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

net/publication/334398780

Study simulation of shallow foundation behavior using different finite

element models

Article in Journal of Advanced Civil Engineering Practice and Research · June 2019

CITATIONS READS

12 2,041

2 authors:

Mohanned Qahtan Waheed Noor Moutaz Asmael

University of Technology, Iraq Mustansiriyah University
31 PUBLICATIONS 97 CITATIONS 63 PUBLICATIONS 221 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Noor Moutaz Asmael on 11 July 2019.

The user has requested enhancement of the downloaded file.

 Journal of Advanced Civil Engineering Practice and Research
2550-1933 © Ababil Publishers
www.ababilpub.com/jacepr
http://dx.doi.org/xx

Research Article

Study simulation of shallow foundation behavior using different finite element models
1,* 2
M. Waheed , N. Asmael
1
Civil Engineering Department, University of Technology, Baghdad, Iraq
2
Highway and Transportation Engineering Department, Faculty of Engineering, Al-Mustansiriyah University, Baghdad, Iraq
* Corresponding author. Tel. +9647700708488; Email: [email protected], [email protected] (N Asmael).

HIGHLIGHTS

• Simulation of shallow foundation using different FEM was studied.

• Behavior of small and large scale shallow foundation in clayey soil was determined.
• For a small scale, the hardening soil model is more realistic.
• In case of large scale, the three simulation models can be used.
• The hardening soil model is the best for a large scale.

KEYWORDS ABSTRACT

Shallow foundation The behavior of the shallow foundation depends mainly on the soil supporting
Finite element models (FEM) it. The numerical programs provide simulation of soil behavior in large scale
Plaxis-3D modeling that cannot be conducted practically. It represents a great benefit of numerical
Hardening soil model analysis. The computer program should include the appropriate material model
Clayey soil to produce realistic and logical results. This paper aims to study simulation of
Behavior shallow foundation using different models of finite elements and comparing the
numerical result with experimental one. The Plaxis-3D program is used for
ARTICLE HISTORY developing the numerical models. Three models are utilized during modeling
soil behavior, i.e. Mohr-Coulomb, Hardening and soft soil model. The behavior
JACEPR-2018008 of small and large scale shallow foundation under axial loading in clayey soil
Received 17 October 2018 was studied. The purpose of this study is to assess and compare the result of
Received revision 29 January 2019 modeling with the experimental one to find which model can better predict the
Accepted 5 February 2019 soil behavior. It is found that for the small scale foundation, the simulation of
Available online 23 June 2019 the Mohr-Coulomb model matches the load-settlement curve at the initial
Volume: 8, Issue: June 2019 straight portion and then over predicts, so a great care must be taken to use this
(30 June 2019) model. The soft soil (creep) model gives the straight line response to a wide
range which is invalid (compared with the experimental curve). The
representation using the hardening soil model is more realistic and gives better
results. In case of the large scale foundation, these three models can be used for
simulation, where the simulation of the hardening soil model is the best.
© 2019 Ababil Publishers. All rights reserved.

1. Introduction for the researcher [1-4].

Shallow foundation is a part of a structure that transmits the The complex soil nature makes the development of
load directly to the underlying soil. The behavior of the constitutive models for simulating the soil behavior represent
foundation depends mostly on the soil that supports it, therefore, the key aspect in the analysis of geotechnical structures,
it is necessary to understand and study the behavior of the soil however, despite the complex soil nature there was a recent
under loads and the factors that affect it. development of new constitutive relations [1-4].

The behavior of soil compared with other materials is more There is a great effort, rising daily, to develop more accurate
complex because of the nature of its components; the trend model in business codes to help designers to solve geotechnical
toward analyzing soil behavior using different models expanded problems by taking into account more realistic behavior of soil
due to development of computer technology and finite [5].
component programs.
The task of this paper is studying simulation the behavior of
The computer program used should include the appropriate shallow foundation using different models of finite elements
material model to produce more realistic and logical results, and comparing the results with experimental and knowledge of
choosing the appropriate model for analysis is the most the most representative of soil behavior.
important factor for the successful representation of behavior

Copyright © 2019 Ababil Publishers. All rights reserved.

http://www.ababilpub.com/download/jacepr8-2/
Journal of Advanced Civil Engineering Practice and Research, 8:4-9, 2019
Waheed and Asmael / Study simulation of shallow foundation behavior using different finite element models 5

2. Modeling of soil behavior case study is selected from the experimental field tests
performed by Ornek et al. [6] of natural clay deposits, where the
In the present paper, a numerical modeling of the behavior footing diameter equal to (0.9 m). The test was performed at the
of a shallow foundation is performed using the Plaxis - 3D Adana Metropolitan Municipality's located in the western part
program. Plaxis is a computer package and it is designed of Adana, Turkey. The soil information is obtained from the
specially to analyze geotechnical problems. It is a results of the investigations after conducting four boreholes.
three-dimensional program used to simulate deformation
analyzes of different types of foundations and geotechnical Three layers of soil are outlined in the profile area, as shown
problems. This program provides the automatic construction of in Fig. 1. The first layer is 0.8 m thick and has been removed
two and three dimension meshes, which allows the user to before test. The second layer is6.2 m thick of high plasticity
create quickly a final three-dimension model according to the silty clay (CH). The third layer is 3.5 m thick of silty clay with
properties of materials. sand of low plasticity (CL). The distribution of standard
penetration test values with depth conducted during excavation
of borehole and soil profile of the site can be seen in Fig. 1.

According to the laboratory test performed on the soil, the

clay content, water content and the unconfined shear strength
(qu) of the soil along the depths are shown in Fig. 2. It can be
seen the values of the undrained cohesion (cu) of clay a bout (70
kPa).

3. Models for simulating soil behavior

The simulation of the experimental case study is conducted

using Mohr-Coulomb, hardening, and soft soil model. A
summary of each model is explained in the following.

Fig. 1 SPT values and soil profile of the site [6]. 3.1 Mohr-Coulomb material model

This model assumes soil behavior of elastic-perfectly

plastic, it was considered most commonly used than other
models. This model is sufficient for some problems in
geotechnical area, especially when it was used by experienced
engineers.

This model is relatively simple and can be easily used as

five input parameters are required; elasticity modulus (E) and
Poisson’s ratio (ν) for the elasticity, cohesion (c) and angle of
friction (Ø) for plasticity and angle (ψ) for the dilatancy. The
model is isotropic and do not take into account
stress-dependency of soil.

The main important strengths of this model include:

• The parameters used to define stiffness are well known
(i. e. modulus of elasticity and Poisson's ratio).
• Input parameters can be easily obtained from various soil
tests.
• Gives acceptable results in the elastic range of stress
state.

3.2 Hardening soil model

Fig. 2 Properties of natural clay soil with depth (a) Clay Hardening soil model is an advanced model that behaves as
content, (b) Water content and (c) Unconfined shear strength isotropic and hardening plastic. In this model the soil's stiffness,
[6]. hardening process and plasticity is simulated more accurate than
the Mohr-Coulomb model, however, assuming a hyperbolic
In order to verify the reliability of the modeling using the shape of the stress-strain relationship, due to the primary
program for predicting soil behavior, the load-settlement curves loading.
from the models are compared with the experimental results . A

Journal of Advanced Civil Engineering Practice and Research, 8:4-9, 2019

Waheed and Asmael / Study simulation of shallow foundation behavior using different finite element models 6

The hardening is assumed to be isotropic, according to the will be over-predicted in the late stage of loading.
plastic shear and volumetric strains. In contrast to the elastic
perfectly-plastic models, the Hardening soil model has a yield
surface that expands due to plastic strain, thereby describing the
plasticity more realistic.

A plastic volumetric strain was observed in case of soil

exhibit to isotropic compression [7]. This model was initially
suggested for sand, but at present it has been used for
representation the behavior of several types of soil, such as soft
and stiff clay.

3.3 Soft soil creep model

In general, the hardening model is appropriate for all types

of soils, but it does not take into account the viscous effect, i.e. (a)
creep and stress relaxation. In facts, all soils exhibit some creep
and primary compression and followed by some amount of
secondary compression. This model is suggested in the
application of settlement problems of a foundation,
embankment, etc.

4. Results of simulation models

The results of finite element simulation of the behavior of

shallow foundation under loading using three different material
models are given below. The foundation is represented as floor
element with a linear elastic model. It is of interest to note that
it was used a footing of square shape of width 0.8 m in
numerical modeling equivalent to 0.9 m circular footing of the (b)
case study according to Lee and Salgado [8] that used square
width equal to 0.9 m diameter of circular shape. Fig. 3 Geometry of footing by Mohr-Coulomb model and the
distribution of vertical displacement under loading.
4.1 Mohr-Coulomb model

The input properties of the soil used in the simulation of the

case study are shown in Table 1. Fig. 3 shows the geometry of
footing models by Plaxis and the distribution of vertical
displacement of the case study. The load-settlement response
obtained from the numerical analysis is compared with the
experimental test, shown in Fig. 4, where a node is selected in
the center of the foundation.

Table 1 Input properties of the soil used in the analysis by the

Mohr-Coulomb model.
Description Symbol Unit Value
F.E. Model MC - Mohr-Coulomb
Type of model
- - Drained
behavior
Youngˋs modulus E MPa 10
Cohesion c kPa 70
Friction angle Ø ᴼ 10
Poisson’s Ratio ν - 0.4
Fig. 4 Load-settlement response of the case study and
It can be seen that the results of the Mohr-Coulomb model Mohr-Coulomb model.
appear to be linear until it reaches the failure level; this is in the
trend with the assumption of model, this makes the simulation 4.2 Hardening soil model
matches the curve at the initial straight portion and then be over
predict, therefore, a care must be taken with this model because The input properties of the soil used in the analysis of the
the prediction by this model may be misleading and the results case study by this model are shown in Table 2 and the

Journal of Advanced Civil Engineering Practice and Research, 8:4-9, 2019

Waheed and Asmael / Study simulation of shallow foundation behavior using different finite element models 7

comparison between numerical and experimental plastic failure occurred which make the response using this
load-settlement response is shown in Fig. 5. It was observed model is the worse in this case.
that the simulation by Hardening-soil model gives good results
compared to the experimental results. Table 3 Input properties of the soil used in the analysis
by soft soil creep model.
The curve obtained from this model is matched to some Description Symbol Unit Value
extent and then there was a tendency to overestimate to be about
F.E. Model SC - Mohr-Coulomb
(25%) at the final stages. The same is observed in the results of
Waheed [9] while comparing the results of the piled-raft model Type of model - - Drained
in a clayey soil with those obtained through the hardening-soil behavior
model. Lambda λ - 0.097
Kappa К - 0.03
Table 2 Input properties of the soil used in the analysis by Mu m - 0.015
Hardening soil model.
Poisson’s Ratio ν - 0.4
Description Symbol Unit Value
F.E. Model HS - Hardening Cohesion c kPa 70
Type of model - - Drained Friction angle Ø ᴼ 10
behavior
Youngˋs modulus Eref50 MPa 10
Oedometer modulus Erefoed MPa 29.4
(Auto. calculate)
Unloading modulus Erefur MPa 30
ref
(Default = 3 E 50)
Cohesion c kPa 70
Friction angle Ø ᴼ 10
Poisson’s Ratio ν - 0.4
Exponential Power m - 1

Fig. 6 Load-settlement response of the case study and the soft

soil creep model.

Fig. 5 Load-settlement response of the case study and

hardening soil model.

4.3 Soft soil creep model

The input properties of the soil used in the simulation of the

case study by soft soil creep model are shown in Table 3 and
the comparison between numerical and experimental
load-settlement response is shown in Fig. 6.

The soft model appears to give a stiffer load-settlement

response compared to Mohr-Coulomb model, however, the Fig. 7 Load-settlement response of various soil models.
behavior appears to be linear from the start of loading until a

Journal of Advanced Civil Engineering Practice and Research, 8:4-9, 2019

Waheed and Asmael / Study simulation of shallow foundation behavior using different finite element models 8

From the comparison of the model results conducted in Fig. foundation to its width (B).
7, it can be observed clearly that the simulation is better by
hardening model than the other two models, however, these two In Fig. 11, bearing ratio (q/cu) is plotted against settlement
models gives the straight line response to a wide range which is ratio (S/B) for the same cases, where the bearing ratio
invalid compared with the experimental curve. It appears that represents the ratio of bearing pressure to the undrained
the representation using this model is more realistic. cohesion of clay.

5. Results of large scale foundation It can be observed that there is a clear convergence of
simulation using the three models, unlike the case of small-scale
For the purpose of studying the simulation of the same foundation, which means, in general, the three models can be
models in case of large scale foundations, a square footing was used for simulation, however, it can be said that in both cases,
considered by a width of 10 m instead of 0.8 m. The same input the hardening model gives better results to predict the behavior
parameters are used for each model; Figs. 8-9 explain the of shallow foundation in clay.
geometry of footing models by hardening model and the
distribution of vertical displacement for the large scale
foundation respectively.

Fig. 8 Geometry of footing by hardening model and the Fig. 10 Bearing pressure versus settlement ratio for different
distribution of vertical displacement under load for the large models used.
scale foundation.

Fig. 9 The distribution of vertical displacement under load for

the large scale foundation using hardening model. Fig. 11 Bearing ratio versus settlement ratio curve using
different models.
For comparison purposes, the results of the three models for
large scale foundations in addition to the experimental case The bearing ratio represents the bearing capacity factor (Nc)
study and their hardening model analysis have been and it was affected by the settlement criterion used for defining
demonstrated as a relation of the bearing pressure (q) and the the ultimate capacity, moreover, according to the observation of
settlement ratio (S/B), as can be seen in Fig. 10, where the Ornek et al. [6] that the natural clay soil deposit collapsed and
settlement ratio represents the ratio of a settlement of cannot apply extra load after settlement ratio (3%), therefore,

Journal of Advanced Civil Engineering Practice and Research, 8:4-9, 2019

 Waheed and Asmael / Study simulation of shallow foundation behavior using different finite element models 9

the (Nc) is 6, where this value is within the acceptable range (4 References
to 6.28) for saturated clay [10].
[1] Darve F. Incrementally non-linear constitutive
relationships in Geomaterials: constitutive equations and
6. Conclusions
modeling. CRC press, UK; 2014:229–254.
[2] Chambon R. Uniqueness, second order work and
According to the results of this study, the following bifurcation in hypoplasticity in constitutive modelling of
conclusions are made: granular materials. Springer, UK; 2000;147–165.
[3] Dafalias YF, Manzari MT. Simple plasticity sand model
• The simulation of the Mohr-Coulomb model for small accounting for fabric change effects. Journal of
Engineering Mechanics 2004;130(6):622–634.
scale foundation matches the curve at the initial straight
[4] Wood DM. Geotechnical modelling. CRC press, UK;
portion and then over predicts, so a great care must be 2014.
taken to use this model. [5] Abate G, Caruso C, Massimino MR, Maugeri M.
• For small scale foundation, the soft soil creep model gives Evaluation of shallow foundation settlements by an
the straight line response to a wide range which is invalid elasto-plastic kinematic-isotropic hardening numerical
(compared with the experimental curve). model for granular soil. Geomechanics and
• The representation using the hardening soil model is more Geoengineering 2008;3(1):27–40.
[6] Ornek M, Laman M, Demir A, Yildiz A. Prediction of
realistic and gives better results to predict the behavior of bearing capacity of circular footings on soft clay stabilized
small scale shallow foundation in clayey soil. with granular soil. Soils and Foundations
• In case of the large scale foundation, these three models 2012;52(1):69–80.
can be used for simulation, where the simulation of the [7] Brinkgreve RBJ, Broere W, Waterman D. PLAXIS 2D -
hardening soil model is the best. Full Manual, Netherlands; 2008.
• It is found the bearing capacity factor (Nc) is 6, which is [8] Lee J, Salgado R. Estimation of footing settlement in sand.
International Journal of Geomechanics 2002;2(1):1–28.
within the acceptable range.
[9] Waheed MQ. Assessment of piled raft system in clayey
soil. Ph.D. Thesis, University of Technology, Iraq; 2016.
[10] Bowles LE. Foundation analysis and design. McGraw-Hill,
UK; 1996.

To cite this article: Waheed M, Asmael N. Study simulation of shallow foundation behavior using different finite
element models. Journal of Advanced Civil Engineering Practice and Research 2019;8:4-9.

To link this article: http://ababilpub.com/download/jacepr8-2/

Journal of Advanced Civil Engineering Practice and Research, 8:4-9, 2019

View publication stats