Anda di halaman 1dari 15

Computers and Geotechnics 59 (2014) 112–126

Contents lists available at ScienceDirect

Computers and Geotechnics

journal homepage:

Proposed nonlinear 3-D analytical method for piled raft foundations

Sangseom Jeong, Jaeyeon Cho ⇑
Department of Civil Engineering, Yonsei University, Seoul 120-749, Republic of Korea

a r t i c l e i n f o a b s t r a c t

Article history: The load distribution and deformation of piled raft foundations subjected to axial and lateral loads were
Received 1 February 2013 investigated by a numerical analysis and field case studies. Special attention is given to the improved ana-
Received in revised form 5 February 2014 lytical method (YSPR) proposed by considering raft flexibility and soil nonlinearity. A load transfer
Accepted 25 February 2014
approach using p–y, t–z and q–z curves is used for the analysis of piles. An analytical method of the
Available online 28 March 2014
soil–structure interaction is developed by taking into account the soil spring coupling effects based on
the Filonenko-Borodich model. The proposed method has been verified by comparing the results with
other numerical methods and field case studies on piled raft. Through comparative studies, it is found
Piled raft
Soil–structure interaction
that the proposed method in the present study is in good agreement with general trend observed by field
Numerical analysis measurements and, thus, represents a significant improvement in the prediction of piled raft load sharing
Field measurement and settlement behavior.
Load transfer approach Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction approximate computer-based methods [5,9,14,15,37] and (3) more

rigorous computer-based methods [12,17,18,45,48].
In recent years, a number of huge construction projects, such as The first type of method is based on the linear elastic analysis of
high-rise buildings and long span bridges, are being undertaken. piled raft subjected to axial loading. Generally, the simplified
The piled raft foundations are especially being recognized as an calculation methods are most commonly used procedure for the
economical foundation system for high-rise buildings. Here, piles preliminary design of a piled raft foundation. However, it is noted
as settlement reducers have been discussed for over a quarter of that these analytical methods are limited to elastic problem.
a century [2] and some significant applications have been reported Because this calculation procedure is developed for rigid raft and
[12,38,42]. Optimized design strategy is a major importance for an is assumed that the soil is perfectly elastic. Thus, it may not repre-
economic construction to be achieved. An optimized design of a sent the nonlinear behavior of actual piled raft in the field: it does
piled raft can therefore be defined as a design with minimum costs not take into account the actual behavior of finite flexible raft and
for the installation of the foundation and satisfactory bearing pile–soil interaction, etc.
behavior for a given geometry and raft loading [35]. The piled raft The second type of method has been used to investigate the
is a composite foundation system consisting of three bearing ele- piled raft system, which is analyzed as a continuous elastic
ments: raft, piles and subsoil. Therefore, the behavior of a piled raft medium using finite element formulation. In these methods, the
is affected by the 3D interaction between the soil, piles and raft, research by Poulos [29], Clancy and Randolph [5], Poulos [30]
thus, a simple and convenient analytical method is needed to eval- and Russo [37] also have some disadvantages. It did not predict
uate these interactions. the membrane behavior of raft because the raft is generally mod-
Much work has been done to study load sharing and settlement eled as plate element. Therefore, the raft used in these methods
behavior of piled raft by many researchers. Numerical methods may not reflect the displacement due to membrane action of large
have been developed widely in the last two decades because size raft foundations for high-rise buildings. In addition, most of
numerical methods are less costly and may be used to consider the previous research is related to piled rafts subjected to vertical
many kinds of different soil and foundation geometries compared loading and only semi-infinite homogeneous single soil layer was
to field and model tests. Although these methods make slightly considered. The consideration of various loading condition and soil
different modeling techniques, they can generally be classified into layer will be more realistic in design practice.
three groups: (1) simplified calculation methods [30,32], (2) The third type of method is based on the three-dimensional
finite-element or finite-difference techniques. Poulos [31] noted
⇑ Corresponding author. that the most feasible method of analysis was the three-
E-mail address: (J. Cho). dimensional linear/nonlinear FE method. However, a rigorous
0266-352X/Ó 2014 Elsevier Ltd. All rights reserved.
S. Jeong, J. Cho / Computers and Geotechnics 59 (2014) 112–126 113

Fig. 1. Flat-shell element.

Fig. 2. Modeling of pile element.

numerical approach of the piled raft system is computationally to examine the validity of the proposed method, the analysis re-
expensive and requires extensive training because of the three- sults are compared with the available solutions from previous re-
dimensional and nonlinear nature of the problem. Therefore, a fi- searches. In the field case study, comparative analyses between
nite element analysis is more suitable for obtaining benchmark YSPR and a field measurement data are carried out for the pile load
solutions against which to compare simpler analysis methods, or and settlement behavior.
for obtaining solutions of a detailed analysis for the final design
of a foundation, rather than as a preliminary routine design tool
[15]. 2. Method of analysis
In this study, an improved analytical method (YSPR) for the de-
sign of piled raft has been proposed to overcome some limitations 2.1. Modeling of flexible raft
of the existing methods. It is intermediate in complexity and theo-
retical accuracy between the second and third type of method. In Finite element techniques have often been used for the analysis
the present method, a numerical technique is used to combine of raft by different researchers such as Clancy and Randolph [5],
the soil and pile head stiffness with the stiffness of the raft. In order Zhang and Small [49], Kitiyodom and Matsumoto [14]. According
114 S. Jeong, J. Cho / Computers and Geotechnics 59 (2014) 112–126

Fig. 3. Soil–structure interactions in piled raft foundation [13].

Fig. 4. Interactions between raft, piles, and subsoil in present method.

to the former methods [5,49], the raft can be treated as a plate and action, therefore it can be considered separately. For the bending
the soil can be treated as a series of interactive springs by using a action, the displacement field for an individual element can be de-
Mindlin’s solutions [22], in which the contact pressure at any point scribed in terms of the vertical nodal displacement and the rota-
on the base of the raft is proportional to the deformation of the soil tions about the x and y axes. For the membrane action, the
at that point or as an elastic half-space in which the behavior of the displacement field can be described in terms of the nodal displace-
soil can be obtained from a number of closed-form solutions. In the ments in the x and y directions.
later method, the raft is modeled as thin plates and the piles as
elastic beams and the soil is treated as interactive springs [14]. 2.2. Modeling of single and pile groups
The interactions between structural members are made by the
use of Mindlin’s solutions. The primary limitation of these methods In this study, piles are treated as beam-column elements. The
is that the membrane behavior of the flexible raft cannot be con- behavior of soil surrounding the individual piles is represented
sidered because the nodal displacements (in the x- and y-direction) by load–transfer curves (t–z, q–z, and p–y curves), and the interac-
for the membrane action are not included. This limitation can be tion between piles is represented by p-multiplier (fm) and group
overcome by using a flat-shell element. An improved four-node efficiency factor (Ge). The load–deformation relationship of individ-
flat-shell element proposed by the authors [48], which combines ual pile heads may be derived by a single pile analysis based on
a Mindlin’s plate element and a membrane element with torsional beam-column method. In this method, a pile member is described
degrees of freedom, is adopted in this study. The flat-shell element as a series of beam column elements with discrete springs to rep-
can be subjected to the membrane and bending actions that are resent the soil support condition as shown in Fig. 2. The governing
shown in Fig. 1. The displacement due to the membrane action is differential equations for the axially loaded and laterally loaded
considered independent of the displacement due to the bending pile can be expressed as:
S. Jeong, J. Cho / Computers and Geotechnics 59 (2014) 112–126 115

P d w
Axially loaded pile : EA 2
 Cbz w ¼ 0 ð1Þ
Iteration dz
2 2’
Load increment

4 2
d y d y
(k t) 2
Laterally loaded pile : EI 4
þQ 2
þ q  Ksy ¼ 0 ð2Þ
ΔP2 dz dz
(k s) 2

P1 1 where EA, EI are the axial stiffness and the flexural rigidity the pile,
1’ w is the vertical deflection of the pile at point z, bz is the stiffness/
(kt) i : tangential slope
(k t) 1 circumference for the axial reaction represented by the modulus
(ks) i : secant slope ΔP1
of the soil-response (t–z or q–z or both), which depends on the
depth z and pile movement w, and C is circumference of the pile
0 at point z. Q is the axial load on the pile, q is the distributed load
u1 u’1 u2 u’2 u
along the length of the pile, and KS is the stiffness for the lateral soil
(a) reaction represented by the modulus of the soil-response (p–y)
(k i) j : i = load increment j=1 : tangential stiffness In the next step, finite difference technique is used to solve the
j = Iteration number j>1 : secant stiffness
differential equations governing the compatibility between the pile
displacement and the load transfer along a pile. These techniques
(k i) 1
are generally based on load tests on full-scale and parametric finite
f((u i) i) element analyses of pile–soil interactions, which are represented
Fu=f(u ) by load–transfer curves (t–z, q–z, and p–y curves).
(ki) j
f((u) i- 1)
2.3. Soil–structure interaction

The load-bearing behavior of a piled raft is characterized by
complex soil–structure interaction between the piles, raft and the
subsoil, as shown in Fig. 3 [13]. The present method makes use
(u) i- 1 (ui) j u of pile–soil–pile and raft–soil–pile interaction to simulate the real

(b) piled raft–soil response under lateral and vertical loadings. Addi-
tionally, for the raft–soil–raft interaction, this study uses a semi-
Fig. 5. Increment secant modulus method [48]. (a) Concept of increment secant empirical parameters proposed by many researcher [7,39,40] as
modulus method. (b) Estimating stiffness at ith load increment. the modulus of soil reaction below the raft. The use of these
parameters as assumed in the derivation procedure, may be a lim-
itation. However, these interactions are incorporated in a calcula-
tion procedure that is computationally very efficient.

Fig. 6. Modeling of piled raft (YSPR).

116 S. Jeong, J. Cho / Computers and Geotechnics 59 (2014) 112–126

Fig. 7. Flow chart of YSPR.

Piles in such groups interact with one another through the sur- concerning the efficiency of the piles in a group is derived by many
rounding soil, resulting in the pile–soil–pile interactions. In this researchers [11,21,41]. In this study, load–transfer curves in side
study, a set of nonlinear p–y curves which can be modified by resistance (t–z curve) and in end bearing resistance (q–w curve)
reducing all of the p-values on each curve by a p-multiplier (fm) which can be modified by reducing all of the t- and q-values on
are used as input to study the behavior of the laterally loaded piles. each curve by a group efficiency factor (Ge) are used as input to
The p-multiplier can be calculated for each pile in the group study the behavior of the vertically loaded piles.
[3,6,19]. For each pile i in the group, the p-multiplier can be ex- In classical solution, the Winkler model [46] is used for analyz-
pressed as: ing raft foundation. However, the Winkler model could not predict
fmi ¼ b1i b2i b3i    bji ð3Þ accurately the displacement of some solids, e.g. soil. The Winkler
model ignores the important interaction existing between adjacent
where bji is the p-reduction factor due to the effect of pile j on pile i. points in the soil continuum. In other words, the soil springs are
In a group of closely-spaced piles, the axial capacity of group is considered as isolated foundation elements. In order to overcome
also dominated by variation in settlement behavior of individual a limitation, much work has been performed to propose some
piles due to pile–soil–pile interaction. The most reliable data improved or refined models [8,10,27,43]. For the raft–soil–pile
S. Jeong, J. Cho / Computers and Geotechnics 59 (2014) 112–126 117

present method proposed an improved raft–soil–pile system by

connecting the top ends of soil springs and pile elements with an
elastic flat-shell element including membrane action. By using
flat-shell element, a realistic representation of the subgrade reac-
tion can be established directly in terms of coupled soil resistance
in which the response at any point on the interface affects other
points. The authors believe that a combination of the soil spring
and the elastic flat-shell element may be used to overcome the
restrictions associated with conventional methods, and thereby
also used to analyze appropriately axially loaded piled raft, in soil
deposits. Consequently, the proposed analytical method should be
based on the concept of soil–structure interaction under the lateral
and vertical loadings.

2.4. Global stiffness matrix

The stiffness matrix of a flat-shell element (Kflat-shell=raft) in local

coordinate system was constructed through combining separately
the stiffness matrix of a plate element (Kplate) and that of a mem-
brane element (Kmembrane) as followings:
K plate 0
K flat-shell ¼ ð4Þ
0 K membrane
The stiffness matrix of a plate element Kplate is represented in
the following form:
K plate ¼ BTb Db Bb dV þ BTs Ds Bs dV ð5Þ

where Bb is the bending strain matrix and Bs is the shear strain ma-
trix. For an isotropic material, Db and Ds are given as follows:
2 3
1 m 0
Et 6 7
Db ¼ 4m 1 5 0 ð6aÞ
12ð1  m2 Þ
0 0 ð1  mÞ=2
WEt 1 0 5
Fig. 8. Schematic diagram of vertical and lateral loaded piled raft. (a) Pile Ds ¼ ; W¼ ð6bÞ
configuration. (b) Section-view. 2ð1 þ mÞ 0 1 6
where E is Young’s modulus, m is Poisson’s ratio, and t is constant
interaction, in this study a membrane-spring system originally
thickness of the plate. On the other hand, the stiffness matrix of a
proposed by Filonenko-Borodich [8] was incorporated to involve
membrane element Kmembrane is represented in the following form:
the soil spring-coupling effects. This system can provide a mechan-
ical interaction between the individual soil spring and pile 1 T
K membrane ¼ ½Bm GRT  C  ½Bm GRdV þ hh ð7aÞ
elements by using the flat-shell element. As shown in Fig. 4, the v cV

Fig. 9. Soil spring constant for linear analysis of a single pile.

118 S. Jeong, J. Cho / Computers and Geotechnics 59 (2014) 112–126

0.02 0.03 0.04 0.05
-0.1 0 0.1 0.2

Depth from G.L.(m)

Depth from G.L.(m)

4 4

6 6

FEM (K&M, 2003) FEM (K&M, 2003)
10 10
(a) (c)
-0.1 0 0.1 0.2 -0.4 -0.2 0 0.2
0 0

2 2
Depth from G.L.(m)

Depth from G.L.(m)

4 4

6 6

FEM (K&M, 2003) FEM (K&M, 2003)
10 10
(b) (d)
Fig. 10. Comparison of analysis result for piled raft: (a) Settlement and (b) lateral displacement, (c) shear force; and (d) bending moment.
g dV; T for pile groups can be formed by sum of n single pile stiffness ma-
h¼ ½bg b ð7bÞ
v trix (Eq. (10)).
2 3
E K 11 0 0 0 K 15 0
c¼ ð7cÞ 6 0 K 22 0 K 24 0 0 7
2ð1 þ mÞ 6 7
6 7
6 0 0 K 33 0 0 0 7
where C is the constitutive modulus, c is taken as the shear modu- K pile ¼6
6 0
7 ð8Þ
6 K 42 0 K 44 0 0 7
lus. Bm, G, R are the strain matrices representing the relationship be- 6 7
4 K 51 0 0 0 K 55 0 5
tween the displacements (the membrane displacement, the
rotation, and midside incompatible displacement respectively) 0 0 0 0 0 K 66 i
and the strains. b, g, b, and g are also the strain matrices for the
infinitesimal rotation fields. ½KpileðiÞ fdgi ¼ fF i g ð9Þ
The pile head stiffness (K11  K66) is assumed to be constant
within each load increment and each iteration and then superposi-
tion can be applied in order to develop a pile head stiffness matrix K pilegroups ¼ ½K pileðiÞ  ð10Þ
(Eq. (8)) in individual piles. Using load–displacement relationships i¼1
representing pile behaviors according to pile head movements
[34], the relationship between the nodal force and nodal displace- where [K]pile(i) is an individual pile head stiffness matrix, {di} a dis-
ments can be expressed in Eq. (9). In addition, the stiffness matrix placement or rotation, and {Fi} force or moment at the ith pile head.
S. Jeong, J. Cho / Computers and Geotechnics 59 (2014) 112–126 119

22 mm 2.5D and 5.0D

2.5D, 5.0D,
22 mm
and 7.5D

Fh Pile cap Fh Pile cap

(variable) (variable)

15 mm 15 mm

600 mm 600 mm
Sand Sand

Pile Pile

Rock Rock
(a) (b)
Fig. 11. Test pile group configurations [4] (a) 2  2 pile groups (b) 3  3 pile groups.

A component (K11  K66) of pile head stiffness matrix is changed at The procedure for nonlinear solution in this study includes the
each load increment and iteration stage. following step. In total, 10 (ten) load–displacement curves (axial 1;
The soil support at various nodes of raft foundation is simulated lateral 8; torsional 1) are estimated per each pile head. Fig. 5(b)
by a series of equivalent and independent springs in three direc- shows the estimation method of stiffness at an ith load increment.
tions (x, y and z directions). The spring behavior can be linear or In this method, external forces are first divided by N (number of
nonlinear. In linear case, soil behavior is defined by soil stiffness load increment). The stiffness at ith load increment and jth itera-
(K11  K33) which is assumed to be constant within each load tion is represented (ki)j. In each load increment, tangential slope
increment and each iteration. The soil reactions at any point can is adopted at first iteration (j = 1) and the secant modulus at j > 1
be expressed as for the stiffness of pile head, which is expressed as Eqs. (14) and
2 38 9 8 9 (15), respectively.
k11 0 0 0 0 0 > du > > Fu >
> >
> >
> >
6 07 > dv > > Fv >
6 0 k22 0 0 0 7>> >
> >
> >
6 7>>
> df ðuÞ
6 0 0 k33 0 0 0 7 dw Fw = ðki Þj ¼ ðj ¼ 1Þ ð14Þ
6 7 ¼ ð11Þ du u¼ðuÞi1
6 07
6 0 0 0 0 0 7>> >
> au >
> Mu >
> >
6 7> > > >
4 0 0 0 0 0 05 >>
> a
> ;v
> M v
: : ;
0 0 0 0 0 0 i aw i Mw i f ððui Þj Þ  f ððuÞi1 Þ
ðki Þj ¼ ðj > 1Þ ð15Þ
ðui Þj  ðuÞi1
½KsoilðiÞ fdgi ¼ fF i g ð12Þ

where [K]soil(i) = individual soil stiffness matrix, {di} = displacement

ðui Þj ¼ ðuÞi1 þ Duj ð16Þ
or rotation, and {Fi} = force of soil at point i. In nonlinear case, spring
behavior is defined by giving pairs of load–relative displacement
values. At this point, soil stiffness is calculated by nonlinear solution where (u)i1 is an accumulated final displacement at a previous
procedure. load increment and (ui)j is an accumulated displacement at the ith
Finally, the stiffness matrix of a piled raft can be defined by the load increment and jth iteration.
combination of the foundation system and the supporting soil. At each load increment, displacements (Duj) are calculated
Therefore, the stiffness matrix formulations of a piled raft system through structural analysis and then accumulated displacements
can be written as the following: (ui)j are estimated using Eq. (16). If the convergence criteria,
Duj–Duj1 < e is satisfied, the accumulated final displacements
½K piled raft  ¼ ½K raft  þ ½K soil  þ ½K pilegroups  ð13Þ
(u)i are calculated and continue to the next load increment. This
process iterates until the load increment number reaches N. In
2.5. Nonlinear solution procedure the structure analyses, the tangential slope (df(u)/du) and load
(f(u)) of individual piles are estimated using cubic spline method
To consider the nonlinear load–displacement relationship at [1]. The procedure described above is iterated until the error
each pile head and soil (below the raft), an incremental secant between the assumed and calculated displacements falls within a
modulus method developed by Won et al. [48] is used. When this tolerance limit.
‘‘incremental secant modulus method’’ is used, the displacement u2 As a final outcome, an improved numerical method (YSPR) was
corresponding to load P2 is increased to u02 as shown in Fig. 5(a), so proposed to analyze the response of a raft and a piled raft consid-
that point (P2, u02 ) will be located on the curve and consequently ering raft flexibility and soil nonlinearity (Fig. 6). Fig. 7 shows the
the displacement will be close to the exact solutions. flow chart of present method.
120 S. Jeong, J. Cho / Computers and Geotechnics 59 (2014) 112–126

0.04 3. Verification of proposed method with previous studies

3.1. Kitiyodom and Matsumoto [14]

0.03 A series of linear piled raft analyses were performed to verify

Lateral load (kN)

the present method by comparison with other numerical methods

which have been used in the preliminary design of piled raft. A
schematic diagram of a 2  2 piled raft is shown in Fig. 8. This
structure consists of a raft, and four identical vertical piles, which
are spaced by 1.5 m (=3.75D, where D is the pile diameter). The
piles have an embedded length of 10 m, a diameter of 0.4 m. Pile
head conditions are fixed. A square raft of size 3  3 m with a
measured (2.5D)
measured (5.0D)
thickness of 0.9 m is rested on a homogeneous soil. The Young’s
0.01 modulus and Poisson’s ratio of the soil are 12,500 MPa and 0.3.
measured (7.5D)
predicted (2.5D) The raft and piles, with a Young’s modulus and Poisson’s ratio of
predicted (5.0D) 125,000 MPa and 0.3 respectively, is subjected to a vertical and
predicted (7.5D) lateral load. Fig. 9 shows the spring constants were used for the
0 linear soil condition. The same axial spring constants were used
0 0.004 0.008 0.012 0.016 along the pile depth, with a constant value of 7,527,867 kN/m2,
which includes the pile perimeter. The end-bearing spring was
Displacement (m)
8,692,180 kN/m2, and the tension part was neglected. The con-
(a) stants of the horizontal springs were increase from 0 to
4,682,274 kN/m2 along the pile depth. Since the soil is assumed
to be an elastic model, the p-reduction and group efficiency factor
of unity were used [6,18,41].
The response of piled raft is presented in settlement, lateral dis-
placement of pile, and in shear force and bending moment distri-
0.06 bution at various depths. Fig. 10(a–d) shows representative
Lateral load (kN)

results from the proposed method. In addition, these results were

tested by comparing them with well-known three existing numer-
ical methods: the PRAB [15]; the finite element method performed
0.04 by Kitiyodom and Matsumoto and PLAXIS 3D [28]. The results are
shown in terms of dimensionless parameters of IwV for the settle-
ment, IuH for the lateral displacement of a pile respectively, CsH,
CbH for the shear force, and the bending moment along the pile
0.02 respectively. These parameters can be calculated by Eqs. (17)–(20).
measured (2.5D)
measured (5.0D) Es Dw
IwV ¼ ð17Þ
predicted (2.5D) qz Br Lr
predicted (5.0D)
0 Es Du
IuH ¼ ð18Þ
0 0.004 0.008 0.012 0.016 0.02 qx B r L r
Displacement (m)
(b) C sH ¼
qx Br Lr

Fig. 12. Lateral load–displacement curves at pile head. (a) 2  2 pile groups. (b)
3  3 pile groups.
C bH ¼ ð20Þ
qx DBr Lr

Table 1
Material parameters used for this study (case studies).

Case Material properties

Type Depth (m) E (MPa) m c (kN/m3) / (°) c (kPa) Modela
Japan case [15] Pile Steel pipe 0 to 5.5 2.1E08 0.2 75 – – L.E.
Raft Concrete 0 to 2.2 30,000 0.2 25 – – L.E.
Soil Sandy silt 0 to 1.7 13 0.3 18 0 25 M.C.
Silty clay 1.7 to 13.5 15 0.3 18 0 29.64 M.C.
Germany case [34] Pile Concrete 5.5 to 25.5 23,500 0.2 25 – – L.E.
Raft Concrete 3 to 5.5 34,000 0.2 25 – – L.E.
Soil Sand 3 to 8 75 0.25 18 32.5 0 M.C.
Frankfurt clay 8 to 113 47a 0.15 19 20 20 M.C.
Korea case Pile Concrete 0 to 30 28,000 0.2 – – – YSPR
Raft Concrete 0 to 6.0 33,234 0.15 – –
Soil Gneiss Soil spring stiffness (kPa/m)
0 to 204,250
Note: M.C. is Mohr Coulomb elasto-plastic model, L.E. is linear elastic model used in PLAXIS 3D Foundation Frankfurt clay: E = 45 + [tanh((z  30)/15) + 1]  0.7z.
S. Jeong, J. Cho / Computers and Geotechnics 59 (2014) 112–126 121

Fig. 13. Field test of piled raft [16]. (a) Plan-view and (b) section-view.

Table 2
Load (MN)
Properties used for estimating load transfer curves (Japan case).
0 0.5 1 1.5 2 2.5
Contents Sandy silt Silty
clay 0
t–z, q–z curves [39] Ultimate skin friction, s 40 40
Initial shear modulus, Gi 5000 5769
Poisson’s ratio, m 0.3 0.3
Settlement (mm)

Ultimate bearing – 250

capacity, Qf (kN)
p–y curves [19,20] Undrained shear strength 25 29.64
Unit weights (kN/m3) 18.0 18.0
p–y modulus, k (kN/m3) 27,150 27,150
Subgrade reaction Kx, Ky (kN/m3) 27,150 –
modulus Kz (kN/m3) 5291 –

where w, u are the settlement and lateral displacement at the pile 40 Measured (K&I, 1967)
head, qz and qx are uniform vertical and lateral load, the breadth, Calculated (R & E, 2006)
Br and length, Lr, S and B are the shear force and the bending mo- YSPR
ment along the pile. PLAXIS 3D
The calculated results of the proposed analysis method closely
approach the computed data from the other numerical methods. Fig. 14. Computed and measured response of piled raft settlement.
It should be noted that the present method provides a very satisfac-
tory prediction of the shear force and the bending moment in indi-
Jeong [4], a series of small scale model tests were carried out to
vidual piles, when the flexibility of the raft is considered by using
study the behavior of pile groups subjected to lateral loadings on
the combination of the membrane and bending actions. Although
sand. The test soil used in this study was: the unit weight
a reasonably good agreement between the proposed and the exist-
15.3 kN/m3, cohesion 0 kN/m2 and drained friction angle 37°. The
ing methods was obtained, the proposed method has a larger settle-
model piles made from PVC tubes were 0.6 m in embedded length,
ment those of the existing methods at the same load. Conclusively,
22 mm in diameter and 2.5 mm wall thickness and 28,265 kN m2
it is thought that YSPR can be used with some confidence in the pre-
flexural rigidity(EI). Fig. 11 shows an idealization of the subsurface
liminary design of axially and laterally loaded piled raft.
profile and pile embedment for test piles.
Using present method the behavior of pile groups are predicted
3.2. Chung and Jeong [4] with different group configurations and different center-to-center
pile spacing: 2.5D, 5.0D, and 7.0D. Back-fitted hyperbolic p–y
In this section, the verification of lateral response of the present curves that are calculated at 5, 10, and 20 cm along the pile depth
method against laboratory load test is discussed. By Chung and in model test of single pile are implemented. Initial tangent

Table 3
Calculated stiffness of single pile and piled raft (Japan case).

K11 (kN/m) K22 (kN/m) K33 (kN/m) K44 (kN/rad) K55 (kN/rad) K66 (kN/rad)
Single pile 0.4052E+02 0.4052E+02 0.3877E+05 0.3434E+03 0.3434E+03 0
Piled raft (w/o Ge) 0.2735E+05 0.2735E+05 0.3453E+06 0.2730E+06 0.2730E+06 0
Piled raft (w/Ge) 0.2735E+05 0.2735E+05 0.2492E+06 0.2208E+06 0.2208E+06 0
122 S. Jeong, J. Cho / Computers and Geotechnics 59 (2014) 112–126

Fig. 16. Pile load: (a) pile 1, 2, 3 and (b) pile 4, 5, 6.

Fig. 15. Torhaus Der Messe: (a) profile view and (b) configuration of pile. stiffnesses (Ks) of the p–y curves at the depths of 0.05, 0.1, and
0.2 m are 11, 14.3, and 50 kN/m2, respectively. Also ultimate capac-
ities (Pu) of the p–y curves at the same depths are 0.0011, 0.0033,
Table 4 and 0.0033 kN/m, respectively.
Properties used for estimating load transfer curves (Germany case). To consider the detailed group effect, p-multipliers calculated
Contents Quaternary Frankfurt
from the Chung’s experiment are implemented. For the 2  2
silt clay group, p-multipliers are 0.86 for lead row and 0.45 for trail row
at 2.5D pile spacing; 0.95 for lead row and 0.67 for trail row at
t–z, q–z curves [39] Ultimate skin friction, sf 143 91.6
(kPa) 5.0D; 1.0, 0.83 for lead, trail row at 7.5D. For the 3  3 group,
Initial shear modulus, Gi 30,000 20,434 p-multipliers are 0.8, 0.3 and 0.4 for lead, middle, and trail rows
(kPa) at 2.5D pile spacing; 0.93, 0.48, and 0.6 at 5.0D pile spacing.
Poisson’s ratio, m 0.25 0.15
Fig. 12 shows the predicted and observed lateral load–settle-
Ultimate bearing – 90
capacity, Qf (kN)
ment curves. The analysis of pile groups was performed for a fixed
p–y curves [24,33] Internal friction angle (°) 32.5 20 head condition and spacing-to-diameter ratios varying from 2.5 to
Unit weights (kN/m3) 18 19 7.5. The present method considering pile–soil–pile interaction rel-
p–y modulus, k (kN/m3) 16,300 136,000 atively well predicts the general trend of the measured lateral
Subgrade reaction Kx, Ky (kN/m3) 16,300 136,000
loads for the pile groups studied if the measured deflections are
modulus Kz (kN/m3) 294,000 –
relatively small (say less than 15 mm).

Table 5
Calculated stiffness of single pile and piled raft (Germany case).

K11 (kN/m) K22 (kN/m) K33 (kN/m) K44 (kN/rad) K55 (kN/rad) K66 (kN/rad)
Single pile 0.3979E+03 0.3979E+03 0.3020E+06 0.4482E+05 0.4482E+05 0
Piled raft (w/o Ge) 0.1118E+08 0.1138E+08 0.1300E+08 0.2583E+09 0.2115E+09 0
Piled raft (w/Ge) 0.1117E+08 0.1137E+08 0.1242E+08 0.2548E+09 0.2078E+09 0
S. Jeong, J. Cho / Computers and Geotechnics 59 (2014) 112–126 123

Load (MN) predicted the general trend of the measured values reasonably
well. However, the calculated results by Roberto and Enrico [36]
0 50 100 150 200 250
have a relatively smaller settlement as the applied load increased
0 than the results of the proposed solution. This clearly demonstrates
PLAXIS 3D that for analysis result, YSPR gives more flexible results for nonlin-
YSPR (w/o Ge) ear behavior of soil, because the Roberto and Enrico [36] use soil
YSPR (w/ Ge) flexibility matrix(based on linear elastic analysis of pile groups)
Measured for soil–pile interaction and the proposed method does so using
40 settlement nonlinear load transfer curves and solution algorithm. These dis-
Settlement (mm)

crepancies between predicted and measured behavior at the high

60 load levels are because the assumptions of raft–soil relative stiff-
ness and group efficiency factor are influenced on the settlement
behavior of piled raft. In addition, computational time to run this
80 case saves 57 min of computer time, and is about 20 times faster
than the 3D FE analysis.
4.2. Germany case

120 The settlement and load sharing behavior of instrumented,

Smax = 124mm large, piled raft installed in stiff clay was compared with the pre-
140 dicted values of the proposed and the FE analyses. Constructed be-
tween 1983 and 1986, the 130 m high Torhaus was the first
Fig. 17. Settlement behavior of large piled raft foundation. building in Germany with a foundation designed as a piled raft. A
total number of 84 bored piles with a length of 20 m and diameter
of 0.9 m are located under two 17.5  24.5 m large rafts. The bot-
4. Comparison with other case histories tom of the 2.5 m thick raft lies just 3 m below ground level
(Fig. 15(a)). The subsoil comprises quaternary sand and gravel up
The validity of the proposed method was examined by compar- to 2.5 m below the bottom of the rafts, followed by the Frankfurt
ing the results from the present approach with some of the field- clay [34]. And a schematic diagram of 7  6 piled raft structure is
measured results. The pile and soil properties employed with the shown in Fig. 15(b). The maximum load of Peff = 200 MN for each
YSPR and PLAXIS 3D analyses for the case histories were the same raft [37] minus the weight of the raft is successively applied by
properties mentioned in their research. In the field, the soil stiff- means of a uniform load over the whole raft area. In the present
ness significantly depends on the stress level, indicating that the method (YSPR), the soil around individual pile is modeled with
stiffness generally increases with depth. To account for the in- nonlinear load transfer curves. The axial load transfer curves (t–z,
crease of the stiffness with the depth, the Young’s modulus of soil q–z curves) are estimated using the equation developed by Wang
(Eincrement) value which is the increment of stiffness per unit of and Reese [44], the lateral load transfer curve (p–y curve) is used
depth was used in FE analyses. Table 1 summarizes the material as an API model [25,32]. The group efficiency factor, Ge, was set
properties used in the case studies. at 0.73 for the average value of pile spacing: 3D  4D [23,47].
The input parameter of soil used to generate the load transfer curve
and soil-spring are summarized in Table 4.
4.1. Japan case Table 5 summarizes the calculated stiffness for the single pile
and the piled raft foundation. A decrease in the group efficiency
The settlement behavior of axially loaded piled raft reported by factor from 1.0 to 0.73 results in about 4.5% decrease in stiffness
Koizumi and Ito [16] are compared with the predicted values of the of piled raft. It is also noted that the stiffness of piles inside the
proposed method. This test site was located near the 1-chome, Ote- group varied with a group effect. Fig. 16(a–b) shows a comparison
machi in Tokyo. A fully instrumented piled raft was installed in the of the measured and calculated pile loads. The prediction of the
clay soil, which consists of sandy silt with gravel and organic silty present method is much more conservative than that of 3D FE
clay. Fig. 13 shows the subsurface profile and pile configurations of analyses and the measured one. However the proposed method
the test piled raft. All of the test piles are 300 mm in dia. and 5.5 m is in good agreement with general trend of pile load which increase
in length. The soil and material properties were determined by from a center pile (pile1) to the edge (piles 2, 4 and 6) and to the
back-analysis of field load test results using PLAXIS 3D Foundation. corner pile (piles 3 and 5). The computed results for the center,
From full-scale tests in clay soil presented by O’Neill [23] and Whi- side, and corner piles show that the load distribution of the indi-
taker [47], the group efficiency factor, Ge, was set at 0.7 for the vidual piles in a group is highly influenced by the flexibility of
reduction of side resistance (t–z curve) and end bearing resistance the raft. This finding was similar to what Won et al. [48] discussed
(q–w curve) of piles. The input parameter of soil used to generate about correlation between the pile member force and the flexibil-
the load transfer curve and soil-spring are summarized in Table 2. ity of pile cap for a pile groups.
Table 3 shows the estimated stiffness of single pile and piled Fig. 17 shows a settlement behavior of the piled raft. The mea-
raft when a vertical load of unity is applied. Compared to the stiff- sured maximum settlement is about 124 mm, the calculated set-
ness in which the group efficiency factor was 1.0, the stiffness of tlements using YSPR and PLAXIS 3D are 106.7 (with Ge; 111.5)
piled raft showed a significant decrease in K33 of about 28%. This mm and 117 mm respectively. This curve demonstrates the effect
is because the decrease of the pile resistance due to the pile– of pile–soil–pile interaction by considering group efficiency. The
soil–pile interaction (i.e. group efficiency factor), change the global proposed method with an interaction factor is more appropriate
stiffness of piled raft. and realistic to represent a pile–soil–pile interaction for closely-
The proposed analysis method (YSPR) and a finite element spaced piles than on that of no-interaction analysis. In Both values
program analysis (PLAXIS 3D) results were compared with the of YSPR and 3D FE analyses are smaller than the measured one.
measured load–settlement curves in Fig. 14. All the methods However, these two numerical methods provide an acceptable
124 S. Jeong, J. Cho / Computers and Geotechnics 59 (2014) 112–126

Fig. 18. Preliminary design case of large piled raft: (a) plan view and (b) profile view.

design prediction. Despite the approximate assumptions involved represent the stiffness of the ground. A schematic diagram of a raft
(i.e., loading condition, construction process, consolidation of clay), foundation with piles is shown in Fig. 18(b). This structure consists
the present method when used in nonlinear analysis is useful for of a raft, and 112 of ground strengthen piles. The piles have an
predicting the settlement behavior of a piled raft foundation taking embedded length of 30 m, a diameter of 1.0 m. A large raft size
account of soil nonlinearity, the flexibility of the large raft, and the 71.7  71.7 m with a thickness of 6.0 m is resting on a banded
pile arrangement. The time taken for the computer to run this case gneiss. The raft and ground strengthen piles, with a Young’s mod-
saves 115 min of computer time, and is about 24 times faster than ulus of 30 GPa and 28 GPa respectively, is subjected to a vertical
the 3D FE analysis. For large problems this computational saving load (Ptotal = 6,701 MN).
can be very significant. Fig. 19(a–d) shows the raft settlement at different sections pre-
dicted by GSRaft [26] and YSPR. Agreement between the GSRaft
4.3. Korea case and YSPR of settlement is generally good; however there is a slight
difference in prediction of settlement in the faulting zone where
As shown in Fig. 18, preliminary design case of a piled raft (OO the sudden drop of the magnitudes were occurred. This can be
super tower) conducted at high-rise building construction sites in attributed to the inappropriate assumption of material properties
Korea were representatively selected for the design application. due to no accurate ground investigation data on this section. The
The construction site is comprised mainly of normally banded calculated raft settlement has some difference between the
gneiss, brecciated gneiss and fault core zones. Based on the results proposed method and the existing solution, based on the same
of pressure meter, Goodman Jack and plate load tests carried out in analysis conditions. This is because the conceptual methodology
the field, a nonlinear elastic modulus design line is established to of the present method is completely different from that of general
S. Jeong, J. Cho / Computers and Geotechnics 59 (2014) 112–126 125

Fig. 19. Raft settlement distribution: (a) section 1, (b) section 2, (c) section 3; (d) section 4.

structural models. The raft is modeled as a grillage and the piles are representation of the settlement and load sharing behavior of
treated as bar element with axial stiffness only in GSRaft while piled raft foundation. It provides results that are in good agree-
YSPR is adopted flat-shell element and 6  6 pile head stiffness. ment with the field measurement and numerical analyses.
Although there are no measured profiles of raft settlement, the 2. Proposed analytical method produces a considerably larger set-
proposed analysis method showed reasonably good correspon- tlement of piled raft than the results obtained by the linear elas-
dence with well-known in-house program. tic analysis. Additionally, the analytical method is intermediate
in theoretical accuracy between general three-dimensional FE
5. Conclusions analysis and the linear elastic numerical method. The settle-
ment of piled raft obtained by the present method is similar
The primary objective of this study was to propose an improved to that obtained by the PLAXIS 3D, while it shows smaller val-
analytical method for a pile raft foundations. The conceptual meth- ues than those obtained by existing method based on linear
odology of the proposed method is completely different from that elastic analysis of pile groups.
of general continuum method. A series of analytical studies were 3. From the example case histories, the proposed method is shown
conducted. Through comparisons with case histories, it is clearly to be capable of predicting the behavior of a large piled raft.
demonstrated that the proposed method was found to be in good Nonlinear load–transfer curve and flat-shell element can over-
agreement with measurement data. From the findings of this come the limitations of existing numerical methods, to some
study, the following conclusions can be drawn: extent, by considering the realistic nonlinear behavior of soil
and membrane action of flexible raft.
1. By taking into account the raft flexibility and soil nonlinearity, 4. Additionally, the comparative studies demonstrated that the
the proposed analytical method is an appropriate and realistic present method, when used in analysis of large scale piled raft,
126 S. Jeong, J. Cho / Computers and Geotechnics 59 (2014) 112–126

is useful for computational saving and improving performance [20] Matlock H. Correlation for design of laterally loaded piles in soft clay. In: Proc.
Offshore technology conference, OTC 1204; 1970.
in engineering practice.
[21] Meyerhof GG. Ultimate bearing capacity of footing on sand layer overlaying
clay. CGJ 1974;11(2):223–9.
[22] Mindlin RD. Force at a point in the interior of a semi-in-finite solid. Physics
Acknowledgements [23] O’Neill MW. Group action in offshore piles. In: Proc specialty conference on
geotechnical engineering in offshore practice. ASCE; 1984.
This work was supported by the National Research Foundation [24] O’Neill MW, Dunnavant TW. A study of effect of scale, velocity, and cyclic
degradability on laterally loaded single piles in overconsolidated clay. Rep. No.
of Korea (NRF) grant funded by the Korea government (MSIP) (No. UHCE 84-7, Dept of Civil Engineering, Univ of Houston, Houston, TX; 1984.
2011-0030040). [25] O’Neill MW, Murchison JM. An evaluation of p–y relationship in sands. A report
to the American Petroleum Institute, PRAC 82-41-1, University of Houston,
Texas; 1983.
References [26] Arup Ove et al. GSRAFT as part of GSA user manual. London: Oasys Ltd.; 1996.
[27] Pasternak PL. On a new method of analysis of an elastic foundation by means
[1] Allen MB, Isaacson EL. Numerical analysis for applied science. John Wiley & of two constants. Gosudarstvennoe Izdatelstvo Literaturi po Stroitelstvui
Sons; 1998. Arkhitekture, Moscow; 1954 [in Russian].
[2] Burland JB, Broms BB, De Mello VFB. Behaviour of foundations and structures. [28] PLAXIS 3D Foundation. PLAXIS 3D foundation user manual, version 2.0.
In: State-of-the-Art Rep., Proc., IX Int. conf. of soil mechanics and foundation Brinkgreve, R.B., Swolfs, W.M., PLAXIS Inc.; 2008.
engineering (ICSMFE). Rotterdam, The Netherlands: Balkema; 1977. p. [29] Poulos HG. Analysis of piled strip foundations. In: Proceedings of conference
495–546. on computer methods and advances in geomechanics. Rotterdam: Balkema;
[3] Brown DA, Reese LC, O’Neill MW. Cyclic lateral loading of a large-scale pile 1991. p. 183–91.
group. J Geotech Eng, ASCE 1987;113(11):1326–43. [30] Poulos HG. An approximate numerical analysis of pile–raft interaction. Int J
[4] Chung SH, Jeong SS. Analysis of pile groups considering pile-cap interaction. Numer Anal Meth Geomech, London 1994;18(2):73–92.
M.S. thesis. Yonsei Univ; 2001. [31] Poulos HG. Piled raft foundations: design and applications. Geotechnique
[5] Clancy P, Randolph MF. An approximate analysis procedure for piled raft 2001;51(2):95–113.
foundations. Int J Numer Anal Meth Geomech 1993;17(12):849–69. [32] Randolph MF. Design of piled foundations. Research Report Soils TR143,
[6] Cox WR, Dixon DA, Murphy BS. Lateral load test of 25.4 mm diameter piles in Cambridge: Cambridge University Engineering Department; 1983.
very soft clay in side-by-side and in-line groups. Laterally loaded deep [33] Reese LC, Cox WR. Field testing and analysis of laterally loaded piles in stiff
foundations: analysis and performance, ASTM, SPT835; 1984. clay. In: Proc. offshore technology conference, OTC 2312; 1975.
[7] Daloglu AT, Vallabhan CVG. Values of k for slab on Winkler foundation. J [34] Reese LC, O’Neill MW, Smith RE. Generalized analysis of pile foundations. J Soil
Geotech Geoenviron Eng, ASCE 2000;126(5):463–71. Mech Found Div, ASCE 1970;96(1):235–50.
[8] Filonenko-Borodich M. Some approximate theories of the elastic foundation. [35] Reul O, Randolph MF. Design strategies for piled rafts subjected to nonuniform
Uchenyie Zapiski Moskovskogo Gosudarstvennoho Universiteta Mekhanica, vertical loading. J Geotech Geoenviron Eng, ASCE 2004;130(1):1–13.
vol. 46; 1940. p. 3–18. [36] Roberto C, Enrico C. Settlement analysis of pile groups in layered soils. Can
[9] Hain SJ, Lee IK. The analysis of flexible raft–pile systems. Geotechnique Geotech J 2006;43:788–801.
1978;28(1):65–83. [37] Russo G. Numerical analysis of piled rafts. Int J Numer Anal Meth Geomech
[10] Hetenyi M. Beams on elastic foundations. Ann Arbor, Michigan: University of 1998;22(6):477–93.
Michigan Press; 1946. [38] Sommer H. Entwicklung der Hochhausgru€ndungen in Frankfurt/Main Festkoll
[11] Jeong SS, Kim SI, Briaud JL. Analysis of downdrag on pile groups by finite oquium 20 Jahre Grundbauinstitut. In: Prof. Dr. -Ing. H. Sommer und Partner,
element method. Comput Geotech 1997;21(2):143–61. Germany; 1991. p. 47–62.
[12] Katzenbach R, Arslan U, Gutwald J, Holzhauser J, Quick H. Soil–structure [39] Terzaghi K. Evaluation of coefficients of subgrade reaction. Geotechnique
interaction of the 300-m-high Commerzbank Tower in Frankfurt am Main. 1955;5:297–326.
Measurements and numerical studies. In: Proc, 14th ICSMFE, vol. 2; 1997. p. [40] Vesic AS. Bending of beams resting on isotropic elastic solid. J Eng Mech Div,
1081–4. ASCE 1961;87:35–53.
[13] Katzenbach R, Arslan U, Moormann C. Piled raft foundations projects in [41] Vesic AS. Experiments with instrumented pile groups in sand. Performance of
Germany, design applications of raft foundations. In: Hemsley JA, editor, deep foundation. ASTM, special technical publication; 1969. 444, p. 172–222.
Thomas Telford; 2000. p. 323–92. [42] Viggiani C. Pali come riduttori di cedimento; un esempio. In: Proc., Atti
[14] Kitiyodom P, Matsumoto T. A simplified analysis method for piled raft and pile XIX Convegno Nazionale Geotecnica, 2, Pavia, Italy, Pàtron, Bologna; 1995.
group foundations with batter piles. Int J Numer Anal Meth Geomech p. 523–6.
2002;26:1349–69. [43] Vlasov VZ, Leontiev UN. Beams, plates, and shells on elastic foundation. Israel
[15] Kitiyodom P, Matsumoto T. A simplified analysis method for piled raft Program for Scientific Translations, Jerusalem (translated from Russian); 1966.
foundations in non-homogeneous soils. Int J Numer Anal Meth Geomech [44] Wang ST, Reese LC. COM624P – laterally loaded pile analysis for the
2003;27:85–109. microcomputer.ver. 2.0, FHWA-SA-91-048, Springfield, VA; 1993.
[16] Koizumi Y, Ito K. Field tests with regard to pile driving and bearing capacity of [45] Wang A. Three dimensional finite element analysis of pile groups and piled -
piled foundations. Soils Found 1967;7(3):30–53. raft, Ph.D. dissertation, University of Manchester, U.K., 1996.
[17] Lee IK. Analysis and performance of raft and raft–pile foundations in a [46] Winkler E. Die Lehre von der Elasizitat und Festigkeit. Dominicus; 1867.
homogeneous soil. In: Proceedings of 3rd international conference on case [47] Whitaker T. Experiments with model piles in groups. Geotechnique
history in geotechnical engineering, St Louis (also Research Report R133, 1957;7(4):147–67.
ADFA, University of New South Wales, Australia); 1993. [48] Won JO, Jeong SS, Lee JH, Jang SY. Nonlinear three-dimensional analysis of pile
[18] Lee JH, Kim YH, Jeong SS. Three-dimensional analysis of bearing behavior of group supported columns considering pile cap flexibility. Comput Geotech
piled raft on soft clay. Comput Geotech 2010;37:103–14. 2006;33:355–70.
[19] Lieng JT. Behavior of laterally loaded piles in sand-large scale model test. Ph.D. [49] Zhang HH, Small JC. Analysis of axially and laterally loaded pile groups
thesis, Department of civil engineering, Norwegian institute of technology; embedded in layered soils. In: Proceedings of 8th Australia NewZealand Conf.
1988. on Geomechanics, vol. 1. Hobart; 2000. p. 475–483.