Strip footing behavior on pile and sheet pile-stabilized sand slope

Mostafa A. El Sawwaf
Structural Eng. Dept., Faculty of Eng., Tanta University, Tanta, Egypt

This paper presents the results of laboratory model tests on the behavior of a strip footing
supported on a row of piles and sheet pile-stabilized earth slope. A comparison between
the bearing capacity improvements in the two cases was made to study the most efficient
case. The parameters varied in the study include pile diameter, pile length, pile spacing
and location of pile row, height of sheet pile, location of sheet pile and location of the
footing relative to the slope crest. Initially the bearing capacity of non-stabilized cases
were determined and then compared with those of stabilized slopes. The results were
then analyzed to study the effect of each parameter. The results indicate that stabilizing
earth slope using a row of piles or sheet pile has a significant effect in improving the
bearing capacity of the strip footing. This improvement in bearing capacity increases
when pile spacing decreases and pile length increases with further improvement with
increasing pile diameter. However, the overall improvement when using sheet pile to
stabilize earth slope is much better than that when using a row of piles. Finally, a series
of finite element analysis was performed on a prototype slope and a comparison between
the laboratory model tests and the finite element analysis was presented.

 
      
    
 

  
 ! "! !#$ %
!   & %    
! '! & '  (!)! & *+ !) (,%,! - % 
 - ), &! 
. &! % #.  (!)! *+  
! '! !)!    
!
  ()!  $  &! 
.
! !  ! 
 
 12! !! $ !   ./%
 & $  
! ! 34! 
! !   (!)! *+ & !  (!)!   ()!

 6#% 
 
! !!    .
! !  % ! &. 15 #.
! !
& %  7 
  
 ! +%!  !) 

 !   -   
!! 15 
  8%%  ! +  
! 9! :'% %  ; 7 

!  
! :'%!
. 4
)!
Keywords: Pile, Sheet pile, Stabilized sand slope, Strip footing, Finite element analysis

1. Introduction
There are many situations where footings
are constructed on sloping surfaces or adjacent to a slope crest such as footings for
bridge abutments on sloping embankments.
When a footing is located on a sloping ground,
the bearing capacity of the footing may be significantly reduced, depending on the location
of the footing with respect to the slope. Therefore it may not be possible to use shallow foundation and using uneconomic foundation
types (piles or caissons) becomes the only
suitable solution of the problem. Therefore,
over years, the subject of stabilizing earth
slope has become one of the most interesting
areas for scientific research and attracted a
great deal of attention. Slope stability can be
Alexandria Engineering Journal, Vol. 43 (2004), No. 1, 41-54
Faculty of Engineering, Alexandria University, Egypt.

increased in different ways such as: modifying

the slope surface geometry, using soil reinforcement, or installing continuous or discrete
retaining structures such as walls or piles.
There have been numerous studies on the use
of slope reinforcement to improve load bearing
capacity of a footing on the slope (Selvadurai
et al. [1], Sawicki et al. [2], Mandal et al. [3],
Huang et al. [4], Zornberg et al. [5] and Yoo C.
[6]). These investigations have demonstrated
that not only the slope stability can be
increased but also both the ultimate bearing
capacity and the settlement characteristics of
the foundation can be significantly improved
by the inclusion of reinforcements (layers of
geogrid, strips or geotextile) in the earth slope.
On the other way, the use of stabilizing
piles to support active earth slope has been

41

M. A. El Sawwaf / Strip footing behavior

considered to be one of the important slope

reinforcement techniques in the last few decades. These piles, which can be driven at the
crest or within the slope itself act as resisting
members and are usually subjected to lateral
forces by the horizontal movements of the surrounding soil. The lateral force acting on each
pile may be obtained in an approximate manner multiplying the resisting force per unit
width of the pile row by the center-to-center
spacing between the piles. However, to evaluate the force acting on the piles more accurately, arching between adjacent piles should
be considered. Several studies reported the
successful use of piles in many situations in
order to improve slope stability (De Beer and
Wallays [7], Viggiani [8], Poulos [9], Lee et al.
[10], Hong and Han [11], Chen et al. [12], Hassiotis et al. [13], Ausilio, et al. [14] and Hull
and Poulos [18]).
Poulos [9] described an approach for the
design of piles to reinforce slopes, in which
both the total shear force needed to increase
the safety factor and the maximum shear force
that each pile can provide were evaluated and
therefore the type, number of piles, and the
most suitable locations of these piles within
the slope can be selected. Hassiotis et al. [13]
proposed a method for the design of slopes
reinforced with a single row of piles based on
the theory of plasticity to find the lateral forces
acting on the pile section above the critical
surface. Chen et al. [12] presented a theoretical procedure for analyzing the lateral response of vertical piles subjected to lateral soil
movements via a simplified boundary-element
analysis. Ausilio et al. [14] reported that when
a row of piles is inserted in a slope, the additional resistance provided by these piles
changes both the slope safety factor and potential failure mechanism with respect to the
case without piles.
Several studies have been conducted in order to find out the best location of the stabilizing piles row within a slope. This was achieved
by determining the position of the piles row
that gives the maximum resistance force and
factor of safety. However, the recommended
results are rather different and contrasting. Ito
and Matsui [15] in their analytical studies
demonstrated that a row of piles installed
closer to the top of the slope gives the best

42

factor of safety but in (1979) Ito et al. [16]

showed that the best location of piles row,
which has the maximum effect on slope
stability, is the upper-middle part of the slope.
Hassiotis et al. [13] concluded that the piles
should be located close to the top of the slope
to achieve the maximum safety factor. Lee et
al. [10] analyzed cohesive soil slope and found
that when the piles are installed into a homogeneous soil the most effective pile positions
are the toe and crest of the slope and in
contrast to other researchers, stated that the
piles have little effect on slope stability when
they are located close to the middle of the
slope. Cai and Ugai [17], using the finite
element method, have pointed out that the
maximum safety factor for the slope can be
achieved when the piles are located in the
middle of the slope.
It should be mentioned that most of the
previous studies concerning the pile-stabilized
slope have aimed at the stability analysis of
the slope itself. However, the improvement in
the load bearing capacity of a footing supported on sand bed adjacent to stabilized
slope hasnt yet been investigated. Therefore
the aim of this study is to gain more understanding about the mechanical behaviour and
the failure mechanism of a strip footing supported on sand bed adjacent to pile or sheet
pile stabilized earth slope. The main objective
was to determine and establish the relationship between the variable parameters of a row
of piles or sheet pile and the bearing capacity
of the footing. Also, to find out the best location of the piles row or sheet pile that gives the
best improvement in the footing bearing capacity. So, series of experimental model tests
were carried out and the obtained results are
presented and discussed. Also, a series of
finite element analysis was performed on a
prototype slope to ascertain the validity of the
findings from the experimental model tests
and understand the deformation patterns of
soil particles underneath the footing.
2. Laboratory model tests
2.1. Model box and footing
The experimental work aimed to study the
effects of stabilizing an earth slope on the

Alexandria Engineering Journal, Vol. 43, No. 1, January 2004

M. A. El Sawwaf / Strip footing behavior

smooth to minimize the end friction effects. A

rough base condition was achieved by fixing a
thin layer of sand onto the base of the model
footing with epoxy glue. The load is
transferred to the footing through a bearing
ball as shown in fig. 1. Such an arrangement
produced a hinge, which allowed the footing to
rotate freely as it approached failure and
eliminated any potential moment transfer from
the loading fixture.
2.2. Test material
The sand used in this research is medium
to coarse sand, washed, dried and sorted by
particle size. It is composed of rounded to subrounded particles. The specific gravity of the
soil particles was determined by the gas jar
method. Three tests were carried out and the
average value was obtained. The maximum
and the minimum dry densities of the sand
were measured and the corresponding values
of the minimum and the maximum void ratios
2
3

4
5
225.0

1
10

50.0 cm800

bearing capacity of a strip footing adjacent to

the slope crest. A plane strain footing was
used in the study while a row of piles and
sheet pile wall were used to stabilize the earth
slope. The main elements of the used
apparatus are a tank, a horizontal steel beam
over the tank, and a sand-raining box. The
test box, having inside dimensions of 1.00 m *
0.50 m in plan and 0.5 m in depth is made
from steel with the front wall made of 20 mm
thickness glass and is supported directly on
two steel columns as shown in fig. 1. The
glass side allows the sample to be seen during
preparation and sand particle deformations to
be observed during testing. The tank box was
built sufficiently rigid to maintain plane strain
conditions by minimizing the out of plane
displacement. To ensure the rigidity of the
tank, the back wall of the tank was braced on
the outer surface with two steel beams fitted
horizontally at equal spacing. The inside walls
of the tank are polished smooth to reduce
friction with the sand as much as possible by
attaching fiber glass onto the inside walls.
The loading system consists of a handoperated hydraulic jack and pre-calibrated
load ring. Since the sand raining technique is
used to deposit the sand inside the tank, the
beam was designed to swing about one end.
Therefore, the beam can be swung out during
sand deposition and returned back, when
sand set up completed, to the original loading
position. The sand raining box is made from
wood and is 0.85 m * 0.48 m in plane and
0.10 m depth. The sand particles rain from
the box through a square grid of holes (4 mm
diameter and 20 mm spacing) in the base
plate. The height of sand raining, measured
from the bottom of the box to sand surface in
the tank, could be changed up or down by
using a manual winch.
A strip model footing made of steel with a
hole at its top center to accommodate bearing
ball was used. The footing is 498 mm in
length, 80 mm in width and 20 mm in
thickness. The footing was positioned on the
sand bed with the length of the footing
running the full width of the tank. The length
of the footing was made almost equal to the
width of the tank in order to prevail plane
strain conditions within the test set-up. The
two ends of the footing plate were polished

1- Manual wench

2- Wench wire
805
cm
90.0

3- Raining box
60

960

60

4- Hydraulic jack
5- Proving ring
6- Model footing

7- Soil

8- Test tank
9- Loading frame
10- Dial gauges

Dimensions are in centimeters

10.0

10.0
125.0

Fig. 1. Schematic view of the experimental apparatus.

Alexandria Engineering Journal, Vol. 43, No. 1, January 2004

43

M. A. El Sawwaf / Strip footing behavior

Table 1
The general physical characteristics of sand
Coefficient of uniformity
Effective diameter (m)
Coefficient of curvature (Cc)
Specific gravity, (Gs)
Maximum unit weight (kN/m3)
Minimum unit weight (kN/m3)
Maximum void ratio
Minimum void ratio

4.071
0.152
0.771
2.654
19.95
16.34
0.593
0.305

percent finer

100
90

2.3. The experimental setup and test program

80
70
60
50
40
30
20
10
0
0.01

0.1

10

100

Grain size, mm
Fig. 2. Grain size distribution of the used sand.

were calculated. The particle size distribution

was determined using the dry sieving method
and the results are shown in fig. 2. Table 1
summarizes the general physical characteristics of the used sand.
Sand beds were placed in -50 mm in
height- layers by raining technique in which
sand is allowed to rain through air at
controlled discharge rate and height of fall to
give uniform densities. The relative density
achieved during the tests was monitored by
collecting samples in small cans of known
volume placed at different locations in the test
tank. The raining technique adopted in this
study provided a uniform relative density of
approximately 75.8 % with a unit weight of
18.94 kN/m3. A series of direct shear tests
was performed to evaluate the shear strength
properties of the model ground using
specimens prepared by dry tamping. The
estimated internal friction angle at the same
relative density used in the model tests was
42.
Model piles made of steel with different
lengths and diameters were used in the study.

44

The piles were 6.0, 8.0, and 12.0 mm in

diameters and 80, 100, 160, 240 mm in
length. Sheet piles were 3.0 mm in thickness
and made of wood with different heights. All
the tests were performed with piles or sheet
pile installed vertically after preparing sand
samples. The difference in the relative densities of the samples which occurs during installing pile or sheet pile wall due to the difference in the pile lengths or sheet pile heights
was considered to be small and neglected.

The model soil was constructed by raining

technique with the bed level and slope
observed through the front glass wall. Then
the sand slope was set up to form a slope of
3(H): 2 (V). In raining the last layer, the sand
surface was approximately horizontal and the
sand slope was about the required inclination.
Great care was given to level the top surface of
the sand and the slope face using special
rulers so that the relative density of the top
layer was not affected. After setting up the
model slope, a row of piles or the sheet pile
was driven vertically at the design place and
spacing. The footing was placed on position
and the load was applied incrementally by the
hydraulic jack until reaching failure. Each
load increment was maintained constant till
the footing settlement had stabilized. This
settlement was measured using two 0.001 mm
accuracy dial gauges, placed on opposite sides
across the center of the footing.
A total of 80 tests in two different test
programs were carried out. Initially, the
response of the model footing supported on
the non-stabilized slopes was determined (2
tests, each one was repeated 3 times). Then,
15 series of tests (60 tests) were performed to
study the effect of the different parameters of
a row of piles on the footing behaviour as
shown in fig. 3.a. Each series was carried out
to study the effect of one parameter while the
other variables were kept constant. The varied
conditions include the pile diameter D, pile
length L, pile spacing x and the horizontal
distance between pile row and the slope crest
d as illustrated in table 2. Finally, in the sheet
pile test program, 4 series of tests (18 tests)
were performed to study the footing response

Alexandria Engineering Journal, Vol. 43, No. 1, January 2004

M. A. El Sawwaf / Strip footing behavior

x
x
x

Footing
d

Footing

H = 4.5 B

Pile

H = 4.5 B
Sheet Pile

0.5 B

0.5 B

0.5 B

(a)

0.5 B

(b)

Fig. 3. Geometric parameters of pile and sheet pile-stabilized sand slope model.
Table 2
Model tests program
Series
A
B
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
I
II
III
IV

Constant parameters
Tests on non-stabilized sand slope.
b = 0.0
Tests on non-stabilized sand slope.
b = B
D/B = 0.075 & d/B = 0.0 & L/B = 1.00
D/B = 0.075 & d/B = 0.0 & L/B = 1.25
D/B = 0.075 & d/B = 0.0 & L/B = 2.00
D/B = 0.075 & d/B = 0.0 & L/B = 3.00
D/B = 0.10 & d/B = 0.0 & L/B = 1.00
D/B = 0.10 & d/B = 0.0 & L/B = 1.25
D/B = 0.10 & d/B = 0.0 & L/B = 2.00
D/B = 0.10 & d/B = 0.0 & L/B = 3.00
D/B = 0.15 & d/B = 0.0 & L/B = 1.00
D/B = 0.15 & d/B = 0.0 & L/B = 1.25
D/B = 0.15 & d/B = 0.0 & L/B = 2.00
D/B = 0.15 & d/B = 0.0 & L/B = 3.00
D/B = 0.075 & x/B = 0.5 & L/B = 2.00
D/B = 0.10 & x/B = 0.5 & L/B = 2.00
D/B = 0.15 & x/B = 0.5 & L/B = 2.00
b/B = 0.0
& d/B = 0.0
b/B = 1.0
& d/B = 0.0
b/B = 0.0
& h/B = 2.0
b/B = 1.0
& h/B = 2.0

for the two cases when the footing is placed

exactly on the slope crest and when the
footing is placed away of the slope crest by a
distance b. For each case, two series of tests
were conducted to find out the best location of
the sheet pile that gives the maximum bearing
capacity improvement and the effects of sheet

Variable parameters

x/B = 0.5, 1.00, 1.25, 2.5

x/B = 0.5, 1.00, 1.25, 2.5
x/B = 0.5, 1.00, 1.25, 2.5
x/B = 0.5, 1.00, 1.25, 2.5
x/B = 0.5, 1.00, 1.25, 2.5
x/B = 0.5, 1.00, 1.25, 2.5
x/B = 0.5, 1.00, 1.25, 2.5
x/B = 0.5, 1.00, 1.25, 2.5
x/B = 0.5, 1.00, 1.25, 2.5
x/B = 0.5, 1.00, 1.25, 2.5
x/B = 0.5, 1.00, 1.25, 2.5
x/B = 0.5, 1.00, 1.25, 2.5
d/B = 0.0, 1.0, 1.5, 2.0, 2.5
d/B = 0.0, 1.0, 1.5, 2.0, 2.5
d/B = 0.0, 1.0, 1.5, 2.0, 2.5
h/B = 0.5, 1.0, 1.5, 2.0, 3.0
h/B = 0.5, 1.0, 1.5, 2.0, 3.0
d/B = 0.0, 1.0, 1.5, 2.0, 2.5
d/B = 0.0, 1.0, 1.5, 2.0, 2.5

Notes
3 times
3 times
piles row
piles row
piles row
piles row
piles row
piles row
piles row
piles row
piles row
piles row
piles row
piles row
piles row
piles row
piles row
sheet pile
sheet pile
sheet pile
sheet pile

pile height h as shown in fig. 3-b. Table 2

summaries all the tests programs with both
the constant and varied parameters illustrated. Several tests were repeated at least
twice to verify the repeatability and the consistency of the test data.

Alexandria Engineering Journal, Vol. 43, No. 1, January 2004

45

M. A. El Sawwaf / Strip footing behavior

3. Finite element analysis

A series of numerical analysis on a prototype footing-slope system was carried out
using finite element method. The object of this
series is to verify the laboratory model tests
results and to understand the particles deformations within the soil mass. The analysis
was performed using the PLAXIS software
package (professional version 7, Bringkgreve
and Vermeer [19]). The geometry of the prototype footing-slope system was assumed to be
10 times the laboratory model (B = 0.8 m, H =
3.60 m, the thickness of sheet pile wall = 0.30
m). The modeled boundary conditions were assumed such that the vertical boundaries are
free vertically and constrained horizontally
while the bottom horizontal boundary is fully
fixed. The same inclination of model test
slopes, 3 (H): 2 (V), and the material of sheet
pile wall (wood) were used in the prototype
study. The software allows the automatic generation of six or fifteen node triangle plane
strain elements for the soil, and three or five
node beam elements for the footing and the
sheet pile wall. The analyzed prototype slope
geometry, generated mesh, and the boundary
conditions are shown in fig. 4.
The non-linear behavior of sand was modeled using hardening soil model, which is an
elastoplastic hyperbolic stressstrain model,
formulated in the framework of friction hardening plasticity. A basic feature of the hyperbolic model is the stress dependency of soil
stiffness. The limiting state of stress are described by means of the secant Youngs
ref

modulus E 50 , tangent stiffness modulus for

ref

primary compression E oed , Poissons ratio ,

power for stress-level dependency of stiffness
m, effective cohesion c, angle of internal
friction , angle of dilatancy , failure ratio Rf
and interface reduction factor Rint. The sheet
pile wall and the footing were modeled using
elastic beam elements based on Mindlins
beam theory with significant flexural rigidity
and normal stiffness. The interaction between
the sheet pile wall and soil is modeled at both
sides by means of interface elements, which
allow for the specification of a reduced wall
friction compared to the friction of the soil.
Load control method was used to apply a

46

prescribed load in increments accompanied by

iterative analysis up to failure. As the slope
surface is not horizontal, the gravity loading
was applied to calculate the initial stress field
of the soil in steps with the sheet pile wall in
place. An internal angle of friction and secant
Youngs modulus

ref
E50

of 40 and 40000

kN/m2

derived from a series of drained triaxial

compression tests were used for the soil
material along with the hyperbolic parameters
for the sand taken from data base provided by
the software manual as shown in table 3.
4. Results and discussion
The bearing capacity improvement of the
footing due to slope stabilization is represented using a non-dimensional factor, called
Bearing Capacity Improvement factor (BCI).
This factor is defined as the ratio of the footing
ultimate pressure with the slope stabilized qu
stabilized to the footing ultimate pressure in tests
without slope stabilizing qu. The footing
settlement S is also expressed in non-dimensional form in terms of the footing width B as
the ratio S/B %. The ultimate bearing capacities for the footing-soil systems are determined
from the loaddisplacement curve as the pronounced peaks, after which the footing collapses and the load comes down. The measured ultimate bearing capacity for non-stabilized cases when the footing was placed at distances d = 0.0 and d = B from the slope crest
are 17.30 and 36.69 kPa, respectively.
4.1. Pile-stabilized slopes
Fig. 5 presents typical variations of BCI with
settlement ratio S/B for a strip footing
supported on sand bed adjacent to both stabilized and non-stabilized earth slope (series 1).
In this series all the variable were kept constant but pile spacing varied. It is clear that
the installation of a row of piles much improves both the initial stiffness (initial slope of
the loadsettlement curves) and the bearing
load at the same settlement level for stabilized
case compared to the non-stabilized earth
slope. Also, the bearing capacity improvements at failure are significantly dependent on
the pile spacing. However, the curves show
that these improvements in bearing capacity

Alexandria Engineering Journal, Vol. 43, No. 1, January 2004

M. A. El Sawwaf / Strip footing behavior

The footing
The sheet pile wall

Fig. 4. The prototype slope geometry, generated mesh, and

boundary conditions .
Table 3
Hardening soil model parameters used in the finite
element analysis

ref

Primary loading stiffness, ( E50 ), kN/m2

kN/m2

Cohesion (c),
Friction angle,
Dilatancy angle,
Soil unit weight (), kN/m3
Power in stiffness law, m
Failure ratio Rf
Interface reduction factor (Rint)

BCI
0

Value
40000

4.1.1. The effect of pile diameters

In order to study the effect of pile
diameter, a series of tests using a row of piles
installed at slope crest with the same
normalized pile length L/B=3 with pile
diameters D of 6.0, 8.0, and 12.0 mm were
carried out. Fig. 6 illustrates the variations of
bearing ca BCI with settlement ratio S/B for a
footing placed near to the crest b/B=0. The
figure clearly shows that the footing performance much improves with the increase in pile
diameter. It can be seen that an improvement
as high as 2.03 % more than that without
piles can be obtained when using normalized

0.5

0.75

1.25

1.5

1.75

pile diameter

= 6 mm

0.00
40.0
10
18.90
0.70
0.90
0.8

are accompanied with an increase in the

settlement ratio. Comparing the curves across
the dotted line for the same level of settlement
ratio, the figure illustrates that using a row of
piles with x/B = 0.5 at the slope crest brings
out an increase in the bearing capacity of 50%
more than that of non-stabilized case. The
ultimate bearing capacities for different cases
of earth slopes stabilized by a row of piles are
given in table 4-a and 4-b. These results for
each parameter are discussed in the following
sections.

0.25

2
3
4
5
6

S/ B, %

Parameter

pile diameter D/B of 0.15. This can be

illustrated that as the pile diameter increases
its resistance to the lateral movement of soils
under the footing increases and hence the
bearing capacity ratio improves. Also, for the
same number of piles in a row, increasing the
pile diameter decrease the clear distance between piles. Therefore, not only less soil
particles are likely to move through but also
the pile resistance for lateral movement by
arching effect is getting higher leading to
better performance of the pile row as retaining
system of the lateral movements.

no piles
x/B = 2.5
x/B = 1.5
x/B = 1.0
x/B = 0.5

Fig. 5. Typical variations of BCI with S/B for different pile

spacing (Series 1).

4.1.2. The effect of pile length

Fig. 7 shows the variation of bearing
capacity improvement factor (BCI) with
normalized pile length (L/B) for a footing
placed near to the slope crest (b/B=0) and
using a row of piles installed at slope crest.
The three curves demonstrate the same
pattern of footing response for the different
pile diameters. As the pile length increases,
the improvement in bearing capacity is
becoming higher. Using a row of piles with
x/B = 0.5 and D/B =0.15 with normalized pile
length of L/B = 3 instead of L/B = 1 increases
the improvement in bearing capacity by 20
%(rom 1.75 to 2.1).

Alexandria Engineering Journal, Vol. 43, No. 1, January 2004

47

M. A. El Sawwaf / Strip footing behavior

2.25

BCI

Table 4-a
Ultimate bearing capacity (kPa ) from model test results of
pile stabilized slope program

Pile
diameter

L/B

X/B
0.5

1.0

1.25

2.5

1.0

27.25

23.33

22.01

18.08

1.25

28.83

24.11

23.06

18.61

2.0

29.88

25.16

24.64

19.40

3.0

33.02

28.31

26.21

20.44

1.0

28.83

24.64

23.06

18.35

1.25

29.88

25.69

24.11

18.87

1.75

1.5

6 mm
x/B
x/B
x/B
x/B

1.25

=
=
=
=

0.50
1.00
1.25
2.50

1
0

8 mm
2.0

31.45

27.25

26.21

19.92

3.0

34.60

29.88

27.26

21.49

1.0

30.40

26.52

24.90

19.24

1.25

31.45

28.30

25.69

19.92

2.0

33.02

29.36

26.73

20.97

3.0

35.12

31.45

28.83

22.02

0.05

0.1

0.15

0.2

(D/B)
Fig. 6. Variations of (BCI) with normalized pile diameter
(D/B) for different pile spacing (b/B=0, L/B=3.0).

12 mm

Table 4-b
Ultimate bearing capacity (kPa ) from model test results
of the series of pile row location (Series 13,14 and 15)

BCI

2.25

1.75

1.5

d/B

D/B = 0.15

Pile
diameter

0.0

1.0

1.5

2.0

2.5

6 mm

29.88

27.78

25.69

22.54

20.44

8 mm

31.45

28.83

26.21

23.59

20.97

12 mm

33.02

29.88

26.73

24.64

22.02

This increase in pile length results in

increasing the part of pile embedded in the
stable underlying soils leading to more
stability for the pile and higher resistance for
the lateral movement of soil particles under
the footing.
4.1.3. The effect of pile spacing
In order to investigate the effect of pile
spacing on the footing performance, a row of
piles with four different pile spacing of 40.0,
80.0, 100.0 and 200 mm were used. Fig. 8
shows the variations of BCI with normalized
pile spacing x/B for different pile lengths. A
significant increase in the bearing capacity of
the model footing supported on stabilized sand

48

1.25

D/B = 0.10
D/B = 0.075

1
0

0.5

1.5

2.5

3.5

(L/B)
Fig. 7. Variations of (BCI) with normalized pile length (L/B) for
different pile diameters (d/B = 0 and x/B = 0.5).

slope with the decrease of normalized pile

spacing x/B can be noticed. Using a row of
piles with normalized pile spacing x/B of 0.5
and normalized pile length L/B =3 could bring
out an improvement in bearing capacity as
high as 2.12 times more than that without
slope stabilizing. However, for the same row of
piles with normalized pile spacing x/B of 2.5,
the improvement in bearing capacity is only
1.28. Therefore, decreasing the pile spacing
from x/B of 2.5 to x/B of 0.5 increases the
improvement in the bearing capacity by about
65 %.

Alexandria Engineering Journal, Vol. 43, No. 1, January 2004

M. A. El Sawwaf / Strip footing behavior

2.25

BCI

L/B = 3.00
L/B = 2.00
2

L/B = 1.25
L/B = 1.00

1.75

1.5

1.25

1
0

0.5

1.5

2.5

(x/B)
Fig. 8. Variations of (BCI) with normalized pile spacing
(x/B) for different pile lengths (d/B = 0 and D/B = 0.15).

This expected increase in the bearing

capacity of the footing can be explained for the
following two reasons. The first is that when
pile spacing is increased, the clear distance
between piles increases allowing more soil to
move through and larger lateral movement of
soil to occur under the footing. The second is
that for small pile spacing the component of
pile resistance for lateral movement by arching effect is higher leading to a decrease in the
lateral movement of the soil and hence increasing both the slope stability and the footing
bearing capacity.
4.1.4. The effect of the location of pile row
It was shown that there was some confusion among researchers about the optimal location of piles row that gives the best stability
of a slope and the maximum factor of safety.
Different locations were reported such as the
slope crest and the slope toe while others
stated that the row placed in the middle part
is the best location. However, it was of interest
to find out experimentally the best location of
pile row from the point of view of the footing
bearing capacity rather than the overall
stability of the slope. So, series of tests using
the same row of piles (x/B =0.5 and L/B = 2)
at five different locations relative to the slope
crest (d = 0.0, 80.0, 120.0, 160 and 200.0

mm) were carried out using different pile

diameters. Fig. 9 shows the variations of BCI
with normalized pile row location d/B for
different pile diameters. It can be seen that as
the pile row is placed nearer to the slope crest
the response of the footing is getting much
better in terms of bearing capacity than anywhere else. The same trend is confirmed by
the different series carried out using different
pile diameters.
This conclusion that the most effective pile
location is at the slope crest can be referred to
the fact that the passive wedge under the
footing is relatively shallow and hence the mobilized passive resistance is getting much
higher when the pile row are placed at the
crest. Any other position far from that location
may increase the overall stability of the slope
but cant prevent or decrease the lateral deformations of soil particles under the footing and
near to the slope.
In order to understand the mechanism of
footing failure and whether or not it was accompanied with slope failure, additional increments of loads were applied after failure point
(the pronounced peak in the load - settlement
readings) and both the footing and the slope
were observed through the front glass wall. It
was noticed that, as the footing approached
failure the vertical settlements were accompanied with horizontal movements and rotations
(toward the slope). It is very important to mention that in all the experimental tests, there
wasnt any failure observed in the slope itself
even after footing settlement S/B ratio of 50%.
4.2. Sheet pile stabilized sand slope
The improvement in footing response due
to slope stabilization by sheet piles with
different heights h and locations relative to the
slope crest d were investigated. The bearing
capacity improvement factor BCI along with
the footing settlement ratio S/B were used to
present the test results. Table 5-a and 5-b
gives the ultimate bearing loads for the
different studied conditions.
4.2.1. The effect of sheet pile height
Two series of tests were performed on sand
slope stabilized with sheet pile placed at slope

Alexandria Engineering Journal, Vol. 43, No. 1, January 2004

49

BCI

BCI

M. A. El Sawwaf / Strip footing behavior

D/B = 0.15

1.9

D/B = 0.10
1.8

2.8
b/B = 0.0
2.6
b/B = 1.0

D/B = 0.075

2.4

1.7

2.2

1.6
2

1.5
1.8

1.4
1.3

1.6

1.2

1.4

1.1

1.2

0.5

1.5

2.5

(d/B)
Fig. 9. Variations of BCI with normalized pile row location
d/B for different pile diameters (x/B = 0.5 and L/B = 2.0).
Table 5-a
Ultimate bearing capacity (kPa ) from model test results
of the series of sheet pile height (d/B=0.0)
h/B
b/B

0.5

1.0

1.5

2.0

3.0

0.0

29.88

32.50

34.34

37.48

45.34

1.0

38.27

40.62

42.20

45.34

49.54

Table 5-b
Ultimate bearing capacity (kPa ) from model test results
of the series of sheet pile location (h/B=2.0)
d/B
b/B

0.0

1.0

1.5

2.0

2.5

0.0

37.48

34.07

30.93

26.73

20.97

1.0

45.34

43.51

42.46

40.89

39.31

crest (d/B = 0.0) to find out the effects of sheet

pile height h on the footing behaviour for the
two cases when the footing is constructed at (b
=0) and away of the slope crest distance b=B.
In Fig. 10, the variations of BCI with normalized sheet pile height h/B for different footing
position are plotted. The figure shows that the
larger is the sheet pile height, the higher is the
improvement in bearing capacity. Using a
sheet pile with h/B = 3, placed at slope crest
brings out improvements in the bearing
capacity as high as 2.62 and 1.35 times those
of non-stabilized cases for the two footing
positions (b=0 and b=B), respectively.

50

0.5

1.5

2.5

3.5

(h/B)
Fig. 10. Variations of BCI with h/B for the two footing
positions (d/B = 0.0).

This significant effect of sheet pile height

can be explained by the particles lateral deformations resistance mobilized by the installation of sheet pile compared with that of the
non-stabilized case. The longer is the sheet
pile, the higher is the embedded part in the
stable underlying soils, the less rotation of the
sheet pile would be and hence the less lateral
deformations. However, for the footing placed
at (b/B = 1) the effect of the sheet pile is not
such as that when the footing placed at the
crest as the bearing capacity of the nonstabilized case is relatively higher. When the
footing is placed far from the crest, the passive
resistance is relatively large and the contribution of the sheet pile on the bearing capacity
improvement seems to be lower.
4.2.2. The effect of sheet pile location
Two series of tests were conducted to
investigate the best location of the sheet pile
that gives the maximum bearing capacity
improvement for the two footing positions.
Tests using the same sheet pile (h/B = 2.0) at
five different locations relative to the slope
crest (d = 0.0, 80.0, 120.0, 160 and 200.0
mm) were carried out and the test results are
plotted in fig. 11. The figure clearly shows the
same pattern of footing behaviour when using
sheet pile as using row of piles. The most
effective location of the sheet pile in terms of
footing response is at the slope crest. Also, as
the distance between the sheet pile and the
slope crest increases the footing bearing cap-

Alexandria Engineering Journal, Vol. 43, No. 1, January 2004

M. A. El Sawwaf / Strip footing behavior

In order to find out the best method to

stabilize a sand slope in terms of the footing
response, a comparison between the bearing
capacity improvements of the footing for the
two methods is drawn in fig. 12. It was taken
into account to make the comparison for the
most similar conditions; the location of piles
or sheet pile and the position of the footing are
the same (d/B =0 and b/B = 0). However for
piles row, the behaviour of the maximum diameter is considered in the comparison. The effect of pile stiffness (made of steel) and the
sheet pile stiffness (made of wood) was considered relatively low as Lee et al. [10] stated that
pile stiffness appears to have little effect on
the overall pile-slope stability. The figure
clearly confirms the expected trend that the
sheet pile has much better effect on the footing behaviour than the row of piles has. This
effect is much pronounced for larger heights of
sheet piles as the resistance of the sheet pile
increases due to the increased embedded part
of the sheet pile in stable layers of soils.

2.4

BCI

4.3. Comparison of pile and sheet pile

stabilized slope

and 1.0 are 90 and 160 kN/m respectively.

Figs. 13 and 14 present the failure pattern
and deformed mesh for a footing placed at the
crest of non stabilized slope and a footing
placed at the crest of sheet pile stabilized
slope with h/B = 3. Fig. 13 clearly shows the
tendency of ground movement and the footing
rotation, which has significant effect on the
bearing capacity, toward the slope face. Also,
the observed behaviour from the experimental
tests that the improvement in the bearing
capacity is accompanied with an increase in
the settlement ratio is confirmed by the FEA
results as can be seen in fig. 14. The improvement obtained in the ultimate load is about

b/B = 0.0
2.2
b/B = 1.0
2
1.8
1.6
1.4
1.2
1
0

0.5

1.5

2.5

(d/B)
Fig. 11.Variations of BCI with d/B for the two footing
positions (h/B = 2.0).
2.8

BCI

acity decreases. The sheet pile strengthens the

soil layer underlying the footing and resists
the soil lateral movements leading to a substantial increase in bearing capacity. This effect
is much pronounced with the sheet pile placed
at slope crest and decreases as the sheet pile
is placed far away of the crest. Also, the installation of the sheet pile is significant for the
footing response when the footing is placed
near to the crest and that effect decreases as
the distance between the footing and the slope
crest increases.

2.6
2.4
2.2

5. Theoretical analysis
Numerical study on the effect of the sheet
pile height and location relevant to the slope
crest using finite element method was performed. The ultimate bearing capacity for both
non-stabilized and stabilized prototype slopes
were determined and then investigated using
the normalized ultimate bearing capacity
improvement BCI. The ultimate bearing capacity of the prototype footing for non-stabilized
cases for the two footing positions b/B = 0.0

2
1.8
1.6
1.4
sheet pile, d/B=0.0, b/B=0.00
1.2

pile, d/B=0.0, b/B=0.0, D/B=0.15

1
0

0.5

1.5

2.5

3.5

(h/B) or L/B
Fig. 12. Comparison of the BCI with sheet pile height h/B
or pile length L/B.

Alexandria Engineering Journal, Vol. 43, No. 1, January 2004

51

M. A. El Sawwaf / Strip footing behavior

60% with an increase in the S/B from 8.03 %

to 11.34%.
Figs. 15 and 16 show typical plots of the
displacement vectors obtained from the finite
element analysis for both the non-stabilized
and stabilized earth slope respectively. As
seen in fig. 15, the observed displacement
vectors at failure for non stabilized slope are
concentrated underneath the footing toward
the slope face, while for the stabilized slope as
shown in fig. 16, the displacement vectors
spread underneath the footing along the sheet
pile face for deeper depth than that in the non
stabilized case. It is clear that the sheet pile
wall acts as a barrier for the particle deformations toward the slope face and pushes them
downward for deeper depth and hence spreads
the footing load deeper into the soil which in
turn means a longer failure surface and
greater bearing capacity.
Fig. 17 presents comparisons of the measured BCI from the model tests (exper.) and the
calculated BCI from the Finite Element Analysis FEA of prototype slope for the different

Deformed Mesh
Extreme total displacement 64.26*10-3 m
(displacements scaled up 10.00 times)

sheet pile heights and footing positions. The

curves clearly demonstrate good agreement in
the general trend of behaviour of both the experimental and numerical analysis. The BCI
increases with the increase of h/B for the two
footing positions. However, the agreement between the measured and calculated values is
much better when the footing is placed away
of the slope crest b/B=1.0 than that when it is
placed at the crest where the difference in the
BCI values appear to be high. Fig. 18 shows
the variations of BCI with the location of the
sheet pile wall for both model footing and
prototype slope. As illustrated, although the
BCI from the FEA appears to be smaller than
that for the model slope (specially for b/B=0),
the general trends of the manner in which BCI
varies with the location of the sheet pile wall
are in good agreement with those from the
model tests. The difference between model test
results and FEA can be illustrated as follow.
The physical model used in this study is
small scale while the problem analyzed by the
FEA is a prototype footing-slope system encountered in the field. Furthermore, it is well
known that due to scale effects soil may not
play the same role in the laboratory models as
in the prototype. However the comparison
shows a close agreement in the general trend
of behaviour which should be considered encouraging to rely on the model scale test results. So, it is recommended to carry out
further investigation using full-scall tests or
centrifugal model tests in order to ascertain
the obtained results.

Fig. 13. Typical deformed finite element. mesh for non

stabilized earth slope.

Fig. 15. Typical plots of displacement vectors for non

stabilized earth slope.

6. Conclusions
Deformed Mesh
Extreme total displacement 90.71*10-3 m
(displacements scaled up 10.00 times)

Fig. 14. Typical deformed finite element mesh for sheet

pile stabilized earth slope (h/B = 3.0).

52

Based on the experimental and theoretical

analysis, the following conclusions can be
drawn:

Alexandria Engineering Journal, Vol. 43, No. 1, January 2004

M. A. El Sawwaf / Strip footing behavior

Fig. 16. Typical plots of displacement vectors for sheet pile

stabilized earth slope (h/B = 3.0).
BCI

2.8
b/B
b/B
b/B
b/B

2.6
2.4

=
=
=
=

0.0 & Exper.

0.0 & FEA
1.0 & Exper.
1.0 & FEA

2.2
2
1.8
1.6
1.4
1.2
1
0

0.5

1.5

2.5

3.5

(h/B)

Fig. 17. Comparisons between the experimental and the

numerical results for the different sheet pile heights h/B
for the two footing positions (d/B = 0.0).
BCI

2.4
b/B
b/B
b/B
b/B

2.2
2

=
=
=
=

0.0 & Exper.

0.0 & FEA
1.0 & Exper.
1.0 & FEA.

1.8

improves the bearing capacity performance of

the footing. However pile spacing has much
significant
effect
on
bearing
capacity
improvement than that of pile length or pile
diameter.
3. The overall improvement when using sheet
piles to stabilize earth slope is much better
than that when using piles. The longer is the
sheet pile, the higher would be the bearing
capacity improvement.
4. In terms of bearing capacity improvements
rather than the overall stability of the slope,
the optimal location of a row of piles or a sheet
pile is at the slope crest.
5. For
the studied problem geometry
condition, the lateral deformations of soils
under the footing toward the slope are the
controlling factor of the footing failure rather
than the stability of the slope itself. The
footing suffers not only vertical settlement but
also rotation towards the slope without any
potential effect on the slope stability.
6. The general trend of behaviour of the
measured experimental tests is similar to that
of the calculated numerical results. The
difference in the bearing capacities values
between the two cases may be due to scall
effects. However the close agreement in the
trend is encouraging to rely on the model and
further investigation using full-scall tests or
centrifugal model tests is recalled in order to
ascertain the obtained results.
Acknowledgements

1.6
1.4
1.2
1
0

0.5

1.5

2.5

(d/B)

Fig. 18. Comparisons between the experimental and the

numerical results for the different sheet pile locations
(d/B) for the two footing positions (h/B = 2.0).

1. Stabilizing earth slope using a row of piles

or sheet pile has a significant effect on improving the bearing capacity of a strip footing
supported on granular soil adjacent to the
slope crest.
2. The maximum bearing capacity is obtained
when pile spacing is minimum and pile length
is maximum. Increasing pile diameter further

The tests were performed in Soil Mechanics Laboratory of Structural Engineering Department, University of Tanta which is acknowledged. The author would like to appreciate the valuable comments of Prof. Dr. M. A.
Mahmoud. Also, the support provided by Prof.
Dr. M. M. Bahloul is gratefully acknowledged.
References
[1]

[2]

A. Selvadurai and C. Gnanendran, An

Experimental Study of a Footing Located
on a Sloped Fill: Influence of a Soil
reinforcement
layer,
Canadian
Geotechnical Journal, Vol. 26 (3) (1989).
A. Sawicki and D. Lesniewska Stability
of Fabric Reinforced Cohesive Soil

Alexandria Engineering Journal, Vol. 43, No. 1, January 2004

53

M. A. El Sawwaf / Strip footing behavior

Slopes, Geotextiles and Geomembranes,

Vol. 10 pp. 125146 (1991).
[3] J. Mandal and L. Labhane, A procedure
for the Design and Analysis of
Geosynthetic Reinforced Soil Slopes,
Geotechnical
and
Geological
Engineering, Vol. 10 (4) pp. 291319
(1992).
[4] C. Huang, F. Tatsuoka and Y. Sato,
Failure Mechanisms of Reinforced Sand
Slopes Loaded with a Footing, Soils and
Foundations, Vol.
24 (2) pp. 2740
(1994).
[5] J. Zornberg, N. Sitar, and J. Mitchell,
Performance of Geosynthetic Reinforced
Slopes
at
Failure,
Journal
of
Geotechnical
and
Geoenvironmental
Engineering, Vol. 124 (8) pp. 670683
(1998).
[6] C. Yoo, Laboratory Investigation of
Bearing Capacity Behavior of Strip
Footing on Geogrid-Reinforced Sand
Slope, Geotextiles and Geomembranes,
Vol. 19 pp. 279298 (2001).
[7] E. De Beer and M. Wallays, Stabilization
of a Slope in Schist by Means of Bored
Piles Reinforced with Steel Beams, In:
Proc. 2nd International Congress on
Rock Mechanics, Beograd (1970).
[8] C. Viggiani, Ultimate Lateral Load on
Piles Used to Stabilize Landslides, In:
Proc. 10th International Conference on
Soil
Mechanics
and
Foundation
Engineering, Stockholm (1981).
[9] H. G. Poulos, Design of Reinforcing Piles
to Increase Slope Stability, Canadian
Geotechnical Journal, Vol. 32 (5), pp.
808818 (1995).
[10] C. Lee, T. Hull and H. Poulos, Simplified
Pile-Slope Stability Analysis, Computers
and Geotechnics, Vol. 17 pp. 116
(1995).
[11] WP Hong and JG. Han, The Behavior of
Stabilizing Piles Installed in Slopes, In:
Proc. 7th International Symposium on
Landslides, Rotterdam, pp. 1709714
(1996).
[12] L.T. Chen and H.G. Poulos, Piles
Subjected to Lateral Soil Movements,
Journal
of
Geotechnical
and
Geoenvironmental Engineering, Vol. 123
(9) pp. 802 - 811 (1997).

54

[13] S. Hassiotis, J. Chameau and M.

Gunaratne,
Design
Method
for
Stabilization of Slopes with Piles,
Journal
of
Geotechnical
and
Geoenvironmental Engineering, ASCE,
Vol. 123 (4) (1997).
[14] S. Ausilio, E. Conte, G. Dente, Stability
Analysis of Slopes Reinforced with Piles,
Computers and Geotechnics, Vol. 28 pp.
591611 (2001).
[15] T. Ito and T. Matsui, Methods to
Estimate Lateral Force Acting on
Stabilizing Piles, Soils and Foundations,
Vol. 15 (4) pp. 4359 (1975).
[16] T. Ito, T. Matsui and WP. Hong,
Design Method for the Stability
Analysis of the Slope with Landing
Pier Soils and Foundations, Vol.
19 (4) pp. 4357 (1979).
[17] F. Cai, and K. Ugai, Numerical
Analysis of the Stability of a Slope
Reinforced with Piles, Soils and
Foundations, Vol. 40 (1) pp. 7384
(2000).
[18] T. Hull and H. Poulos, Design Method
for Stabilization of Slopes with Piles,
Journal
of
Geotechnical
and
Geoenvironmental Engineering, ASCE,
Vol. 125 (10) pp. 911913 (1999)
[19] R. B. J. Bringkgreve and P. A. Vermeer,
PLAXIS-Finite Element Code for Soil
and Rock Analysis, Version 7 Plaxis B.
V., the Netherlands (1998).
[20] G.
Gudehus
and
W.
Schwarz,
Stabilization of Creeping Slopes by
Dowels, In: Proc. 11th International
Conference on Soil Mechanics and
Foundation Engineering, San Francisco
(1985).
[21] T. Ito, T. Matsui and WP. Hong, Design
Method for Stabilizing Piles Against
Landslide - One Row of Piles, Soils and
Foundations, Vol. 21 (1) pp. 2137
(1981).
[22] G. Poulos, Analysis of Piles in Soil
Undergoing Lateral Movement Journal
of Soil Mechanics and Foundation
Division, ASCE, Vol. 99 (5), pp. 391 406
(1973).
[23] D. W. Taylor, Fundamentals of soil
mechanics New York: John Wiley and
Sons (1948).

Alexandria Engineering Journal, Vol. 43, No. 1, January 2004

M. A. El Sawwaf / Strip footing behavior

Received August 4, 2003

Accepted November 30, 2003

Alexandria Engineering Journal, Vol. 43, No. 1, January 2004

55