, a slope or a raft subjected to uplift)
(Fig. 8.1).
Clause 8.2(1)P The anchorage has to be designed to sustain the anchorage load, Pd, to prevent (Fig. 8.2):
• Geotechnical failure: for grouted anchorages, failure at the interface between the
ground and the body of grout, or failure by excessive displacement of the anchor head or
by creep; for deadman anchorages, failure by insufficient resistance of the ground in
front of the deadman.
• Structural failure: failure of the tendon or anchor head; for grouted anchorages, failure
of the bond between the tendon and the grout.

Clause 8.3(2)P The anchorage load Pd is thus an unfavourable action for the geotechnical and structural
design of the anchorage.
Clause 8.3(1)P It should be noted that forces can be applied to the anchorage during prestressing which
may exceed the forces required for the design of the structure, e.g. during acceptance tests.

8.2.2. Design value of the anchorage load

Clause 8.5.5(1)P The design value of the anchorage load, Pd, is obtained from:
• the design of the supported structure in ULSs and, if relevant,
• the design of the supported structure in SLSs.
When using a calculation model for the supported structure where the earth pressures are
introduced as known actions in the calculations (e.g. see Chapter 9, where the design
of embedded walls using equilibrium methods or an equivalent beam is discussed), an
anchorage is usually modelled as a support. In these models, the lock-off force P0 is not
introduced in the calculation. The calculated reaction in the support is the design value of
the anchorage load Pd in the design situation considered (ULS or SLS). The prestress
(lock-off) force may be taken as a fraction, e.g. 50 to 70% or even more of the SLS reaction.
When using a (numerical) calculation model simulating the interaction between the soil
and the supported structure, the anchorages are usually introduced as (linear) springs with
appropriate stiffness (e.g. see Chapter 9, where the design of embedded walls as beams on
non-linear springs and finite elements is discussed); in such models the earth pressures are
not known at the start of the calculations. Prestress forces, P0, are introduced as favourable
actions (with γG = 1.0). The reactions calculated in the anchorages are design values of the
anchorage loads in the (ULS or SLS) design situation considered.
The choice of calculation model will depend on the required accuracy of the results.
Models where earth pressures are introduced as known actions are usually well suited to
ULS design for simply supported walls. They are sometimes used in simplified methods for
the SLS check, using appropriate stresses (e.g. pressures at rest) and stress distributions. An
interaction model is often used in ULS design of embedded walls supported by several
rows of anchorages or when the length of the wall deviates significantly from the length

134

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 8. ANCHORAGES

qd

P0

jdcd or cud

Fig. 8.1. Design of the retaining wall. The prestress force (lock-off), P0, is a favourable action in the
design of the wall and for overall stability

Pd
Structural resistance of tendon

Resistance at grout–ground

Fig. 8.2. The design value, Pd, obtained from the wall design has to be sustained by the anchor: it is
the design value of the action on the anchorage

corresponding to free earth support conditions, and when the SLS check requires the
calculation of a value of the displacement. These aspects are considered in more detail in
Chapter 9 of this guide.
ULS design calculations for the supported structure will deliver the value of the
anchorage load Pd for whichever of the three Design Approaches is adopted. For STR/GEO
ULS design checks, the partial factors will be taken from Tables A.3 and A.4. For UPL design
situations, the partial factors will be taken from Tables A.15 and A.16.
SLS design calculations will deliver the value of the anchorage load PSLS complying with
the serviceability criteria. This value may be more severe for the design of the anchorage
than the value obtained from the ULS design, especially for stiff walls or stiff anchorages in
heavily over-consolidated soils or in cases where very strict serviceability (displacement)
criteria apply. In such cases, the earth pressures may not drop to their active values, and the
design of the anchorage can be governed by the serviceability criteria for the retaining wall.
The design value of the anchorage load Pd should then be assessed for the anchorage load
obtained in the serviceability design condition.
The following subsections present examples of the assessment of ULS anchorage loads for
ULS and SLS design of the supported structure.

Determination of the design anchorage load starting from the ULS check of the structure
When non-prestressed anchorages are used, or when the serviceability requirements are not
stringent, the first step in the design is usually a check on the ULSs of the geotechnical

135

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

structure (retaining wall, reinforced slope, anchored raft etc.). For STR/GEO ULS design
checks, the partial factors will be taken from Tables A.3 and A.4 according to the adopted
Design Approach; for UPL design situations, the partial factors will be taken from Tables
A.15 and A.16. The ULS calculation delivers design values of the anchorage loads Pd (as the
design value of the ‘reactions’ of the supported structure), which will be the action used for
checking the ULS of the anchorage according to clause 8.5.
For prestressed anchorages, the prestress (lock-off) force will be introduced as an action
in the calculations using the partial factors for favourable actions (γF = 1.0). The soil–
structure interaction models calculate a ‘reaction’ (e.g. the anchorage load to restrain an
embedded wall or to tie down a raft), since the anchorage prestress force is a favourable
action for the geotechnical structure. As for non-prestressed anchorages, the ULS calculation
will deliver the design values of the anchorage loads necessary to ensure the stability of the
structure in ULS design conditions.
Figure 8.3 illustrates this design method.
In overall ULS stability calculations of the type illustrated in Fig. 8.4, when using the
‘assumed failure surface method’ (see Section 11.5 of this guide), the design value of
the anchorage resistance (or of the part of the anchorage outside the failure surface) is
introduced into the stability calculation as a favourable action.

Determination of the ULS design anchorage load starting from an SLS check of the structure
When prestressed anchorages are used, the minimum prestress (lock-off) force P0 and the
Clause 8.5.5(1)P corresponding anchorage reaction in SLS are often determined by SLS calculations, to avoid
excessive displacements of the supported structure. In such cases the ULS check of the
anchorage is performed using the following procedures.
(1) When a soil–structure interaction model is applied, e.g. an elastic, prestressed spring
model for the anchorage and a non-linear spring model for the soil interacting with the
retaining structure (see Chapter 9), the lock-off force P0, SLS is calculated, in a first step,
by trial and error until the SLS displacement criteria are met. The ULS check of the
structure is performed in a second step using the partial factors for ULS design. In this
ULS calculation, the prestress (lock-off) force P0, SLS is a favourable action on the
structure, and should be treated accordingly (partial action factor γF = 1.0). The ULS
calculation will deliver the design value Pd of the anchorage load. Figure 8.5 illustrates
the two-step procedure.

qd
Lock-off force P0, j
introduced as a
favourable action

Reaction from ULS

calculation Pd

j ¢d, c ¢d or cu, d
Anchorage to be
Excavation level
designed for action Pd
for ULS design

j ¢d, c ¢d or cu, d

Fig. 8.3. ULS design model (STR plus GEO) for an embedded wall in which the prestress force is
introduced as a favourable action; the calculation delivers a design value of the anchorage load

136

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 8. ANCHORAGES

qd

Rd

jd , c d

Fig. 8.4. Design model for overall stability using assumed failure surface. The design value of the
anchorage resistance, Rd, is entered into the calculation

Step 1: SLS calculation Step 2: ULS calculation

qk qd
Lock-off force
P0, SLS to comply
with SLS criteria P0, d = 1.0P0, SLS

Pd

j ¢k, c ¢k or cu, k j ¢d, c ¢d or cu, d

Input: P0 by trial and error until the displacement (SLS) Input: P0d = 1.0P0, SLS as a favourable action
requirements are met

Output: lock-off prestress force P0, SLS to comply with Output: Pd, design value of anchorage load in ULS
required wall deformation and characteristic value of
anchorage load Pk in serviceability conditions

Fig. 8.5. Example of the sequence of calculations to obtain the design ULS anchorage load starting
from an SLS check of the structure, using a soil–structure interaction model. The ULS design of the
anchorage is based on the greater of 1.35Pk and Pd

137

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

It is important to note that in Design Approaches 1 (DA-1) and 2 (DA-2), Pd is at least

equal to γG Pk due to the use of the partial action factor γG in Table A.3 (recommended
value γG =1.35) applied in DA-1 Combination 1 and DA-2 (see Chapter 9 of this guide
for more details). In Design Approach 3 (DA-3), values of Pd /Pk as low as 1.1 or 1.2 can
be found, e.g. for stiff walls and highly prestressed anchorages. As the recommended
values of partial resistance factors γa for ULS design of the anchorages have been
established on the assumption that Pd ≥ 1.35Pk, it is advisable to perform the ULS design
of the anchorages with a design value of the anchor load equal to the more severe of Pd
and γmodel Pk, where Pd results from the ULS design of the structure and Pk results from
Clause 8.6(4) the SLS design of the structure. γmodel is the model factor permitted in clause 8.6(4). A
value for γmodel equal to the value of γG for unfavourable actions is recommended in this
guide (γG = 1.35 itself being recommended by EN 1997-1).
Clause 8.6(3) When using calculations based on spring models, the most adverse combination of the
low and high value of the spring constant of the anchorage and of the prestress (lock-off)
force, respectively, should be selected. An estimate of the maximum anchorage stiffness
may be made from the total length of the anchorage minus the tendon bond length; an
estimate of the minimum anchorage stiffness is made from the total length of the
anchorage minus half the tendon bond length.
(2) When the SLS calculations are performed using limit equilibrium methods or equivalent
beam methods (i.e. methods where the earth pressures acting on the wall are assumed to
be known before the start of the calculations), the calculated SLS reaction should be
Clause 8.6(4) multiplied by a model factor to obtain the design value of the anchorage load. The value
of the model factor may be set in the National Annex. It is recommended that a value of
not less than the value of γG for unfavourable actions be used (γG = 1.35 itself being
recommended by EN 1997-1).

8.2.3. Design value of the anchorage resistance

Clause 8.5.1(2) The design value of pull-out resistance may be determined from the results of tests or by
calculations.

Clause 8.5.2 Pull-out resistance determined from the results of tests

Clause 8.4(10)P For grouted anchorages and screw anchorages, the characteristic value of the pull-out
resistance Ra, k shall be determined on the basis of suitability tests. The design value of the
anchorage resistance Ra, d is then obtained from the characteristic value by applying a partial
factor γa:
Ra, d = Ra, k /γa (8.2)
Clause 8.5.2(2) Recommended values of γa are given in Table A.12 for STR/GEO and may be changed in
the National Annex. The recommended value γa =1.0 for DA-3 (see column R3) does not
apply for calculating the design value of anchorage resistance from pull-out tests since, in
DA-3, the resistance of the anchorages is calculated using the shear strength parameters of
the ground. This comment also applies in the case of the resistance of piles obtained from
load tests (see Chapter 7). For UPL design situations, recommended values of γa are given in
Table A.16.
Clause 8.5.2(3) The characteristic value Ra, k should be related to the test results through a correlation
factor ξa that accounts for the number of tests and the variability of the test results. If a ξa
value is introduced, it may relate, for example, the minimum of the measured resistances
Ra, measured, min and mean value of the measured tests resistances Ra, measured, mean to the
characteristic value Ra, k by a relationship similar to the one introduced for piles:
Ra, k = Min{Ra, measured, mean /ξa, 1; Ra, measured, min /ξa, 2}
Recommended values of the correlation factors ξa are not given in EN 1997-1; they may be
set by the National Annex. Values for ξa, 1 could range from 1.20 (one test) to 1.10 (three or
more tests); ξa, 2 would then range from 1.20 (one test) to 1.05 (three or more tests).

138

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 8. ANCHORAGES

Combining inequality (8.1) and equation (8.2) yields

Pd £ Ra, d = Ra, k /γa or Ra, k ≥ Pd γa (D8.1) Clause 8.5.2(1)P
If the anchorage load Pd is known, e.g. from the calculation of the wall, the requirement
becomes that the pull-out resistance of the anchorages is the more stringent of
Ra, measured, min ≥ Pd γa ξa, 1
Ra, measured, mean ≥ Pd γa ξa, 2

Pull-out resistance determined by calculations Clause 8.5.3

A calculated design value of the pull-out resistance has to be determined according to the Clause 8.5.3(1)P
principles of clauses 2.4.7 and 2.4.8.

Deadman anchorages
A deadman anchorage uses the passive resistance of the soil in front of it. The design value of
this passive resistance is calculated from the soil shear strength parameters ck¢ and ϕk¢ (or cuk)
by applying the partial factors in Table A.4 and Table A.13 as follows:
• DA-1 and DA-3: obtain design values of the soil shear strength parameters by applying
the partial factors of set M1 (DA-1 Combination 1) or M2 (DA-1 Combination 2 and
DA-3) according to Table A.4. The values of γR, e are equal to 1.0 in both approaches.
• DA-2: divide the characteristic resistance of the deadman anchorage (calculated using
characteristic values of soil shear strength parameters) by the partial factor γR, e of set R2
in Table A.13.

Grouted or screw anchorages

The characteristic resistance of grouted anchorages may be assessed by using charts or
other empirical rules and correlations. Such charts should have been established from
experimental data and should deliver characteristic values of the anchorage pull-out
resistance. EN 1997-1 is unclear on a method to assess the design value of the pull-out
resistance when such charts are used. Due to the direct correlation between these charts and
the results of load tests, it is logical to determine the design value of the pull-out resistance by
applying equation (8.2):
Rd = Ra, k /γa
where γa is taken from Table A.12 (STR/GEO design situation) or Table A.16 (UPL design
situation). A complementary model factor may be used to account for the uncertainty of
such charts.

Lock-off force of prestressed anchorages

EN 1997-1 gives no recommendations about the lock-off force, but it is advisable to select a
value sufficiently lower than the characteristic resistance Ra, k to avoid creep deformation.
EN 1537 recommends that the lock-off force should not be greater than 0.65 times the
characteristic resistance Ra, k of the grouted anchorage.

8.3. Structural design of anchorages Clause 8.5.4

The structural design of the anchorage is performed using EN 1992, EN 1993 and EN 1537, Clause 8.5.4(2)P
with the design value of the anchorage load Pd as obtained above.
The requirement is that a ULS is first attained in the soil (e.g. bond failure between the
grout and the soil or passive failure of a deadman anchorage), before a ULS is reached in the
material of the anchorage. The reason for this is to avoid brittle failure in the anchorage
material, as failure in the soil is often more ductile and gives ‘early warning’. For grouted Clause 8.5.4(1)P
anchorages, inequality (8.3) may be rewritten as

139

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Rt, d = Rt, k /γstructural, a ≥ Ra, d = Ra, k /γa (D8.2)

or
Rt, k ≥ Ra, k(γstructural, a/γa)
where Rt, k is the characteristic material resistance of the anchorage and γstructural, a is the
partial factor for the anchorage material.
In contrast to this, EN 1537 recommends, in its Annex D (informative), clause D.5.3, that
Ra, k > Rt, k. The EN 1997-1 and EN 1537 requirements are both fulfilled only if (γstructural, a/γa)
is smaller than 1.0. The EN 1537 requirement in fact ensures that an SLS failure occurs first
in the tendon and then in the bond between grout and soil. In a case of conflict between these
requirements, EN 1997-1 supersedes EN 1537.

8.4. Load testing of ground anchorages

Clause 8.8 8.4.1. Acceptance tests
Clause 8.1.2.3 These tests are performed on all working anchorages to confirm that each meets the design
Clause 8.4(10)P requirements. Clause 8.4(10)P, second sentence, requires the design resistance to be checked
by acceptance tests, in accordance with EN 1537. EN 1537 states in its Annex E (informative)
that the proof load, PP, is the greater of 1.25P0 and Ra, d when the anchor is loaded in
incremental steps (‘test method 3’), or PP = 1.25P0 when the anchor is loaded in cycles (‘test
methods 1 and 2’) from a datum load to the test load; P0 is the calculated prestress force
(lock-off) and Ra, d is the design value of the anchorage resistance. Applying the test load in
incremental steps is the most common method. The acceptance criterion is expressed as a
maximum deformation per logarithmic time unit.
As the design value of the resistance of a grouted anchorage Ra, d is close to its characteristic
resistance Ra, k (recommended value of partial factor γa= 1.1, see Annex A, Table A.12) and to
the failure load, the combination of the requirements of EN 1997-1 and EN 1537 leads to
very high values of the proof load compared with the working load of the anchorage; on
average, a proof load Pp = Ra, d (where Ra, d ≥ Pd) would then become as much as 1.4–1.7
times the working load under service conditions, Pk, because the ratio between anchorage
load in a ULS and in an SLS, Pd /Pk, usually ranges from 1.4 to 1.7. The acceptance criteria set
out in EN 1537, when applied to a proof load equal to Ra, d (‘method 3’), may be severe. It
should be remembered in this respect that, in EN 1537, a value of γa ≥ 1.35 is applied
(EN 1537, Annex D), whereas in EN 1997-1 the recommended value is 1.1. This guide
therefore recommends a broad interpretation of the wording ‘shall comply with EN 1537’ and
‘meets the design requirement’; the consequences of running acceptance tests up to the design
resistance, Rd, of the anchorage should be carefully evaluated. The acceptance test should be
planned in such a way that the proof load remains far enough from pull-out failure to avoid
adverse affects on the anchorage, but close enough to prove that the as-built anchorage will
meet the design requirements. It may be preferable to apply a proof load equal to the more
severe of 1.1–1.3 times the load under serviceability conditions (the ‘working load’) or 1.25
times the lock-off force, and to put an acceptance criterion on a measured deformation in a
defined time interval.

Clause 8.7 8.4.2. Suitability tests

Clause 8.1.2.4 These tests are performed on selected anchors to confirm that a particular anchor design is
adequate or to determine the characteristic resistance of the anchorage. Suitability tests
Clause 8.4(10)P should preferably be run to failure in order to assess the characteristic pull-out resistance of
the anchorage, Ra, k, from the measured pull-out resistance, Ra. The pull-out resistance Ra is
defined as the load corresponding to the vertical asymptote of the creep slope curve α in
millimetres per log cycle time (time in minutes) plotted against the applied load. If the
asymptote cannot be defined, Ra is the load corresponding to α = 5 mm (see EN 1537,
Annex E.4.4 (informative), Fig. E5).

140

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 8. ANCHORAGES

The characteristic value of the pull–out resistance may be related to the results of the Clause 8.5.2(3)
suitability tests by applying a ξ factor (see Section 8.2.3 above).
If a certain design resistance Ra, d ≥ Pd has to be proven, the proof load should be at least
equal to the expected resistance of the anchorage, i.e. to
Pp ≥ ξa Ra, k = ξa Ra, d γa ≥ Pd γa ξa (D8.3)
where Pd is the calculated design value of the anchorage load.
At this value of the proof load, PP, the anchorage is adequate if it does not reach a state of
failure between grout and ground, as defined above, e.g. α < 5 mm.
The establishment of test standards for anchorages is one of the tasks of CEN Technical
Committee 341 on ground investigation and testing.

8.4.3. Investigation tests Clause 8.1.2.5

Investigation tests are necessary for new types of anchorage or for ground or load conditions Clause 8.4(8)P
where no experience is available for the considered anchorage system. These tests are
required to proceed to failure between the grout and the soil, or are performed as creep
tests.

8.4.4. Proof load as an action to the structure

The proof load is an unfavourable (characteristic, monotonic) action applied to the structure.
Thus, the structure (the wall, waling, etc.) has to be checked for this unfavourable action
according to the principles of clause 2.4.7.

Example 8.1: assessment of proof load for suitability and acceptance tests
The lock-off force and design values of the anchorage load per metre of wall in an SLS and
a ULS are obtained from the SLS and ULS calculations given in Chapter 9 (see Example
9.2: Table 9.6, for DA-1, DA-2 and DA-3 using the ‘limit equilibrium model’ (LEM);
Table 9.7, for DA-2 using a spring model; and Table 9.9 for SLSs). The anchorages are
spaced at 2.4 m intervals, and inclined at 10° to the horizontal. The design value of the
action to be sustained by the anchorage, Pd, in the ULS, and the SLS is obtained by
multiplying the design value of the horizontal component of the anchor force Fh by the
distance between the anchors (2.4 m) and correcting for the anchorage inclination by a
factor 1/cos 10° = 1.015, as summarized in Table 8.1.
In this example, the proof load to be applied to the anchorages in suitability and
acceptance tests will be assessed. At least three suitability tests will be performed.
The design value of the anchorage load in ULSs is greater than 1.35 times the
anchorage load in SLSs; thus, the ULS design of the anchorage will start from the ULS
anchorage loads.
Table 8.2 indicates:

• the required characteristic anchor resistances,

Ra, k = Ra, d γa ≥ Pd γa
• the required resistances to be proved by the suitability tests,
Ra, measured, min ≥ Ra, k ξa, 2
for a ξa, 2 value equal to 1.05, or Ra, measured, mean ≥ Ra, k ξa, 1 for ξa, 1 = 1.10 (three tests). The
values of ξa, 1 and ξa, 2 are chosen for the purposes of this example. As the partial factors of
DA-3 cannot be used to check the suitability of anchorages from the results of a load test,
the partial factors of DA-2 are used to meet the requirements for the proof loads in the
suitability tests. Of course, the value Pd from the wall calculations of DA-3 are used; the
required proof loads for the acceptance test on each anchor are determined according to

141

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

142

DESIGNERS’ GUIDE TO EN 1997-1

Table 8.1. Design value of anchorage load Pd obtained from SLS and ULS design of the retaining wall using DA-1, DA-2 and DA-3

ULS and SLS design value of horizontal anchorage load Fh from wall
design (kN/m) ULS and SLS design value of anchorage load Pd (kN/anchorage)

DA-2(S) DA-2(S)
Lock-off DA-1 DA-2(L) Spring DA-3 Lock-off DA-1 DA-2(L) spring DA-3
force (kN/m) SLS LEM LEM model LEM force (kN) SLS LEM LEM model LEM

100 112 172 228 157 172 244 273 419 557 383 419

Table 8.2. Overview of required characteristic anchor resistance and proof loads for suitability and acceptance tests (all values are axial forces in the anchorage, in
kN)

Required characteristic anchor Required proof load for suitability tests

resistance, Ra, k (γa = 1.1) (ξa = 1.1 and 1.05) Required proof load Pp for acceptance tests
Lock-off Anchor load
force, P0 in SLSs, PSLS DA-1 DA- 2(L) DA-2 (S) DA-3a DA-1 DA-2 (L) DA-2 (S) DA-3a DA-1 DA-2 (L) DA-2 (S) DA-3

244 273 461 611 421 461 Mean 484 642 442 484 EN 1537 419 557 383 419
Minimum 507 672 463 507 Realistic 313 313 313 313
a
Suitability to be checked for Pd calculated in DA-3 by applying the partial factors of DA-2

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 8. ANCHORAGES

the requirement of EN 1537, i.e. the greater of 1.25P0 and Ra, d (see the first row of proof
loads for acceptance tests).
Table 8.2 shows that, in all Design Approaches, the proof load for acceptance tests is
very high compared with the lock-off force and to the SLS anchorage load. These high
proof loads, Pp, for acceptance tests can lead to creep of the anchors during testing
because there is only a very small margin between the proof load and the anchor
resistance (compare the values indicated for the proof load for suitability tests, keeping in
mind that proof loads for the suitability tests are failure loads). Table 8.2 proposes a more
realistic proof load (second row, in bold) obtained from
PP = max{1.25P0, 1.15RSLS}

143

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 9

Retaining structures

This chapter is concerned with the design of structures which retain ground (soil, rock or
backfill) and water. The material is covered in Section 9 of EN 1997-1 and in Annex C
(‘Sample procedures to determine limit values of earth pressures on vertical walls’). The
structure of this chapter follows that of Section 9:
9.1. General Clause 9.1
9.2. Limit States Clause 9.2
9.3. Actions, geometrical data and design situations Clause 9.3
9.4. Design and construction considerations Clause 9.4
9.5. Determination of earth pressures Clause 9.5
9.6. Water pressures Clause 9.6
9.7. Ultimate limit state design Clause 9.7
9.8. Serviceability limit state design Clause 9.8
The following three main types of retaining structures can be distinguished: Clause 9.1.2(1)
• Gravity walls, in which the weight of the wall, sometimes including stabilizing masses of
ground (stem walls), plays a significant role in supporting the retained material.
• Embedded walls, which are relatively thin walls of steel, reinforced concrete or timber.
These walls either rely for stability solely on the earth resistance due to the passive earth
pressure in front of the walls (cantilever walls) or are supported by anchorages or struts
and by the resistance in the area of their toes (supported walls). The bending resistance
of an embedded wall plays a significant role in the support of the retained material
compared with the weight of the wall.
• Composite retaining structures, which include walls combining elements of the previous
two types. Typical examples are cofferdams and reinforced earth and nailed structures.
Section 9 applies the principles of Sections 1–4 in the design and construction of gravity
and embedded walls. Composite retaining structures are not discussed, although many of the
basic design principles are still applicable.
Section 9 relies on Section 6 for the design of gravity retaining walls against bearing
resistance failure, Section 8 for the design of anchorages against pull-out failure, Section 10
for design against hydraulic failure and Section 11 for the design of retaining structures
against overall stability failure. Certain principles of Section 7 may also be used, where
relevant, when checking the vertical stability of embedded walls.
In common with other sections of EN 1997-1, Section 9 gives only the basic requirements
for the design of retaining structures without describing or specifying particular calculation
methods. To assist the designer in this task, the present chapter describes typical calculation
methods for ultimate limit state (ULS) and serviceability limit state (SLS) designs (see
clauses 9.7 and 9.8), and includes the following worked examples:

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

• Example 9.1: ULS design of the foundation of a gravity (stem) wall against sliding failure
and bearing failure.
• Example 9.2: ULS and SLS design of an embedded wall supported by a single row of
anchorages, including its design against hydraulic failure by heave. The design of the
anchorage is presented in Chapter 8 (see Example 8.1).

9.1. General
Clause 9.1.1(1)P Section 9 applies to structures which retain ground materials and water. Ground material is
retained if it is kept at a slope steeper than it would eventually adopt if no structure were
present. Retained ground materials exert actions (earth pressures) on retaining structures.
Earth pressures on gravity and embedded walls are within the scope of Section 9. Earth
Clause 9.1.1(2)P pressures on composite retaining structures are not discussed, while pressures from granular
materials stored in silos are discussed in EN 1991-4 (Actions in Silos and Tanks).
Clause 2.1(19) Retaining structures of usual size typically belong to Geotechnical Category 2. Examples
are given in the note to Clause 2.1(19).

9.2. Limit states

Clause 9.2(1)P When designing a retaining structure, all possible limit states should be considered. Typical
limit states include the following.

Ultimate limit states

(1) Loss of overall stability (GEO ULS). It should be demonstrated that the occurrence of
an overall stability failure is sufficiently unlikely. As a minimum, limit modes of the types
illustrated in Fig. 9.1 should be considered, taking into account progressive failure and
possible strength degradation (e.g. due to excess pore pressure build-up).
(2) Foundation failure of gravity and composite walls, i.e. exceedence of the sliding or the
bearing resistance (GEO ULS). Examples of foundation failures are illustrated in Fig. 9.2.
The principles of Section 6 should be applied as appropriate, especially when designing
Clause 6.5.4 against bearing failure of the ground below the base of a wall under loads with large
Clause 6.5.3 eccentricities and inclinations and when designing against sliding failure. Uplift pressures
under the foundation due to water seepage should also be included in the analyses.
(3) Foundation failure of gravity and composite walls by toppling (EQU ULS). EQU-type
failure (i.e. loss of equilibrium of the wall considered as a rigid body) is usually limited to
gravity walls founded on rock or propped at the toe.
(4) Failure of embedded walls by rotation or horizontal translation or by lack of vertical
equilibrium (GEO ULS). Examples of such failures are illustrated in Figs 9.3 and 9.4.
(5) Failure of a structural element such as a wall, anchorage, wale or strut, including failure of
the connection between such elements (STR ULS). Failure of an anchorage includes
structural failure (STR ULS) and pull-out failure in the ground (GEO ULS). Examples of
such failures are illustrated in Fig. 9.5 (structural failure) and Fig. 9.6 (geotechnical failure).
(6) Failure of a retaining structure by hydraulic heave, internal erosion or piping, unacceptable
leakage of water, or transport of soil particles through or beneath the wall, all caused by
excessive hydraulic gradients (HYD ULS). This condition can arise at the bottom of
deep excavations when there is a significant difference between the groundwater levels
at the opposite sides of the retaining wall (see also Section 10).

Serviceability limit states

(1) Unacceptable movement of the retaining structure, which may affect the appearance or
functionality of the structure itself, or other structures or utilities influenced by these
movements.
(2) Unacceptable change in the groundwater regime.

146

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 9. RETAINING STRUCTURES

9.3. Actions, geometrical data and design situations

9.3.1. Actions
Earth pressures on the active side of retaining structures are unfavourable geotechnical Clause 2.4.2(4)
actions, since their magnitude depends on the properties of the ground, namely its weight
density, cohesion, angle of shearing resistance and wall–ground interface resistance (see
EN 1990, clause 1.5.3.7). Groundwater pressures are also geotechnical actions.
Although clause 2.4.2(4) states that ‘earth pressures should be considered for inclusion as
actions’, without differentiating between earth pressures on the active and passive sides, the
earth pressures on the passive side of embedded retaining walls in designs against GEO and
STR ULSs:

• may be treated either as favourable geotechnical actions or as earth resistances in

Design Approaches 1 (DA-1) and 3 (DA-3) with identical results (since the partial
factors for favourable actions and resistances are unity in both cases)
• should be treated as earth resistances in Design Approach 2 (DA-2), since this approach
includes a non-unit partial factor for ‘earth resistance’ (recommended value γR = 1.4,
see Table A.13).

Earth pressures on the passive side of gravity retaining walls should be treated as
favourable actions when designing against bearing failure (see inequality (6.1) and clause Clause 6.5.2.1(3)P
6.5.2.1(3)P) and as resistances when designing against base sliding (see inequality (6.2) – Clause 6.5.3(2)P
term Rp, d). Actually, in DA-1 and DA-3 the earth pressure on the passive side of gravity
retaining walls may be treated either as a favourable geotechnical action or as an earth
resistance with identical results; in DA-2, however, the results are different.
In supported retaining walls, the forces from propping elements should be considered as
follows (see further discussion in Chapter 8):

(1) Struts are usually modelled as kinematic constraints (fixed points), and thus the
corresponding prop forces are calculated as reactions due to these constraints.
(2) In certain GEO ULS designs, where wall movements are sufficiently large to mobilize
fully the resistance of the anchorages, prestressed or deadman anchorages may be
modelled as permanent favourable actions with design values equal to the design
resistance of the anchorages (i.e. equal to the pull-out resistance, which should be lower
than the structural resistance of the tendon according to Section 8).
(3) In SLS and certain ULS designs, wall movements may not be sufficiently large to
mobilize fully the resistance of the anchorages. Such cases may include stiff pile or
diaphragm walls, where a plastic hinge in the concrete section (STR ULS) can develop
at small wall deflections which may be insufficient to cause pull-out or structural
collapse of the anchorages (especially if they are long and flexible). In these cases,
the forces from prestressed or deadman anchorages generally result from interaction
between the wall and the ground, as follows:
(a) Forces from struts, deadman anchors (and other non-prestressed anchorages) are
calculated as the reactions of springs having an appropriate stiffness (estimated
from the stiffness of the tendon and the deformability of the deadman anchor as the
earth resistance is mobilized).
(b) Forces from prestressed anchorages consist of an action (equal to the known
prestressing lock-off force chosen by the designer) and a spring reaction (with
stiffness calculated from the stiffness of the tendon). In practice, many prestressing
systems are fairly extensible, so the ‘reaction’ component may be neglected in
comparison with the prestress action.
Generally, the forces exerted on a retaining structure with values assumed known at the
beginning of the calculation are considered as ‘actions’, while forces with initially
unknown values, to be determined by the interaction of the retaining structure with
support elements (ground springs, anchorages, struts, etc.), are considered as ‘reactions’.

147

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

9.3.2. Geometrical data

Clause 9.3.2 Clause 9.3.2 deals with uncertainties in the geometrical data (excavation and water levels). In
most cases, small variations in geometrical data are considered to be accommodated by
the safety elements included in the calculations (mainly the characteristic values of the
geometrical data and the partial factors). However, because the design of retaining structures
may be very sensitive to ground and water levels (especially for embedded walls in soils with
high shear strength or high angles of shearing resistance), special requirements are included
in this clause, mainly for unforeseen overdig in front of the wall and groundwater levels on
both sides of the wall.

Unforeseen overdig in front of the wall

Clause 9.3.2.2(2) In ULS design calculations, where the wall stability depends on the earth resistance in front
of the wall, the level of the resisting soil should be lowered below the nominally expected
level by an amount ∆a, which depends on the degree of control on the excavation level. With
a normal degree of control, ∆a should:
• for cantilever walls, be equal to 10% of the wall height above excavation level (up to a
maximum of 0.5 m)
• for supported walls, be equal to 10% of the distance between the lowest support and the
excavation level (up to a maximum of 0.5 m).
Smaller values of the overdig (even equal to zero) may be used when the surface level is
controlled reliably during the construction of the works, while larger values may be used
when the surface level is particularly uncertain.
The ground level reductions recommended in clause 9.3.2.2(2) are critical in the design,
and should only be disregarded with great caution, particularly where embedded walls rely
heavily on relatively short penetrations into the restraining ground. Simpson and Driscoll
(1998) showed that unforeseen overdig can have a large influence on the safety against
passive failure and on the calculated bending moments.

Groundwater levels in front of and behind the wall

Clause 9.3.2.3 The selection of the levels of the phreatic surfaces in front of and behind the wall must
consider long-term variations of the groundwater regime and/or the ground permeability,
the presence of perched or artesian aquifers and the possibility that drainage behind the wall
may cease to function with time.

9.4. Design and construction considerations

Clause 9.4.1(1)P According to clause 9.4.1(1)P, the design of earth retaining structures requires consideration
of all relevant ULSs and SLSs.
When checking against the occurrence of a ULS, two sets of calculations should be
performed for all relevant design situations:
(1) GEO ULS calculations, which consist of a set of limit equilibrium calculations to
determine the proportions and geometry of the structure (the width of the base of a
gravity wall, the toe penetration depth of an embedded wall, the number and location of
anchorages or struts in a propped wall, etc.). These calculations should examine the
horizontal and vertical force equilibrium and the moment equilibrium of the wall under
appropriate lateral earth pressure on the active side (and the passive side, if relevant),
other external actions (e.g. the self weight of the wall, hydraulic actions, anchor and strut
forces) and resistances (e.g. the sliding resistance at the base of a gravity wall).
(2) STR ULS calculations, which consist of a set of structural design calculations to determine
the size and properties of the structural sections necessary to resist the bending moments
and shear forces determined from the limit equilibrium calculations.

148

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 9. RETAINING STRUCTURES

The determination of appropriate earth pressures for ULS designs is discussed in clause Clause 9.5
9.5, and the design methodology is described in clause 9.7. Clause 9.7
SLS designs require estimates (and in some cases rigorous calculations) of the
displacements of the wall and the retained ground, with the objective of ensuring that these
do not exceed the serviceability requirements of influenced structures and utilities. For
retaining structures without strict serviceability requirements, the geometry is usually
determined by ULS design calculations and checked by SLS calculations (if relevant). For
retaining structures with strict serviceability requirements, the SLS requirements often
govern the design. The determination of appropriate earth pressures for use in SLS designs Clause 9.5
is discussed in clause 9.5, and the design methodology is described in clause 9.8. Clause 9.8

9.5. Determination of earth pressures

The determination of the appropriate earth pressures is a major issue in the design of Clause 9.5.1
earth retaining structures, since earth pressures depend not only on the values of ground
parameters but also on the movement and/or deflection that the wall can accommodate,
due to kinematic constraints or serviceability requirements. Examples of kinematic
constraints resulting in different wall movements and/or deflections, and hence different
earth pressures, include walls with rigid connections to basement structures, cantilever and
anchored embedded walls, gravity walls with surface foundations and walls founded on piles.
An example of a serviceability requirement is the need to avoid cracking of structures
founded on retained soil. The earth pressure values used in designs to EN 1997-1 are
intended to give conservative estimates of the earth pressure on the active and passive
sides of the wall (overestimates and underestimates, respectively) for the anticipated wall
movement/deflection at the relevant limit state and design situation.
According to clause 9.5.1(10), the magnitude of the appropriate earth pressures is generally Clause 9.5.1(10)
different in SLS and ULS calculations.
In principle, if a wall is installed without any disturbance to the ground and no subsequent
movement occurs, the earth pressures on both sides correspond to the at-rest conditions.
Clause 9.5.2 gives equations for estimating the at-rest earth pressure coefficient (K0) for Clause 9.5.2
normally consolidated and over-consolidated soils and for ground sloping upwards from the
wall. However, earth pressures corresponding to the at-rest conditions are unlikely in most
retaining structures since the wall construction and excavation sequence will certainly induce
movements sufficient to modify the at-rest conditions.
Clauses 9.5.3 and 9.5.4 discuss the dependence of the earth pressure on wall movement. As Clause 9.5.3
the wall moves and/or deflects, the earth pressure on the passive side increases, and the Clause 9.5.4
maximum value (limiting passive earth pressure) usually occurs at the ULS. On the active
side of a wall, the earth pressure generally decreases with wall movement, but the actually
mobilized value of the earth pressure for use in SLS and ULS designs depends on the
stiffness of the wall, the stiffness of any support elements (struts, anchorages, etc.) and the
stress–strain characteristics of the ground. Annex C gives guidance on the magnitude of wall
movement required for mobilization of the limiting active and passive earth pressures.
The methodology for determining the appropriate earth pressures for use in ULS and SLS
designs is described below.

ULS designs
As ULS designs usually involve large wall deformations/deflections causing collapse of the
system, limiting values of the earth pressures on both the active and passive sides of the wall
can be estimated by the usual methods of plastic analysis using the earth pressure equations Clause C.1(1)
given in Annex C, which may be applied either in terms of total or effective stresses, as
appropriate. The values of the limiting earth pressure coefficients for ϕ¢ ranging from 10 to 45°
and for different wall–ground interface parameters and ground inclinations are presented as
graphs in Annex C (based on Caquot et al., 1973). Account must be taken of increased Clause 9.5.5(1)P
pressures due to compaction and due to possible water-filled tension or shrinkage cracks. Clause 9.6(5)P

149

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

In walls which cannot move sufficiently to mobilize the limiting earth pressures, even in
ULSs (e.g. due to kinematic constraints), ULS design calculations should be performed with
Clause 9.5.4 intermediate earth pressures compatible with the kinematic constraints, i.e. using higher
values than the limiting active pressure on the retained side and lower values than the
limiting passive earth pressure on the resisting side. Appropriate earth pressures may
be calculated using numerical analyses such as finite-element analyses, one-dimensional
models of walls supported on linear (Winkler-type) or non-linear (elasto-plastic) springs,
or other methods which take account of earth pressure redistribution, or even by using
empirical earth pressure distributions, such as those described by Terzaghi and Peck (1967)
and EAU (1980). The use of limiting earth pressures in such cases is usually not conservative,
especially with regard to the calculated strut or anchor forces.

SLS designs
If analyses are performed which include interaction between the wall and the ground (e.g.
using non-linear spring models or finite elements), a priori earth pressure determination is
not required since the appropriate earth pressures are automatically generated through
the numerical model. When such analyses are considered impractical, the serviceability
requirements may be checked using simplified beam models of the wall loaded with
appropriate earth pressures and supported by linear (Winkler-type) or non-linear soil
springs (see Section 9.7 for more details). In such cases, appropriate earth pressures should
be calculated using the fraction of the soil strength mobilized at strains compatible with the
serviceability requirements.

Wall–ground friction and adhesion

Clause 9.5.1(5) The maximum shear stress (τ) which can be mobilized at the wall–ground interface is
determined from the equation
τ = a + σn¢ tan δ
where a is the adhesion between the wall and the ground, σn¢ is the normal effective stress
acting on the wall, and δ is the interface parameter between the wall and the ground (angle of
shearing resistance).
Clauses 9.5.1(5) Clauses 9.5.1(5) to 9.5.1(8) recommend the following design values of the wall–ground
to 9.5.1(8) adhesion, a, and the interface parameter, δ, for ULS designs:
• For a completely smooth wall: δ = 0 and a = 0. For a completely rough wall, values of
a = c¢ and δ = ϕ¢ may be appropriate.
• For steel sheet piles in clay, immediately after driving: δ = 0. An increase in this value
may take place over a period of time.
Clause 9.5.1(7) • For concrete walls cast against ground in sand or gravel: δ = ϕ¢cv .
• For pre-cast concrete walls, or driven steel sheet pile walls, in sand or gravel: 23 ϕ¢cv .
where c¢, ϕ¢ and ϕ¢cv are, respectively, the design values of the effective cohesion, angle of
shearing resistance and critical state angle of shearing resistance of the ground adjacent to
the wall.
The above values of the wall–ground interaction parameters (a, δ) are maximum (limiting)
values since they correspond to the maximum shear stress which can be mobilized at the
wall–ground interface. Lower values of the interaction parameters may actually be
mobilized, depending on the vertical component of the ground displacement relative to the
wall (on either of its two sides) and the requirement to ensure vertical equilibrium of the
wall. This means that different values of the interaction parameters may be required on
either side of the wall to achieve vertical equilibrium. For example, in high embedded walls
with multiple levels of prestressed anchorages (especially when the anchorages are strongly
inclined towards the vertical), use of the maximum value of the δ parameter on the active
side, as recommended above, may have the consequence that vertical equilibrium of the wall

150

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 9. RETAINING STRUCTURES

cannot be satisfied. In such cases, the actual value of d should not be assumed, but should be
calculated from vertical equilibrium of the wall; this value is used subsequently in horizontal
equilibrium calculations.
The vertical equilibrium requirement implies that the wall friction on the passive side
(acting upwards) may not exceed the wall friction on the active side (acting downwards),
taking into account any vertical forces on the wall (e.g. the vertical component of anchor
forces). This requirement results in a reduction of the assumed maximum (limiting) friction
on the passive side or on the active side of the wall, depending on the need to reduce the
vertical upward force or the vertical downward force.
Clause 9.5.5 discusses the determination of earth pressures on walls with compacted Clause 9.5.5
backfill. A method for calculating earth pressures due to compaction was published by
Ingold (1979). Compaction pressures should only be included in STR ULS design
calculations (structural design of the wall) and in SLS design calculations. In GEO ULS
design calculations, compaction pressures should not be considered since they are normally
relieved with relatively little horizontal movement of the wall.

9.6. Water pressures

The determination of water pressures on retaining structures is very important since their
magnitude often exceeds that of the earth pressures. If the equilibrium level of the water
table is well defined and measures are taken to prevent variations during heavy rain or
flooding, the design water pressures can be calculated from the equilibrium water table,
making due allowance for possible seasonal variations; otherwise, the most adverse water
pressure conditions should be used in design. Where a difference in water pressures is likely
to exist on opposite sides of the wall, account should be taken of seepage flow occurring
around the structure. A simplified method for calculating the water pressures on a wall due
to steady state seepage flow around it is based on the assumption of uniform head loss
along a flow path around the wall in the relatively less permeable ground layer (see
Example 9.2).
Since the actions derived from water pressures are considered as geotechnical actions, the
appropriate partial factors for actions are applied to them. This is particularly significant in
DA-3, where the partial factors on structural and geotechnical actions are different. The
partial factors for actions are applied to the net water pressure acting on the wall, i.e. on the
difference between the water pressures acting on the two sides of the wall.

9.7. Ultimate limit state design

The provisions in clause 9.7 concern both permanent and temporary retaining structures. Clause 9.7
According to clause 2.4.7.1(2)P, the partial factors for use in ULS designs of retaining Clause 2.4.7.1(2)P
structures for persistent and transient situations should be those defined in Annex A. These
partial factors are included in the following tables:

• Table A.3 – partial factors on actions or the effects of actions

• Table A.4 – partial factors on soil parameters
• Table A.12 – partial resistance factors for pre-stressed anchorages
• Table A.13 – partial resistance factors for retaining structures.

Table 9.1 summarizes the values of the partial factors recommended in Annex A.
ULS designs of retaining structures for persistent and transient design situations are
carried out using appropriate earth pressures (see Section 9.5) and ensuring that the design
values of the effects of actions do not exceed the corresponding design values of the
resistances: Ed £ Rd. Either of the three Design Approaches (DA) described in clause Clause 2.4.7.3.4
2.4.7.3.4 may be used.

151

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Table 9.1.
(a) Recommended values of the partial factors on actions and effects of actions (persistent and
transient situations) according to Annex A

Partial factors on actions

Unfavourable permanent, Unfavourable variable,

Design Approach (sets of partial factors) γGa γQb

DA-1 Combination 1 (A1 ‘+’ M1 ‘+’ R1) 1.35 1.5

DA-1 Combination 2 (A2 ‘+’ M2 ‘+’ R1) 1.0 1.3
DA-2 (A1 ‘+’ M1 ‘+’ R2) 1.35 1.5
DA-3 (A2 or A1 ‘+’ M2 ‘+’ R3)
Geotechnical actionc 1.0 1.3
Structural actiond 1.35 1.5
a
For favourable permanent action: γG = 1.0
b
For favourable variable action: γQ = 0.0
c
Geotechnical action: action transmitted to the wall through the ground
d
Structural action: action from a supported structure applied directly to the wall

(b) Recommended values of the partial factors on soil parameters and resistances (persistent and
transient situations) according to Annex A

Soil parameters, γM Resistances

Design Approach (sets of Weight

partial factors) density tan ϕ¢: γϕ c¢: γc cu: γcu Passive, γR, e Anchor, γa

DA-1 Combination 1 1.0 1.0 1.0 1.0 1.0 1.1

(A1 ‘+’ M1 ‘+’ R1)
DA-1 Combination 2 1.0 1.25 1.25 1.4 1.0 1.1
(A2 ‘+’ M2 ‘+’ R1)
DA-2 (A1 ‘+’ M1 ‘+’ R2) 1.0 1.0 1.0 1.0 1.4 1.1
DA-3 (A2 or A1 ‘+’ M2 ‘+’ R3) 1.0 1.25 1.25 1.4 1.0 1.0

The design values of actions are calculated as follows (see also Chapter 2):
(1) In DA-1 (Combination 1) and DA-2, by applying appropriate partial factors on:
(a) the characteristic values of the actions, i.e. Fd = γF Fk, for non-geotechnical actions
(e.g. the self-weight of the wall)
(b) the values of the actions calculated using the characteristic values of the ground
parameters, i.e. Fd = γF F(Xk), for geotechnical actions (e.g. earth pressure).
Alternatively, design values of actions may be calculated by applying appropriate partial
factors on the calculated effects of actions, i.e. Ed = γE E(Fk, Xk, ad), for example on the
bending moments (M) in the wall, which are calculated using the characteristic actions
and ground parameters.
(2) In DA-1 (Combination 2) and DA-3, design values of non-geotechnical actions are
calculated from the equation Fd = γF Fk while design values of geotechnical actions are
calculated using design (factored) values of ground parameters, i.e. Fd = γF F(Xk /γM). In
both cases, γF = 1.0, except for unfavourable variable actions (where the recommended
value is γF = 1.30 – see Table A.3).
Clause 2.4.7.3.2(2) According to clause 2.4.7.3.2(2), in some design situations the application of partial
factors to geotechnical actions (i.e. earth pressures on the ‘active’ side of walls and net water
pressures) may lead to design values of the effects of the actions which are unreasonable or
even physically impossible. A typical example is the calculation of the design value of the
water pressure on the ‘active’ side of a wall with a high groundwater table (near the ground

152

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 9. RETAINING STRUCTURES

surface), where the application of the relevant partial factor for actions (γF = 1.35
in persistent and transient situations) may lead to unrealistic design water pressures
corresponding to groundwater levels above the crest of the wall. In these situations, the
partial factors for actions may be applied directly on the effects of the actions derived from
the characteristic values of the actions, i.e. using the equation
Ed = γE E(Fk, Xk, ad)
For example, the appropriate partial factor for actions (γF) may be applied directly to the
shear force and to the bending moment at a critical section of a cantilever retaining wall,
and to other key design quantities (effects of actions), where these quantities have been
calculated using characteristic values of the actions and ground parameters (active earth
pressures, water pressures, surcharges, etc.).

ULS design of gravity walls

A list of the ULSs that need to be considered in the case of gravity walls is provided in clause Clause 9.2
9.2, with the most important being bearing and sliding failure of the foundation (GEO ULS) Clause 9.7.3
and exceedence of the structural resistance in critical sections of stem walls (STR ULS).
The bases of gravity walls are usually spread foundations, often subjected to markedly
eccentric loads. EN 1997-1 includes no ‘middle third rule’ to limit highly eccentric loads on
the foundations of gravity walls, but requires design against bearing failure under eccentric Clause 6.5.4
loading (see clause 6.5.4). In walls founded below the water table, uplift forces on the
foundations due to water pressure should be included in the design.
In stem walls with significant heels, the earth pressure is usually assumed to act on the
‘virtual back’ of the wall (i.e. a vertical plane from the rear extremity of the heel up to the
ground surface), provided that the length of the heel is large enough to permit the
development of the conjugate (i.e. symmetrical with respect to the vertical) ‘failure surface’
within the soil mass above the heel. In such cases, it can be shown that the active earth
pressure is parallel to the ground surface behind the wall regardless of the angle of frictional
resistance of the ground (e.g. if the ground surface is horizontal the active earth pressure is
also horizontal).
The passive earth pressure force (Pp) acting on the resisting side of the wall should be
considered as an earth resistance (with partial factors from Table A.13) when designing
against base sliding, and as a favourable geotechnical action (with partial factors from
Table A.3) when designing against bearing failure of the foundation (see discussion in
Section 9.3.1).
Example 9.1 presents the methodology for the ULS design of a stem gravity wall according
to the three Design Approaches.

ULS design of embedded walls

The aim in the ULS design of embedded walls is to determine the minimum wall penetration
required to prevent rotational failure and ensure vertical equilibrium and, for a given wall Clause 9.7.4
penetration, to determine the distribution of the effects of the actions (e.g. bending Clause 9.7.5
moments and shear forces) along the height of the wall and the magnitude of the support
reactions (anchor or strut forces), in the case of supported walls.
The ULS design of embedded walls is usually more complicated than the design of other
geotechnical structures, not only because embedded walls require an interaction analysis but
also mainly because the fundamental assumption of any geotechnical ULS design (i.e. a limit
state occurring in the ground) usually cannot be satisfied simultaneously at all locations in
the ground affecting the wall, due to the kinematic constraints imposed at the toe of the wall
and at the positions of the supports. Consequently, the ‘simple’ ULS assumption that
limiting earth pressures develop simultaneously on both the active and passive sides of the
wall is not always realistic and cannot be used in walls with long toe penetrations and walls
supported at multiple levels. In these cases, calculation models should include wall–ground
interaction for the analysis of earth pressure development, thus complicating significantly

153

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

the application of the partial factors in the three Design Approaches (mainly because of
the interplay between the stiffness-controlled wall–ground interaction and the strength-
controlled earth pressures caused by the interaction). Such calculation models are discussed
below.
In general, depending on the magnitude of toe penetration, embedded walls can be
designed for any condition in the range between ‘free’- and ‘fixed’-earth support at their
toes. In the free-earth support condition, wall penetration is the minimum required to
ensure toe stability, with forward movement and rotation of the toe permitted to a degree
sufficient to mobilize the limiting passive earth pressure on the resisting side of the wall. In
the fixed-earth support condition, wall penetration is sufficiently longer than the minimum
to ensure complete toe fixation (zero forward movement and zero rotation). Walls designed
for free-earth support conditions obviously have the shortest length, but suffer larger
bending moments, anchor forces and deflections compared to similar walls designed for
fixed-earth support conditions.
The ULS design of embedded walls usually consists of the following steps:
(1) The geometrical characteristics of the model are established, taking into account an
Clause 9.3.2.2 unforeseen overdig at the toe of the excavation, in accordance with clause 9.3.2.2.
(2) In cases with a hydraulic head difference between the two sides of the wall, the pore
pressure distributions on the active and passive sides are determined as discussed in
Section 9.6. Their difference is equal to the net water pressure acting on the wall.
(3) GEO ULS analyses, usually involving horizontal force and moment equilibrium, are
normally used to determine the minimum wall penetration. Vertical equilibrium is
normally used to determine the appropriate values of the wall–ground interface parameter
(δ) on the active and passive sides of the wall. In supported walls, the free-earth support
condition provides the minimum wall penetration satisfying the ULS requirement of
rotational stability, i.e. adequate safety against failure in the passive zone.
(4) The wall penetration may be increased beyond the required minimum, in order to
reduce the bending moments or certain support reactions, or to reduce the movement in
the supported ground, or for hydraulic reasons (e.g. to reduce water seepage or prevent
hydraulic heave/uplift). In such cases, the above analyses should be repeated using the
new wall penetration depth and appropriate earth pressures (usually different from the
limiting values) compatible with the kinematic constraints imposed at the toe of the
wall and at the positions of the supports. Calculation models for these analyses are
recommended below.
(5) The structural design of the wall and the design of the anchorages are performed using
the design values of the effects of the actions (bending moments, shear forces) and the
anchor forces calculated using the previous analyses.
(6) Other ULSs are analysed as relevant. Such limit states may include overall stability
failure, stability of anchorages at a lower failure plane (e.g. using a Kranz-type method
as described in EAU (1996)), vertical failure of the wall, hydraulic failure, etc. These
analyses may result in longer wall penetration, higher capacity of the wall and the
support elements and/or increased length of the anchorage system.
The ULS design of embedded walls can be performed with several types of calculation
model, including:
• assumed earth pressure models, such as limit equilibrium models (LEMs) and models
with other assumed earth pressures (e.g. the Terzaghi and Peck earth pressures)
• wall–ground interaction models, such as models of beams on ground spring supports and
finite-element models.
Each type of model involves different assumptions concerning the magnitude of the earth
pressures acting on the wall, thus limiting its applicability to those situations where these
assumptions are fulfilled. The following subsections describe the application of the three
Design Approaches in the ULS design of embedded walls for persistent and transient

154

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 9. RETAINING STRUCTURES

situations, using the above types of models. It is pointed out that wall–ground interaction
models can also be used in SLS design calculations.

Limit equilibrium models

LEMs assume that the limiting earth pressures develop along the full height of the wall
(limiting active pressure on the back of the wall and limiting passive earth pressure on the
front, i.e. resisting, side). Because of this assumption, LEMs can be used to determine the
minimum wall penetration required for rotational stability in cantilever walls and in walls
supported at a single level. These analyses also provide the design values of all the effects of
the actions on the wall (bending moments and shear forces) and the anchorage forces for this
minimum wall penetration. For walls with longer penetrations than the above minimum
value and for walls with multiple support levels, LEMs are not appropriate because, in
these cases, the mobilized earth pressures (on the active or the resisting side) can deviate
significantly from the limiting values.
ULS calculations with LEMs in persistent and transient situations are performed using
design values of the limiting earth pressures, which are calculated as follows in each of the
three Design Approaches:
• In DA-1 Combination 2 and in DA-3, the design values of the active and passive earth
pressures are obtained using the partial factors for soil parameters γM from Table A.4
(set M2).
• In DA-2 and in DA-1 Combination 1 the design value of the active earth pressure is
obtained by multiplying the characteristic values of the active pressure components by
the appropriate partial factor for actions (γG or γQ from Table A.3, set A1). The design
value of the passive earth pressure is obtained by dividing the characteristic value by the
partial factor for earth resistance γR, e from Table A.13: set R1 for DA-1 Combination 1
(recommended value: 1.0) and set R2 for DA-2 (recommended value: 1.4).
Figure 9.1 illustrates the implementation of LEMs in the case of a simply supported wall
loaded with active earth pressure (σa), passive earth pressure (σp) and net water pressure
(σw).
In DA-2, all partial factors are applied at the very end of the calculations (see p. 35), i.e.
the total driving action (active earth pressure force plus the net water pressure force minus
any support reaction) is multiplied by the partial factor on actions, and the result is
compared with the resultant resistance (passive earth pressure force) divided by the partial
factor on resistances, i.e.
FkγF £ Rk/γRe
Implementation of this procedure is complicated by the fact that the various components
of the driving actions (favourable, unfavourable, permanent and variable) require different
values of the partial factor for actions (γF), and separation of the various components of the
driving actions is complicated (and results in inaccuracies in non-linear problems).
If the behaviour of the wall is linear (i.e. plastic hinges do not develop), DA-2
can be applied using the following equivalent alternative procedure, called DA-2* for
convenience:
• Actions on the active side of the wall (active earth pressure, net water pressure and any
other permanent actions) enter the calculations with values equal to their characteristic
values. Surface variable loads enter the calculation with the value q = qk(γQ /γG), where
qk is the characteristic value (in order to account for the difference between the partial
factor of variable and permanent actions).
• Earth resistances on the passive side of the wall enter the calculations with their
characteristic value factored by 1/(γReγG), in order to account for the partial factor of
earth resistance (γRe) and the partial factor of permanent actions (γG) missing from the
actions on the active side.

155

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

qd = 1.3qk

s ¢a, h, d = Ka(j¢d, c ¢d)(1.0g ¢ha + 1.3qk)

s ¢a, h, d = Ka(j¢d, c ¢d)(1.0g ¢ha + 1.3qk)

s ¢p, h, d = Kp(j¢d, c ¢d)g ¢hp /1.0

sw, d = 1.0sw, k
(a)

qd = 1.5qk

s ¢a, h, d = 1.35Ka(j¢k, c ¢k)g ¢ha + 1.5Ka(j¢k, c ¢k)qk

s ¢a, h, d = 1.35Ka(j¢k, c ¢k)g ¢ha + 1.5Ka(j¢k, c ¢k)qk

s ¢p, h, d = Kp(j¢k, c ¢k)g ¢hp /1.4 sw, d = 1.35sw, k

(b)

qd = 1.1qk

s ¢a, h, d = Ka(j¢k, c ¢k)g ¢ha + 1.1Ka(j¢k, c ¢k)qk

s ¢a, h, d = Ka(j¢k, c ¢k)g ¢ha + 1.1Ka(j¢k, c ¢k)qk

s ¢p, h, d = Kp(j¢k, c ¢k)g ¢hp /(1.4 × 1.35) sw = 1.0sw, k

(c)

Fig. 9.1. (a) Application of an LEM in DA-1 Combination 2 and in DA-3. (b) Application of an
LEM in DA-2; the same model may be used in DA-1 Combination 1 with γR, e = 1 (instead of 1.4).
(c) Application of an LEM using the equivalent alternative procedure DA-2* described in the text

156

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 9. RETAINING STRUCTURES

Reactons;
resistance Actions

s ¢h, d according to
the DA chosen

R4, d

Fig. 9.2. Beam model of a multi-supported wall with assumed earth pressure loads

• Equilibrium calculations are performed as for any other Design Approach giving the
minimum wall penetration, the effects of the actions (e.g. bending moments) and the
support reaction. The design values of the effects of the actions and the support reaction
are obtained by multiplying the calculated values by the partial factor γG (since this
factor was not applied to the actions on the active side).
If the behaviour of the wall is linear, DA-2* gives exactly the same results as DA-2.
Figure 9.1c illustrates the implementation of LEMs with DA-2* (as described above) in
the case of a simply supported wall.

Beam models with assumed earth pressure loads

In beam models with assumed earth pressure loads, the wall is modelled as a beam resting on
supports at the locations of the anchors (or struts) and at the point of application of the
resultant earth resistance on the passive side. The beam is loaded by the net water pressure
and the a priori known earth pressure distributions on the active side (e.g. the Terzaghi and
Peck earth pressures for multi-strutted walls). Usually, the shape of the earth pressure
distribution on the resisting side of the wall is also assumed in order to estimate the location
of the point of application of the resultant earth resistance. For a given wall penetration,
equilibrium calculations provide the design values of the effects of the actions on the wall,
the support reactions and the earth resistance on the passive side required for equilibrium.
The sufficiency of the wall penetration is ensured by checking that the design value of the
earth resistance on the passive side calculated from equilibrium does not exceed the design
value of the limiting earth resistance (limiting passive earth pressure force). LEMs belong to
the class of assumed earth pressure models, since in LEMs the earth pressures are known a
priori (they are equal to the limiting values).
Figure 9.2 illustrates a beam model of a multi-supported wall with assumed earth pressure
loads. In beam models with assumed earth pressurs, the three Design Approaches are
applied exactly in the same way as LEMs (see above).
The accuracy of beam models with assumed earth pressures depends on the accuracy of
the assumed earth pressure distributions. As these pressure distributions strongly depend on
the lateral deflection of the wall (i.e. on the stiffness of the ground, the wall and the supports
and on the detailed construction sequence), more sophisticated models including wall–
ground interaction are usually more appropriate. Such models, based on ground springs and
on finite elements, are described below.

Beam models on ground spring supports (spring models)

In spring models, the wall is modelled as a one-dimensional beam. Wall–ground interaction
is treated in a simplified way, namely by independent non-linear spring supports arranged on

157

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

the active and passive sides of the wall, either continuously distributed or lumped at
specific locations along the wall. Spring models are appropriate in cases where wall–ground
interaction effects are important, i.e. for multi-propped walls, for walls with lengths exceeding
the minimum required for toe stability, for very stiff walls retaining overconsolidated soil
and/or embedded in very stiff ground (e.g. rock), and for walls supported by very stiff
elements (struts or anchorages) installed early and highly prestressed (see Fig. 9.5). Spring
models are also appropriate in the case of staged excavations and in cases involving gradual
(multi-stage) prestressing of anchorages.
The stress–strain behaviour of the springs simulates as closely as possible the reaction
of the ground as it is displaced from the initial K0 condition to the limiting condition
(active or passive). The ground springs yield at pressures corresponding to the limiting
active and passive earth pressures, and their pre-yield stiffness can be either constant or
multi-linear. In this way, the earth pressures mobilized on the active and resisting sides of
the wall are calculated from the ground spring reactions. The mobilized pressures are also
compatible with the wall deflections, which are controlled by the stiffness of the wall and its
supports.
Spring models can be used with the three Eurocode Design Approaches as follows:

(1) DA-1 Combination 2 and DA-3. The design values of the ground strength parameters
(obtained via the partial factors γM from Table A.4, set M2, are used to determine the
yield (limiting) pressures of the ground springs. Thus, the calculations ensure that the
mobilized earth resistance on the passive side of the wall (i.e. the sum of the calculated
ground spring reactions on the passive side of the wall) does not exceed the limiting
earth resistance. The calculations provide directly the design values of all the effects of
the actions on the wall (e.g. the design values of the support reactions). If the behaviour
of the wall is taken as elasto-plastic, the yield value of the wall bending moment entering
the calculations (Mu) should be taken as being equal to the design value of the wall
ultimate (plastic) moment of resistance, i.e. Mu = Mu, d = Mu, k/γM, where Mu, k is the
characteristic moment of resistance of the wall section and γM is the partial factor for the
wall material.
(2) DA-2 and DA-1 Combination 1. DA-2 and DA-1 Combination 1 normally involve
factoring the actions (earth and net water pressures) on the active side of the wall with
partial factors greater than unity (γG and γQ from Table A.3, set A1). However, such
factoring is incompatible with wall–ground interaction analyses as it introduces artificial
yielding in the ground springs, resulting in unrealistic stress redistributions along the
wall. Thus, the following procedure can be adopted:
(a) All ground parameters, net water pressures and other permanent actions enter the
calculations with values equal to their characteristic values. Surface variable loads
enter the calculations with a value q = qk(γQ /γG), where qk is the characteristic
value, in order to account for the difference between the partial factors on variable
and permanent actions. If the behaviour of the wall is taken as elasto-plastic, the
yield value of the wall bending moment entering the calculations (Mu) should be
taken as being equal to the design value of the wall ultimate (plastic) moment of
resistance divided by γG, i.e.

Mu = Mu, d /γG = Mu, k/(γMγG)

(b) Equilibrium calculations provide ‘unfactored’ values of the support reactions,
effects of actions (e.g. bending moments) and the mobilized earth resistance (Rmob)
on the passive side of the wall. The design values of these quantities are obtained by
multiplying the calculated values by the partial factor on unfavourable permanent
actions (γG from Table A.3, set A1, recommended value: 1.35). For example, the
design value of the mobilized earth resistance is

Rmob, d = γG Rmob

158

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 9. RETAINING STRUCTURES

(c) The stability of the base of the wall against rotational failure is checked by ensuring
that the calculated design value of the mobilized earth resistance (Rmob, d) does not
exceed the design value of the limiting earth resistance (Rp, d), i.e.
Rmob, d £ Rp, d

fi γG Rmob £ Rp, k/γR, e

fi Rmob £ Rp, k/γGγR, e

where the partial factor on earth resistance γR, e is obtained from Table A.13, set
R1 for DA-1 Combination 1 (recommended value: 1.0) and set R2 for DA-2
(recommended value: 1.4).
In DA-2, this method ensures that the actually mobilized earth resistance (Rmob)
does not exceed about 50% of the limiting earth resistance (Rp, k), since
1/γGγR, e = 1/(1.35 × 1.4) = 1/1.89 = 0.53
i.e. it ensures a minimum overall factor of safety (OFS) ≥ 1.89 against rotational
failure of the base of the wall. In DA-1 Combination 1, the available OFS is lower
(≥ 1.35), but Combination 2 is usually more critical with regard to safety against
rotational failure of the base of the wall (thus providing a larger OFS).
In addition to the limiting values of the earth pressures, spring models also require the
coefficient of earth pressure at rest (K0), as it affects the initial loading of the wall prior to any
deflection, and other parameters such as the ground spring stiffness, the bending stiffness
(EI) of the wall and the axial stiffness of the supports (struts or anchorages). Empirical
rules have been established to derive the ground spring stiffness from ground test results
(pressuremeter and cone penetrometer tests). It is pointed out that the information included
in Annex C2 is related to the movements required to mobilize the limiting earth pressures
over the total length of the wall and is not appropriate for estimating the ground stiffness in
spring models.
As ground displacements are crucial in ULS calculations with spring models, important
interaction parameters are the initial ground stresses, the stress–strain relationship of the
ground from the initial K0 condition up to the limiting pressures (active and passive), the
stiffness of the wall and its supports and, finally, the detailed construction sequence.
Although EN 1997-1 does not provide guidance on the design values of these parameters, it
is recommended that the design values of the ground spring stiffness, the K0 coefficient and
other non-strength-related model parameters are set equal to their characteristic values.
Usually, a safe design is obtained with a low characteristic value of the spring stiffness and a
high characteristic value of K0. However, as it is often difficult to obtain reliable values of the
spring stiffness and the K0 coefficient, it is recommended that the sensitivity of the design is
checked with regard to variations of these parameters. It is also recommended that, where
relevant, the design is checked with upper and lower values of the wall and support
stiffnesses.

Finite-element models
Finite-element models include wall–ground interaction by treating the ground as a continuum
in contact with the wall and by imposing displacement compatibility at the wall–ground
interface. Thus, finite-element models are appropriate in the design of any type of embedded
wall, but they are especially useful in cases where wall–ground interaction effects are
important, i.e. in those cases where spring models are also appropriate (see above).
The advantage of finite-element models compared with spring models is that finite-
element models avoid the interpretation of the ground stress–strain curve in the pressure–
displacement relationship (as required by the spring models), thus eliminating the significant
uncertainty involved in such interpretation.

159

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Finite-element models can be used with the three Eurocode Design Approaches as
follows:
(1) DA-1 Combination 2 and DA-3. The procedure described above to determine the wall
length when using the spring models for DA-1 Combination 1 and DA-2 with design
values of the ground strength parameters can also be used with finite-element models.
Alternatively, numerical calculations can be performed using the characteristic values of
the ground strength parameters. In this case, when ULS requirements need to be
checked (i.e. the adequacy of the available margin of safety against toe failure at each
excavation stage), the values of the ground strength parameters are gradually reduced
down to their design values (obtained via the partial factors γM from Table A.4, set
M2) while ensuring that model stability is not compromised. The method of strength
reduction is discussed in more detail in Chapter 11, on overall stability.
Both procedures can provide the design values of the effects of the actions on the wall
and the support reactions, but the second procedure is preferable when the ground
stress–strain relationship depends strongly on the loading history.
(2) DA-2 and DA-1 Combination 1. The procedure for DA-2 and DA-1 Combination 1
described for the corresponding spring models can be used (see above), i.e. all ground
parameters, net water pressures and other permanent actions enter the calculations with
values equal to their characteristic values.

Specific comments on the application of DA-1 in all calculation models

In DA-1, which requires two calculations, the minimum wall penetration required for toe
stability is usually governed by the Combination 2 analysis. It is therefore recommended that
initially, in the first calculation, the Combination 2 analysis is performed to determine the
required penetration length and then, in the second calculation, the Combination 1 analysis
is performed to check that this length is in fact larger than the minimum length required by
the Combination 1 analysis. A spring (or finite-element) model is more appropriate for this
second calculation (Combination 1 analysis), because the wall penetration calculated using
Combination 2 is most probably longer than the Combination 1 requirement, and thus the
mobilized earth pressure on the resisting side of the wall in the Combination 1 analysis will
be smaller than the limiting value. The wall moment of resistance in the second calculation
may be deduced from the maximum bending moment obtained in the first calculation. The
design of the wall section and the anchorage in DA-1 should be performed using the
larger of the maximum design bending moments and anchor forces obtained from the two
combinations using the same design length of the wall, i.e. the longer length obtained by the
two previous analyses.
Example 9.2 presents the methodology for the ULS design of an embedded wall supported
with a single row of anchorages, using the three Design Approaches. This example also
provides calculations for checking against base failure by hydraulic heave.

9.8. Serviceability limit state design

9.8.1. General
Clause 9.8.1(5) Under serviceability conditions, wall movements are usually not sufficiently large to mobilize
the limiting earth pressures. This is especially relevant where the supported soil is
overconsolidated or heavily compacted, the retaining structure and its foundation are fairly
rigid (e.g. gravity walls founded on piles), or the displacement tolerance of the structures
behind the wall is small. In these circumstances, the appropriate earth pressures are usually
larger than the limiting earth pressures on the ‘active’ side of the wall and lower than the
limiting earth pressures on the passive side of the wall. The determination of the appropriate
earth pressures for SLS design calculations is discussed in Section 9.5.
Clause 2.4.8(2) In SLS design calculations, as stated clause 2.4.8(2), the values of the partial factors should
normally be taken equal to unity so that the design values of all actions, resistances and

160

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 9. RETAINING STRUCTURES

ground parameters are equal to their characteristic values. Hence, the design values of earth
pressures for SLS calculations should normally be calculated using design values of all
ground parameters equal to their characteristic values.

9.8.2. Displacements
EN 1997-1 requires the following approach when performing SLS designs of retaining Clause 9.8.2
structures:
(1) Determination of limiting acceptable values for the displacements of the wall and any Clause 9.8.2(1)P
supported structures and services.
(2) Cautious estimate of the wall deflection and displacements and their effects on supported Clause 9.8.2(2)P
structures and services on the basis of comparable experience.
(3) More detailed investigation, including rigorous displacement calculations, should be Clause 9.8.2(3)P
undertaken in any of the following situations: Clause 9.8.2(5)P
(a) the initial cautious estimate exceeds the allowable limiting values
(b) comparable experience is not well established
(c) nearby structures and services are unusually sensitive to displacement.

Clause 9.8.2(6) recommends that displacement calculations are considered in the case of Clause 9.8.2(6)
walls retaining more than 6 m of cohesive soil of low plasticity or 3 m of high-plasticity soils
or in cases where the wall is supported by soft clay within its height or beneath its base.
Wall deflections and ground displacements under SLS conditions can be calculated using
numerical analyses (e.g. finite-element models) of the ground–structure system, including
the complete wall construction and, where relevant, the support installation sequence. The
main uncertainty in these analyses is the estimation of the appropriate ground stiffness, i.e. a
stiffness compatible with the level of strain, mode of deformation, ground anisotropy, etc.
If such analyses are considered impractical, deflections of embedded walls can also be
calculated by modelling the structure as a beam supported on elasto-plastic Winkler-type
springs, as discussed in ULS design (see Section 9.7). One of the limitations of these models
is that displacements in the supported ground (and thus in influenced structures and
utilities) are not calculated but need to be assessed indirectly.
Overdig is usually not introduced in SLS design. The characteristic resistance and ground
parameter values should be selected as cautious estimates of the values governing the
serviceability state considered. According to clause 4.2(8) of EN 1990 (Basis of Structural
Design), characteristic values of stiffness parameters are equal to their mean values. This is
accepted for the structural stiffness but is over-ridden for the ground stiffness by EN 1997-1, Clause 2.4.5.2(2)P
which requires that the characteristic value of the ground stiffness is selected as a cautious
estimate of the mean value (and not the mean value). Finally, it is pointed out that
since ground stiffness depends on strain, the value of ground stiffness used in SLS design
calculations may be different from the corresponding ‘elastic’ value used in ULS design
calculations performed via finite elements (where strains are usually larger than those in SLS
designs).
Example 9.2 presents the methodology for the SLS design of an embedded wall supported
by a single row of anchorages.

Example 9.1: ULS design of a stem (gravity) wall

This example presents the ULS design of a concrete stem (gravity) wall according to
the three Design Approaches of EN 1997-1, and compares the results with those of
the conventional OFS method. Figure 9.3 shows the wall supporting a cohesionless
medium-dense to dense gravelly sand sloping at an angle β = 20°. The ground beneath
the wall base and the ground in front of the wall base providing earth resistance have the
same properties as the retained ground. The underside of the wall base is considered

161

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

rough (concrete cast against the ground), while the vertical face of the wall base on the
passive side is considered to be smooth (concrete cast against a smooth formwork). For
simplicity, hydraulic effects are irrelevant by assuming that the water table is well below
the zone of influence of the wall.
The following parameters are used in the design:
• Geometrical data (see Fig. 9.3):
– Wall stem height: h = 6 m.
– Toe embedment depth: h1 = 0.80 m. This includes some accidental overdig in
accordance with clause 9.3.2.2. An overdig less than 10% of h may be justified in
this case, since the wall relies only slightly on the earth resistance for stability and
the ground level on the passive side is usually well controlled.
– Height of base: h2 = 0.80 m.
– Width of stem: top, b1 = 0.50 m; bottom, b3 = 0.70 m.
– Width of toe: b2 = 0.95 m.
• Characteristic values of material parameters and actions:
– Characteristic angle of shearing resistance of the backfill and the ground
(beneath the wall base and on the passive side): ϕk¢ = 32°. For simplicity, it is
assumed that the peak strength angle is equal to the critical state angle.
– Weight density of the backfill and ground: γk = 20 kN/m3.
– Wall–ground interface parameter (friction angle) on the underside of the
wall base (assumed rough; concrete cast on the ground; see clause 6.5.3(10):
δk = ϕk¢ = 32°.
– Wall–ground interface parameter (friction angle) on the passive side (assumed
smooth; concrete cast against a smooth side-wall formwork): δpk = 20°.
– Weight density of the concrete: γbk = 24 kN/m3.
– Surcharge at ground surface (variable action): qk = 10 kPa.

qk = 10 kPa
b1 = 0.5 m

b = 20˚

jk = 32˚
gk = 20 kN/m3
'Virtual back'
h = 6.00 m

b2 b3 b
eB
h1 = h2 = 0.80 m B/2
O
B
0.95 0.7 b

Fig. 9.3. Geometrical data for a concrete stem (gravity) wall

162

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 9. RETAINING STRUCTURES

q q

b = 20° b = 20°

Gs2

Pa, q + Pa, g Pq + Pg

b b

Gb1 Gb1
Gs1
Base of stem

dp
Gb2
Pp Rh
RV

(a) (b)

Fig. 9.4. Actions and resistances in (a) GEO and (b) STR ULSs

The objective of the design is to determine the required minimum width, b, of the heel
for stability and calculate the bending moments and shear forces at critical sections of the
wall. The minimum width of the back heel is determined by checking the GEO ULSs in
the ground, namely exceedence of the sliding resistance on the wall base and bearing
failure of the foundation. The structural design of the wall (calculation of bending
moments and shear forces in critical sections) is performed by checking STR ULSs at the
bottom of the stem only; a similar procedure can be used for the design of other critical
sections (see Fig. 9.4).
For conciseness, an SLS design check has not been carried out in this example.
Serviceability requirements relate to the displacements of the wall and the retained
ground and to the performance of the concrete sections (especially with regard to
cracking). The SLS design procedure is described in Section 9.8.2. It involves first making
a cautious estimate of the wall deflection and the resulting ground displacements on the
basis of comparable experience, and, if the limiting values are exceeded, performing a
more detailed analysis using rigorous calculations (e.g. finite elements). In the present
example, the ground behind and beneath the wall is relatively dense, and there are no
important structures behind the wall. It thus appears that, on the basis of comparable
experience, the anticipated wall and ground displacements are acceptable.
In GEO ULSs, failure occurs in the ground, and thus the wall movement is sufficiently
large to mobilize the active earth pressure (Pa) along the ‘virtual back’ of the wall (see Fig.
9.4). This pressure is inclined at an angle equal to the slope of the ground surface
(β = 20°), regardless of the angle of shearing resistance of the ground (see Section 9.7).
Note that the use of the active earth pressure on the virtual back of the wall implies
that the width of the heel is large enough to permit the development of a conjugate
Coulomb-type ‘failure surface’ (inclined at 45 + ϕk¢/2 to the horizontal) within the soil
mass above the heel, i.e. the length of the heel should not be much smaller than
bmin = h/tan(45 + ϕk¢/2) = 3.32 m
In GEO ULSs, the wall movement is also sufficient to mobilize the limiting earth
resistance, i.e. the limiting passive earth pressure force (Pp) in front of the wall base.
Usually, these ULSs are checked by ensuring that the design value of the effect of the
action does not exceed the design value of the corresponding resistance, i.e. Ed £ Rd. The
actions and resistances acting on the wall are shown in Fig. 9.4a. Note that, in the
following equations, the subscript ‘d’ (indicating design values) has been dropped for
simplicity.

163

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

To check against base sliding: Eh £ Rh, where

Eh = Pah = (Pa, q + Pa, g)cos β
is the horizontal action on virtual back of the wall (= Rh), and
Rh = Pph = (Gb + Gs + Pa, v – Pp, v)tan β
is the horizontal resistance of wall base to sliding. Note that, in this design calculation, the
limiting passive earth pressure force (Pph) in front of the wall is treated as an earth
resistance (see the discussion in Section 9.3.1 and clause 6.5.3(5)).
To check against bearing resistance failure of the foundation: Ev £ Rv, where
Ev = Gb + Gs + Pav – Ppv
is the vertical action on the foundation,
Eh = Pah – Pph
is the horizontal action on the foundation, and Rv is the ultimate bearing resistance of the
foundation for inclined and eccentric loads (calculated using the method described in
Annex D). Note that, in this design calculation, the limiting passive earth pressure force
(Pp) in front of the wall is treated as a favourable action (see the discussion in Section 9.3.1
and clause 6.5.2.1(3)P).
Here, Pah and Pav are, respectively, the horizontal and vertical components of the active
earth pressure force acting on the ‘virtual back’ of the wall (height H = h2 + h + b tan β):
Pah = Pah, q + Pah, g Pah, q = KahqH Pah, g = 12 KahγH 2
Pav = Pav, q + Pav, g Pav, q = Pah, q tan β Pav, g = Pah, g tan β
Note: the surface surcharge (q), as a variable action, is only included on the surface of the
slope beyond the ‘virtual back’ of the wall where it is unfavourable, i.e. not above the heel,
where it is favourable.
Pph and Ppv are, respectively, the horizontal and vertical components of the limiting
passive earth pressure force acting on the front of the wall base (height h1):
Pph = 12 KphγH12 Ppv = Pph tan δp
In this example, it is assumed that the limiting passive earth pressure (inclined at an angle
δp) develops in the GEO ULSs because it is considered that the wall can move sufficiently
to mobilize the limiting passive earth pressure. It is pointed out that often conservative
assumptions are made for the magnitude of the earth pressure on the passive side of
gravity walls (to account for limited mobility of the wall, e.g. see clause 6.5.3(5)) like
δp = 0, or Kp = 1, or even neglecting the passive earth pressure completely.
Gb is the weight of the concrete wall:
Gb = Gb1 + Gb2 Gb2 = 12 (b1 + b3)hγb Gb2 = (b2 + b3 + b)h2γb
Gs is the weight of the backfill above the heel of the wall (between the wall and the
‘virtual back’):
Gs = Gs1 + Gs2 Gs1 = bhγ Gs2 = 12 b2γ tan β
When designing against bearing resistance failure of the foundation, the eccentricity
(eB) of the vertical action (Ev) is calculated as follows (see Figs 9.3 and 9.4):
(1) Calculation of the overturning moment (Mo) of actions with respect to the front of the
wall base (point O), positive if anticlockwise (see Fig. 9.3):
Mo = Â Fh Yo - Â Fv X o
using the following actions (F) and corresponding lever arms:

164

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 9. RETAINING STRUCTURES

– Action Pah, q with lever arm Y = H/2, action Pah, g with lever arm Y = H/3.
– Action Pav, q and Pav, g with lever arm X = b2 + b3 + b.
– Action Gb1 with lever arm X = b2 + b3 /2, action Gb2 with lever arm
X = (b2 + b3 + b)/2, assuming that the stem of the wall is inclined symmetrically
at the front and back. The slight inclination of the back of the wall (0.95° to the
vertical) is ignored in calculating the earth pressure in the STR-type ULS design.
– Action Gs1 with lever arm X = b2 + b3 + b/2.
– Action Gs2 with lever arm X = b2 + b3 + 23 b.
– Action (–Pph) with lever arm Y = h1/3.
(2) Calculation of the eccentricity (eB) of the vertical action (Ev):
ÊB M ˆ
eb = - Á + o ˜
Ë 2 Ev ¯
where B = b2 + b3 + b is the total width of the wall base. The eccentricity is positive if
the vertical action is applied to the right of the centre of the wall base (see Fig. 9.3).

In STR ULSs, failure occurs in a structural element (e.g. the base of the wall stem) by
the development of a plastic hinge at this location. The structural design of the wall
section at the base of the stem involves calculating the design bending moment and shear
force at this location due to the horizontal component of the appropriate force (Ph)
resulting from the limiting active earth pressure acting over the height h of the wall stem
(see Fig. 9.4b):
Ph = Ph, q + Ph, g Ph, q = Khqh Ph, g = 12 KhγH2
The surface surcharge (q) is considered to act everywhere on the ground surface, since it is
an unfavourable (geotechnical) action with respect to the wall stem.
In STR ULSs, the appropriate earth pressure coefficient (Kh) should be compatible
with the wall rotation when a plastic hinge develops at the base of the stem. As the
rotation of the wall stem under this condition may not be sufficient to mobilize fully the
active earth pressure in the backfill, two alternative values of the earth pressure are
considered in this example (other assumptions may also be appropriate, depending on the
kinematic constraints of the wall):

(1) The active earth pressure, which is relevant if the wall movement is sufficiently large
to cause a ULS condition in the retained ground when a ULS condition occurs in the
structure (base of stem). Then, Kh = Kah.
(2) An intermediate earth pressure, corresponding to the earth pressure coefficient
Kh = 0.5(Kah + K0), i.e. the average of the horizontal earth pressure at-rest coefficient
(K0) and the horizontal active earth pressure coefficient (Kah). An earth pressure of
about this magnitude may be relevant (as an empirical rule) if the wall and the
foundation are stiff and a ULS condition develops in the concrete section prior to the
development of a ULS in the retained ground. (see clause 9.5.4).

The design value of the ground angle of shearing resistance (ϕd¢) is calculated using the
equation
Ê 1 ˆ
ϕd¢ = arctan Á tan ϕ ¢k ˜
Ë γϕ ¢ ¯
where ϕk¢ is the characteristic angle of shearing resistance of the ground, and the partial
factor γϕ ¢ is obtained from Table A.4.
In DA-1 Combination 1 and DA-2, γϕ ¢ = 1 (Table A.4), and thus ϕd¢ = ϕk¢ = 32°, and, on
the front of the wall, δd = δk = 20°, while, on the virtual back of the wall, δd = β = 20°. The
design (and characteristic) value of the active earth pressure coefficient (Kah) for the

165

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

backfill is determined from Fig. C.1.3 for ϕd¢ = 32°, β = 20° and δd /ϕd¢ = β/ϕd¢ = 0.625.
Hence, Kah = 0.35. Note that the inclination of the earth pressure on the virtual back of
the wall is assumed to be equal to the inclination of the ground surface, regardless of the
ground parameters (see Section 9.7). The design value of the at-rest horizontal earth
pressure coefficient of the backfill is calculated from equation (9.2) with OCR = 1:
K0 = (1– sin ϕd¢)(1 + sin β) = 0.631
The design (and characteristic) value of the limiting passive earth pressure coefficient is
obtained from Fig. C.2.1 for ϕd¢ = 32°, β = 0° and δd /ϕd¢ = 20/32 = 0.625. Hence, Kph = 5.5.
In DA-1 Combination 2 and DA-3, γϕ ¢ = 1.25 (Table A.4), and thus ϕd¢ = 26.6°, and, on
the front of the wall, δd = tan–1(tan ϕk¢/1.25) = 16.2°, while, on the virtual back of the wall,
δd = β = 20°. The design value of the active earth pressure coefficient (Kah) of the backfill
is obtained from Fig. C.1.3 of Annex C for ϕd¢ = 26.6°, β = 20° and δd /ϕd¢ = β/ϕd¢ = 0.75.
Hence, Kah = 0.48. The design value of the at-rest horizontal earth pressure coefficient of
the backfill is calculated from equation (9.2) with OCR = 1:
K0 = (1 – sin ϕ¢)(1 + sin β) = 0.741
The design value of the limiting passive earth pressure coefficient is obtained from Fig.
C.2.1 of Annex C for ϕd¢ = 26.6°, β = 0° and δd /ϕd¢ = 16.2/26.6 = 0.62. Hence, Kph = 3.9.
The sequence of the calculations and the main results for the three Design Approaches
and for the conventional OFS method are listed in the following tables. The appropriate
partial factors (taken from Annex A) are also included in the tables for ease of reference.
Note that in DA-1 the width of the footing is determined from the more critical of the two
combinations (in this example, Combination 2). In the OFS method, the ‘design values’ of
the various quantities are identical to their ‘characteristic values’, since this method
does not include partial factors; the ‘characteristic values’ are calculated using the
characteristic ground parameters of the Eurocode Design Approaches.
The characteristic and design values of the geotechnical (Pa, g, Pa, q) and non-geotechnical
(Gb, Gs) actions are listed in Table 9.2, together with the design values of the effects of the
actions (net vertical and horizontal actions on the foundation) for GEO ULS calculations.
The application of the partial factors for calculating the design values of the actions is
described in Section 9.7. For example, the design value of the unfavourable variable
geotechnical action Pah, q is calculated using the equation
Fd = γF F(Xk /γM)
i.e.
Pah, q = Kah(γF q)H
with partial factors given in Table 9.2, as follows:
• In DA-1 Combination 1:
Pah, q = 0.35 × (1.5 × 10) × 7.70 = 40.43 kN/m
• In DA-1 Combination 2 and DA-3:
Pah, q = 0.48 × (1.3 × 10) × 8.21 = 51.23 kN/m
• In DA-2:
Pah, q = 0.35 × (1.5 × 10) × 7.84 = 41.16 kN/m
The different values of the height H (see Fig. 9.3) correspond to the different values of
the heel width b in each Design Approach. Note that the earth pressure force on the
passive side of the wall (Pp) is considered as a favourable geotechnical action when
designing against bearing resistance failure, and as an earth resistance when designing
against base sliding (see the discussion in Section 9.3.1).

166

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 9. RETAINING STRUCTURES

Table 9.2. Example 9.1 – parameters for design of concrete stem (gravity) wall

DA-1

Comb. 1 Comb. 2 DA-2 DA-2* DA-3 OFSa

Required width of heel of wall base, 2.47 3.87 2.86 2.17 3.87 2.79
b (m)
Total width of wall base, B (m) 4.12 5.52 4.51 3.82 5.52 4.44
Height of virtual wall, H (m) 7.70 8.21 7.84 7.59 8.21 7.82
I. GEO ULSs
Characteristic value of soil weight, 318.61 518.91 372.97 277.54 518.91 363.13
Gs1 + Gs2 (kN/m)
Characteristic value of concrete weight, 165.50 192.38 172.99 159.74 192.38 171.65
Gb1 + Gb2 (kN/m)
Characteristic value of Pah, q (kN/m) 26.95 – 27.44 26.56 – 27.35
Characteristic value of Pah, g (kN/m) 207.46 – 215.18 201.62 – 213.79
Characteristic value of Pav, q (kN/m) 9.81 – 9.99 9.67 – 9.96
Characteristic value of Pav, g (kN/m) 75.51 – 78.32 73.38 – 77.81
Characteristic value of Pph (kN/m) 35.20 – 35.20 35.20 – 35.20
Characteristic value of Ppv (kN/m) 12.81 – 12.81 12.81 – 12.81
Partial factor on soil shearing resistance 1.0 1.25 1.0 1.0 1.25 1.0
Design value of angle of shearing
resistance, ϕ (°)
Backfill 32 26.6 32 32 26.6 32
Passive side of footing 32 26.6 32 32 26.6 32
Below footing 32 26.6 32 32 26.6 32
Design value of wall–ground interface
parameter, d (°)
Backfill (equal to surface slope d = b) 20 20 20 20 20 20
Passive side of footing, δp 20 16.23 20 20 16.23 20
Below footing (rough footing), d = ϕ 32 26.6 32 32 26.6 32
Design value of horizontal active earth 0.35 0.48 0.35 0.35 0.48 0.35
pressure coefficient, Kah
Design value of horizontal at-rest earth 0.631 0.741 0.631 0.631 0.741 0.631
pressure coefficient, K0
Design value of intermediate earth 0.49 0.61 0.49 0.49 0.61 0.49
pressure coefficient, Kh
Design value of horizontal passive earth 5.50 3.90 5.50 5.50 3.90 5.50
pressure coefficient, Kph
Partial factor on vertical actions, Gs + Gb
Favourable permanent for base sliding 1.0 1.0 1.0 1.0 1.0 1.0
Favourable permanent for bearing 1.0 1.0 1.0 1.0 1.0 1.0
resistance failure
Partial factor on geotechnical actions Pav, q 1.5 1.3 1.5 1.5 1.3 1.0
and Pah.q: unfavourable variable
Partial factor on geotechnical actions Pav, g 1.35 1.0 1.35 1.35 1.0 1.0
and Pah, g: unfavourable permanent
Partial factor on geotechnical actions Ppv 1.0 1.0 1.0 1.0 1.0 1.0
and Pph (passive earth pressure is a
favourable permanent action)
Design value of horizontal action, Pah 320.49 374.65 331.66 312.03 374.65 241.14
(kN/m)
Design value of vertical passive earth 12.81 7.27 12.81 12.81 7.27 12.80
pressure force, Ppv (kN/m)

167

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Table 9.2. Contd

DA-1

Comb. 1 Comb. 2 DA-2 DA-2* DA-3 OFSa

Design value of horizontal passive earth 35.20 24.96 35.20 35.20 24.96 35.20
pressure force, Pph (kN/m)
Design value of resultant vertical action
For base sliding, Gs + Gb + Pav (kN/m) 600.76 847.66 666.68 520.33 847.66 622.55
For bearing resistance failure, 587.95 840.39 653.87 538.04 840.39 609.74
Gs + Gb + Pav – Ppv (kN/m)
Design value of resultant horizontal
action
For base sliding, Pah (kN/m) 320.49 374.65 331.66 312.03 374.65 241.10
For bearing resistance failure, Pah – Pph 285.29 349.69 296.46 276.83 349.69 205.94
(kN/m)
a
In the OFS method, all values are ‘nominal’ (characteristic) values

The design values of the effects of the actions are calculated from the design
values of the relevant actions using the following equations (the index ‘d’ is omitted for
conciseness):

• For base sliding (net horizontal action):

Eh = Pah = (Pa, q + Pa, g)cos β = Pah, q + Pah, g
• For bearing failure (resultant vertical action):
Ev = Gb + Gs + Pav – Ppv
Finally, note that the calculated width, b of the heel of the wall, in DA-1 and DA-3 is
greater than the minimum value, and in DA-2 is not much smaller than the minimum
value,
bmin = h/tan(45 + ϕk¢/2) = 3.32 m
which permits the development of a conjugate Coulomb-type ‘failure surface’ (inclined at
45 + ϕk¢/2 to the horizontal) within the soil mass above the heel. Thus these calculated
widths validate the assumption of the active earth pressure acting on the ‘virtual back’ of
the wall.
Note that in this example, the weight of the wall and soil (Gb + Gs) is a favourable
action when designing against bearing resistance failure, because its favourable effect in
reducing the load eccentricity on the foundation (favourable effect) is more significant
with regard to the bearing resistance than its unfavourable effect in increasing the
vertical load. In different wall configurations the significance of these effects may be
reversed. Thus, it is recommended that two calculations are performed, one with partial
factors for ‘favourable’ and one for ‘unfavourable’ actions on the weight of the wall and
the soil.
The design values of the resistances required in the ULS analyses of the wall against
base sliding and bearing resistance failure (GEO ULSs) are presented in Table 9.3. The
horizontal resistance against base sliding is calculated using the equation
Rh = Pph + (Gb + Gs +Pav – Ppv)tan δ¢

168

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 9. RETAINING STRUCTURES

Table 9.3. Example 9.1 – results of the GEO design for the concrete stem (gravity) wall

DA-1

Comb. 1 Comb. 2 DA-2 DA-2* DA-3 OFSa

Width of heel of wall base, b (m) 2.47 3.87 2.86 2.17 3.87 2.79
Total width of wall base, B (m) 4.12 5.52 4.51 3.82 5.52 4.44
A. Design against sliding
(GEO ULS)
Character value of horizontal – – – 352.3 – 416.2
sliding resistance Rh, k (kN/m)
Partial factor on sliding resistance 1.0 1.0 1.1 1.1 1.0 1.0
(on Rv tan δ)
Partial factor on earth resistance 1.0 1.0 1.4 1.4 1.0 1.0
(on Pph)
Design value of horizontal sliding 402.6 445.1 396.6 313.4 445.1 416.2
resistance Rh, d (kN/m)
Design value of net horizontal 320.5 374.6 331.7 312.0 374.6 241.1
effect of the actions Eh, d (kN/m)
Over-design factor for Design 1.26 1.19 1.20 1.00 1.19 1.73
Approaches or OFS, Rh, d /Eh, d
Minimum required value of 1.0 1.0 1.0 1.0 1.0 1.5
Rh, d /Eh, d for stability
Result for design: CR, critical; NCR NCR NCR CR NCR NCR
NCR, not critical
B. Design against bearing
resistance failure (GEO ULS)
Design value of overturning –840.8 –1986.1 –1125.3 –755.0 –1986.1 –1177.3
moment with respect to O,
Mdo (kN m/m) (positive if
anticlockwise – see Fig. 9.3)
Design value of total vertical 587.9 840.4 653.9 538.0 840.4 609.7
effect of the actions, Ev, d
(kN/m)
Design value of total horizontal 285.3 349.7 296.5 193.0 349.7 205.9
effect of the actions, Eh, d
(kN/m)
Design value of the overturning 370.4 333.4 349.2 214.3 333.4 176.3
moment with respect to the
centre of the wall base, Md
(kN m/m) (positive if
anticlockwise – see
Fig. 9.3)
Design eccentricity of Ev, with –0.63 –0.40 –0.53 –0.42 –0.40 –0.29
respect to the centre of the
base, eB (m) (positive if action
is to the right of centre – see
Fig. 9.3)
Effective foundation width: 2.86 4.73 3.44 2.98 4.73 3.86
B¢ = B – 2 × abs(eB) (m)
Partial factor on embedment 1.0 1.0 1.0 1.0 1.0 1.0
pressure (p): favourable
permanent action
Design value of embedment 16.00 16.00 16.00 16.00 16.00 16.00
pressure, p = γ h1 (kPa)

169

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Table 9.3. Continued

DA-1

Comb. 1 Comb. 2 DA-2 DA-2* DA-3 OFSa

Bearing resistance factors

Nq 23.18 12.59 23.18 23.18 12.59 23.18
Nγ 27.72 11.59 27.72 27.72 11.59 27.72
iq 0.265 0.341 0.299 0.384 0.341 0.439
iγ 0.136 0.199 0.163 0.238 0.199 0.290
Characteristic value of bearing – – – 1007.8 – 1828.5
resistance, Rvk (kN/m)
Partial factor on bearing 1.0 1.0 1.4 1.4 1.0 1.0
resistance, γR
Design value of bearing 590.3 839.8 655.4 719.9 839.8 1828.5
resistance, Rvd (kN/m)
Design value of total vertical 588.0 840.0 653.9 538.0 840.0 609.7
action, Ev, d (kN/m)
Over-design factor for Design 1.00 1.00 1.00 1.34 1.00 3.00
Approaches or OFS, Rv, d/Ev, d
Minimum required value of 1.00 1.00 1.00 1.00 1.00 3.00
Rv, d/Ev, d for stability
Result for design: CR, critical; NCR CR CR NCR CR CR
NCR, not critical
a
In the OFS method, all values are ‘nominal’ (characteristic) values

The vertical component of the bearing resistance (Rv) is calculated using the procedure
recommended in Annex D for surface foundations. In general, design values of resistances
are calculated using the equation
Rd = R[γF F(Xk /γM)]/γR
as follows (see example below):
(1) In DA-1 Combination 1, using design values of the actions, and design values of the
soil parameters (ϕ¢, δ¢) equal to the characteristic values (since γM = 1) and γR = 1.
(2) In DA-1 Combination 2 and DA-3, using design values of soil parameters and actions
and γR = 1.
(3) In DA-2, using design values of the soil parameters (ϕ¢, δ¢) equal to the characteristic
values (since γM = 1), and design values of actions and appropriate partial factors for
resistances (γR = 1.1 and 1.4 for sliding and γR = 1.4 for bearing resistance). An
alternative calculation (DA-2*) is also performed, which is similar to DA-2 but uses
the characteristic (and not the design) values of the actions in the calculation of
design resistances.
The characteristic value of the resistance is not required in any of the Eurocode Design
Approaches (except in the variation DA-2*). It is only calculated in the OFS method. For
example, the design value of the horizontal resistance against base sliding (Rh, d) is
calculated as follows:
• In DA-1 Combination 1:
Rh, d = 35.20/1.0 + [(484.11 + 116.65–12.81)tan 32]/1.0 = 402.6 kN/m
• In DA-1 Combination 2 and DA-3:
Rh, d = 24.96/1.0 + [(711.3 + 136.36–7.27)tan 26.6]/1.0 = 445.1 kN/m

170

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 9. RETAINING STRUCTURES

• In DA-2:
Rh, d = 35.20/1.4 + [(545.96 + 120.71–12.8)tan 32]/1.1 = 396.6 kN/m
• In DA-2*, using the width of heel of the wall, which is critical in DA-2 (b = 2.86 m),
for comparison:
Rh, d = Pph, d + [(Gb + Gs +Pav – Ppv)tan δ¢]d
fi Rh, d = Pph, k /γR, e + [(Gb, k + Gs, k +Pav, k – Ppv, k)tan δ¢k]/γR, h
fi Rh, d = 35.20/1.4 + [(172.99 + 372.97 + 88.31–12.81)tan 32]/1.1 = 378.2 kN/m
The corresponding effect of the actions is
Ed = Pah, d = Pagh, d + Paqh, d = γG Pagh, d + γQ Paqh, d
fi Ed = 1.35 × 215.18 + 1.5 × 27.44 = 331.66
The wall is more than safe against base sliding since
Ed = 331.66 £ 378.2 = Rh, d
The over-design factor is
Rh, d /Ed = 378.2/331.66 = 1.14 ≥ 1
• In the OFS method:
Rh, d = 35.20 + (534.78 + 87.77–12.8) tan 32 = 416.2 kN/m
Note that DA-2 and DA-2* give slightly different results for Rh, d (396.6 and
378.2 kN/m, respectively) when the same value of the width of the heel of the wall b is
used. The difference is due to the unfavourable action Pav, which is factored by γF >1 in
DA-2 (Pav = 120.71 kN/m) and factored by γF = 1 in DA-2* (Pav = 88.31 kN/m). All
the other actions involved in the calculation of the sliding resistance are favourable
permanent actions, and thus are factored by γF = 1 in both variations of DA-2.
The cells in Table 9.3 denoted CR (critical) or NCR (not critical) indicate the
calculation which is critical in determining the width of the wall base. In this example,
bearing resistance failure is found to be critical in all the Design Approaches. In DA-1,
Combination 2 controls the width of the wall base.
Comparison of DA-2 with the alternative approach DA-2* indicates that DA-2* is less
conservative, since the required foundation width for DA-2* (3.82 m) is smaller than that
required for DA-2 (4.51 m); in fact, DA-2* is the least conservative of all the Design
Approaches, including the OFS method with minimum required overall factors of safety
of 1.5 against sliding and 3.0 against bearing failure.
The design values of the bending moment and the shear force at the bottom of the stem
of the wall for structural (STR) ULSs are presented in Table 9.4, together with the
coefficients of earth pressure, partial action factors and earth pressure forces used
to calculate these. The calculations are performed for two alternative earth pressure
magnitudes: the active value and an earth pressure between the K0 and the active values,
to illustrate the effect of the relative flexibility of the wall with respect to the retained
ground, i.e. the occurrence of the structural ULS prior to, or simultaneously with, the
ULS in the retained ground (see the discussion on intermediate earth pressures for ULS
design calculations in Section 9.5).
Bending moments and shear forces are the effects of actions and thus their design
values are calculated using the design values of the relevant actions (the earth pressure, in
the case of the effects of actions at the bottom of the stem of the wall), i.e. using the
equation Ed = E(Fd). In DA-1, the critical wall base width (obtained from Combination 2)
is used.

171

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Table 9.4. Example 9.1 – Results of STR design of concrete stem (gravity) wall

DA-1

Comb. 1 Comb. 2 DA-2 DA-2* DA-3 OFSa

Width of heel of wall base, b (m) 3.87 2.86 2.17 3.87 2.79
Total width of footing, B (m) 5.52 4.51 3.82 5.52 4.44
II. STR ULS
Structural design of wall (at bottom
of stem)
(a) Design using limiting active earth
pressure
Characteristic value of Kah coefficient 0.35 – 0.35 0.35 – 0.35
Design value of Kah coefficient 0.35 0.48 0.35 0.35 0.48 0.35
Partial factor on action Pah, q 1.5 1.3 1.5 1.5 1.3 1.0
Partial factor on action Pah, g 1.35 1.0 1.35 1.35 1.0 1.0
Design value of Pah, q (kN/m) 31.5 37.4 31.5 31.5 37.4 21.0
Design value of Pah, g (kN/m) 170.1 172.8 170.1 170.1 172.8 126.0
Design value of bending moment, 434.7 457.9 434.7 434.7 457.9 315.0
Md (kN m/m)
Design value of shear force, Vd (kN/m) 201.6 210.2 201.6 201.6 210.2 147.0
(b) Design using intermediate earth
pressure: K = 0.5(Ka + K0)
Characteristic value of Kh coefficient 0.49 – 0.49 0.49 – 0.49
Design value of Kh coefficient 0.49 0.61 0.49 0.49 0.61 0.49
Partial factor on action Ph, q 1.5 1.3 1.5 1.5 1.3 1. 0
Partial factor on action Ph, g 1.35 1.0 1.35 1.35 1.0 1.0
Design value of Ph, q (kN/m) 44.1 47.7 44.1 44.1 47.7 29.4
Design value of Ph, g (kN/m) 238.3 220.0 238.3 238.3 220.0 176.6
Design value of bending moment, 609.1 582.9 609.1 609.1 582.9 441.4
Md (kN m/m)
Design value of shear force, Vd (kN/m) 282.4 267.6 282.4 282.4 267.6 206.0
a
In the OFS method, all values are ‘nominal’ (characteristic) values

The bending moments and shear forces of the Eurocode Design Approaches are
factored, while those of the conventional OFS method are unfactored, and thus are not
directly comparable with the Eurocode 7 results. Specifically, the Eurocode 7 results are
ULS design values of the effects of actions for use directly in the structural design of the
concrete section (according to Eurocode 2). The bending moments calculated according
to the OFS method may be used in the structural design of the concrete section (according
to Eurocode 2) after application of the appropriate partial factors for actions. For
example, the bending moment may be multiplied by a factor of 1.4 as an approximate
equivalent partial factor weighting, corresponding to the Eurocode partial factors of 1.35
and 1.5 for permanent and variable actions, respectively, in structural designs for STR
ULSs.
The following conclusions can be drawn from the results of the above analyses:
(1) For GEO ULSs, DA-1 and DA-3 give identical results since:
(a) In DA-1 Combination 2 is critical in the design; this combination uses the same
material partial factors as DA-3.
(b) All actions in the system are geotechnical (there are no structural actions);
geotechnical actions have the same partial factors in DA-1 Combination 2 and
in DA-3.

172

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 9. RETAINING STRUCTURES

(2) For GEO ULSs, DA-2 and DA-2* are less conservative than DA-1 and DA-3, since
the required width of the wall base (B = 4.51 m in DA-2 and B = 3.82 m in DA-2*) is
18–31% smaller than that required by DA-1 and DA-3 (B = 5.52 m).
(3) For the structural design of the stem (STR ULSs), DA-2 and DA-2* are slightly less
conservative by about 5% (Md = 434.7 kN m/m) compared with DA-1 and DA-3
(which give identical results: Md = 457.9 kN m/m).
(4) For STR-type limit states, the design values of the effects of actions (bending moment
and shear force) at the bottom of the stem of the wall depend strongly on the assumed
magnitude of the earth pressure. For example, in DA-1 and DA-3, the effects of
actions calculated with the assumed intermediate earth pressure are 27% larger than
those calculated using the active earth pressure (582.9 versus 457.9 kN m/m). This
conclusion illustrates the importance of ensuring that the structural design of the wall
allows some lateral movement.
(5) In DA-1, the Combination 2 set of partial factors is critical in the geotechnical design
(GEO ULS). In the structural design (STR ULS) of the stem, Combination 2 is
critical if active earth pressures are used, while Combination 1 is critical if intermediate
earth pressures are used.
(6) The OFS method is the least conservative in terms of the required width of the wall
base (B = 4.44 m compared with B = 5.52 m for DA-1 and DA-3, respectively, and
B = 4.51 m for DA-2) with the exception of DA-2* (where B = 3.82 m). A possible
explanation is that the OFS method makes no allowance for uncertainty in the active
force on the virtual back of the wall. The wisdom of this is, however, questionable.
(7) In the structural design of the wall stem, the OFS method gives roughly the same
results as the Eurocode Design Approaches since, using the multiplication
factor of 1.4 (as mentioned above), the comparable ‘design’ bending moment of the
conventional method is 315.0 × 1.4 = 441.0 kN m/m, i.e. about the same as the
design values of all the Eurocode Design Approaches (457.9–434.7 kN m/m).

It should be pointed out that the above conclusions are applicable in the case of this
particular example only. In other examples, involving different wall geometries and
ground parameters, the corresponding conclusions may be different.

Example 9.2: ULS and SLS design of an embedded sheet pile wall
This example presents the ULS and SLS design of an embedded sheet pile wall. The wall,
shown in Fig. 9.5, has a nominal excavation depth of 5.0 m; an additional excavation depth
to allow for accidental overdig (see Section 9.3.2) equal to 10% × 4.0 m = 0.40 m is
added, giving a design value of the excavation depth H = 5.40 m. The wall is supported by
one row of anchorages at an elevation of –1.0 m (anchorage inclination β = 10°). For
excavations in stiff clays, drained conditions are usually critical for stability; thus, effective
stress analyses are performed using effective stress ground parameters and steady state
hydraulic conditions.
The ground profile consists of two layers (with the interface at an elevation of –4.0 m)
having the following parameters for use in the ULS design (all values are characteristic):

• Layer A (medium-dense to dense gravelly sand):

– Weight density above the water table: γk = 18 kN/m3.
– Saturated weight density below the water table: γsk = 20 kN/m3.
– Angle of shearing resistance: ϕk¢ = 35°.
– Maximum value of wall–ground interface parameter used on active side
(clause 9.5.1(6)): δk = 23 ϕk¢ = 23°.
• Layer B (stiff clay):
– Saturated weight density below the water table: γsk = 20 kN/m3.
– Angle of shearing resistance: ϕk¢ = 24°.

173

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

qk = 10 kPa
0.0

WT = –1.0 m
–1.0 m

Layer A

H = 5.4 m
–4.0 m

–5.0 m
WT = –5.4 m
0.4 m

Layer B Layer B

Fig. 9.5. Geometrical data for an embedded wall

– Cohesion intercept: ck¢ = 5 kPa.

– Maximum value of wall–ground interface parameter (indicative values for
long-term conditions used; other values may also be appropriate – see clause
9.5.1(8)):
Active side: δak = 23 ϕk¢ = 16°.
Passive side: δpk = 12 ϕk¢ = 12°.

The water table in the sand layer is assumed to remain at an elevation of –1.0 m, where
the ground level behind the wall is at an elevation of 0.0 m. Due to the excavation, the
water table in the pit is lowered to an elevation of –5.4 m. The gravelly sand (layer A) is
assumed to maintain hydrostatic conditions, while the total hydraulic head difference
(equal to 4.4 m) develops within the stiff clay; linear head loss is assumed along an
idealized flow path starting from an elevation of –4.0 m (behind the wall), going around
the toe of the wall and exiting at an elevation of –5.4 m in front of the wall (total length of
path: 2L + 1.4 m, where L is the wall penetration below the excavation level). This
method ensures that the water pressure at the toe of the wall (both sides) is in equilibrium,
as anticipated. Thus, the water pressure on the active side of the wall is lower than the
hydrostatic pressure (for the water table at an elevation of –1.0 m), while on the passive
side of the wall the water pressure is higher than hydrostatic (for the water table at an
elevation of –5.4 m). Vertical effective stresses (σv¢) on both sides of the wall are
calculated using these water pressures (u) from the equation σv¢ = σv – u, where σv = γ z is
the total vertical (geostatic) stress (z is measured from the ground surface behind the wall
and from the bottom of the excavation in front of the wall). Hydraulic failure of the
bottom of the excavation by heave is examined according to the principles of Section 10.
Other input parameters are:

• Weight density of water: γk = 10 kN/m3.

• Surcharge at ground surface (variable action): qk = 10 kPa.
• Steel sheet pile (Frodingham 3N): Wel = 1688 cm3/m, Wpl ª 1900 cm3/m, steel grade
S355 GP, yield strength 355 MPa.

The design value of the ultimate moment of resistance (Mu, d) of the sheet pile
(Frodingham 3N) is calculated according to EN 1993-5:

174

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 9. RETAINING STRUCTURES

Mu, d = Wpl fy /1.0 = 675 kN m/m

The value of the material partial factor for steel (1.0) is taken from EN 1993-5, and may be
set by national determination.
In addition to the classical LEM , a spring model is also used below, in the ULS design
with DA-1 Combination 1 and DA-2 and in the SLS design. Spring models can be used in
DA-1 Combination 2 and DA-3 designs following the recommendations given in Section
9.7. In the spring model, the calculations use the characteristic values of all stiffness
parameters (soil, anchor and wall). This model requires soil stiffness parameters, the wall
bending stiffness (EI), and the anchor axial stiffness and prestress force. The soil is
simulated by an elastic–perfectly plastic spring yielding at the limiting active and passive
earth pressures. The characteristic value of the pre-yield spring stiffness (horizontal
subgrade modulus) is assumed equal to k = 10 000 kN/m3 in the sand and
k = 6000 kN/m3 in the clay layer. The characteristic value of the coefficient of earth
pressure at rest is K0, k = 0.50 in the sand and K0, k = 0.95 in the clay layer. The
characteristic anchor pre-yield axial stiffness K = EA/L is calculated from its free length
(L = 11 m), cross-section (A = 1 cm2/m) and the Young modulus of steel (E = 210 GPa).
The anchor prestress force (lock-off force) P0 = 100 kN/m was obtained by trial and
error. The sheet pile wall pre-yield bending stiffness is EI = 5 × 104 kN m2/m.
In the LEM the design values of ground parameters and the corresponding calculated
horizontal earth pressure coefficients for limiting active (Kah) and passive (Kph) conditions
are given in Table 9.5. The design values of Kah and Kph are obtained from Figs C.1.1 and
C.2.1.

ULS design of the wall

In the ULS design of the embedded wall, the aim is to determine:
(1) The minimum embedment depth. This is calculated by assuming free-earth support
conditions at the toe, i.e. assuming that the wall moves sufficiently to mobilize the
limiting active and passive earth pressures on the retained and the resisting sides,
respectively (LEM).
(2) The bending moment distribution on the wall and the anchor force. These may be
determined by two methods:
(a) Using known earth pressure distributions on the wall, namely the limiting earth
pressures (i.e. using a LEM without earth pressure redistribution).
(b) Using a spring model, i.e. a beam model of the wall supported on soil springs with
appropriate stiffness. This model includes wall–ground interaction and earth
pressure redistribution. Details concerning the application of spring models are
presented in Section 9.7.
The following horizontal actions are applied to the wall in the LEM:
(1) The horizontal component of the active earth pressure on the retained side:
p¢ah = p¢ah, g + p¢ah, q
at a depth z from ground surface, where
p¢ah, g = Kahσv¢ – 2c¢÷Kah
is an unfavourable permanent geotechnical action and σv¢ is the vertical effective
stress at depth z, and
p¢ah, q = Kahq
is an unfavourable variable geotechnical action.
(2) The net water pressure (pressure on the retained side minus pressure on the resisting
side), which is an unfavourable permanent geotechnical action. The net water

175

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Table 9.5. Example 9.2 – LEM

DA-1 DA-2 DA-3 OFSa

Comb. 1 Comb. 2

Partial factor on ground strength 1.0 1.25 1.0 1.25 1.0

parameters, γM
Design value of the angle of shearing
resistance, ϕ¢ (°)
Layer A 35 29.3 35 29.3 35
Layer B 24 19.6 24 19.6 24
Design value of cohesion intercept,
c¢ (kPa)
Layer B 5 4 5 4 5
Max. design value of wall–ground
interface parameter, d = 23 ϕ¢ (°)
Layer A – active side 23 19.6 23 19.6 23
Layer B – active side 16 13.1 16 13.1 16
Layer B – passive side 12 9.8 12 9.8 12
Design value of horizontal active
earth pressure coefficient, Kah
Layer A 0.23 0.30 0.23 0.30 0.23
Layer B 0.37 0.45 0.37 0.45 0.37
Design value of horizontal passive
earth pressure coefficient, Kph
Layer B 3.20 2.50 3.20 2.50 3.20
Partial factors for actions, γF
Unfavourable permanent 1.35 1.0 1.35 1.0 1.0
Unfavourable variable 1.5 1.3 1.5 1.3 1.0
Favourable permanent 1.0 1.0 1.0 1.0 1.0
Partial factor for earth resistance on 1.0 1.0 1.4 1.0 1.0
passive side, γR
a
In the OFS method, all values are ‘nominal’ (characteristic) values

pressure is zero at the toe of the wall, since at this point water pressures on both sides
are in equilibrium. Thus, the characteristic value of the net water pressure increases
linearly from zero at an elevation of –1.0 m to 30 kPa at an elevation of –4.0 m,
reaches a maximum value at an elevation of –5.4 m (depending on the wall
penetration) and then reduces linearly to zero at the toe of the wall (see Table 9.6).
The above horizontal actions are resisted by the horizontal component of the anchor
force and by the horizontal component of the limiting passive earth pressure on the
retained side. The limiting passive earth pressure at a depth z¢ below the excavation base is
calculated from the equation
p¢ph = Kphσv¢ + 2c¢÷Kph
where σv¢ is the corresponding vertical effective stress. The limiting passive earth pressure
is treated as a resistance (see the discussion in Section 9.3.1).
Design values of actions are calculated according to the method described in Section
9.7. For example, the design value of the unfavourable variable geotechnical action p¢ah, q in

F = γ F(X /γ )
d F k M

i.e.

176

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 9. RETAINING STRUCTURES

p¢ah, q = γFKahq
and with partial factors from Table 9.5:

• In DA-1 Combination 1, step 1, and DA-2:

p¢ah, q = 1.5 × 0.23 × 10 = 3.45 kPa
• In DA-1 Combination 2 and DA-3:
p¢ah, q = 1.3 × 0.30 × 10 = 3.90 kPa
The design value of the limiting passive earth pressure (resistance) is calculated using
the equation:
Rd = R[F(Xk/γM)]γR
i.e.
pph = (Kph σv¢ + 2 c¢÷Kph )/γR
with partial factors from Table 9.5. For example, at a depth z¢ = 3 m below the excavation
base, where the characteristic vertical effective stress σv¢ = 30 kPa and assuming
hydrostatic pore pressures (adjusted for equality at either side of the toe), the design value
of the limiting passive earth pressure is:

• In DA-1 Combination 1, step 1:

p¢ph = [3.2 × (1.0 × 30) + 2 × 5 × ÷(3.2)]/1.0 = 113.9 kPa
• In DA-2:
p¢ph = [3.2 × (1.0 × 30) + 2 × 5 × ÷(3.2)]/1.4 = 81.3 kPa
• In DA-1 Combination 2 and DA-3:
p¢ph = [2.5 × (1.0 × 30) + 2 × 4 × ÷(2.5)]/1.0 = 87.6 kPa
In fact, due to the upward water flow, the actual vertical effective stress (σv¢) is
somewhat lower than the values obtained above, assuming hydrostatic water pressures.
Note that the vertical effective stress at a particular depth varies in each Design Approach
as the wall penetration is different. The correct value of the vertical effective stress was
used in the worked example (see Table 9.6), while the same hydrostatic pore pressure was
used in the above calculation in order to compare the three Design Approaches.
In the LEM, the required wall penetration is calculated by ensuring bending moment
equilibrium of the design values of actions and resistances around the anchorage point
(since free-earth support conditions are selected). Then, the design value of the
horizontal component of the anchor force (P0h) is calculated by ensuring equilibrium of
the design values of horizontal actions and resistances.
The vertical equilibrium of the wall is finally checked by adjusting the design value of
the wall–ground interface parameter (δa) on the active side, i.e. by calculating a factor (m)
such that
¢ tan mδa + Fh tan β = P ph
P ah ¢ tan δp
¢ and P ph
where P ah ¢ are the resultant horizontal active and passive earth pressure forces on
the wall (all quantities in the equation enter with their design values). It is noted that the
wall–ground interface parameter (δp) on the passive side is not adjusted since the wall
tends to move downwards and thus tends to reduce the friction on the active side only. In
the unlikely case that the calculated m value is greater than 1, then m is set equal to 1, and
the parameter δp is reduced until equilibrium is verified (since in this case the wall moves
upwards and reduces the friction on the passive side).

177

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Table 9.6 lists the main results of the ULS calculations based on the Eurocode Design
Approaches and the OFS method with an OFS of 2.0 on passive earth pressure (i.e.
passive earth pressure is reduced to 50% of its characteristic value). The design values of
bending moments and shear forces listed in this table were determined from the design
values of the earth and water pressures and support reactions.

Table 9.6. Example 9.2 – ULS design by LEM (except for step 2 of DA-1 Combination 1)

DA-1

Comb. 1a Comb. 2

Step 1 Step 2 DA-2 DA-3 OFSb

Required wall embedment, d, below 5.89 6.62 6.62 7.89 6.62 8.27
maximum excavation (m)
Required length of sheet pile wall (m) 11.29 12.02 12.02 13.29 12.02 13.67
Characteristic value of pore water
pressure (kPa)
Elevation –1.0 m (active side) 0 0 0 0 0 0
and –5.40 m (passive side)
Elevation –4.0 m (active side) 30 30 30 30 30 30
Elevation –5.4 m (active side) 39.3 39.3 39.8 40.4 39.8 40.6
Toe of wall (both sides) 78.6 78.6 86.1 99.1 86.1 103.0
Characteristic value of the net water
pressure (kPa)
From elevation 0.0 to –1.0 m 0 0 0 0 0 0
Elevation –4.0 m 30 30 30 30 30 30
Elevation –5.4 m 39.3 39.3 39.8 40.4 39.8 40.6
Toe of wall 0 0 0 0 0 0
Design value of total horizontal thrust 662.4 638.9 858.6 638.9 660.3
on the active side, Pah¢ + Pw (kN/m)
Design value of total horizontal resistance 475.1 466.9 630.1 466.9 486.9
on the passive side Pph ¢ (kN/m)
Design value of the horizontal component 167.0 172.0 228.5 172.0 173.4
of the anchor force, P0h (kN/m)
Design value of the anchor force, 169.6 174.6 232.1 174.6 176.1
P0 = P0h/cos b (kN/m)
Design value of the overturning (and the 4021.1 5951.9 4021.1 4722.8
resisting) moments with respect to the
anchorage level (kN m/m)
Design value of total vertical thrust on 42.3 83.2 42.3 65.2
the active side, Pav¢ (kN/m)
Design value of total vertical resistance 72.3 123.2 72.3 95.6
on the passive side Ppv¢ (kN/m)
Maximum bending moment in wall
Design value, Mmax, d (kN m/m) 347.0 446.7 649.5 446.7 706.4
Elevation (m) –5.40 –5.40 –5.40 –5.40 –6.04
Maximum shear force in wall
Design value, Sd (kN/m) 159.1 215.7 159.1 164.2
Elevation (m) –1.00 –1.00 –1.00 –1.00
a
The two calculation steps for DA-1 Combination 1 are first to determine the minimum wall
penetration (embedment) depth and then, in a separate calculation (using a soil spring model), to
determine the internal forces and the anchor force
b
In the OFS method, all values are ‘nominal’ (characteristic) values

178

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 9. RETAINING STRUCTURES

DA-1 Combination 1 – step 2

In DA-1, as the wall penetration obtained from Combination 2 is larger than for
Combination 1 (as is usually the case), a separate calculation (step 2) is performed with
the Combination 1 partial factors and the wall penetration d = 6.62 m, determined with
the Combination 2 partial factors as shown in Table 9.6. As the wall penetration is now
greater than that required for the full mobilization of the limiting earth pressures with the
Combination 1 partial factors (d = 6.62 m > 5.89 m), a wall–ground interaction model
is required to calculate the mobilized fraction of the limiting earth pressures. The
interaction model used in this example is a spring model, as discussed in Section 9.7. The
spring model gives the following results:

• Design value of the horizontal component of the anchor force:

P0h = 1.35 × 123.7 = 167.0 kN/m
• Design value of the maximum bending moment in the wall for DA-1 Combination 1:
Mmax, d = 1.35 × 257 = 347 kN m/m < Mu, d = 675 kN m/m
Note that when it is obvious that step 1 of DA-1 requires a smaller wall penetration
than DA-1 Combination 2, step 2 can be performed readily after calculating the wall
penetration using DA-1 Combination 2.
Table 9.6 leads to the following conclusions:

(1) In DA-1, Combination 2 provides the wall penetration depth (6.62 m). Combination
2 is also critical for the structural design of the wall (maximum bending
moment = 446.7 kN m/m).
(2) DA-3 gives identical results to DA-1 Combination 2.
(3) DA-2 is the most conservative of the Eurocode Design Approaches as it gives
the greatest wall penetration (7.89 m) and the largest design bending moment
(649.5 kN m/m). The OFS method gives even deeper wall penetration (8.27 m) and a
slightly larger comparable design bending moment (1.4 × 504.6 = 706.4 kN m/m).
(4) The Frodingham 3N sheet pile section has sufficient ultimate moment of resistance
(Mu, d = 675 kN m/m) to resist the calculated design bending moments in DA-1 and
DA-3; a slightly heavier sheet pile section is required in DA-2 and in the OFS method
(Mu, d ≥ 649.5 kN m/m and Mu, d ≥ 1.4 × 504.6 = 707 kN m/m, respectively).

DA-2 – use of an LEM

In the above application of DA-2 using the LEM, partial factors for actions are applied at
the source, i.e. on the limiting active earth pressure and the net water pressure.
The following equivalent alternative procedure can be used, called DA-2*:

(1) Permanent actions (e.g. limiting active earth pressure due to the soil weight and net
water pressure) enter the calculation with values equal to their characteristic values
(i.e. using unit partial factors). Soil parameters (i.e. shear strength) also enter the
calculation with values equal to their characteristic values, as usual for DA-2.
(2) Unfavourable variable actions (e.g. limiting active earth pressure due to the surface
surcharge) enter the calculation multiplied by 1.5/1.35 = 1.11, i.e. qd = 1.11 and
qk = 11.1 kPa.
(3) The overturning moment (ME) is calculated, which is the sum of the moments of the
active earth pressure and the net water pressure (sum = Ek) with respect to the
anchorage point. The corresponding design value is then calculated: MEd = 1.35ME.
This is the design value of the effect of the actions for use in the ULS design.
(4) The horizontal component of the limiting earth resistance (Rp, k) is calculated, as
is the stabilizing moment (MR) of Rp, k with respect to the anchorage point. The

179

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

corresponding design value is then calculated: MRd = MR/1.4. This is the design value
of the stabilizing moment for use in the ULS design calculation.
(5) The wall embedment is determined from the ULS requirement that MEd £ MRd, or,
equivalently, ME £ MR/(1.35 × 1.4) = MR/1.89.
(6) After calculation of the required wall embedment, the characteristic value of the
anchor force is determined by checking the horizontal equilibrium of the actions and
resistances.
(7) Finally, the characteristic values of the bending moments along the wall are
calculated from the known actions, anchor force and earth pressure forces. The
design values of the anchor force and bending moments are obtained by multiplying
the corresponding characteristic values by 1.35.
In this example, DA-2* gives the following results:

• Embedment depth d = 7.89 m.

• Resultant active earth pressure and net water pressure force Ek = 636.0 kN/m.
• Overturning moment of Ek with respect to the anchorage point ME = 4408.8 kN m/m.
• Design value of the overturning moment MEd = 1.35 × 4408.8 = 5951.9 kN m/m.
• Characteristic value of the horizontal component of the earth resistance Rk = 882.1 kN/m.
• Stabilizing moment due to the earth resistance MR = 8335.3 kN m/m.
• Design value of the stabilizing moment MRd = 8335.3/1.4 = 5953.8 kN m/m.
• Thus, MEd = 5952 kN m/m £ MRd = 5954 kN m/m.

Calculation of the horizontal component of the anchor force (P0h):

– Design value of P0h = 1.35 × Ek – Rk/1.4 = 1.35 × 636.0 – 882.1/1.4 = 228.5 kN/m.
– Characteristic value of Fh = 228.5/1.35 = 169.3 kN/m.

• Characteristic value of the maximum bending moment = 481.14 kN m/m.

• Design value of the maximum bending moment Md = 1.35 × 481.14 = 649.5 kN m/m.
Thus, DA-2 and DA-2* when using the LEM give the same value of the bending
moments (see Table 9.6).
In conclusion, DA-2 and DA-2* give exactly the same results in this worked example.
This is because the calculations are linear.
In the previous analyses (DA-1 Combination 2, DA-2, DA-2* and DA-3), the minimum
required wall penetration depth, bending moments and shear forces along the wall are
calculated with the LEM, i.e. from the limiting earth pressure on the active side of the
wall, without any pressure redistribution to account for wall deflections possibly smaller
than those required for the full mobilization of the limiting earth pressure. Such pressure
redistribution can be analysed with interaction models (finite-element or beam with
spring models) as discussed in Section 9.7.
Spring models, as discussed in Section 9.7, can be used for all three Design Approaches.
The application of a spring model (with the input parameters given above) is illustrated in
this example for DA-2, and DA-1 Combination 1 as follows.

DA-2 – use of a spring model

(1) The minimum embedment depth of the wall is determined with the standard
calculation presented above using limiting earth pressures (LEM). This calculation
gives a design which is safe against failure on the passive side of the wall. The
calculated minimum embedment for DA-2 is d = 7.89 m (see above). Alternatively,
one can search by trial and error to determine the embedment of the wall
leading to mobilization of 53% of the characteristic earth resistance (since
1/γGγRe = 1/(1.35 × 1.4) = 0.53 using the recommended values).
(2) The spring model is employed to analyse the wall with the above embedment
depth and determine the internal forces and the anchor force and to check that

180

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 9. RETAINING STRUCTURES

Table 9.7. Example 9.2 – ULS design: results for DA-2 using the spring model

Design value = 1.35 ×

Calculated value calculated value

Embedment depth From step 1: d = 7.89 m –

Mobilized earth resistance 0.53a £ 1/(1.35 × 1.4) = 0.53 –
Bending moment 209.5 kN m/m 1.35 × 209.5 = 283 kN m/m
Anchor force 116.0 kN/m 1.35 × 116.0 = 157 kN/m
a
Corrected anchor force accounting for active pressures on the back of the wall

Table 9.8. Example 9.2 – ULS design: results for DA-1 Combination 1 using the spring model (wall
length from DA-1 Combination 2)

Design value = 1.35 ×

Calculated value calculated value

Embedment depth From step 1: d = 6.62 m –

Mobilized earth resistance 0.60a < 1/1.35 = 0.74 –
Bending moment 257 kN m/m 1.35 × 257 = 347 kN m/m
Anchor force 126 kN/m 1.35 × 126 = 170 kN/m
a
Corrected accounting for active pressures at the back of the wall

the mobilization of the characteristic earth resistance does not exceed 53%
(1/(1.35 × 1.4) = 0.53). It should be pointed out that greater wall penetration could
also be used in this analysis (e.g. if the wall has to penetrate to an impervious layer).
The characteristic values of the stiffness and strength parameters for the ground
and the wall are used in the analysis. The surface load (q) is input with a
value q = 1.5 × 10/1.35 = 11.1 kPa. The calculated bending moments and anchor
forces are then multiplied by the partial factor for permanent unfavourable actions
(1.35) to determine their design values. The main calculation results are presented in
Table 9.7.
The design values of bending moments and anchor reactions obtained using the
spring model for DA-2 are much lower than the values obtained using the LEM (see
Table 9.6). This is due to the fact that the spring model uses characteristic values
of the earth resistance, and the partial factors on the earth resistance are only
introduced at the end to check the mobilization level.

DA-1 Combination 1 – use of a spring model

(1) The minimum embedment depth of the wall is determined with the standard
calculation (LEM) for DA-1 Combination 2. The calculated minimum embedment
for DA-1 Combination 2 is d = 6.62 m (see above).
(2) The spring model is employed to analyse the wall with the above embedment depth
and determine the internal forces and the anchor force and to check whether the
mobilization of the characteristic earth resistance does not exceed 0.74 = 1/1.35.
The surface load (q) is introduced with a value q = 1.5 × 10/1.35 = 11.1 kPa. The
calculated bending moments and anchor forces are then multiplied by the partial
factor for permanent unfavourable actions (1.35) to determine their design values.
The main calculation results are presented in Table 9.8.
The equivalent OFS of DA-1 against passive failure is about 1.6, and about 1.9
in DA-2. The bending moments and anchorage reactions calculated using the

181

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

spring model for DA-1 Combination 1 are larger than for DA-2 because the wall is
somewhat longer in DA-2.
Note that, in all interaction models, the calculated bending moments and support
reactions depend on the selected value of the soil stiffness (spring stiffness) relative to the
wall and anchor stiffness. As soil stiffness involves appreciable uncertainty, the sensitivity
of the design to variations of the stiffness parameters should be checked. For a given wall
length, low values of the soil spring stiffness combined with high values of the wall
stiffness result in higher bending moments in the wall.
It should be pointed out that the above conclusions are relevant for this particular
example only. In other examples, involving different wall geometries and ground
parameters, the corresponding conclusions may be different.

ULS design of excavation against failure by hydraulic heave

Upward groundwater seepage towards the bottom of a deep excavation may cause failure
by hydraulic heave in soils with relatively uniform permeability, and failure by hydraulic
uplift in cases where a low permeability layer overlies a high permeability layer. Although
failure by hydraulic heave is more relevant in high-permeability soils, as steady state
seepage is established relatively quickly, designing against heave is performed in the
present example (stiff clay), mainly for illustrative purposes. The methodology adopted to
design against hydraulic heave is discussed in Chapter 10.
The hydraulic heave calculation model usually involves examining a thin ground
column between the bottom of the excavation and the elevation of the toe of the wall,
neglecting the lateral frictional forces (conservative assumption). In this example, the
design against failure by hydraulic heave is performed for the shortest embedment depth
(d = 6.62 m) determined by DA-1 and DA-3.
According to clause 2.4.7.5, one inequality for designing against failure by hydraulic
heave involves checking that, at the bottom of any ground column, the design value of the
destabilizing pore water pressure udst, d is less than the design value of the stabilizing total
vertical stress σstb, d (see Fig. 9.6 for a definition of the symbols), i.e.
udst, d £ σstb, d (2.9a)
Another inequality for designing against failure by hydraulic heave involves checking
that the design value of the destabilizing seepage force Sdst, d in the ground column is less
than the design value of the stabilizing buoyant (i.e. effective) weight G¢stb, d of the ground
column, i.e.
Sdst, d £ G¢stb, d (2.9b)

Bottom of excavation

i
d

dgwater
u = dgwater + idgwater

Fig. 9.6. Definition of symbols in the design against hydraulic heave

182

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 9. RETAINING STRUCTURES

Inequality (2.9b) gives

Sdst, d = γG, dstγwater id £ γG, stb(γk – γwater)d = G¢stb, d
where i is the characteristic value of the hydraulic gradient, d is the height of the ground
column up to the level where heave failure is checked (Fig. 9.6), γwater is the characteristic
value of the weight density of the water, γk is the characteristic value of the total weight
density of soil layer B, γ ¢k = γk – γwater is the characteristic value of the buoyant (effective)
weight density of soil layer B, and γG, stb and γG, dst are appropriate partial factors for
stabilizing and destabilizing actions (see Table A.17).
The pore water pressure at the elevation of the toe of the wall (elevation –5.40 – 6.62 =
–12.02 m) is calculated assuming a linear hydraulic head loss in layer B, as explained
previously (length of seepage path: L = 2d + 1.40 = 2 × 6.62 + 1.40 = 14.64 m). This
assumption gives an upward hydraulic gradient:
i = ∆H/L = (–1.0 + 5.40)/14.64 = 0.30
The recommended values of the partial factors for hydraulic heave are (Table A.17)
γG, stb = 0.9 γG, dst = 1.35
Application of the above methodology of inequality (2.9b) gives
γG, dstγwater id £ γG, stb(γk – γwater)d
fi 1.35 × 10 × 0.30 × 6.62 £ 0.90 × (20 – 10) × 6.62
fi 26.8 kPa £ 59.6 kPa
indicating that safety against failure by hydraulic heave is sufficient.
Note: inequality (2.9a) gives a result which is not the same as that given by inequality
(2.9b):
udst, d £ σstb, d fi γG, dstuk £ γG, stbσk
udst, d £ σstb, d fi γG, dst(γwater d + γwater id) £ γG, stbγkd
udst, d £ σstb, d fi γG, dstγwater id £ (γG, stbγk – γG, dstγwater )d
udst, d £ σstb, d fi 1.35 × 10 × 0.30 × 6.62 £ (0.9 × 20 – 1.35 × 10) × 6.62
udst, d £ σstb, d fi 26.8 kPa £ 29.8 kPa
In this example, inequality (2.9a) also indicates sufficient safety against failure by
hydraulic heave (but with greater safety, because of the more stringent requirement on
the design excess pore water pressure γG, dstγwater id = 26.8 kPa). For a discussion about the
differences between inequalities (2.9a) and (2.9b), see Chapter 10.

SLS design against wall deflections

The procedure for the SLS design against wall deflections is briefly described in clause
9.8.2. It consists of obtaining a cautious estimate of the wall deflection and the resulting
ground displacements on the basis of comparable experience, and, if the limiting
acceptable values are exceeded, a more detailed investigation is performed by carrying
out rigorous calculations (see also the discussion in Section 9.8.2). This implies that, in
many SLS designs, calculations of wall deflections and ground displacements may be
avoided if comparable experience is available.
In this example, wall deflections in the SLS are calculated using a one-dimensional
analytical model that includes the ground interaction with the wall. The wall is considered
as a vertical beam, with the properties of the sheet pile and the ground (on both sides of
the wall) modelled as a series of elasto-plastic springs yielding at the active and passive
earth pressures, as shown in Fig. 9.7. The stiffness of the soil springs is obtained by

183

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Earth pressure on wall

KPs ¢v

K0s ¢v

Kas ¢v

Wall displacement

Fig. 9.7. Model for the SLS design of an embedded wall with wall–ground interaction. The earth
pressure distribution on the active side develops by elasto-plastic soil springs

Table 9.9. Example 9.2 – SLS design (all values are equal to the characteristic values)

DA-1 Combination 2 DA-2 DA-3

Wall penetration below the maximum excavation 6.62 7.89 6.62

base (m)
Total length of the sheet pile wall (m) 12.02 13.29 12.02
Horizontal component of the prestressing 100 100 100
(lock-off) force (kN/m)
Final horizontal component of the anchor force 112 112 112
(kN/m) after completion of the excavation
Calculated wall deflection (mm)
Elevation 0.0 m –2.6 2.6 –2.6
Elevation –1.0 m 6.1 5.8 6.1
Elevation –3.0 m 22 21 22
Elevation –5.0 m 27.7 26 27.7
Elevation –7.0 m 21 21 21
Elevation –9.0 m 13 13 13
Toe 5.5 7 5.5
Calculated maximum bending moments (kN m/m) 172 168 172

considering the wall movement required to mobilize the active and passive earth pressures
(see clause C.3 in Annex C). The design values of all actions, resistances and ground
parameters, and wall and anchor stiffnesses are equal to their characteristic values. As the
model includes soil springs on the active side, no assumption is required about the earth
pressure distribution on the active side. The net water pressure difference is considered as
a known action. The anchorage is simulated as a prestressed elastic spring with the
stiffness determined by the ULS design. The main calculation results are summarized in
Table 9.9.
As the model is one-dimensional, it does not allow the ground displacements behind
the wall to be calculated directly in order to check the serviceability conditions of
any supported structures and utilities. Such estimates would require a two- or three-
dimensional numerical model (e.g. using finite elements), as discussed in Section 9.8.2.
Alternatively, indirect methods may be used to convert the calculated horizontal wall
deflections in ground surface settlements.
The main calculation results are summarized in Table 9.9.

184

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 10

Hydraulic failure

This chapter deals with the designs against the different types of hydraulic failure. The
material described in this chapter is covered by Section 10 of EN 1997-1, together with the
partial factors in Annex A. The structure of this chapter follows that of Section 10:
10.1. General Clause 10.1
10.2. Failure by uplift (UPL) Clause 10.2
10.3. Failure by heave (HYD) Clause 10.3
10.4. Internal erosion Clause 10.4
10.5. Failure by piping Clause 10.5

10.1. General
Section 10 of Eurocode 7 begins by listing the modes of hydraulic failure covered by the code, Clause 10.1(1)P
which must be checked:
• failure by uplift
• failure by heave
• failure by internal erosion
• failure by piping.
Definitions are then provided for these hydraulic failure modes, as the terms used for
them and their meaning can differ from country to country.
Table 10.1 summarizes the types of failure and how they are designed against.
Hydraulic gradients, pore pressures and seepage forces are the predominant actions to be Clause 10.1(3)P
considered in designs against hydraulic failures. For this reason, special attention has to be
given to those parameters which have the greatest influence on the actions due to the water,
which are:

Table 10.1. Summary of the design procedures for different types of hydraulic failure

Table of recommended Equivalent global

Type of failure ULS Basic inequality partial factor values safety factor

Uplift UPL Expression (2.8) Tables A.15 and 16 1.10

Heave HYD Expression (2.9) Table A.17 1.50
Internal erosion – Filter criteria
Piping – Determine flownet or the distribution of the hydraulic gradient for
possible adverse ground conditions and check for failure by hydraulic
heave (for horizontal ground) or by slope failure (for inclined ground)

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

• the variation of soil permeability in time and space

• variations in water levels and pore water pressure in time
• modifications in the geometry.

An example illustrating that different soil conditions can result in different failure
mechanisms is shown in Fig. 10.1. When a stratum of, for example, varved clay of low
permeability is situated below the level of excavation, there will be practically no
groundwater flow, and it must be verified that there is sufficient safety against uplift of the
stratum of low permeability. If, instead, the ground below the excavation is permeable (e.g.
sand), groundwater will flow upwards into the excavation, and it must be checked that failure
by heave of the ground below the excavation is avoided.
Design situations involving uplift, where there is no seepage of water, should be analysed
using the partial factors for the UPL ultimate limit state in Tables A.15 and A.16 in Annex
A.4. Design situations where there is seepage due to hydraulic gradients should be analysed
using the partial factors for the HYD ultimate limit state in Table A.17 in Annex A.5.

10.2. Failure by uplift (UPL)

10.2.1. General
Clause 10.2(1)P Failure by uplift is checked by comparing the sum, Gdst, d and Qdst, d, of the design values of the
destabilizing permanent and variable vertical actions, i.e. the sum of the water pressures
under the structure (permanent and variable parts) and any other upwards forces, with the
sum, Gstb, d and Rd, of the design values of the stabilizing permanent vertical actions and the
design value of any additional resistance to uplift provided, for example, by tension piles,
ground anchorages or friction forces on the side of a buried structure:
Clause 2.4.7.4(1)P Gdst, d + Qdst, d £ Gstb, d + Rd (2.8)
This general inequality is applied when checking the stability of submerged structures
against uplift failure and the stability against uplift failure of impermeable layers in
excavations. The design procedure for a structure where tension piles are required to
provide additional stabilizing resistance is presented in a worked example in this guide (see
Example 7.5). For persistent and transient situations, the values of the partial factors on the
Clause 2.4.7.4(3)P destabilizing and stabilizing actions should be selected from Table A.15, and the values of the
partial factors on the ground strength parameters or the resistance of geotechnical elements
should be selected from Table A.16.
Clause 2.4.7.4(2) According to clause 2.4.7.4(2), the additional resistance to uplift due to tension piles,
ground anchors or friction forces may be treated as a stabilizing permanent vertical action

Varved clay

Sand

(a) (b)

Fig. 10.1. Examples of hydraulic failure mechanisms due to different soil conditions: (a) failure by
uplift and (b) failure by heave

186

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 10. HYDRAULIC FAILURE

Gsoil, d

H1

Td Td
Gstr, d H2
H

Ud

Fig. 10.2. Tunnel below the groundwater table

rather than as a resistance, and hence the design value will be obtained by applying the
partial factor on permanent favourable actions to it (of which the recommended value is 0.9).
If this procedure is adopted and the partial factor values on actions recommended in Table
A.15 are applied, the resulting UPL design will be less conservative than if the partial soil
parameter values in Table A.16 are applied to the additional resistance. This is because
applying the partial factor in Table A.16 to the additional tensile pile resistance is equivalent
to multiplying the resistance by 0.71 (or, in the case of friction forces, to multiplying the
ground strength parameters by 0.8). It is clear from this that, if clause 2.4.7.4(2) is applied in
the case of the resistance from tension piles, ground anchors or friction, then the GEO
ultimate limit state should also be checked.

10.2.2. Submerged structures

As an example of a design against failure by uplift, it may be useful to consider a tunnel
completely below the groundwater level, as illustrated in Fig. 10.2. With the design values of
the destabilizing uplift action Ud of the water and the design values of the stabilizing actions
of the weight Gstr, d of the structure, the weight Gsoil, d of the ground on top of the structure
and the side friction force Td on the vertical walls, inequality (2.8) becomes
Ud £ Gstr, d + Gsoil, d + Td (D10.1)
If a structure is completely below the groundwater level, the water pressure acting on the
top of the structure could be regarded as a stabilizing action and the water pressure acting
on the bottom as a destabilizing action. As the stabilizing and destabilizing actions are
multiplied by different partial factor values, the safety against uplift would then depend on
the water depth above the structure. Therefore, the principle described in the note to clause Clause 2.4.2(9)P
2.4.2(9)P may be applied in this situation: that is, a single partial factor may be applied to the
difference between these actions, i.e. to the difference between the characteristic permanent
actions of the water pressure acting on the top and the bottom of the structure. The
difference between the characteristic destabilizing actions due to the water pressures is
UK = γw(H2 – H1)A = γw HA (D10.2)
where A is the base area of the tunnel and the other symbols are as illustrated in Fig. 10.2.
The design value of the destabilizing actions is then

187

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

Water

Hk

Impermeable Permeable
layer layer
d

Fig. 10.3. Uplift of an impermeable layer

Ud = γG, dstγw HA (D10.3)

It should be noted that, when calculating the stabilizing action, the weight of the soil below
the groundwater level should be calculated using the effective weight density γ¢ of the soil.
As indicated in inequality (D10.1), friction forces, e.g. on the walls of submerged structures,
may be taken into account. They can be derived from
Td /A = Kσ¢v tan δd (D10.4)
The wall friction angle δd should be determined by dividing tan δk by γϕ from Table A.16 to
obtain a conservative low friction force. If K, the coefficient of earth pressure, is calculated
from the angle of shearing resistance ϕ¢, an appropriately cautious value should be selected
for ϕ¢.

10.2.3. Design against uplift of an impermeable layer

If friction forces are neglected, the design against uplift of an impermeable layer where there
is no seepage through the layer, e.g. at the bottom of, or below, an excavated building pit
(Fig. 10.3), can use stresses instead of forces. In this case, the design value of the destabilizing
total water pressure ud acting at the interface between the two layers must be less than or
equal to the stabilizing total vertical stress σstb, d due to the total weight of soil above the
interface
udst, d £ σstb, d (D10.5)
Using the partial factors specified in Table A.15 and the symbols of Fig. 10.3, inequality
(2.8) for design failure by uplift becomes
γG, dstγw Hk £ γG, stbγd (D10.6)
Using the values for the partial factors given in Annex A.4, the method of EN 1997-1 is
equivalent to an overall factor of safety (OFS) against uplift given by
OFS = γG, dst /γG, stb = 1.00/0.90 = 1.11 (D10.7)

10.2.4. Worked example of a design against uplift

An example of a design against uplift of a structure with tension piles is presented in
Example 7.5 of this guide.

188

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 10. HYDRAULIC FAILURE

10.3. Failure by heave (HYD)

10.3.1. General
Failure by heave occurs when upward seepage forces acting against the weight of the soil Clause 10.1(1)P
become so strong that the vertical effective stress is reduced to zero. According to clause Clause 2.4.7.5(1)P
2.4.7.5(1)P the stability of soil against heave must be checked for every relevant soil column
either by a calculation in terms of pore water pressures and total stresses (inequality (2.9a))
or by a calculation in terms of seepage forces and submerged weight (inequality (2.9b)). The
left side of the retaining wall in Fig. 10.4 illustrates an example of failure by heave. In the
following, any vertical friction forces (stabilizing) have been neglected.

10.3.2. Design using total stresses

When designing against seepage using total stresses it is necessary to check that the design Clause 10.3(1)
value of the destabilizing total pore pressure ud, dst at the bottom of the column is less than or
equal to the design value of the stabilizing total vertical stress σd, stb:

ud, dst £ σd, stb (2.9a)

For the conditions on the left side in Fig. 10.4, the characteristic value of the destabilizing
water pressure at the toe of the wall is

uk, dst = γw(d + dw + ∆h) (D10.8)

The characteristic value of the stabilizing total stress at the toe of the wall is

σk, stb = (γ¢ + γw)d + γw dw (D10.9)

Applying the partial factors γG, dst and γG, stb to the destabilizing and stabilizing permanent
stresses, inequality (2.9a) becomes

γG, dst uk, dst £ γG, stbσk, stb (D10.10)

Water

Dh

dw

Sand
d

Fig. 10.4. Hydraulic heave under hydrodynamic conditions

189

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

i.e.
γG, dstγw(d + dw + ∆h) £ γG, stb[(γ¢ + γw)d + γw dw] (D10.11)
and, hence, using inequality (2.9a), the design excess pore water pressure is
γG, dstγw ∆h £ γG, stbγ¢d – (γG, dst – γG, stb)γw(d + dw) (D10.12)
Using the recommended values of γG, dst and γG, stb, equal to 1.35 and 0.9, respectively, in the
second component of the right-hand side of the above inequality yields
γG, dstγw ∆h £ γG, stbγ¢d – 0.45γw(d + dw) (D10.13)

10.3.3. Design using submerged weight

When designing against seepage using submerged (effective) weights, it is necessary to check
that the design value of the destabilizing seepage force Sdst, d acting within the soil column is
less than or equal to the design value of the stabilizing submerged weight G¢stb, d of the soil
column:
Sdst, d £ G¢stb, d (2.9b)
For the hydraulic situation shown on the left in Fig. 10.4, the design value of destabilizing
seepage force is
Sdst, d = γdst, dγw id (D10.14)
where i is the hydraulic gradient in the soil column. With i = ∆h/d, the design value of the
destabilizing seepage force becomes
Sdst, d = γG, dstγw ∆h (D10.15)
With the design value of stabilizing submerged weight,
G¢stb, d = γG, stbγ¢d (D10.16)
inequality (2.9b) becomes
γG, dstγw ∆h £ γG, stbγ¢d (D10.17)
Comparing inequalities (D10.13) and (D10.17), it is seen that the design, i.e. allowable,
excess pore water pressure, γG, dstγw ∆h, calculated using inequality (2.9a) is always smaller
than the value calculated using inequality (2.9b) by the amount 0.45γw(d + dw). Hence,
inequality (2.9a) provides greater safety than inequality (2.9b). This is shown by the
calculations in Example 9.2, where the two inequalities are used to analyse heave. Inequality
(2.9b) is the inequality normally used when designing against hydraulic failure; however, in
some situations, inequality (2.9a) may be preferred.

10.3.4. Determination of the relevant pore water pressure

Clause 10.3(2)P The pore water pressure distribution is influenced very strongly by the permeability
conditions of the ground. Small changes in the permeability, e.g. thin layers of low
permeability soil, may change the pore water distribution dramatically. As a consequence of
this, a realistic and safe assessment of the hydraulic conditions and the subsequent
determination of the pore water pressure distribution form the most important part of the
design. The most reliable way to assess the hydraulic conditions is to check the assumed
distribution by taking pore pressure measurements. If no measurements can be performed,
the assumed hydraulic conditions should be on the safe side, which may often result in a very
conservative and costly design.
Clause 10.3(2)P Spatial effects originating from, for example, the edges of rectangular excavations or
impermeable structures in dams should be considered very carefully since only
three-dimensional numerical simulations of the groundwater flow combined with
measurements can give a realistic and safe assessment of the situation in such cases.

190

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 10. HYDRAULIC FAILURE

10.3.5. Worked example of a design against failure by heave

A worked example of a design against failure by heave is given in Example 9.2 of this guide.

10.3.6. Discussion on failure by uplift and failure by heave

When comparing the partial factors for failure by uplift and failure by heave in Tables A.15
and A.17, it is found that the value of the partial factor on unfavourable permanent actions
for uplift (1.0) is considerably smaller than that for heave (1.35). Looking at Fig. 10.3, where
an example of a design against uplift of an impermeable layer is shown, the question arises as
to when is the layer ‘sufficiently’ impermeable that design against failure by uplift (UPL),
with a smaller factor on unfavourable permanent actions, is relevant and when must the
design be against failure by hydraulic heave (HYD), using the higher partial factor on
unfavourable permanent actions.
With respect to their hydraulic properties, cohesive soils may be regarded as
impermeable. When subjected to pore water pressure in excess of hydrostatic, a stiff
cohesive layer will not disintegrate into small grains. Therefore, a cohesive layer may be
regarded as a compact impermeable block, and the partial factors for the UPL limit state
may be applied to the favourable and unfavourable actions on this block The basis for
deciding which type of limit state, UPL or HYD, is relevant in this situation depends on
whether or not the layer can be regarded as being an impermeable block due to its low
permeability and cohesive strength. In cases of doubt, both limit states should be investigated.

10.4. Internal erosion

10.4.1. Filter criteria and hydraulic criteria
The procedure for designing against failure due to internal erosion, which is material Clause 10.4(1)P
transport caused by seepage, will normally consist of a sequence of checks. Initially, it is
necessary to check if the geometry of the grain structure and the voids will prevent the
transport of fine particles by seepage. This can be investigated by checking if the filter
criteria are satisfied, either by the soil layer itself (Cistin, 1967) or at the interface between
two soil layers (e.g. Sherard et al., 1963). Erosion within a non-uniform, non-cohesive layer is
checked by dividing its grain size distribution into a fine and a coarse part, each of which is
then checked by applying filter criteria. If the filter criteria are satisfied, no material
transport will occur, and the ground will be safe against internal erosion.
If the filter criteria cannot be satisfied, the safety against internal erosion can be checked Clause 10.4(5)P
by investigating if the hydraulic gradient causing seepage is high enough to move the soil
particles. This check is based on determining the flownet for each case under consideration.
In simple cases this can be achieved by plotting a flownet ‘by hand’, but generally speaking, a
numerical simulation must be performed to find out where the hydraulic gradient may
become critical and how high its value will become. The computed hydraulic gradient, ical at
crucial points within the ground, for example at the toe of a sheet pile wall or at the ground
surface, will then be compared with the critical gradient, icrit, required to transport soil
particles within the ground. For internal erosion, Busch and Luckner (1974) established
formulae for determining the critical gradient icrit as a function of the grain size distribution
of a soil.
For erosion at the interface between two layers, the critical gradient, icrit, is dependent on Clause 10.4(6)P
the direction of the groundwater flow with respect to gravity and the orientation of the
layers. The critical hydraulic gradient icrit for nine combinations of fine to coarse soil layers
and directions of flow is given in by Armbruster and Tröger (1993).
The filter criteria published in the geotechnical literature will normally incorporate some
sort of safety margin, but which, however, is normally unquantified and therefore unknown.
In addition, such criteria are often established on the basis of experience with certain
regional or local types of soils. Care should therefore be taken when such criteria are used in
design, especially when their use is not based on comparable experience.

191

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

10.4.2. Effects of material transport

If neither the filter criteria nor the hydraulic criteria can be satisfied, it is likely that material
transport will take place. A method for estimating the quantities of fine soil particles
transported by groundwater seepage is given by Busch and Luckner (1974). According to
their empirical research, a material transport of 3% of the soil will not be critical. If, for
example, the serviceability of a structure or the stability of a dam is jeopardized by internal
erosion, either different materials should be used for the dam or else structural measures
should be taken to prevent the seepage causing internal erosion.

10.5. Failure by piping

10.5.1. General
Clause 10.5(1)P Failure by piping is a particular form of failure by internal erosion where transport of soil
material begins at the ground surface, and erosion then regresses until a pipe-shaped
discharge tunnel is formed in the soil mass or between the soil and a foundation, or at the
interface between cohesive and non-cohesive soil strata. Failure occurs as soon as the
upstream end of the eroded tunnel reaches the bottom of the reservoir. A typical situation
where piping may occur is illustrated in Fig. 10.5. Here, piping may start in the ditch behind a
dam, where the impermeable top layer will crack due to failure by hydraulic heave caused by
the high water pressure in the subsoil during a flood. The crack will develop into a well with a
high hydraulic gradient, where seepage forces will transport soil particles from the subsoil
into the ditch. Protected by the cohesive top layer, a pipe is formed by the continual transport
of soil. As the pipe develops upstream, the hydraulic gradient in the subsoil increases, thus
accelerating the piping process.

10.5.2. Design against failure by piping

Most of the procedures to asses the danger of piping due to water percolating below a dam or
a structure are based on global or weighted hydraulic gradients assuming homogeneous
ground (see Bligh, 1910; Lane, 1935). The effects of the presence of different strata cannot
be taken into account. Moreover, most failures due to piping are caused by heterogeneities
at the interface between the ground and the structure. Therefore, the procedures referred to
above should only be used to provide an initial rough assessment.
As there is no direct procedure for checking that there is sufficient resistance against
piping, Armbruster et al. (1999) have proposed an indirect procedure. The basis of this
procedure is a sequence of investigations to check that water seepage cannot transport soil
or soil particles from the ground to the ground surface. If this can be checked, then the
structure and the ground will be safe against piping, although transport of soil particles may
occur within the soil mass. This may produce settlements that could have an adverse effect

Piezometric level in the

Flood permeable subsoil

Low permeability

Possible pipe
Possible well,
Permeable subsoil starting point
for piping

Fig.10.5. Piping below an impermeable dam and top layer

192

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 10. HYDRAULIC FAILURE

on the serviceability, but a disastrous collapse due to piping will be avoided as long as there is
no transport of soil particles out of the ground.
Two checks are necessary to provide sufficient safety against failure by piping:
• Where the ground surface is horizontal, it should be checked if, even under the most Clause 10.5(4)
unfavourable hydraulic conditions, there is sufficient safety against hydraulic heave (see
Section 10.3 of this guide); where the ground surface is inclined, it should be checked if
the slope has an adequate level of stability, taking into account the seepage force in the
slope (see Chapter 11).
• Furthermore, it should then be checked if the top layer at the ground surface has a Clause 10.4(1)P
sufficient level of safety against internal erosion (see Section 10.4 of this guide). Since
soil will be loosened by internal erosion, this should be prevented, as loose soil can
facilitate the process of piping.
The flownet or the distribution of the hydraulic gradient in the ground has to be
determined for both checks. The flownet is not only influenced by the permeability of the
ground and its possible anisotropy but also by the geometry of a structure, its
three-dimensional effects and by the way it was built. Interfaces between the ground and the
structure are very often preferred seepage paths along which the water can percolate without
resistance, and the hydraulic gradient will therefore drop to zero in these places. Interfaces
should therefore be investigated carefully to determine if they are preferred flow paths and
are likely to lead to piping before determining the flownet.
When checking safety against piping, the following stages should be performed:
• In the first stage, it is necessary to assess where preferred flow paths either already exist
or may develop due to distinctive features of the ground, the geometry of the structure
or the way it was built. These areas or interfaces should be modelled by layers of high
permeability. It is also necessary to assess whether drainage systems and sealings will
perform properly or not.
• In the second stage, the flownet is determined, taking into account the hydraulic
assumptions and the boundary conditions assessed in the first stage. Preferably a
finite-element analysis is performed for layered ground conditions with different
permeabilities or where the three-dimensional effects of a structure have to be taken
into account. It may be necessary to perform several analyses in order to determine the
most unfavourable flow conditions in the ground or near the structure.
• In the third stage, those areas where the water comes to the ground surface have to be
investigated. In the case of slopes, the force of the seeping groundwater must be taken
into account when checking overall stability and local stability. If necessary, safety can be
increased by adding drainage layers or by flattening the slope. If the ground surface is
horizontal, the resistance to failure by hydraulic heave should be checked. In such cases,
safety can also be increased by adding drainage layers at the surface or by using
prescriptive measures to increase the length of the seepage path.
• In the final stage, the susceptibility of the ground, and in particular the susceptibility of
the top layer, to internal erosion should be checked. If the ground does not satisfy the
filter criteria, it must be replaced or covered by a drainage layer.

193

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 11

Overall stability

This chapter is concerned with the overall stability of natural slopes, embankments,
excavations and retained ground, and with ground movement around foundations on sloping
ground and near excavations or coasts. The material is covered in Section 11 of EN 1997-1.
The structure of the chapter follows that of Section 11:
11.1. General Clause 11.1
11.2. Limit states Clause 11.2
11.3. Actions and design situations Clause 11.3
11.4. Design and construction considerations Clause 11.4
11.5. Ultimate limit state design Clause 11.5
11.6. Serviceability limit state design Clause 11.6
11.7. Monitoring Clause 11.7
Example 11.1 applies the provisions of EN 1997-1 in checking the avoidance of instability
of a cutting in stiff clay.

11.1. General
In addition to the provisions of Section 11, overall stability issues are also discussed in
Sections 6–10 and 12, which refer to specific structures. Clause 11.1.(2)

11.2. Limit states

Checking against overall stability failure should include all possible limit states. Typically,
these include:
(1) Ultimate limit states (ULSs):
(a) GEO-type limit states, where failure occurs in the ground only, such as failure of a
natural slope or a road embankment on soft clay.
(b) STR-type limit states, with combined failure or large movement in the ground and
in certain structural members, such as failure of a deep excavation supported by an
anchored sheet pile wall, where the failure surface cuts through the anchors (or,
rarely, the sheet pile).
(2) Serviceability limit states (SLSs), such as excessive movements in the ground (e.g. in a
natural or artificial slope), due to shear deformations and settlements (e.g. in retained
ground next to a deep excavation), vibration (e.g. ground densification due to the
operation of vibrating machinery) or heave in the supported structure (e.g. in swelling
clays). The term ‘excessive movements’ includes movements causing loss of serviceability
or damage in neighbouring structures, roads or utilities.

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

The methods for analysing overall stability failure in ULSs and SLSs are discussed below,
in Sections 11.5 and 11.6 respectively.

11.3. Actions and design situations

A list of possible actions and situations in geotechnical design is given in clause 2.4.2.4(P).
Clause 11.3(2)P Clause 11.3(2)P provides some additional design considerations, mainly applicable in overall
stability considerations. In addition to the above, future excavations that may trigger a global
instability should be taken into account.
In checking ULSs, the most unfavourable loading, hydraulic and material parameter
conditions that could occur in the relevant design situation should be considered. In
checking SLSs, less severe conditions may be used, as appropriate.

11.4. Design and construction considerations

Overall stability should be checked during the design by calculations, taking into account
comparable experience according to clause 1.5.2.2 of EN 1997-1. If checking indicates
instability or unacceptable movements, the site should be judged to be unsuitable without
stabilizing measures. In cases where overall stability is not certain, additional investigations,
Clause 11.7 monitoring and analysis should be specified, according to the provisions of clause 11.7.

11.5. Ultimate limit state design

Clause 11.5.1(1)P The overall stability of slopes (including any existing or planned future structures on them)
must be checked in all relevant ULSs (GEO and STR) with design values of actions, material
parameters and resistances. The partial factors should be selected from Tables A.3, A.4 and
Clause 11.5.2 A.14 of EN 1997-1 (Annex A). The same provisions are also applicable to slopes and cuts in
Clause 11.5.3 rock masses and for excavations.
Clause 11.5.1(4) Overall ULS stability checking is usually performed by one of the following calculation
methods:

(1) ‘Assumed failure surface’ methods. These methods are usually implemented
numerically via ‘methods of slices’ (mainly in soil materials) or wedge-type methods
(mainly in rock or rock-like materials). Assumed failure surface methods are the most
common methods for overall stability checking.
(2) Limit analysis methods. These methods give approximate lower- or upper-bound
solutions, i.e. solutions where the occurrence of instability is estimated conservatively or
unconservatively, respectively. Available closed-form solutions are limited to simple
geometrical situations such as vertical cuts and infinite slopes.
(3) Advanced numerical methods (e.g. finite elements). These methods are more versatile
in checking overall stability compared with the ‘assumed failure surface’ and limit
Clause 11.5.1(11)P analysis methods. They are most appropriate in cases where instability includes combined
failure of structural members and the ground, such as failure surfaces intersecting
flexible walls, piles, ground anchorages or nails. In such cases, ground–structure interaction
should be considered by allowing for the difference in their relative stiffness. Such
effects are investigated more accurately with numerical methods including deformation
analysis rather than simplified analyses of ULSs. Advanced numerical models can also
permit the mobilization of different fractions of the strength of the ground and of the
structural member for a specific movement. Effectively, this means that the peak soil
strength and the peak resistance in the structural member need not be assumed to occur
simultaneously.

196

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 11. OVERALL STABILITY

‘Assumed failure surface’ and limit analysis methods are simplified methods which do not
satisfy stress equilibrium and strain compatibility conditions simultaneously, but give
priority to the failure criterion (usually Mohr–Coulomb). Advanced numerical methods
(e.g. classical finite-element methods) satisfy both of these conditions globally (though not
necessarily at all points), but they are more complicated to use.
In ‘assumed failure surface’ methods, the location of a potential failure surface is
assumed, and the mass of ground bounded by this surface is treated as a rigid body or
as several rigid bodies moving simultaneously. Failure surfaces and interfaces between
rigid bodies may have a variety of shapes including planar, circular or more complicated
shapes. The shape of the assumed failure surface may be selected using the following Clauses 11.5.1(5)
recommendations: to 11.5.1(9)
• In relatively homogeneous and isotropic soil materials, circular failure surfaces are
normally used.
• In layered or anisotropic soils with considerable variations of strength, failure surfaces
normally follow layers or directions with lower shear strength.
• In jointed rocks or intensely fissured stiff soils the shape of the failure surface is usually
governed by the discontinuities. In such cases, three-dimensional wedge-type failure
surfaces are normally assumed. The stability of slopes and cuts in rock masses should
also be checked against translational and rotational modes of failure involving isolated
rock blocks or large portions of the rock mass, and also against rockfalls, toppling and
sliding. Particular attention should be given to the hydraulic pressure caused by confined
seepage water in joints and fissures.
• In slopes with pre-existing failure surfaces which may be reactivated, the assumed failure
surface should follow as closely as possible the geometry of the pre-existing failure
surface, which may be non-circular. If the failure surface deviates significantly from
plane-strain geometry, analysis of a three-dimensional failure surface may be required.

In ‘assumed failure surface’ methods, stability is usually examined by dividing the sliding
mass into a number of vertical slices and checking the equilibrium of each slice separately.
Various assumptions are made about the direction of the inter-slice forces, leading to
various numerical models (e.g. Fellenius, Bishop, Spencer and Morgenstern–Price,). Clause Clause 11.5.1(10)
11.5.1(10) recommends that when horizontal equilibrium is not checked, inter-slice forces
should be assumed to be horizontal (the Bishop method).
Using any of the above methods, overall stability checking may be performed according to
the following calculation sequence:

(1) The geometrical model is established including external loading, ground layering and
hydraulic conditions.
(2) Undrained or drained stability analysis is selected according to the soil mechanics
principles, depending on the geometry, the ground type and the expected life of the
structure compared with the time required for drainage of the ground. Typically,
undrained analyses are critical in cases involving loading of soft clays (e.g. construction
of an embankment, or construction on top of a natural slope), while drained analyses are
critical in cases involving unloading of stiff clays (e.g. in deep excavations). In cases
where the most critical analysis is not obvious, both undrained and drained analyses
should be performed.
Undrained stability may be checked using effective-stress or total-stress conditions. If
piezometric data are sparse or unreliable, total-stress conditions with an appropriate
initial undrained shear strength (cu) of the ground are recommended. Effective-stress
conditions with appropriate effective shear strength parameters (c¢, ϕ¢) may also be used
if reliable piezometric data are available.
Drained stability analyses should always be performed with effective stress conditions
and steady state piezometric levels.
(3) Appropriate characteristic values of ground strength parameters are selected.

197

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

In cases of ‘first-time failure’, shear strength parameters should be selected,

considering ground deformation compatibility along the failure surface. This is
especially important in cases where the failure surface crosses materials with an
appreciable contrast in stiffness and strength (e.g. a soft clay and a dense sand). In
simple cases involving relatively uniform soils, it may normally be appropriate to use the
peak strength parameters or, in some cases, somewhat lower values such as those
corresponding to the post-rupture shear strength (the post-peak plateau of the stress–
strain curve), but higher than the residual values corresponding to large relative slip
along the failure plane.
When re-activation of a pre-existing failure surface is examined, residual shear
strength (or a value somewhat higher than that) may normally be more appropriate. In
such cases, shear strength parameters should preferably be selected by back-analysis.
Clause 11.5.1(12) Clause 11.5.1(12) recommends that, since a distinction between favourable and
unfavourable gravity loads is not possible in assessing the most adverse slip surface, any
uncertainty about ground weight density should be considered by applying upper and
lower characteristic values, in separate calculations.
(4) One of the three Design Approaches described in Chapter 2 of this guide and in Annex B
of EN 1997-1 is adopted, in accordance with the National Annex applying in the country
in which the structure is located.
Although not explicitly stated in EN 1997-1, in Design Approach 1 (DA-1), combination
2 is normally the recommended method for overall stability checking in problems where
ground is the main element providing resistance (i.e. in mainly GEO-type limit states);
in such cases, Combination 1 is not relevant.
Treatment of permanent actions due to gravity loads and water is often difficult in
Design Approach 2 (DA-2) (and in DA-1 Combination 1 if relevant), since these loads
are usually unfavourable (i.e. they contribute to a positive driving moment) in part of the
sliding mass but favourable (i.e. they contribute to a negative driving moment) in
another part. As the limit between the two parts varies with the location of the point
about which moment equilibrium is checked, it is often difficult to assess the appropriate
partial factors for such permanent actions. Finally, note that the partial factors on
actions due to gravity loads and water are not the same as the partial factors on weight
densities, γγ (which are equal to unity – see Table A.4 in Annex A of EN 1997-1).
The authors recommend the following procedure for the selection of partial factors
for actions in DA-2 (and in DA-1 Combination 1, if relevant):
(a) Partial factors for all permanent actions (favourable and unfavourable), both
structural and geotechnical, including gravity loads due to ground and water, are set
equal to unity (instead of γG = 1.35 as recommended in Table A.3 of EN 1997-1 for
the unfavourable actions).
(b) The partial factor for variable unfavourable actions (e.g. traffic loads on the crest of
a slope) is set equal to γQ = 1.50/1.35 = 1.11 (instead of 1.5 as recommended in
Table A.3). The partial factor for variable favourable actions is set equal to zero, as
usual (see Table A.3).
(c) The missing partial factor (γG = 1.35) is accounted for at the end of the calculations,
as described in items 5 and 6 below.
In DA-1 Combination 2 and DA-3, partial factors on geotechnical actions are obtained
from set A2 of Table A.3, as usual.
Partial factors for ground strength parameters are selected from Table A.4 of
Clause 11.5.1(8) EN 1997-1. Clause 11.5.1(8) mentions that, in cases where slopes with pre-existing
failure surfaces are analysed for potential re-activation of the instability, partial factors
normally used for overall stability analyses may not be appropriate. The meaning of this
clause is that:
(a) Unit values of all partial factors (γF = γM = γRe = 1) should be used in the
determination of ground strength parameters along pre-existing failure surfaces by

198

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 11. OVERALL STABILITY

back-analysis, since a ULS is assumed, and the objective is to determine the actual
mean value of the mobilized shear strength along the ‘known’ failure surface.
(b) The level of confidence in shear strength parameters determined by such
back-analyses is higher than usual (e.g. compared with the usual cases where shear
strength parameters are determined from laboratory or field tests), since the
volume of ground involved is appreciably larger. Thus, lower values of the partial
factors for ground strength parameters (γM), compared with those given in Annex A
of EN 1997-1, may be appropriate in subsequent checks of potential re-activation
of the slope instability, after taking stabilizing measures (e.g. after water table
draw-down).
(5) Overall stability is checked according to the adopted Design Approach and the
appropriate partial factors for actions and ground strength parameters, using any of
the above calculation methods (‘assumed failure surface’, limit analysis or advanced
numerical methods).
If the effects of actions driving instability and the corresponding resistances are
calculated separately, overall stability is checked using expression (2.5) of EN 1997-1
(Ed £ Rd), where Ed is the design value of the effects of actions driving instability (e.g. the
overturning moment of the sliding mass), and Rd is the design value of the corresponding
resistance (e.g. the moment of the appropriate shearing resistance along the assumed
failure surface). In DA-2 (and in DA-1 Combination 1, if relevant), it is required to
account for the missing partial factor of permanent unfavourable actions (γG = 1.35 –
see item 4 above) by the modified expression: γGE £ Rd, where E is the value of the effect
of the actions calculated with the partial factors for actions as recommended in item 4
above.
(6) Many of the present-day computer codes do not provide separately the values of the
effect of actions driving instability (E) and the corresponding resistance (R) but instead
provide only their ratio (factor F ∫ R/E), which is the ‘overall factor of safety’. In such
cases, overall stability can be checked by the following procedure, using the ‘overall
factor of safety’ (F) and an auxiliary factor, called the over-design factor (ODF), defined
below.
In ‘assumed failure surface’ and limit analysis methods, the factor F is provided
directly (as the overall factor of safety) by an analysis with actions and ground strength
parameters factored as discussed in item 4 above. The ODF is then calculated in terms of
F, as described in paragraphs I(f) and II(f) below.
In advanced numerical methods (e.g. the finite element method), the factor F and
the ODF are normally calculated using the ‘strength reduction’ (or ϕ-c reduction)
procedure described below:
I. Method for DA-1 Combination 1, DA-2, and for the application of DA-1 Combination
2 and DA-3 when ground parameters are factored (by the partial factors γM) at the
beginning of the calculations:
(a) At each construction stage where overall stability is checked, the numerical
model is analysed using the appropriate design values of the ground strength
parameters (c¢d, ϕ¢d). The displacement (Dd) of a ‘control point’ is selected as the
control value. Such a ‘control point’ can be the centre of the crest of an
embankment, the edge of the crest of a slope or a retaining wall, the centre of
the footing of a structure, etc.
(b) The ground strength parameters are reduced (or increased, occasionally) by
applying a factor f > 1 (or f < 1, occasionally):
c¢dF = c¢d /f ϕ¢dF = arctan(tanϕ¢d /f)
(c) The model is re-analysed using the new values of the ground strength parameters
(c¢dF, ϕ¢dF) and the new displacement (DdF) is calculated.
(d) The procedure is repeated for several values of ‘f’, until the calculated
displacement (DdF) becomes appreciably large.

199

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

(e) The calculated values of (DdF) are plotted against ‘f’. The required factor F is
the value of ‘f’ (i.e. F = f) at the point along the curve where the calculated
displacement (DdF) starts to increase rapidly, indicating that failure is imminent.
(f) The ODF is obtained from the equation
ODF = F/γGγRe
where:
– γG is the partial factor for permanent unfavourable actions given in Table
A.3 of EN 1997-1. The recommended values of this factor are γG = 1.35
in DA-2 and DA-1 Combination 1, and γG = 1.00 in DA-1 Combination
2 and DA-3.
– γRe is the partial resistance factor for slopes and overall stability given in
Table A.14 of EN 1997-1. The recommended values of this factor are
γRe = 1.00 in DA-1 and DA-3, and γRe = 1.10 in DA-2.
The division of F by the partial factors γG, and γRe is required because these
factors were not used in the calculation of F. Specifically, the division of F by
γG = 1.35 in DA-2 and DA-1 Combination 1 aims to account for the ‘missing’
partial factor of unfavourable permanent actions (see item 4, above).
Based on the values of γG and γRe recommended in Annex A of EN 1997-1, the ODF
is related to the factor F as follows:

In DA-2: ODF = F/(1.35 × 1.10) fi ODF = F/1.485

In DA-3 and DA-1 Combination 2: ODF = F
In DA-1 Combination 1 (when applicable): ODF = F/1.35.

II. Method for the application of DA-1 Combination 2 and DA-3 when calculations
are performed using the characteristic values of ground parameters, with partial
factors (γM) applied at the end:
(a) At each construction stage where overall stability is checked, the numerical
model is analysed using the characteristic values of the ground strength
parameters (c¢k, ϕ¢k). The displacement (Dk) of a ‘control point’ is selected as the
control value. Such a ‘control point’ can be the centre of the crest of an
embankment, the edge of the crest of a slope or a retaining wall, the centre of
the footing of a structure, etc.
(b) The ground strength parameters are reduced by applying a factor f > 1:
c¢kF = c¢k/f ϕ¢kF = arctan(tan(ϕ¢k/f)
(c) The model is re-analysed using the new values of the ground strength parameters
(c¢kF, ϕ¢kF), and the new displacement (DkF) is calculated.
(d) The procedure is repeated for several values of ‘f’, until the calculated
displacement (DkF) becomes appreciably large.
(e) The calculated values of DkF are plotted against ‘f’. The factor F is obtained
from the equation F = f/γM, where γM is the partial factor for the ground
strength parameters and ‘f’ is the value at the point along the curve where the
calculated displacement (DkF) starts to increase rapidly, indicating that failure
is imminent. The reason for dividing by γM is that the factor F aims to provide
the margin of safety beyond that provided by the design values of the shear
strength parameters.
(f) The ODF is calculated by the equation ODF = F.
The above two methods (I and II) of the ‘strength reduction’ procedure do not
provide exactly the same value of the ODF if the stress–strain relationship of the ground
(used in the numerical model) is stress-path dependent and the construction sequence
involves non-linear stress paths (e.g. if undrained loading is followed by consolidation or
if loading is followed by unloading).

200

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 11. OVERALL STABILITY

In all the above methods, a value of the ODF equal to unity indicates that overall
stability is exactly adequate, i.e. the available margin of safety is exactly that required by
EN 1997-1. An ODF > 1 indicates that the available margin of safety is more than
adequate by EN 1997-1, while an ODF < 1 implies that safety is inadequate (but the
structure may not necessarily fail, since there may still exist some margin of safety which
is considered inadequate by EN 1997-1). In fact, the requirement ODF ≥ 1 is equivalent
to the overall stability requirement of EN 1997-1 (Ed £ Rd, expression (2.5)), because:
(a) In DA-1 Combination 1 and DA-2:
γGEk £ Rk/γRe fi (Rk/Ek)/γGγRe ≥ 1 fi F/γGγRe ∫ ODF ≥ 1
(b) In DA-1 Combination 2 and DA-3:
Rd/Ed ≥ 1 fi F ≥ 1 fi F/(1 × 1) ≥ 1 fi F/γGγRe ∫ ODF ≥ 1

Comparisons with the OFS in conventional design:

On the basis of the above arguments, and with the values of the partial factors recommended
in Annex A of EN 1997-1, checking overall stability according to the three Design
Approaches in Eurocode 7 – Part 1 (i.e. the requirement Ed £ Rd or, equivalently, ODF = 1)
corresponds to the following values of the OFS in conventional design, at least in cases
involving circular failure surfaces and no variable loads:
• DA-1 Combination 2 and DA-3:
OFS = γM = 1.25 (effective stress stability analysis)
OFS = γM = 1.40 (total stress stability analysis)
• DA-1 Combination 1 (if applicable):
OFS = γG = 1.35
• DA-2:
OFS = γG γRe = 1.35 × 1.10 = 1.485
The above conclusions indicate that, with the values of the partial factors recommended in
Annex A of EN 1997-1, DA-2 is the most conservative, while DA-1 and DA-3 give identical
results in typical problems. Equivalence between the three Design Approaches may be
achieved either by changing the value of the partial resistance factor (γRe) for slopes and
overall stability in Table A.3 of EN 1997-1, or by introducing a ‘model factor’ (according to
the principles of EN 1990). For example, if the partial resistance factor for slopes and overall
stability in DA-2 is changed to γRe = 1.0, the equivalent value of the overall factor of safety
will be reduced to OFS = 1.35.
Example 11.1 checks the overall instability of a cutting in stiff clay using the ‘assumed
failure surface’ method.

11.6. Serviceability limit state design

SLS design should demonstrate that ground deformations, under design actions and design
ground material parameters with partial factors equal to unity, will not exceed the serviceability
requirements for the structures and infrastructure on, or near, the particular ground. Clause 11.6(1)P
Estimates of ground deformations may be performed by analytical methods (typically for
simple problems) and by numerical methods such as finite elements. In the case of natural
slopes, clause 11.6(3) recommends that, since these methods usually do not provide reliable Clause 11.6(3)
predictions of the deformation, the occurrence of SLSs should be avoided either by limiting
the mobilized shear strength or by monitoring ground movements and specifying procedures
to reduce or prevent them, if necessary.

201

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

11.7. Monitoring
Clause 11.7 Clause 11.7 lists the conditions under which monitoring is required and the objectives of a
monitoring system in relation to overall stability requirements.

Example 11.1: overall stability of a cutting in stiff clay

This example checks the overall stability of a cutting in stiff, over-consolidated clay (Fig.
11.1), of height H = 10 m and slope inclination 1:2 (slope angle β = 26.6° with respect to
the horizontal). A building exerting a uniform pressure with the characteristic value
gk = 35 kPa (permanent unfavourable action) is located at a distance of 2 m from the
edge of the slope. The initial permanent groundwater table was located at a depth
d = 3 m below the original ground surface (before the excavation was made). It is
assumed that, due to the excavation, the water table is gradually lowered and eventually
reaches the configuration shown in Fig. 11.1.
As unloading of the stiff over-consolidated clay is the dominant design situation failure
mode in this case, drained stability failure is most critical. Thus, the stability analysis is
performed using effective-stress ground strength parameters (c¢, ϕ¢), in conjunction with
steady state piezometric levels corresponding to the water table shown in Fig. 11.1. Since a
‘first-time’ stability failure is investigated, ground strength parameters somewhat lower
than the peak values will be used. The characteristic values of the ground parameters are:
• Saturated weight density: γk = 20 kN/m3.
• Effective angle of shearing resistance: ϕ¢k = 28°.
• Effective cohesion intercept: c¢k = 10 kPa.
ULS analysis of the overall stability is performed using the Bishop method of slices. The
Bishop method of slices belongs to the ‘assumed failure surface’ methods, and investigates
circular potential failure surfaces. The computer program examines a multitude of
circular potential failure surfaces (several thousand in this case), and gives the minimum
F value and the corresponding circle.
The stability check is performed for all three Design Approaches of EN 1997-1 and the
conventional OFS method.
Although formally required by EN 1997-1, DA-1 Combination 1 is not relevant in this
case, because:
(1) structural strengths do not provide resistance against overall stability failure
(2) failure is controlled by uncertainty in the ground strength rather than uncertainty in
the actions (see Section 11.5 above).
If, however, an analysis is performed according to DA-1 combination 1, then:

2m 10 m

gk = 35 kPa

WT 3m

H = 10 m 1
Stiff clay
2

Fig. 11.1. Geometrical data of a cutting in stiff clay

202

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 11. OVERALL STABILITY

Table 11.1. Characteristic and design values of actions and ground strength parameters

DA-1 OFS
Combination 2a DA-2 DA-3 method

Characteristic value of ground weight density 20 20 20 20

(kN/m3)
Characteristic value of weight of water (kN/m3) 10 10 10 10
Characteristic value of surface surcharge 35 35 35 35
(permanent unfavourable) (kPa)
Partial factor for actions
For surface surcharge (permanent 1.00 1.00b 1.00 1.00
unfavourable), γG
For gravity (permanent unfavourable), γG 1.00 1.00b 1.00 1.00
Design value of ground weight density (kN/m3) 20 20 20 20
Design value of weight density of water (kN/m3) 10 10 10 10
Design value of surface surcharge, gd = γG gk 35 35 35 35
(kPa)
Characteristic value of angle of shearing 28 28 28 28
resistance, ϕ¢ (°)
Characteristic value of cohesion intercept, 10 10 10 10
c¢ (kPa)
Partial factor for ground strength parameters, 1.25 1.00 1.25 1.00
γM
Design value of angle of shearing resistance, 23 28 23 28
ϕ¢ (°)
Design value of cohesion intercept, c¢ (kPa) 8 10 8 10
Partial factor for earth resistance, γRe 1.00 1.10 1.00 1.00
a
Combination 1 is not relevant in this example
b
Adjusted factors. See discussion in Section 11.5 (item 4) in this chapter

(1) the design value of the effect of the actions will be calculated by applying the partial
factor value 1.35 to the corresponding characteristic value
(2) the design value of the shearing resistance along the failure surface will be calculated
by applying a partial material factor, equal to 1.0, to the corresponding characteristic
value.
SLS analysis of the cutting is not performed, for simplicity. Such analysis would require
the determination of ground deformations due to the excavation and an assessment of
their effects on the building located at the top of the slope, using numerical methods (e.g.
finite elements).
Partial factors for actions are obtained using the recommendations in Section 11.5.
The characteristic and design values of actions and ground strength parameters are
summarized in Table 11.1.
Figure 11.2 shows the critical failure surface and the minimum F value calculated for
DA-2. The ODF is
ODF = F/γGγRe = 1.494/(1.35 × 1.1) = 1.006 ª 1.00
i.e. safety against overall stability failure is exactly adequate (the slope is not over-designed).
In this example, the calculations according to the conventional design (OFS method)
are identical to those of DA-2, since all partial factors are equal to unity. Note that if
variable, unfavourable actions existed (e.g. variable surface loads on the crest of the
slope), DA-2 would use a partial factor
γQ = 1.50/1.35 = 1.11

203

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

1.494

35.00 kN/m

Fig. 11.2. Design Approach 2: the calculated minimum F value is 1.494

1.193

35.00 kN/m

Fig. 11.3. DA-1 Combination 2 and DA-3: the calculated minimum F value is 1.193

for those actions, and thus the results of DA-2 and the OFS method would be somewhat
different.
Figure 11.3 shows the critical failure surface and the minimum F value calculated for
DA-1 Combination 2 and for DA-3. As mentioned above, Combination 1 of DA-1 is not
relevant in this case. The ODF is
ODF = F/γGγRe = 1.193/(1.0 × 1.0) = 1.193 > 1.00

204

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 11. OVERALL STABILITY

Table 11.2. Results of the overall stability calculations

DA-1 OFS
Combination 2 DA-2 DA-3 method

ODF calculated for 1.193 1.00 1.193 1.00a

overall stability
a
In the OFS method, the required minimum OFS is assumed equal to OFSmin = 1.50, although lower
values (OFSmin = 1.30–1.40) are often used in drained analyses. Thus, the equivalent ODF for the
conventional design is ODF = F/OFSmin = 1.494/1.50 = 1.00

i.e. safety against overall stability failure is more than adequate (the slope is over-
designed by about 19%).
Table 11.1.2 summarizes the results of the overall stability calculations for all Design
Approaches and the OFS method.
The above calculations show that:
(1) DA-1 and DA-3 give identical results, as expected (see the discussion in Section 11.5
above).
(2) For the values of the partial factors recommended in Annex A of EN 1997-1, DA-2 is
practically equivalent to the conventional OFS method (for OFSmin = 1.50), and is
the most conservative of the Design Approaches in EN 1997-1.
The above conclusions are applicable to all overall stability calculations, i.e. not only in
the present example, since they result from the choice of the partial factors for actions and
resistances. For example, if the partial resistance factor in DA-2 is changed to γRe = 1.0
(instead of 1.1), the ODF will become
ODF = F/γGγRe = 1.494/(1.35 × 1.0) = 1.11
(instead of 1.00) making DA-2 approach the results of DA-1 and DA-3. It is noted that a
choice of γRe = 1.0 (instead of 1.1) in DA-2 is equivalent to requiring OFSmin = 1.35
(instead of 1.35 × 1.1 = 1.485).

205

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 12

Embankments

This chapter is concerned with the design requirements for embankments for small dams
and for infrastructure, such as road and railway embankments. The material is covered in
Section 12 of EN 1997-1. The structure of this chapter follows that of Section 12:
12.1. General Clause 12.1
12.2. Limit states Clause 12.2
12.3. Actions and design situations Clause 12.3
12.4. Design and construction considerations Clause 12.4
12.5. Ultimate limit state design Clause 12.5
12.6. Serviceability limit state design Clause 12.6
12.7. Supervision and monitoring Clause 12.7
The design requirements for embankments are closely related to the construction of fill
and to overall stability. A general reference is therefore made to Section 5, on fill, dewatering,
ground improvement and reinforcement, and to Section 11, on overall stability.
For readers familiar with the ENV version of Eurocode 7 – Part 1 it may be noted that that
version contained a section on embankments and slopes (Section 9). During the enquiry
period for the ENV version it became evident that combining embankments and slopes in
one section was inappropriate. For the present EN version of Eurocode 7 – Part 1 it was
therefore decided to divide the two items into two separate sections and to rename the
section on slopes to ‘Overall stability’.

12.1. General
Section 12 applies to embankments for small dams and for infrastructure. No definition, Clause 12.1(1)P
however, is given in EN 1997-1 for the word ‘small’. Since Eurocode 7 – Part 1 is aimed
mainly at the design of Geotechnical Category 2 structures, it is probably appropriate to
assume that ‘small dams’ include dams (and embankments for infrastructure) up to a height
of approximately 10 m.

12.2. Limit states

The different limit states that should be considered in the design of embankments are given
as a checklist in clause 12.2(2). It should be noted that limit states involving adjacent Clause 12.2(2)
structures, roads and services are included in this list. This concern for the possible adverse
influence of the embankment on adjacent structures, roads and services is likewise reflected
in the following discussion on clauses of Section 12 on actions and design situations, on
design and construction considerations, and on ultimate and serviceability limit state design.

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

It should also be noted that, of the 12 limit states listed, three are directly concerned with
water:

• failure caused by internal erosion

• failure caused by surface erosion or scour
• deformations caused by hydraulic actions.

However, the presence of water also plays an important role when designing an embankment
against loss of overall stability, failure in the embankment slope or crest, and loss of
serviceability and creep by climatic influences. The fact that so many limit states for
embankments are concerned with water demonstrates the importance of taking all aspects of
the presence of water into account in the design of embankments. This concern is also
reflected in the clauses of Section 12 discussed below.

12.3. Actions and design situations

Clauses 12.3(5)P In clause 12.3 on actions and design situations, it is again stressed how critically important it
to 12.3(7)P is to take the effects of water correctly into account in geotechnical designs. It is also stressed
that the most unfavourable groundwater condition within the embankment and free water
levels in front of the embankment should be selected when assessing the stability.

12.4. Design and construction considerations

Clause 12.4 Clause 12.4 on design and construction considerations contains a long list of items and
options that need to be considered in the design and construction of embankments. It should
Clause 12.4(1)P be noted that the first item listed in this clause is the need to take account of experience with
embankments on similar ground and made of similar fill material. This emphasizes the
importance of visiting the site of a future embankment and observing and evaluating the
stability and conditions of adjacent structures and utilities.

12.5. Ultimate limit state design

Clause 12.5 In clause 12.5, which is concerned with the ultimate limit state design of embankments, the
Clause 12.5(7)P most important clauses, clauses 12.5(7)P and 12.5(1)P, require that the provisions for
Clause12.5(1)P preventing hydraulic failure and fulfilling overall stability must be fulfilled. These provisions
are contained in Sections 10 and 11, and are discussed in Chapters 10 and 11 of this guide.

12.6. Serviceability limit state design

Clause 12.6(1)P Clause 12.6(1)P states that, in the serviceability limit state design of an embankment, it is
necessary to check that the deformation of the embankment does not lead to the occurrence
of a serviceability limit state in either the embankment or any adjacent structures, roads or
Clause 2.4.8(1)P services. It should be noted that the serviceability limit state check is carried out using partial
Clause 2.4.8(2) factor values which are normally taken as equal to unity. Thus, in the serviceability limit state
design of an embankment, the design values of the actions and material parameters, e.g.
stiffness, are normally equal to the characteristic values.

12.7. Supervision and monitoring

Clause 2.7 The design of embankments normally involves many possible combinations of actions – of
which water pressures are often the most important – and many possible failure modes and

208

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

CHAPTER 12. EMBANKMENTS

limit states. Use of the observational method supplementing design calculations is consequently
highly recommended for this type of structure. An essential element in the use of this
method is the reliance on supervision and monitoring. This is especially important in the Clause 5.3.3
selection, placing and compaction of fill for the construction of embankments for small dams
and for infrastructure.
Clause 12.7(4) provides a list of items for which it states records should be included in a Clause 12.7(4)
monitoring programme for an embankment. In the authors’ view, the word ‘should’ in this
clause is not entirely appropriate; a better wording is ‘may contain the following and other
records as relevant’, since which particular items should be included will depend on the
particular design situation.

209

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

References

Alonso, E. E., Josa, A. and Ledesma, A. (1984) Negative skin friction on piles: a simplified
analysis and prediction procedure. Géotechnique, 34, 341–357.
Armbruster, H. and Tröger, I. H. M.(1993) Underground erosion caused by raised water
levels of impounded rivers. In: J. Brauns, M. Heibaum and U. Schuler (eds), Filters in
Geotechnical and Hydraulic Design. Balkema, Rotterdam.
Armbruster, H., Heibaum, M. and Schuppener B. (1999) Principles of the recommendations:
stability of dams for waterways. In: Proceedings of the 12th European Conference on Soil
Mechanics and Foundation Engineering, Amsterdam, Vol. 2.
Baguelin, F., Frank, R. and Jézéquel, J. F. (1982) Parameters for friction piles in marine
soils. In: Proceedings of the 2nd International Conference on Numerical Methods in Offshore
Piling, Austin, pp. 197–214.
Bauduin, C. M. (2001) Design procedure according to Eurocode 7 and analysis of the test
results. In: Proceedings of the Symposium on Screw Piles – Installation and Design in Stiff
Clay, Brussels. Balkema, Rotterdam, pp. 275–303.
Bauduin, C. M. (2002a) Determination of characteristic values. In: U. Smoltczyk (ed.),
Geotechnical Engineering Handbook. Ernst, Berlin, Vol. I, pp. 17–50.
Bauduin, C. M. (2002b) Design of axially loaded piles according to Eurocode 7. In: Proceedings
of the 9th International Conference on Piling and Deep Foundations (DFI 2002), Nice.
Presses de l’ENPC, Paris, pp. 301–312.
Bligh, W. G. (1910) Dams, barrages and weirs in porous foundations. Engineering News, 64,
708–710.
Brinch Hansen, J. (1956) Limit Design and Safety Factors in Soil Mechanics. Danish Geotechnical
Institute, Copenhagen, Bulletin No. 1 [in Danish with an English summary].
Burland, J. B., Broms, B. B. and De Mello, V. F. B. (1977) Behaviour of foundations and
structures. In: Proceedings of the 9th International Conference on Soil Mechanics and
Foundation Engineering, Tokyo, Vol. 2, pp. 495–546.
Busch, K.-F. and Luckner, L. (1974) Geohydraulik. Enke, Stuttgart.
Caquot, A., Kérisel, J. and Absi, E. (1973) Tables de Butée et de Poussée. Gauthier-Villars, Paris.
Cistin, J. (1967) Zum Problem mechanischer Deformationen nichtbindiger Lockergesteine
durch die Sickerwasserströmung in Erddämmen. Wasserwirtschaft, 2.
De Cock, F., Legrand, C. and Huybrechts, N. (2003) Axial static pile load test (ASPLT)
in compression or in tension – Recommendations from ERTC3 – Piles, ISSMGE
Subcommittee. In: Proceedings of the 13th European Conference on Soil Mechanics and
Geothnical Engineering, Prague, Vol. 3, pp. 717–741.
EAU (1980) Recommendations of the Committee for Waterfront Structures. Ernst, Berlin.
EAU (1996) Recommendations of the Committee for Waterfront Structures. Ernst, Berlin.
European Commission (2003a) Guidance Paper L (Concerning the Construction Products
Directive – 99/106/EEC). Application and Use of Eurocodes, Version 27 November 2003.
EC, Brussels.

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

DESIGNERS’ GUIDE TO EN 1997-1

European Commission (2003b) Commission recommendation of 11 December 2003. Document

No. C(2003) 4639. Official Journal of the European Union.
Frank, R. (1999) Calcul des Fondations Superficielles et Profondes. Presses de l’ENPC et
Techniques de l’Ingénieur, Paris.
Frank, R. and Kovarik, J. B. (2005) Comparaison des niveaux de sécurité, calage d’un
coefficient de modèle pour la résistance ultime des pieux sous charges axiales. Revue
Française de Géotechnique, 110 (in press).
Gulvanessian, H., Calgaro, J.-A. and Holický, M. (2002) Designers’ Guide to EN 1990,
Eurocode: Basis of Structural Design. Thomas Telford, London.
Ingold, T.S. (1979) The effects of compaction on retaining walls. Géotechnique, 29, 265–284.
Institution of Structural Engineers (1955) Report on structural safety. Structural Engineer,
33, 141–149.
ISSMGE (1981)Lexicon in 8 Languages, 5th edn. Bryant Press, Ontario.
Lane, E. W. (1935) Security from under-seepage masonry dams on earth foundations.
Transactions of the American Society of Civil Engineers, 100, S.1233–S.1351.
Lumb, P. (1974) Application of statistics in soil mechanics. In: I. K. Lee (ed.), Soil Mechanics –
New Horizons. Butterworth, London, Ch. 3.
Mortensen, K. (1983) Is Limit State Design a Judgement Killer? Sixth Laurits Bjerrum Memorial
Lecture, 1982. Publication 148. Norwegian Geotechnical Institute, Oslo.
Orr, T. L. L. and Farrell, E. R (1999) Geotechnical Design to Eurocode 7. Springer-Verlag,
London.
Poulos, H. G. and Davis, E. H. (1980) Pile Foundation Analysis and Design. Wiley, New York.
Rethati, L. (1988) Probabilistic Solutions in Geotechnics. Developments in Geotechnical
Engineering 46. Elsevier, Amsterdam.
Schneider, H. R. (1999) Determination of characteristic soil properties. In: Proceedings of
the 12th European Conference on Soil Mechanics and Foundation Engineering, Amsterdam.
Balkema, Rotterdam, Vol. 1, pp. 273–281.
Schuppener, B., Walz, B., Weissenbach, A. and Hock-Berghaus, K. (1998) EC7 – a critical
review and a proposal for an improvement: a German perspective. Ground Engineering,
31, 32–35.
Sherard, J. L., Woodward, R. J., Gizienski, S. F. and Clevenger, W. A. (1963) Earth and
Earth–Rock Dams: Engineering Problems of Design and Construction. Wiley, New York.
Simpson, B. (2000) Partial factors: where to apply them? In: Proceedings of LSD2000:
International Workshop on Limit State Design in Geotechnical Engineering, Melbourne.
Simpson, B. and Driscoll, R. (1998) Eurocode 7 – A Commentary. Construction Research
Communications, Watford.
Smoltczyk, U. (1985) Axial pile loading test – part 1: static loading. ISSMFE Subcommittee
on Field and Laboratory Testing. Geotechnical Testing Journal, 8, 79–90.
Terzaghi, K. and Peck, R.B. (1967) Soil Mechanics in Engineering Practice. Wiley, Chichester.
Van Aalboom, G., and Mengé, P. (1999) The derivation of characteristic values of shear
strength parameters according to EC 7. In: Proceedings of the 12th European Conference on
Soil Mechanics and Foundation Engineering, Amsterdam. Balkema, Rotterdam, Vol. 1,
pp. 295–302.
Wastiaux, M., Ducroq, J. and Baguelin, F. (1998) Les fondations maritimes du Pont Vasco
de Gama. Travaux, 743, 33–41.

212

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

INDEX

Index Terms Links

a priori knowledge 29
acceptance tests, anchorage 140
accidental loads 24
actions 24
design effect of 32
embankments 208
embedded walls 147
ground movements 102–3
pile foundations 102–3
stability 196
structural 88 91
active earth pressure, horizontal component 175
allowable stress on the soil 39
alternative procedure 109 110–12 120–1 123–4
128
anchorage load (Pd) 134–8
anchorage resistance 138–9
anchorages 133–43
ground, load testing 140–3
structural design 139–40
ultimate limit state design 134–9
Application Rules 16
assumed failure surface 136 196 197 199
201 202
axial pile resistance (Rk) 116
axially loaded piles 105–14

back-analysis 198 199

barrettes 101
base sliding (Eh) 168
Bayesian techniques 29

This page has been reformatted by Knovel to provide easier navigation.

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

Index Terms Links

beam models
on ground spring supports (spring models) 157–9
with assumed earth pressure loads 157
bearing failure (Ev) 168
bearing resistance (Rv)
retaining structures 170 171
spread foundations 72–8 96
Bishop method of slices 197 202
block failure of group of piles 113–14

characteristic base resistance (Rb, k) 108 110 111

characteristic shaft resistance (Rs, k) 108 110 111 112
characteristic value
of ground property 44–5 49
of structural material 44
coefficient of earth pressure at rest (K0) 159
cofferdams 145
compressive ground resistance 106–11
compressive resistance failure 105–6
computed hydraulic gradient (ical) 191
cone penetrometer test (CPT) 53 56 109
confidence in characteristic value 26
construction checking 63
content of Eurocode 7
control investigations 54 63
Coulomb-type failure surface 163 168
creep piles 101
critical hydraulic gradient (icrit) 191

deadman anchorages 133 147

resistance in 139
definitions 16–17
derived values 56
Design Approaches 3–7 33–7 70 72
151–2 155–7
design by calculation 23–39
design investigations 54

This page has been reformatted by Knovel to provide easier navigation.

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

Index Terms Links

design methods
direct 70
indirect 70
design requirements 19–20
design situations 20–1
destabilizing total water pressure (ud) 188
destabilizing uplift action (Ud) 187
development of Eurocode 7 2–3
dewatering 66
displacements of retaining structures 161
downdrag (negative friction) 103
durability 21
dynamic impact (hammer blow) tests 112
dynamic load tests on piles 105

earth pressure determination 149–51

earth resistance 147
eccentricity loads 80–1 98–9
effective stress analysis 131
effective-stress conditions 197
embankments 207–9
actions 208
design and construction considerations 208
limit states 207–8
serviceability limit state design 208
supervision and monitoring 208–9
ultimate limit state design 208
embedded walls 145
actions 147
design and construction 148–9
earth pressure determination 149–51
geometrical data 148
serviceability limit state 146 160–84
ultimate limit state 146 151–60
EN 1537 133 139 140
Eurocodes 1–2
examples
avoidance of instability of a cutting in stiff
clay 202–5

This page has been reformatted by Knovel to provide easier navigation.

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

Index Terms Links

examples (Cont.)
characteristic value selection 41–2
concrete stem (gravity) wall 161–73
embedded wall supported by single row of
anchorages 173–84
indirect method using pressuremeter test 99–100
pile in compression from in situ test results 118–21
pile in compression from laboratory test
results 121–4
pile in compression from static load test
results 115–17
pile subject to downdrag 125–8
spread foundation for a tower 93–9
square pad foundation on soft clay 86–93
uplift of piled structures 128–32

failure by heave (HYD) 30 189–91

design using total stresses 189–90
design using submerged weight 190
determination of relevant pore water
pressure 190
to seepage of water in the ground 37
failure by uplift (UPL) 30 186–8 191
avoidance of 36–7
failure criterion (pile foundations) 106
failure in the ground (GEO) 30 31–2 32
failure in the structure (STR) 30 31–2
Fellenius method 197
fill construction 66
finite-element analyses 150 159–60 193 196
197
fixed-earth support condition 154
free-earth support condition 154
fundamental requirements 65

Geotechnical Categories (GCs) 63

Geotechnical Design Report 39–44 58 61 62
63 64

This page has been reformatted by Knovel to provide easier navigation.

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

Index Terms Links

geotechnical failure of anchorage 134

geotechnical investigations 53–4
geotechnical parameters
assessment 57
calibration factors 57
characteristic values 24–30 58
correction factors 57
directly obtained and derived values 56–7
evaluation 54–8
measured values and test results 55–6
‘other relevant data’ 57–8
gravity walls 145
ULS design 153
ground conditions, checking 63
ground improvement and reinforcement 66–7
Ground Investigation Report 58–9
ground movements, actions due to 102–3
ground properties 24
characteristic value 44–6
ground resistance 34 35 105
ground stiffness 161
ground tensile resistance 112–14
groundwater effects 66
retaining wall and 148
grouted anchorages 133 138
resistance in 139

hammer blow tests 112

heave 103
homogeneous soil 46–50
hydraulic failure 185–91
failure by heave (HYD) 30 37 189–91
failure by piping 192–3
failure by uplift (UPL) 30 36–7 186–8 191
internal erosion 191–2 193
hydraulic heave, ULS design of excavation
against failure 182–3

This page has been reformatted by Knovel to provide easier navigation.

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

Index Terms Links

impermeable layer, design against uplift by 188

inequality, resistance in 33
informative annexes 7–8
internal erosion 191–2 193
effects of material transport 192
filter criteria and hydraulic criteria 191
investigation tests, anchorage 143

kinematic constraints, retaining walls 147 149

limit analysis 196 197 199

limit designs, definition 2
limit equilibrium models (LEMs) 141–2
retaining structures 154 155–7
limit states 19–20
design procedure 23
embankments 207–8
pile foundations 102
stability 195–6
see also serviceability limit states; ultimate
limit states
limiting earth pressures (LEM) 180 181
limiting earth resistance (Rp, k) 159
limiting earth resistance (Rp, d), design value 159
limiting load, definition 2
limiting passive earth pressure 176 177
load–displacement graph 104
loads 24
load–settlement curve 81–2
load test report, pile 105
local population 28
lock-off force of prestress anchorages 139
log-normal distributions 49

mass–spring model 112

This page has been reformatted by Knovel to provide easier navigation.

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

Index Terms Links

middle-third model 80
mobilization factors 84–5
mobilized earth resistance (Rmob) 158 159
model pile procedure 109–10 118–20 122–3 128
Mohr–Coulomb theory 56 197
Mohr’s envelope 42–3
monitoring 64
Morgenstern–Price method 197

National Annex 5–7

national implementation of Eurocode 7 5–7
national tasks for implementation of EN 1997–1 9
Nationally Determined Parameters (NDPs) 5–6
net water pressure 175–6
non-linear spring models 150

observational method 39
over-design factor (ODF) 199 200–1 202–5
overall factor of safety (OFS) 4
equivalent deterministic 88 89 91
overall stability 202–5
on pile foundations 117
retaining structures 159
overall stability 70–2 195–205
overdig, unforeseen 154
in front of retaining wall 148
overturning moment (ME) 179

packages of EN Eurocode parts 8

partial factors
format of checking using 30
selection of 198
values 30 31
Peck earth pressure 154 157
permanent loads 24
phi-c reduction 199

This page has been reformatted by Knovel to provide easier navigation.

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

Index Terms Links

pile base resistance 27

pile-driving formulae 112
pile foundations
actions 102–3
design considerations 103–4
design methods 103
limit states 102
structural design 114–15
supervision of construction 115
vertical displacements 114
pile–ground displacement 125
pile load tests 104–5
dynamic 105
static 104
pile shaft resistance 27
pile space optimization 131–2
populations with trend 28
populations without trend 28
preliminary investigations 54
prescriptive measures 39
prescriptive method, spread foundations 85
pressuremeter tests (PMTs) 110
prestress anchorages 147
Principles 16
probability theory 2
proof load 143
pull–out resistance 138–9 141
p-y curve 102

references 13–14
regional population 28
resistance 31
design 32–3
in the ground 32
inequality 33
retaining walls 145–84
rock type characterization 55

This page has been reformatted by Knovel to provide easier navigation.

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

Index Terms Links

safety factors on loads 23

see also overall safety factors
schedule of EN 1997–1 8
screw anchorages 138
resistance in 139
serviceability limit state (SLS) 19 37–9
embankments 208
embedded walls 146 160–84
spread foundations 69 70 82–4
single pile tensile failure 129–30
single-source principle 24 131
sliding resistance 78–80 94–5 96 97–8
spread foundations (SLS)
settlement calculations 82–4
simplified 84–5
soil–pile interaction analysis 103
soil–structure interaction model 136
soil type characterization 55
soil volume 26–7
Spencer method 197
spread foundations structural design 85–100
based on indirect method using pressuremeter
test results 99–100
square pad foundation on soft clay 86–93
ULS design of spread foundation for a tower 93–9
stability
actions 196
design and construction 196
of ground mass 70–2
limit states 195–6
monitoring 202–5
overall 70–2 195–205
serviceability limit state design 201
ultimate limit state design 196–201
stabilizing moment (MR) 179
static equilibrium (EQU ) 30 31
static load tests on piles 104
statistical methods 28–30 44–51
strength reduction method 199

This page has been reformatted by Knovel to provide easier navigation.

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

Index Terms Links

structural actions 88 91
structural failure
of anchorage 134
due to foundation movement 81–2
submerged structures, failure by uplift 187–8
suitability tests, anchorage 140–1
supervision
definition 62–3
embankments 208–9
embedded walls 115
pile construction 115–32
symbols 17

tensile resistance of individual piles 113

Terzaghi earth pressure for multi-strutted walls 154 157
Terzaghi’s logarithmic settlement formula 81 82
total stress analysis 130–1
total-stress conditions 197
transverse loading 103
transversely loaded piles 114
t-z curve 102 103 125
t-z load transfer functions 114

ultimate compressive resistance Rc, d 106

from dynamic tests results 111–12
from ground tests results 109
from static load tests 107–8
ultimate limit state (ULS) 19 30–2
anchorages 134–9
avoidance 31
spread foundations 69 70 72–82
analytical calculation 74–8
semi-empirical calculation 78
types 30
ultimate load, definition 2
ultimate moment of resistance (Mu, d) 174–5

This page has been reformatted by Knovel to provide easier navigation.

https://www.geotashfeen.tech/
 https://www.geotashfeen.tech/

Index Terms Links

variable loads 24
vertical bearing resistance 94 96–7
Vx known 29 46–7 49
example 47–8
previous knowledge 49
Vx unknown 29 46–7
example 41–2

wall deflections, SLS design against 183–4

wall–ground friction and adhesion, retaining
walls 150–1
wall–ground interaction model 153–5 179
wall–ground interface resistance 147
water pressures, retaining structures 151
wave equation analyses 112
workmanship, requirements for 62

This page has been reformatted by Knovel to provide easier navigation.

https://www.geotashfeen.tech/