See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/311729864

Behavior of Modeling Piled-Raft System in Sandy Soil under Vertical Load

Conference Paper · October 2015

CITATIONS READS
0 188

3 authors, including:

Hussein Hameed Karim

University of Technology, Iraq
149 PUBLICATIONS   92 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Ground Penetrating Radar (GPR) View project

I am working with Electrical Resistivity Imaging (ERI) and GPR projects. View project

All content following this page was uploaded by Hussein Hameed Karim on 19 December 2016.

The user has requested enhancement of the downloaded file.

2nd Int. Conf. on Buildings, Construction and Environmental Engineering, 2015 PAPER IDENTIFICATION NUMBER 251 1

Behavior of Modeling Piled-Raft System in

Sandy Soil under Vertical Load
Kais T. Shlash1, Hussein H. Karim2, and Hussein H. Hussein3

Abstract—The piled raft system is a geotechnical complex The adoption of piled raft foundations concept in the design of
structure composed of three elements; soil, piles and raft. The pile groups is by no means new, and has been described by
foundations of piled raft provide an economical foundation several authors (including [2] – [7] among many others). In
where the raft isolated does not satisfy the requirements of
design. Under these situations, the addition of a limited number
the first years and because of the limited availability of
of piles may increase the ultimate load capacity and decrease the processing speed and computer memory, the use of numerical
settlement of piled raft system. The study discusses the behavior methods was limited for simple problems. But, with the rapid
of load-settlement of piled-raft foundation on sandy soil under growth in the computer technology, numerical methods are
vertical load by means of both full scale tests and finite element often utilized for full three dimensional techniques to solve
analysis. The effects of number of piles, ground water level, the complex problems [8].
spacing between piles and raft thickness are discussed. The full
scale test (compression loading) for piled raft 1.8m×1.8m × 0.6m
The design of conventional pile philosophy is depended on
contains 4 piles of 6 m length. A vertical load of 500 tons that piles carrying the entire load which are recognized as a
(simulated as a uniform surface load) and a pile spacing of 3 pile group, and no contribution is made through the raft to the
diameters have been used in this study for verification purpose maximum load capacity. The new trend in foundation
using PLAXIS-3D software to show compatibility of the engineering is to combine pile and raft foundations. Combined
numerical analysis with field test results. 9 finite element models system can be based on a different design philosophy [9].
have been prepared and run on the basis of the analyses
performed with verifications. It is noticed that the bearing
The study considers developing a numerical model for
capacity of piled raft system increases with increasing the analyzing piled-raft systems based on the finite element
number of piles and spacing without any significant effect of the program PLAXIS-3D [10]. The validity of the developed
raft thickness. The presence of water is found to increase numerical model is examined by comparing its results with the
settlement. results of full scale tests. This paper deals with studying the
effect of some important design parameters related to piles
Index Terms—Piled raft, Pile foundation, Bearing capacity, and raft dimension.
Settlement, Plaxis-3D

II. SOIL STRATIFICATION OF THE SITE

I. INTRODUCTION
The site soil profile was specified according to the
P ILED raft foundations are composite structures unlike
classical foundation where the building load is either
transferred by the raft or the piles alone. The piles transfer is a
geotechnical investigations results on two boreholes with their
Cone Penetration test (CPT) and standard penetration tests
(SPT). The stratigraphy of soil consists of two major layers.
portion of the building loads to the deeper and more severe
The higher layer consists of a very dense yellowish to brown
than the soil layers, and thus allows reducing the total and
sandy silt soil (from 0 to -5) m and the second layer consisting
differential settlement in areas very economical way. Piles are
of medium dense light brown to brownish yellow sand from
used up to a load level which can be the same order of size
(-5 m to -15). Laboratory tests involved triaxial and direct
and load carrying capacity of a single pile similar to or greater
shear tests were carried out on Shelby tube samples. The main
than that level [1].
properties of the soil of every soil layer, derived through
geotechnical investigations, the evaluation tests of the situ and
laboratory are shown in Table 1.

Manuscript received July 21, 2015; accepted August 23, 2015.

1
Professor, Civil Engineering Department, Al-Esraa University College,
Baghdad, Iraq. (e-mail: [email protected])
2
Professor, Building and Construction Engineering Department, University
of Technology, Baghdad, Iraq. (e-mail: [email protected]).
3
Ph.D in Geotechnical Engineering, Building and Construction Engineering
Department, University of Technology, Baghdad, Iraq. (corresponding author:
e-mail: [email protected])
2nd Int. Conf. on Buildings, Construction and Environmental Engineering, 2015 PAPER IDENTIFICATION NUMBER 251 2

Table 1. A summary of soil properties of the study site.

III. STATIC LOAD TEST Fig. 2. The entire load frame in the present study site.

A full real test was conducted and applied on Karbala soil

by load testing according to [11] in order to obtain real results.
The dimension of the piled raft foundation was (1.8 m×1.8m ×
0.6 m) and the piles are distributed in the same manner
maintaining a typical spacing (three times pile width = 0.9 m)
between every two adjacent piles with 6 m pile length.
The compressive load was applied using from one to three
hydraulic jacks having a capability of 500 tons for each that
placed among the head of cap and the main steel beam. An
appropriate square reinforced concrete cap was casted on the
head of the test foundation to enable the transfer of the applied
load uniformly. In addition, 25 mm thick steel plates were
installed on the cap head. The settlements versus the vertical
applied loads are shown in Fig. 1. This Figure illustrates curve
representing the load settlement behavior of piled raft
Fig. 3. Setting hydraulic jacks dial gages in the present study site.
foundation (PRF). Fig. 2 shows the entire load frame and load
concrete blocks. While, Fig. 3 shows setting the hydraulic
jacks and displacement gages over the cap.
IV. VERIFICATION WITH FIELD PILED RAFT TEST
As explained in static load test in previous section, a real
scale model for foundation of piled raft has been executed and
tested in the Karbala soil. The raft and piles are made of
concrete, while the soil was dry sand. The results of (2×2)
piled raft foundation (Fig. 4) depicts the layout of the piled-
raft foundation considered in this analysis as a reference for
checking the numerical solution implemented by PLAXIS 3D
program. The model sand ground was modeled during
utilizing the (HS small) model having the parameters listed in
the Table 2. Fig. 5 shows the quarter of the problem analyzed
by PLAXIS 3D program. While, Fig. 6 presents the mesh of
the finite element of the vertical loading, taking into account
the elastic behavior of the piled raft and the ealsto-plastic
behavior of sandy soil by incorporating the (HS small) model.
Fig. 1. The load settlement behavior of pile raft foundation.
2nd Int. Conf. on Buildings, Construction and Environmental Engineering, 2015 PAPER IDENTIFICATION NUMBER 251 3

Fig. 6. Mesh of the finite element of piled raft model (2×2) for vertical
Fig. 4. The Problems of piled-raft (2×2) used for verification (all dimensions loading.
are in mm).
By plotting a load-settlement curve for the results, it can
Table 2. Material properties of the sand adopting soil model (HS small). be realized that there is a very good matching (93%) between
the finite element and experimental results. This indicates that
these results agreement. Fig. 7 presents load-settlement curve
behavior for the piled raft foundation with a comparison
between the experimental and PLAXIS-3D results for the case
(2×2).

Fig. 7. Load-settlement curve showing a comparison between experimental

and PLAXIS-3D results for case (2×2) PRF.

Parametric Study
The behavior of piled raft foundation is influenced by
several parameters in the design of piled raft. It is important to
take into consideration these parameters to achieve the
objective of economic construction with a satisfactory
performance. The parametric study presented in this paper was
Fig. 5. Quarter of the problem of piled raft (2×2) as executed by PLAXIS-3D. carried out with a computer program based on the finite
element method. The reliability of the numerical simulation by
PLAXIS-3D was verified by real scale tests as shown in Fig.
8. Based on these results, this work presents the influences of
the parametric study for investigating the optimum design for
a piled raft foundation. The piled raft model is considered to
2nd Int. Conf. on Buildings, Construction and Environmental Engineering, 2015 PAPER IDENTIFICATION NUMBER 251 4

be subjected to vertical load. The study investigates the effects

of changing the parameters on the load-settlement behavior
such as: number of piles, raft thickness, the spacing between
piles and water table depth. This study will help practicing
engineers in recognizing the preeminent and economical
qualities of optimizing pile arrangement in designing piled
raft.

Influence of Pile Spacing

The effect of the pile spacing (3d to 7d) on the piled raft
behavior is studied for the settlement of the piled raft for a
value of intensity of loading as 500 tons. In this analysis, the
raft thickness is 0.6 m and the dimension of the raft will
increase with increasing pile spacing. The piles are
30cm×30cm in dimension and 6m in length. Numerical
Fig. 9. The distribution of deformation of pile spacing 3d between piles.
analysis output of PLAXIS-3D for the pile spacing ranging
from 3d to 7d have been presented in the load settlement
behavior as shown in Fig. 8. The output shows:
1. Increasing the spacing of piles from 3d, 5d and 7d leads
the decrease the total Settlement of the piled raft
foundation by 54.7% and 34.2% respectively.
2. This parameter is not safe because duo the increasing of
spacing between piles, the load transformed the soil
beneath the raft will be increased.

Fig. 10. The distribution of deformation of pile spacing 5d between piles.

Fig. 8. Influence of pile spacing on load settlement behavior.

The present study shows that the maximum spacing of

piled raft system be 5d must because for any increase more
than this distance will cause a differential settlement with
transmission of carrying loads from piles to raft. This will
cause an increase in the displacement at the top of raft with a
decrease around the piles as shown in Figs. 9 to 11.
Fig. 11. The distribution of deformation of pile spacing 7d between piles.

Effects of the Number of Piles

Increasing the number of piles is another option that is
used commonly in design to reduce the total and differential
settlements. This section investigates the effects of increasing
the number of piles on reducing the settlements of piled rafts
subjected to maximum loads 500 tons. For piled raft models,
the raft thickness is 0.6 m and pile length is 6.0 m. The
2nd Int. Conf. on Buildings, Construction and Environmental Engineering, 2015 PAPER IDENTIFICATION NUMBER 251 5

numerical analysis output of PLAXIS-3D for the number of

piles 4, 9 and 25 have been presented in the load settlement
behavior as shown in Fig. 12. From this figure it can be
observed that when the settlement is 20 mm, the ultimate
loads at the number of piles 4, 9 and 25 are 500, 800 and 1450
tons respectively. But this effect is due to an increase number
of piles as well as due to increasing raft dimensions. The
general trend explains that increasing the number of piles will
decrease the settlements of a piled raft. However, increasing
the number of piles also adversely affects the economical
design and it should be carefully minimized.

Fig. 13. Influence of the raft thickness on load settlement behavior applying
uniform load.

Effects of the Water Table Level

A piled raft system with 4 piles supporting rafts with piles
sizes of 30×30 cm length of piles of 6 m, spacing of piles is
3d and varying water depth (0.0, -3, and -10 m) for a
maximum load 500 tons applied as uniform distributed load
on square rafts of 1.8×1.8 m have been investigated to the
water table depth influences on pile raft foundation behavior.
Quarter of the problem is analyzed using the finite element
Fig. 12. Influence of the number of piles on load settlement behavior. program PLAXIS-3D. Fig. 14 shows the maximum piled raft
settlement at center for various water depths. From this study,
Influence of the Raft Thickness it was found that the water table has an effect on the load
A piled raft system with 25 piles supporting rafts have settlement relationship of piled-raft foundations in the case of
been investigated to the raft thickness influences on pile fully saturated. The output shows:
raft foundation behavior with the following characteristics:
• In the absence of water, (dry case) the maximum settlement
varying thicknesses (0.45, 0.6 and 0.9 m) with piles sizes of
is 20.1 mm at total applied load 500 tons. When the water
30×30 cm, length of piles of 6 m, spacing of piles is 3d for a
level is lifted up to half the length of piles (partially
maximum load 3125 tons applied as uniform distributed load
saturated), note the slightly change in the maximum
on square rafts of 22.5×22.5 m with different thicknesses.
settlement to 23.1 mm applying the same load.
Quarter of the problem is analyzed using the finite element
• When the water is on the ground surface (fully saturation),
program PLAXIS. Fig. 13 shows the raft center settlement for
the total and differential settlement of the piled raft
various raft thickness in this study. It was found that the raft
foundation increases by about 53.3% compared with dry
thickness has no effect on the load settlement relationship
condition due to increase in shear stress as shown in Fig. 15.
of piled-raft foundations. Other researchers reported the
similar observations regarding the effect of raft thickness.
“Reference [12] reported that raft thickness has little effect on
the maximum settlement of piled-raft foundations on sand
soil.” While, “reference [13] reported finite element analyses
of piled-raft foundations and showed that the raft thickness
has little effect on maximum settlement in soft cohesive soils.”
From the present results, it can be concluded that increasing
the raft thickness is effective primarily in reducing the
differential settlement.
2nd Int. Conf. on Buildings, Construction and Environmental Engineering, 2015 PAPER IDENTIFICATION NUMBER 251 6

respectively. This slight increase of settlement is due to

the increase of the dead loads of cap.
4. When the water approaches the ground surface, the
total and differential settlement of the piled raft
foundation increases by about 53.3% compared with
dry condition therefore, the piled raft system is not
advisable to be used in the saturated soils and water
treatment plants even if the soil is dry for fear of water
leak in the future.

REFERENCES
[1] F. Hartman, and P. Jahn, “Boundary element analysis of raft
foundations on piles,” Kluwer Academic Publishers, and Netherlands,
Meccanica, vol. 3, 2001, pp. 351–366.
[2] J. Hooper, “Observations on the behavior of a piled-raft foundation on
Fig. 14. Influence of the water table on load settlement behavior. London clay,” Proc. Inst. Civ. Engrs., vol. 55, No. 2, 1973, pp. 855-
877.
[3] E. Davis, and H. Poulos, Pile foundation analysis and design, Wiley,
1980.
[4] J. B. Burland, B. B. Brooms, and V. F. B. De Mello, Behavior of
foundations and structures, in Proc. of the 9th International
Conference on Soil Mechanics and Foundation Engineering, Tokyo,
Japan, vol. 2, 1977, pp. 495-546.
[5] Prakoso, W.A. and Kulhawy, F.H., “Contribution to piled raft
foundation design,” J. Geotechnical and Geoenvironmental
Engineering, vol. 127, No. 1, 2001, pp. 17-24.
[6] O. Reul, and M. F. Randolph, “Piled rafts in over consolidated clay:
Comparison of In situ Measurements and Numerical Analyses,”
Geotechnique, vol. 53, No. 3, 2003, pp. 301-315.
[7] R. Katzenbach, U. Arslan, C. Moormann, and O. Reul, “Piled raft
foundation interaction between piles and raft,” Darmstadt Geotechnics,
Darmstadt Univ. of Technology, No. 4, 1998, pp. 279-296.
[8] M. W. O’Neill,”Piled Rafts in Geotechnical Practice – An international
perspective,” Proc. 10th. International Conference on Piling and Deep
Foundations, Amsterdam, 2006, pp. 197-208.
[9] H. G. Poulos, “Methods of analysis of piled raft foundations,” A report
Fig. 15. Shear stress distribution duo to water level at surface. prepared on behalf of Technical Committee TC18 on piled foundations,
2001.
V. CONCLUSION [10] PLAXIS, Vermeer and Brinkgreve: Manual, 2004.
[11] ASTM D1143: Standard test method for piles under static axial
In the light of full scale tests and finite element analysis, compressive load, American Society of Testing and Materials, 2007.
the following conclusions can be drawn: [12] E.Y. N. Oh, M. Huang, C. Surrak, R. Adamec, and A .S.
Balasurbamaniam, “Finite element modelling for piled raft foundation
1. Increasing the spacing of piles from 3d to 5d and 7d in sand,“ Eleventh East Asia-Pacific Conference on Structural
decreases the maximum settlement of the piled raft Engineering & Construction (EASEC-11), Building Sustainable
foundation by about 54% and 34%, respectively. Environment, Taiwan, 2008, pp. 1-8.
[13] A. K. Singh, and A. N. Singh,“Experimental study of piled raft
Increasing piles spacing > 5d is not safe as the load foundation,” Proc. of Indian Geotechnical Conference, December 15-
transferred with a considerable increase from the pile 17, 2011, Kochi (Paper No. D-378), 2011.
to the raft foundation, therefore, the maximum spacing
between the piles must be not exceed 5d to improve the
piled raft system.
2. The total settlement of the piled raft decreases due to
increasing the number of piles when the settlement is
20 mm, the ultimate loads at the number of piles 4, 9
and 25 are 500, 800 and 1450 tons respectively. This Prof. Dr. Hussein Hameed Karim received the
parameter is considered uneconomic because the gain B.S. and M.S. degrees in Earth Science/ Geophysics from the
in bearing capacity is small. University of Baghdad, in 1978 and 1982 respectively, and the Ph.D.
3. The thickness of the raft (pile cap) has no effect on the degree in Engineering Geophysics from University of Baghdad in
bearing capacity of the piled raft foundation. When the 1996, Iraq.
From 1983 to 1997 he was within the academic staff of
raft thickness is increased from 45, 60, 90 and 180 cm,
University of Basrah and graduated from Asst. lecturer to Asst.
for example, the maximum settlement of the piled raft Professor. Since 1997 he is within the academic staff of University of
foundation increases by 1.6%, 0.2% and 1.25% Technology. Since 2000, he has been a Professor with the Building
& Construction Engineering Department/ University of Technology.
He is the supervisor of more than 45 post graduate (M.Sc. and Ph.D)
2nd Int. Conf. on Buildings, Construction and Environmental Engineering, 2015 PAPER IDENTIFICATION NUMBER 251 7

students, the author of 5 books, and one under preparation, more than
100 scientific researches, more than 100 scientific articles, and two
patents. His research interests include Engineering Geophysics, Soil
& Rock Mechanics, Site Investigation, and Geomatics Engineering.
Prof. Karim was chosen by the Committee of University of
Technology as the 1st Professor in the University of Technology for
the academic year (2010-2011). Prof. Karim’s awards and honors
include training session to Bordeaux University- France (4 months,
1984); Visiting Professor to University of Hodeidah University-
Yemen (3 months, 2002); Visiting Professor (one month, 2010) to
Malaya University- Kuala Lumpur, Malaysia; and 3 Professor Visits
sponsored by DAAD (Germany), Bauhaus University- Weimar 2006,
Rhur University – Bochum 2010 and 2014 (3-months for each).

View publication stats