Validation of geotechnical finite element analysis

Validation d'analyse par lments finis pour la gotechnique

Brinkgreve R.B.J.

Delft University of Technology & Plaxis bv, Delft, The Netherlands

Engin E.

Plaxis bv, Delft, The Netherlands

ABSTRACT: The use of the Finite Element Method for geotechnical analysis and design has become quite popular. It is often the
younger generation of engineers who operate easy-to-use finite element programs and produce colourful results, whilst the responsible
senior engineers find it difficult to validate the outcome. The NAFEMS Geotechnical Committee has concluded that there is a need
for guidelines on validation of geotechnical finite element calculations. The first author is a member of this committee and the main
author of a reference document on validation of numerical modelling in geotechnical engineering. This paper contains the highlights
of the aforementioned document. After defining the term Validation, sources of discrepancies between a real project and its
corresponding finite element model are described. In addition, the paper presents various methods to validate geotechnical finite
element calculations. The paper ends with some conclusions and a list of references for further reading.
RSUM : L'utilisation de la mthode des lments finis pour l'analyse et la conception en gotechnique s'est gnralise. C'est
souvent la plus jeune gnration d'ingnieurs qui utilise des programmes d'lments finis et produit des rsultats avec des figures
pleines de couleurs, quand les ingnieurs seniors trouvent difficile la validation de ces rsultats. Le comit gotechnique NAFEMS a
conclu que des recommandations pour la validation des calculs gotechniques utilisant les lments finis sont ncessaires. Le premier
auteur est un membre de ce comit et l'auteur principal d'un document de rfrence sur la validation des modlisations numriques en
gotechnique. Cet article contient les points principaux de ce document. Aprs la description du terme Validation, les sources de
divergence entre un projet concret et le modle lment fini correspondant sont dcrites. De plus, cet article prsente des mthodes
varies pour valider des calculs lments finis en gotechnique. L'article se termine par les conclusions et une liste de rfrences pour
une lecture approfondie.
KEYWORDS: Finite element method (FEM), validation, verification, benchmark, numerical modelling, discrepancies.
1

Considering the use of geotechnical finite element software,

it is often the younger generation of engineers who perform the
numerical modelling and produce colourful results; sometimes
without fully understanding the backgrounds and limitations of
the constitutive models and the numerical methods used in the
software. Supervisors, i.e. project managers or senior engineers,
often find it difficult to validate the outcome, especially when
these do not match with what they would expect based on their
experience. This leads to the conclusion that there is a need for
guidelines on validation of geotechnical finite element
calculations, which was the primary motivation for the
NAFEMS Geotechnical Committee to write a publication on
validation of finite element models for geotechnical engineering
applications. This paper summarizes the main issues addressed
in the NAFEMS publication.
The next chapter 2 starts with a definition of the term
Validation and other related terms. Before elaborating various
methods of validating finite element models for geotechnical
applications in Chapter 4, an overview of possible discrepancies
between a real project and the corresponding finite element
model is presented in Chapter 3. The last chapter contains the
main conclusions of this paper.

INTRODUCTION

In the past decennia the Finite Element Method (FEM) has been
used increasingly for the analysis of geotechnical engineering
applications. Besides developments related to the method itself
the role of the FEM has evolved from a research tool into a
daily engineering tool. It has obtained a position next to
conventional design methods, and offers significant advantages
in complex situations. However, as with every other method, the
FEM also has its limitations. These limitations are not always
recognized by users of finite element software, which can lead
to unreliable designs.
Despite the development of easy-to-use finite element
programs, it is difficult to create a good model that enables a
realistic analysis of the physical processes involved in a real
project and that provides a realistic prediction of design
quantities (i.e. displacements, stresses, pore pressures, structural
forces, bearing capacity, safety factor, drainage capacity,
pumping capacity, etc.). This is particularly true for
geotechnical applications, because the highly non-linear and
heterogeneous character of the soil material is difficult to
capture in numerical models. When using the finite element
method, soil is modelled by means of a constitutive model
(stress-strain relationship) which is formulated in a continuum
framework. The choice of the constitutive model and the
corresponding set of model parameters are the most important
issues to consider when creating a finite element model for a
geotechnical project. It forms the main limitation in the
numerical modelling process, since the model (no matter how
complex) will always be a simplification of the real soil
behaviour. Hence, some features of soil behaviour will not be
captured by the model.

WHAT IS VALIDATION?

Validation is concerned with the accuracy at which a model

represents reality. In order to use the results from a model
reliably in the design process, a proper validation of the model
is required. Another term that is often used in relation to
validation is Verification. To give more insight in the meaning

677

Proceedings of the 18th International Conference on Soil Mechanics and Geotechnical Engineering, Paris 2013

of validation and verification, the modelling process of Reality

is divided into four steps, as visualised in Figure 1.
In the first step (1) the complex physical reality is abstracted
into a simplified conceptual model. The main aim of this
modelling step is to determine the crucial processes and to
reformulate reality in a conceptual model by applying valid
simplifications such that the main phenomena as observed in
reality are retained in the model.

approximations and errors, it is necessary to identify each of

these components and their role in and contribution to the
discrepancy as a whole. Identifying possible individual
discrepancies may result in an improvement of the model and a
possible reduction of the overall modelling error. It may also
enable a quantification of the variation of design quantities by
considering parameter uncertainties and their possible value
ranges. Discrepancies may be divided into the following
categories: Simplifications, Modelling errors, Constitutive
models, Uncertainties, Software and Hardware issues and
Misinterpretation of results.
3.1

Figure 1. Visualisation of the modelling process from Reality to a

Computer model, and the position of Validation and Verification.

The second step (2) is the translation of the conceptual

model into a mathematical model. The mathematical model is
the mathematical formulation of the processes identified in the
conceptual model. Examples of mathematical models are the set
of partial differential equations describing equilibrium in a
continuum, and the constitutive model (stress-strain
relationship) describing the soil deformation behaviour.
The third step (3) is the translation of the mathematical
model into a numerical scheme. This generally requires a
discretisation of the problem in space and/or time.
The fourth step (4) is the implementation of the numerical
scheme into a computer model using a programming language
or using a modelling package.
The process to verify that a model or method has been
properly implemented in a computer program is called
Verification (b). Validation, on the other hand, is the process to
make plausible that a computer model includes the essential
features for a real situation to be analysed and the results
obtained with the model are representative for the situation in
reality (a).
All the above steps may involve differences between the
computer model and reality. Considering existing dedicated
finite element software packages, it is the developers of such
packages who take most of the above steps and decide about the
mathematical formulation, the numerical schemes and the
implementation of models in their software. Therefore, the
responsibility of software developers mainly lies in the
verification of the software and the proper documentation of the
models and methods implemented in the software. For users of
existing software packages the division of the modelling process
into different steps is still relevant, although their position is
different. Starting from a practical engineering problem, users
first decide about the most relevant phenomena to be modelled
(conceptual model). By using an appropriate software package,
they select, apply and combine several of the implemented
models and methods to create a computer model for their
analysis. It is their responsibility to make plausible that the
model is a good representation of reality. Hence, the process of
validation is primarily their responsibility.
3

SOURCES OF DISCREPANCIES

In this chapter we will focus on the sources of discrepancies

between reality and finite element models. Since a numerical
model involves several components that may introduce

678

Simplifications

Simplifications are the results of modelling choices made by the

user of a software package. These are made in different parts of
the modelling process. Examples of simplifications are:
Geometrical simplifications
Selection of model boundaries
Simplifications in material behaviour
Simplifications in the construction process
For every simplification of reality the user needs to be aware
of its consequences.
3.2

Modelling errors

In addition to the aforementioned simplifications there is a

variety of other sources of modelling errors. Some of these can
be reduced when they are recognized; some can even be
completely avoided. Examples of modelling errors are:
Input errors
Discretisation errors (meshing)
Boundary conditions
Time integration
Tolerances (tolerated numerical errors)
Limitations in theories and methods (e.g. smalldeformation theory)
The process of validation can help to identify and quantify
such modelling errors.
3.3

Constitutive modelling

Probably the most important part of the numerical modelling

process is the selection of the constitutive model the
determination of the corresponding model parameters. Real soil
behaviour may involve several features that can be observed
and measured in lab tests and in situ, but which are still difficult
to capture in a homogenized continuum formulation. Apart from
the limitations of the constitituve models themselves with
respect to real soil behaviour, some typical issues related with
different aspects of constitutive modelling are highlighted here:
Non-uniqueness related to non-associated plasticity and
strain-softening
Undrained behaviour
Unsaturated behaviour
Software developers need to properly document the
constitutive models used in their software, whereas users need
to be aware of the typical issues related to constitutive
modelling in general and the possibilities and limitations of the
models used in their applications.
3.4

Uncertainties

In the aforementioned sections it was assumed that the

behaviour in reality would be fully known and that modelling
discrepancies are the result of the modelling process only. The
fact is that there are many aspects in a real project that are not
completely known (yet) or which cannot be measured
accurately. In other words, there are uncertainties about what
we need to model precisely to reflect the real construction
process and the conditions that are applied to the real structure
during its lifetime. Examples of uncertainties are:
Lack of soil data

Technical Committee 103 / Comit technique 103

Spatial variation of soil properties

Loading conditions during an earthquake
Future developments around the project to be designed
Design versus the actual construction
In order to deal with uncertainties, various methods are
available, such as:
Global safety factor approach
Partial factor approach
Probabilistic analysis
Parametric analysis
Users of finite element models in which such methods have
been implemented need to be aware of the possibilities and
limitations of these methods.
3.5

Software and hardware issues

Although some models or processes may seem to be uniquely

described by their mathematical model, the outcome of these
models or processes, when implemented in computer software,
might deviate from their original formulation; either
deliberately or by accident. Here, the focus is on specific
software and hardware issues that might lead to discrepancies in
the outcome of a computer model which cannot immediately be
influenced by users because they are
result of specific implementations made by the
developers of the software
result of the used operating system
result of the used computer configuration.
Examples of such software or hardware issues are:
Bugs (programming flaws in the application software)
Specific implementations of models (for example
rounding-off the corners of the Mohr-Coulomb failure
criterion)
Iterative solvers and their numerical solution tolerances
Parallel solvers (solution differences depending on the
number of threads or cores being used)
3.6

Misinterpretation of results

If the modelling process has been completed, the calculation has

finished successfully and results have been obtained, it is not
the end of the story. It should be realised that the computer
model does not directly provide the answer to the original
engineering problem. Therefore, a translation needs to be made
from the results of the computer model towards the engineering
and design issues. The translation and (mis)interpretation of
results may also lead to discrepancies between the real situation
and the computer model. Examples where misinterpretation of
results might occur are:
Interpretation of safety factors
Structural behaviour (if the structure is too much
simplified)
Overlooking essential details (in particular complex 3D
models)
In general: Insufficient knowledge and understanding of
the modelling software being used.
All this is subject of the validation process. In the next
chapter, various methods of validation and other procedures are
described in order to (im)prove the quality of finite element
models and the modelling results.
4

METHODS OF VALIDATION

In the previous sections several sources of discrepancies

between a real project and its finite element model have been
identified. In order for a particular project to manage the
uncertainties and to reduce the discrepancies, the finite element
model must be validated.
Before considering validation of a computer model for a
practical application, it is relevant to verify that the models and
methods implemented in a software package are reliable. In the

679

first place this is a responsibility of the software developers, but

also users should consider performing a verification of models
and methods that are relevant for the solution of their
engineering problem. Verification is done by comparing the
results of computer models for typical situations with known
solutions. Examples of such solutions are:
Analytical solutions of elasticity problems, plasticity
problems, constitutive models, dynamic problems,
bearing capacity solutions, solutions of flow and
coupled problems.
Limit equilibrium solutions (approximations) for global
safety factors or bearing capacities.
Upper and lower bound solutions (limit analysis).
Benchmarks (see Section 4.5).
After a proper verification of the models and methods to be
used in a finite element model, the finite element model itself
needs to be validated. Validation of the model as a whole will
not be enough to make plausible that the results that are
obtained from the model are representative for the real situation.
In fact, discrepancies in individual components may
accidentally cancel each other out if they are not validated
individually. The validation process should therefore also
comprise the individual components of the modelling process in
addition to validation of the integral model. This also gives
insight in the accuracy of the individual modelling components.
The following sections briefly describe the validation of
individual components of a finite element model.
4.1

Validation of constitutive models and parameters

The selection of a constitutive model should be based on an

evaluation of the capabilities (and limitations) that the model
has to describe the essential features of soil behaviour for the
application at hand. In that respect, the constitutive model
provides the qualitative description of soil behaviour, whereas
the parameters in the model are used to quantify the behaviour.
The composition of the model plus parameters can be regarded
as the artificial soil that is used in the finite element model,
which should be representative for the real soil behaviour in the
application. Before considering the numerical model in full
detail, it makes sense to evaluate the behaviour of the artificial
soil (= model + parameters) separately in particular stress
paths. Therefore it is useful to check the behaviour of the soil in
simplified soil lab tests simulations in element tests or using a
single stress-point algorithm.
The results of the lab test simulations can be compared with
real test data. This provides insight in the possibilities and
limitations of the model to describe particular features of soil
behaviour and the accuracy at which it does so. Moreover,
parameters could be optimised to make a best fit to the test
data. However, it should be noted that the stress paths, stress
levels and strain levels in the real application can be
significantly different than those in the soil lab tests. Hence, a
good fit between the results of a simulated test and the real test
data is not a guarantee that the artificial soil is a good
representation of the real soil in the practical application.
Nevertheless, the numerical simulation of soil lab tests is, in any
case, relevant to qualitatively understand the behaviour of the
artificial soil and should therefore be considered in the
validation process.
In contrast to soil lab tests, in-situ tests cannot be simplified
to a single stress point model. However, some in-situ tests can
still be modelled as a simplified boundary value problem in the
finite element method. The simplified modelling of in-situ tests
can be used to optimise stiffness and strength properties, and
they could be useful as part of the validation process. An
example of such a model test is the pressuremeter test, modelled
as a cavity expansion problem.
The validation of the selected soil model and parameters on
the basis of soil lab tests is not sufficient to make plausible that
the artificial soil will sufficiently resemble the real soil in the

Proceedings of the 18th International Conference on Soil Mechanics and Geotechnical Engineering, Paris 2013

engineering application. Since the soil stiffness and strength

properties are highly dependent on the stress and strain levels
(or void ratio) encountered as well as the loading history and
direction (anisotropy), it is necessary to estimate the stress
levels, the stress paths, the strain levels (or void ratio) and the
movement direction at different locations in the geometry and to
relate these to the conditions for which the model parameters
are deemed to be valid. The estimation may be based on
engineering judgement, but it might also be considered to
perform a preliminary analysis with a preliminary set of model
parameters in order to support the estimation. If necessary, soil
layers can be divided into sub-layers in which representative
values of model parameters are used.
As part of the validation of model parameters for the
engineering application it might also be considered to perform a
preliminary analysis on a semi one-dimensional soil column
representing the ground profile at the project location. In the
case that the project involves mainly vertical loading, the soil
column analysis can be used to check if the calculated
settlements match the expected settlements (based on
engineering judgement or conventional settlement calculations).
Some parameters will have a dominant influence on the
outcome of the numerical analysis whereas other parameters
may have little influence. In order to evaluate which parameters
have a high influence, a parametric analysis may be performed.
In a parametric analysis parameters may be varied individually
in order to evaluate their influence on the results (sensitivity
analysis), or combined in order to evaluate the variations in
results. Parameters with a high influence need to be given most
attention. Additional soil investigation may be required in order
to be able to determine these parameters more accurately in an
attempt to reduce the uncertainties in results.
After the final analysis with definite parameters has been
performed it is necessary to validate the stress levels, stress
paths, strain levels (or void ratio) and loading directions as
obtained from the finite element model and to check whether
these correspond with what has been assumed in the first place
and what is deemed to be valid for the selected parameters.
4.2

Validation of model boundaries

Model boundaries are introduced to limit the extent of the finite

element model and calculation time. It has to be validated
whether the outcome of the finite element model is not
influenced by the particular choice of the model boundaries
(Figure 2a vs. 2b). This can crudely be done by redoing the
analysis with model boundaries taken further away from the
main modelling object and comparing the results, but that may
be a time-consuming way of working. It should at least be
verified after any finite element analysis that changes in stress
and strain near the model boundaries are relatively small. This
is not required near (vertical) symmetry boundaries. However,
in the latter case it should be validated that the symmetry
conditions are properly applied.
In the case of a dynamic analysis, users should check that
there is no spurious reflection at the model boundaries. This is
primarily of interest for the vertical model boundaries. The best
way to check this is by creating an animation of the velocities in
the model. If the bottom boundary is taken at the top of a
bedrock layer, reflections may occur and are not unrealistic.

Figure 2. Generated initial stresses in a slope problem. a. Proper

distribution based on Gravity procedure. b. Gravity procedure with
inappropriate boundaries. c. Wrong distribution using K0-procedure.

680

4.3

Validation of initial conditions

In order to make an accurate prediction, it is necessary to

initialise the stress in the model as much as possible in
correspondence with the situation in reality (Figure 2a vs. 2c).
The initial situation in the model may involve total or effective
stress components, pore water pressures, pre-consolidation
stress, void ratio and other state parameters, depending on the
constitutive model(s) being used. Most soil constitutive models
involve at least some sort of stress-dependency. Moreover, the
initial stress state directly influences the forces in soil retaining
structures. In the case of time-dependent behaviour, the initial
state may have influence on the settlement rate. Therefore, the
validation of the initial conditions is a necessary part of the
validation process.
In an effective stress analysis, it is essential to create a
realistic distribution of initial pore water pressures. Simple
hydrostatic pore pressure distributions may be generated on the
basis of a phreatic level (Figure 3b), whereas more complicated
situations may require a separate groundwater flow calculation
to be performed (Figure 3c). In the latter case, realistic
hydraulic conductivities (permeabilities) are required, which are
often difficult to obtain from soil investigation data. That is why
modellers often abuse the phreatic level tool to create more
complicated pore pressure distributions based on non-horizontal
level sections. Care has to be taken with such an approach, since
in reality non-horizontal levels imply groundwater flow and
possibly non-hydrostatic pore pressure distributions. A jump
in the phreatic level should definitely be avoided, since this
would cause a similar jump in pore pressure all the way down in
the layer, which is highly unrealistic (Figure 3a).

Figure 3. Generated pore water pressure distribution in an excavation

problem. a. Wrong distribution based on a jump in the phreatic level
b. Improved distribution using interpolation between high and low head
under excavation c. Distribution based on groundwater flow calculation
(increased horizontal permeability)

Generated pore pressures should be validated against

measured pore pressure distributions in the field. It should be
validated that the pore pressure distribution is continuous and
smooth; jumps are suspicious and are likely to be the result of
a wrong way of modelling.
4.4

Validation of (the accuracy of) results

The previous sections focused on essential components of the

model that are part of the modelling process. It also needs to be
validated that the finite element mesh is fine enough to produce
sufficiently accurate results. In case of doubt, the model can be
recalculated with a refined mesh. After the individual model
components and the model as a whole have been validated, and
numerical results have been obtained, there are various ways to
validate the results for the practical problem as considered. The
following methods can be used to validate the results of finite
element models:
Comparison with measurements (if the project is
already under construction)
Comparison with design charts.
Comparison with experience and common practice
Comparison with simplified models (e.g. reduced
dimensions; 1D vs. 2D or 2D vs. 3D)
Comparison with other software.
When considering a project in an urban environment,
experiences with previous projects in the neighbourhood can be
of great help in the validation of numerical models, since soil
conditions may be quite similar. Here, it should be realised what

Technical Committee 103 / Comit technique 103

the differences are in loading conditions, duration of the

construction process and influences from adjacent buildings,
between the new project and the existing projects, and how that
affects the numerical model.
Care should be taken to use common practice and experience
to design a new project on a larger scale than the projects on
which common practice and experiences are based. Because of
the high non-linear character of soil behaviour, the design of a
larger system is not simply an extrapolation from a smaller size
project. This principle in the design process should also be
realised when using finite element models.
It might be worth to model a project with different software
packages and compare the results. No doubt, this will lead to
different results, whereby the engineer should realize that the
different packages may use different models and methods, and
there could be differences in the way how these models and
methods have been implemented. However, when it was
intended to create very similar models, the results should be less
than 10% different from each other to conclude that they are
actually similar. This is a necessary condition for a positive
validation, but it is not the only condition, since modelling
errors can still be made in one or both software packages, such
that the results are still within 10% difference from each other.
It could be that two errors in one model accidentally cancel each
other out and still lead to results that are more or less right. It
can also be that errors made in both packages lead to similar
results, which are both wrong. Therefore the validation of
numerical models purely on the basis of a comparison with
other software is not a sufficient validation. In fact, both models
need to be validated individually and other types of validations
(as described above) need to be performed as well.
Considering the modelling of the same situation in different
software packages brings us to the issue of Benchmarking.
4.5

validation process. In summary, benchmarks can serve the

following purposes:
To verify computer software.
To train unexperienced geotechnical engineers to help
them becoming familiar with numerical analysis.
To let modellers prove their competence in numerical
analysis of geotechnical problems.
To make modellers aware of differences in results for a
well-defined problem, irrespective of their origin. This
point highlights the importance of validation of
numerical models.
To highlight the importance to use appropriate
constitutive models.
To identify limitations of the present state of the art in
numerical modelling in practice (Carter et al., 2000).
To date this is still true.
4.6

A checklist of the various sources of discrepancies, as described

in Chapter 3, can be helpful to remind the numerical modellers
of the possible modelling errors that they could make. Thinking
about the various sources of discrepancies will increase the
awareness of possible mistakes and will lead to better computer
models. In the NAFEMS document (Brinkgreve, 2013) an
extensive checklist is given, based on various sources of
discrepancies. Moreover, a list of possible questions that
modellers may ask themselves as part of the validation process
is included in the document. The checklist and the list of
possible questions may also be used by managers and
supervisors to get an impression how well a model has been
validated by the engineer.
5

Benchmarking

A Benchmark, in the framework of validation and verification,

is a well-defined example problem for which a reference
solution exists, whereas the term Benchmarking can be defined
as the process to evaluate the variation in results from different
modellers or different computer software for a well-defined
example problem. Although the latter definition is probably
mostly related with how users translate the example problem
into a computer model and how they interpret the results, it can
also be used to benchmark different software packages against
each other or against the reference solution. According to
NAFEMS, a Benchmark is a standard test designed to probe the
accuracy or efficiency of a finite element system or model
(Baguley, 1994). This definition clearly addresses the role of the
system (hardware + software), but also involves the role of the
user in creating an appropriate and accurate finite element
model.
The solution of a benchmark example is not a theoretical
solution, but a reference solution that is considered to be a right
solution for that particular problem. Most benchmarks are
simplified practical problems for which no analytical solution
exists. Modellers can use a benchmark to check if they obtain a
similar solution with their own software. Since the solution is
obtained using numerical methods, a small deviation (few
percent) from the reference solution is likely to occur and is
quite acceptable. Larger deviations may still be acceptable,
depending on the type of problem and the level of detail that is
provided with the benchmark. Published benchmarks have
shown that quite large differences can occur, which underlines
the need for validation of numerical models.
A number of benchmark examples for geotechnical
engineering have been defined and published (e.g. Jeffries,
1995; Schweiger, 1998, 2002, 2006; Andersen et al., 2005).
In conclusion, benchmarks are not only useful for
verification purposes, but they are also relevant for the

681

Checklists

NON-TECHNICAL ISSUES

In addition to the technical issues related to the validation of

finite element models for geotechnical applications, there are a
number of non-technical issues involved with the validation
process. Such issues include decisions, responsibilities and
organizational issues, which are to be considered primarily by
the management of a company or a project. Nevertheless, it
should be realised by each individual working on numerical
modelling that these issues exist.
5.1

Availability of data

Besides knowledge and experience, another key issue to be able

to make an accurate finite element model is the availability of
data. This involves:
Geometric data
Soil data
Structural data (if structures are involved)
Data of external conditions (loads, water levels,
adjacent structures)
Information about the construction process
Soil data is probably the most important, although this is not
always recognized by project owners. In practice, there is often
a lack of soil investigation data because it costs money. It is
important to convince clients or project owners of the need of
sufficient and good quality soil investigation. It does not only
reduce the uncertainties in ground conditions, but it will also
facilitate the validation of model parameters, thereby reducing
the risk that the design is inadequate because it is based on
insufficient or wrong geotechnical data.
5.2

Responsibilities

Regarding the use of numerical models for geotechnical

engineering applications and the use of its results for
geotechnical engineering and design, four main responsibilities
can be identified.

Proceedings of the 18th International Conference on Soil Mechanics and Geotechnical Engineering, Paris 2013

Responsibility of the engineer (user of finite element

software)
Responsibility of the supervisor (manager)
Responsibility of the organisation (engineering
company)
Responsibility of the software developer (software
company)
It is the primary responsibility of the engineer (user of finite
element software) to create a computer model and to determine
the required parameters such that the model accurately
represents the real project and captures the phenomena that lead
to the quantities that need to be determined or interpreted from
the model (deformations, stresses, structural forces, flow, etc.).
This responsibility includes a proper validation of the model
and its components. It is also the responsibility of the engineer
to report any lack of data and the consequences thereof to his or
her supervisor or client.
It is the primary responsibility of the supervisor (manager)
of the modelling engineer or the project manager to check that
the model created and used by the project engineer is a reliable
model on the basis of which the project can be properly
analysed and/or designed with the required safety level. This
responsibility involves a check on how and to what extent the
model has been validated. For supervisors without advanced
numerical modelling experience themselves this may be
regarded as a difficult task, but it remains their responsibility.
This NAFEMS book is intended to provide at least some
guidelines for managers to discuss key elements of the
numerical modelling process with their engineers. Together
with his/her technical expertise and experience from other
projects, the supervisor should obtain a good impression of the
quality of the results obtained from a numerical model.
It is the primary responsibility of the organisation in which
numerical models are being used to create an environment in
which the importance and complexity of numerical modelling is
realised on all levels. If numerical modelling is part of their
activities, it should be included in their quality procedures. The
organisation should be structured such that there is sufficient
knowledge and room, not only to create numerical models but
also to validate models and to control the process from the early
stage of numerical modelling to the interpretation of the results
towards the geotechnical design. Just like any other subject,
numerical modelling is continuously evolving and new methods
become available. This requires organisations to invest in
facilities (literature, courses) to let their staff acquire the
necessary knowledge to remain up-to-date in order to use
numerical models with state-of-the-art technology in an
appropriate way.
It is the primary responsibility of the software developer to
produce software that has been sufficiently verified and that is
(ideally) free of programming errors. Moreover, it is also the
responsibility of the software developer to properly document
the models and methods that are implemented in the software
and make this documentation available to the user.

CONCLUSIONS

In this paper a summary is given of a NAFEMS document on

validating finite element models for geotechnical engineering
purposes (Brinkgreve, 2013). The document may be used by
engineers who actually build the numerical models and interpret
the results, as well as by supervisors and project managers who
are responsible for the overall design of a project. Validation, in
the context of this document, is the process to make plausible
that a finite element model includes the essential features for a
real situation to be analysed and its results are representative for
the situation in reality.
After defining validation and related terms, the paper first
describes sources of discrepancies between a real project and its
finite element model. Insight in the sources of discrepancies is

682

essential for a proper validation of the model and to reduce

modelling errors.
The main chapter is devoted to the various methods of
validation. The process of validation involves a validation of the
model as a whole, as well as a validation of the various model
components. Particular model components that need to be
validated are the geometry, the model boundaries, the material
(including soil) behaviour, the finite element mesh, the initial
conditions and the calculation phases. Results obtained from the
model should be checked against results obtained from other
analysis methods, design charts, experience, common practice
and measurements, if available.
The last chapter describes some non-technical issues related
with decisions, responsibilities and organisational issues to
control the quality of numerical modelling as part of the
geotechnical engineering and design process. In the first place it
emphasizes on the availability (or lack) of data and the need to
convince the client or project owner of the essence of good
quality soil investigation. In the second place, it highlights the
importance to spend time and money on education and training.
The latter is a common responsibility of the engineer and the
organization in which he or she is employed.
7

ACKNOWLEDGEMENTS

The authors would like to thank the NAFEMS Geotechnical

Committee for their feedback on the original document and for
their consent to publish this summarizing paper.
8

REFERENCES

Andersen K.H., Murff J.D., Randolph M.F., Clukey E.C., Erbrich C.T.,
Jostad H.P., Hansen B., Aubeny C., Sharma P., Supachawarote C.
(2005). Suction anchors for deepwater applications. Proc. Frontiers
in Offshore Geotechnics (eds. Gourvenec & Cassidy). Taylor &
Francis, London, 3-30.
Baguley D., Hose D.R. (1994). How to Understand Finite Element
Jargon, NAFEMS, London.
Brinkgreve R.B.J. (2013). Validating numerical modelling in
geotechnical engineering. NAFEMS, London.
Carter J.P., Desai C.S., Potts D.M., Schweiger H.F. and Sloan S.W.
2000. Computing and computer modelling in geotechnical
engineering. Proc. GeoEng 2000, Int. Conf. on Geotechnical and
Geological Engineering, Melbourne. Technomic, 1157-1252.
Jeffries R.M. (1995). Interclay II project. A coordinated benchmark
exercise on the rheology of clays. Final report. European
Commission, Nuclear science and technology, report EUR 16498.
NAFEMS i.c.w. ASME (2009). What is Verification and Validation?
Leaflet. NAFEMS, London. www.nafems.org.
Schweiger H.F. (2002). Results from numerical benchmark exercises in
geotechnics. Proc. 5th European Conf. Numerical Methods in
Geotechnical Engineering (ed. P. Mestat). Presses Ponts et
chaussees, Paris, 305314.
Schweiger H.F. (2006). Results from the ERTC7 benchmark exercise.
Proceedings NUMGE 2006 (ed. H.F. Schweiger). Taylor &
Francis, London, 3-8.