International Journal of Civil Engineering and Technology (IJCIET)

Volume 7, Issue 4, July-August 2016, pp. 62–80, Article ID: IJCIET_07_04_006

Available online at
http://www.iaeme.com/IJCIET/issues.asp?JType=IJCIET&VType=7&IType=4
ISSN Print: 0976-6308 and ISSN Online: 0976-6316
© IAEME Publication

RELIABILITY BASED ANALYSIS OF

BEARING CAPACITY OF FOOTINGS ON
SOFT SOIL STRENGTHENED BY STONE
COLUMNS
Mohammed Y. Fattah
Professor, Building and Construction Engineering Department, University of
Technology, Baghdad, Iraq.

Mohammed A. Al-Neami
Assistant Professor, Building and Construction Engineering Department, University
of Technology, Baghdad, Iraq.

Ahmed S. H. Al-Suhaily
Former Graduate Student, Building and Construction Engineering Department,
University of Technology, Baghdad, Iraq.

ABSTRACT
Uncertainties in geotechnical engineering are inevitable. The soil
properties may disperse within a significant range over a domain. Thus, the
factor of safety is used in the deterministic approach which account for the
uncertainty associated with the soil properties. Deterministic approach does
not consider the sources and amount of uncertainty associated with the
system. Limit state design of the structure is difficult to estimate using
deterministic methods. So, it is reasonable to study the probability of failure of
the structure.
In this study, a procedure for carrying out reliability analysis of bearing
capacity of foundation resting on soil improved by stone columns is described.
The procedure requires definition of standard deviation of the undrained
shear strength, shear strength of the stone and the surrounding soil.
The procedure is an extension of the point estimate method in which the
expected values of the standard deviation of the capacity and demand
functions are calculated. The probability of failure, the reliability, central
factor of safety and reliability index are calculated as appropriate.
Two non-linear regression models have been made using SPSS V.20
(Statistic Computer Application) program. Data of experimental work from
previous studies are analyzed for building a bearing capacity equation of
floating and end bearing stone column group.

http://www.iaeme.com/IJCIET/index.asp 62 editor@iaeme.com
 Reliability Based Analysis of Bearing Capacity of Footings on Soft Soil Strengthened by
Stone Columns

It was concluded that for a given factor of safety, the probability of failure
is approximately the same whether the stone column was singular or within a
group, and the reliability index increases with the increase of factor of safety
for both stone columns, while the probability of failure decreases with the
increase of factor of safety.
Key words: Uncertainty, factor of safety, stone column, bearing capacity,
reliability index.
Cite this Article: Mohammed Y. Fattah, Mohammed A. Al-Neami, Ahmed S.
H. Al-Suhaily, Reliability Based Analysis of Bearing Capacity of Footings on
Soft Soil Strengthened by Stone Columns. International Journal of Civil
Engineering and Technology, 7(4), 2016, pp.62–80.
http://www.iaeme.com/IJCIET/issues.asp?JType=IJCIET&VType=7&IType=4

1. INTRODUCTION
Foundations are generally failed by two types of failure that the designer should
account for. These possible failures are structural and soil failures. Structural failure
consists of shear failure and moment failure.
Soil failure consists of bearing capacity failure and settlement failure. In addition
to the above, the foundation should meet other requirements. It should be properly
located with respect to any future influence which could adversely affect its
performance such as frost action, high volume change adjacent structure, property
line. Generally, all mentioned requirements are independent of each another, and all
should be satisfied (Ranjan and Rao, 1993).
“Certainty” can be a description for most engineering problems, where their
cause-effect relationships and their involved parameters are known precisely to give
only unique output. The complementary problems are “uncertainty” where several
outputs arise from a solution due to the variability in the determination of the
parameter involved and the incomplete understanding in the totality of the casual
relationships (Rethati, 1988).
The sources of uncertainty are unavoidable and they come from the following
(Bowles, 1996):
• The incomplete knowledge of the subsoil conditions.
• Inherent variability in soil parameters.
• Lack of control over environmental changes after construction.
• The accuracy of the theoretical or empirical methods for calculating bearing capacity.
• Predication of the applied loads such as dead loads, live loads, wind loads,
earthquake, etc.
Accordingly, the design of foundation is uncertain, in general, variability and
randomness cause a difficulty in selecting the suitable design parameter.
During the last few decades, numerous remarks were raised against the factor of
safety, as many authors see disadvantages in disregarding the reliability of the applied
data and the risk reflecting the economical background. In other word, the empirical
choice of a certain value of a safety factor does not convey the safety quantitatively
and its effect can be neglected by presence of large uncertainties in the design
environment.

http://www.iaeme.com/IJCIET/index.asp 63 editor@iaeme.com
 Mohammed Y. Fattah, Mohammed A. Al-Neami and Ahmed S. H. Al-Suhaily

There are many ground improvement techniques to enhance the strength and
compressibility behaviour of these deposits. Important among these techniques are
stone columns, lime columns, granular trenches etc. The choice of a particular
technique depends upon the soil profile at a site, quality of materials used in the
technique, nature of pore fluid and the method of implementing the technique.
Further, the successful application of a ground improvement technique depends also
on the quality assurance tests conducted for the improved behaviour of the ground and
the performance studies spread over certain period by maintaining settlement records
etc. Even though substantial work through research has been carried out on these
techniques, confidence to any field engineer develops mainly based on the successful
field applications (Kameswara et al., 1993).
The stone column technique of ground treatment has successfully proven in
• Improving slope stability of both embankments and natural slopes,
• Increasing bearing capacity,
• Reducing total and differential settlements,
• Reducing the liquefaction potential of loose sands, and
• Increasing the time rate of settlement.
Stone columns are used to support structures overlying both very soft to firm
cohesive soils and also loose silty sands having greater than about 15 percent fines,
(Barksdale and Bachus, 1983).
The high potential for beneficial use of stone columns is mainly as a ground
improvement technique to strengthen weak and soft soil. This includes the area of
highway; railway and also airport applications prompted a comprehensive
investigation to determine how and why the system works so well, and to develop
appropriate design and construction guidelines. This has resulted in many empirical
design concepts to be published for the purpose of designing the stone column.
Phoon and Kulhawy (2004) presented an overview of the development in
structural and geotechnical design practice over the past half a decade or so in relation
to how uncertainties are dealt with, this would provide a valuable historical
perspective of our present status and important outstanding issues that remain to be
resolved. The key elements of reliability-based design (RBD) are briefly discussed
and the availability of statistics to provide empirical support for the development of
simplified RBD equations is highlighted. Reliability-based design, simplified or
otherwise, provides a more consistent means of managing uncertainties, but it is by no
means a perfect solution. Engineering judgment still is indispensable in many aspects
of geotechnical engineering – reliability analysis merely removes the need for
guesswork on how uncertainties affect performance and is comparable to the use of
elasto-plastic theory to remove the guesswork on how loads induce stresses and
deformations.
Dasaka et al. (2005) investigated the probabilistic analysis of bearing capacity of
strip footing resting on cohesion less soil deposit. The calculated factors of safety
corresponding to a target reliability index of 3 are 7.3 and 5.5 respectively for simple
and advanced probabilistic analysis. These factors of safety are generally considered
higher than those adopted in routine foundation designs. The higher values of factors
of safety associated with allowable bearing pressure obtained by probabilistic
approach clearly demonstrates the importance of uncertainty studies in geotechnical
engineering and strongly demands the need to include probabilistic framework in
geotechnical engineering design.

http://www.iaeme.com/IJCIET/index.asp 64 editor@iaeme.com
 Reliability Based Analysis of Bearing Capacity of Footings on Soft Soil Strengthened by
Stone Columns

Fattah (2010) recommended a procedure to estimate the bearing capacity of

axially loaded piles based on reliability calculations. The procedure is an extension of
the point estimate method in which the expected values of the standard deviation of
the capacity and demand functions were calculated. The probability of failure, the
reliability, central factor of safety and reliability index are calculated as appropriate.
The procedure was then applied to two cases where the pile in the first case is driven
in sand while in the second, it is driven in clay. It was found that the proposed
procedure is simple and can be extended to other applications in geotechnical
engineering.
Honjo et al. (2010) proposed scheme of a reliability based design (RBD) for
practicing geotechnical engineers. The essence of the proposed scheme is the
separation of the geotechnical design part from the uncertainty analysis part in
geotechnical RBD. In this way, practical engineers are able to perform RBD in a more
comfortable way compared to the traditional RBD procedure. Based on the results,
some discussions were made to identify the major issues geotechnical RBD is facing.
It was concluded that spatial variability of soil properties is only one of the sources of
uncertainty. In many design problems, statistical estimation error, design calculation
model error and transformation error associated with estimating soil parameters (e.g.
friction angle) from the measured quantities (e.g. SPT N-values) have higher
uncertainty. It is important to recognize these aspects in developing the geotechnical
RBD to next and the higher stage.
Fattah et al. (2013) found an experimental base for the value of the stress
concentration ratio by manufacturing a model of a single stone column with rigid
instrumented loading plates such that the total load applied to the model footing, and
the load applied to the stone column can be measured alone. Model tests of soil
treated with a single stone column and groups of two, three and four stone columns
were tested. In addition, model tests were prepared with 10% cement as additive to
backfill material for the columns. The backfill material was also modified by mixing
30% sand by weight with 70% crushed stone. It was found that the addition of the
sand to the crushed stone indicated a marginal effect on stress concentration ratio in
soil of shear strength cu 5 12 kPa, but no clear effect was noticed during addition of
sand to stone in treated soil having a shear strength cu 5 6 kPa.
Etezad et al. (2015) stated that columns are made of compacted aggregate and are
installed in weak soil as reinforcements to increase the shear resistance of the soil
mass and, accordingly, its bearing capacity. While a single stone column mostly fails
by bulging, a group of stone columns together with the surrounding soil may fail by
general, local, or punching shear mechanism, depending on the
soil/columns/geometry of the system. The mode of failure of the reinforced ground
could be identified based on the ground geometry and strength parameters of both
stone column and soft soil. Etezad et al. (2015) presented an analytical model to
predict the bearing capacity of soft soil reinforced with stone columns under rigid raft
foundation subject to general shear-failure mechanism. The model utilizes limit-
equilibrium method and the concept of composite properties of reinforced soil. The
proposed theory was validated for the case of bearing capacity of footings on
homogenous soil and via the laboratory and numerical results available in the
literature for this case. Design procedure and charts are presented for practicing
purposes.
Factors of safety provide a hedge against uncertainties in calculations, and the fact
that the factor of safety is never possible to be computed with perfect accuracy.

http://www.iaeme.com/IJCIET/index.asp 65 editor@iaeme.com
 Mohammed Y. Fattah, Mohammed A. Al-Neami and Ahmed S. H. Al-Suhaily

Through experience, conventions have developed with regard to what values of factor
of safety are suitable for various situations. Requiring the same factor of safety for all
bearing capacity applications, is a “one size fits all” approach that is certain to result
in inappropriate factors of safety in some cases.
A more logical approach would be considered: A procedure is followed in this
paper to investigate the reliability of bearing capacity equation of foundation on soft
clay reinforced by stone columns based on reliability index rather than conventional
factor of safety.
The main objective of the present work is to show that the factor of safety and the
reliability can be used together as a complementary measure of acceptable design
based on the following conditions using reliability based-design:
• Cohesion (cu), angle of internal friction (φ) and soil unit weight (γ) are considered to
be independent and uncorrelated variables.
• Footing width and dimensions of the stone column are set of deterministic variables.

2. RELIABILITY ANALYSIS OF BEARING CAPACITY OF

FOUNDATIONS ON IMPROVED SOIL
Simple reliability analyses, involving neither complex theory nor unfamiliar terms,
can be used in routine geotechnical engineering practice. These simple reliability
analyses require little effort beyond that involved in conventional geotechnical
analyses. They provide a means of evaluating the combined effects of uncertainties in
the parameters involved in the calculations, and they offer a useful supplement to
conventional analyses. The additional parameters needed for the reliability analyses
standard deviations of the parameters can be evaluated using the same amount of data
and types of correlations that are widely used in geotechnical engineering practice
(Duncan and Honorary, 2000).

2.1. Selection of Reliability Coefficients

The procedure for carrying out reliability analysis of bearing capacity of foundation
resting on soil improved by stone columns requires definition of standard deviations
of the undrained shear strength, angle of internal friction of stone and the unit weight
of soil which are taken from Table 1. The coefficient of variation is defined as the
ratio of standard deviation to the mean value. The coefficient of variation CoV(x),
usually expressed as a percentage:

CoV x = ∗ 100 % (1)

Table 1: Coefficient of Variation of Geotechnical Parameters.

Coefficient
Property of Variation Source
CoV(%)

Unit weight (γ) 3–7% Harr (1984), Kulhawy (1992)

Effective stress friction
2–13% Harr (1984), Kulhawy (1992)
angle (φ')
Undrained shear strength Harr (1984), Kulhawy (1992), Duncan
13–40%
(cu) and Honorary (2000)

http://www.iaeme.com/IJCIET/index.asp 66 editor@iaeme.com
 Reliability Based Analysis of Bearing Capacity of Footings on Soft Soil Strengthened by
Stone Columns

2.2. Reliability Model for Bearing Capacity of Footing on Soil

Strengthened by Stone Columns
The event whose probability is described as the “probability of failure” is not
necessarily a catastrophic failure. In the case of stone column settlement or bulging,
for example, “failure” would not be catastrophic. If the column settles a small
distance, the shear resistance of the soil starts getting mobilized. Subsequently, this
behavior could constitute unsatisfactory performance of the stone column system, but
not catastrophic failure.
In this study, two non-linear regression models have been made using the program
SPSS V.20 (statistic computer application). Data of experimental work from previous
studies are analyzed for estimating the bearing capacity of floating and end bearing
stone column groups.
The following previous studies results were adopted as data base for building the
mathematical model:
• Rahil (2007) conducted tests on models of floating stone columns embedded in soft
clay with undrained shear strength of cu = 9, 16 and 25 kPa.
• Al-Waily (2008) conducted tests on models of floating and end bearing columns
embedded in soft clay with undrained shear strength of cu = 6 , 9 and 12 kPa.
• Abbawi (2010) conducted tests on models of floating columns embedded in soft clay
with undrained shear strength of cu = 9 and 25 kPa.
• Al-Baiaty (2012) conducted tests on models of floating columns embedded in soft
clay with undrained shear strength of cu = 15 kPa.
• Al-Suhaili (2014) conducted 12 model tests soft clays of different values of undrained
shear strength Cu varying from 8 to 18 kPa. The series consisted of twelve models of
stone columns; single column, two column group, three column group, four column
group, five column group and six column group, in addition to one model of
untreated soil.
Bouassida et al., (1995) developed an equation for groups of stone columns. The
ratio between the total cross-sectional area of the reinforcement and the footing area
Af is therefore ( ar = Ns πa2 / Af ) where Ns is the number of stone columns and a is
the diameter of stone column. The equation is:
=4 +2 −2 + (2)
where: cs = the cohesion of granular pile material.
Bowles (1996) gives an approximate formula for the allowable bearing capacity of
stone columns as:
"#
! = &4 + ' ( ) (3)
$%

where: * Is the passive earth pressure coefficient,

0(
* = + ,-&45° + ) (4)
-

1′= drained angle of friction of stone,

c = either drained cohesion (suggested for large areas) or the undrained shear
strength cu,
' ′ = effective radial stress as measured by a pressure meter (but may use 2c if
pressure meter data are not available), and

http://www.iaeme.com/IJCIET/index.asp 67 editor@iaeme.com
 Mohammed Y. Fattah, Mohammed A. Al-Neami and Ahmed S. H. Al-Suhaily

Fs = factor of safety, 1.5 to 2.0.

Equations 14 and 15, which will be derived in the next sections are used instead of
equations 2 and 3 because the last two are general equations and do not take into
consideration the type of stone column and whether it is end bearing or floating. In
addition, the results of the equations 2 and 3 are far from the results from the
experimental work, for example from the work of Al-Suhaily (2014), for four stone
column group with area ratio of 0.126 and undrained shear strength of the soil 11
kN/m2 and the angle of internal friction of the stone columns material is 40o, the
bearing capacity according to Bouassida et al. (1995) will be:
= 4 ∗ 11 + 2 ∗ 0.126 11&4.5989 − 2) + 0 = 50.93 kN/m2
while the measured bearing capacity form experimental work for the same case is
68.6 kN/m2 for floating stone columns and 74.8 kN/m2 for the end bearing stone
columns.

2.3. Probabilistic Preliminaries

The probability of the success of a structure is called its reliability, R, symbolizing the
probability of failure as p (f), the standard deviation, a measure of the dispersion of a
set of data from its mean (Harr, 2002)
R + p (f) = 1 (5)

A more meaningful measure of dispersion of a random variable (x) is the positive

square root of its variance (compare with radius of gyration of mechanics) called the
standard deviation, σ[x], (Harr, 2002):

σx = 67 (6)
It is seen that the standard deviation of the exponential distribution is σ[x] = 1/a.
An extremely useful relative measure of the scatter of a random variable (x) is its
coefficient of variation CoV(x), usually expressed as a percentage:

CoV x = ∗ 100 % (7)

It should be emphasized that a straight line fit can be assumed. The reasonableness
of this assumption is provided by the correlation coefficient, P defined as:
9:; <,>
8 = ? < .? > (8)
where σ [x], and σ [y] are the respective standard deviations and cov[x, y] is
Coefficient of Variation. It is the measure of dispersion of data which is defined as:
B
@6 7, A = C ∑C
GHB 7&E) − 7̅ A&E) − A
I (9)
with analogy to statics, the covariance corresponds to the product of inertia, (Harr,
2002).
The adequacy of a proposed design in geotechnical engineering is generally
determined by comparing the estimated resistance of the system to that of the imposed
loading. The resistance is the capacity C (or strength) and the loading is the induced
demand D imposed on the structure.
In the procedure presented by Harr (2002) for analysis of footings and developed
by Fattah (2010) for piles, a capacity– demand concept will be used. Some common
examples are the bearing capacity of a soil and the column loads, allowable and

http://www.iaeme.com/IJCIET/index.asp 68 editor@iaeme.com
 Reliability Based Analysis of Bearing Capacity of Footings on Soft Soil Strengthened by
Stone Columns

computed maximum stresses, traffic capacity and anticipated traffic flow on a

highway, culvert sizes and the quantity of water to be accommodated, and structural
capacity and earthquake loads.
Conventionally, the designer forms the well-known factor of safety as the ratio of
the single-valued nominal values of capacity C and demand D (Ellingwood et al.,
1980:
L
JK = (10)
M
In general, the demand function will be the resultant of the many uncertain
components of the system under consideration (vehicle loadings, wind loadings,
earthquake accelerations, location of the water table, temperatures, quantities of flow,
runoff, and stress history, to name only a few). Similarly, the capacity function will
depend on the variability of material parameters, testing errors, construction
procedures and inspection supervision, ambient conditions, and so on.
If the maximum demand (Dmax) exceeds the minimum capacity (Cmin), the
distributions overlap (shown shaded), and there is a nonzero probability of failure.
The difference between the capacity and demand functions is called the safety margin
(S); that is (Harr, 2002):

K =N−O (11)
Obviously, the safety margin is itself a random variable. Failure is associated with
that portion of its probability distribution wherein it becomes negative (shaded); that
portion wherein S = C – D ≤ 0. As the shaded area is the probability of failure8&P),
we have:
8&P) = 8 &N − O) ≤ 0 = 8 K ≤ 0 (12)
The procedure of determination of P (f) is presented in Appendix A.
It is seen that a maximum is for a perfect positive correlation and a minimum for a
perfect negative correlation. It can be shown that the sum of difference of two normal
varieties is also a normal variant (Haugen, 1968). Hence, if it is assumed that the
capacity and demand functions are normal variants, it follows that:
B
8&P) = - − R S (13)
where R S is standard normal probability as given in standard normal
probability tables.

2.4. Reliability Estimation for Floating Stone Columns

The following equation was obtained using the SPSS program form previous studies
data with a determination coefficient (R2) = 0.889 to estimate the bearing capacity of
floating stone column group for a range of undrained shear strength between (6-25)
kPa with a diameter of 50 mm and L/D = 6:
= NT U.VWX Y U.-Z[ \Y U.UUX *]Y -.BW^ (14)
In order to validate equation 14, Figure 1 is drawn to make a comparison between
the measured bearing capacity and predicted values by this equation.
The above equation will be used in a step by step reliability analysis of one of the
experiments carried out in this work as illustrated in the following section.

http://www.iaeme.com/IJCIET/index.asp 69 editor@iaeme.com
 Mohammed Y. Fattah, Mohammed A. Al-Neami and Ahmed S. H. Al-Suhaily

Figure 1: Comparison between Measured and Predicted Values of Bearing

2.4.1. Case Study 1

Al-Suhaili (2014) found that for a six floating stone column group under a footing of
250*375 mm dimensions with as = 0.126 having the parameters listed in Table 2:

Table 2: Parameters of Stone Columns and Bed Soil used in Reliability Estimation.

Values form Standard

Parameter X+ X-
Experimental Work Deviation

Angle of friction of stone

40 5 45 35
column, φ (o)

Cohesion of the soil bed, 17 5.78 22.78 11.22

cu (kN/m2)
The bearing capacity is a function of two independent variables, therefore the
bearing capacity will be calculated 22= 4 times.
The respective values of Q ((φ, c) are calculated in (kN) as listed in Table 3.

Table 3: Bearing capacity values Q((φ, c) for Case Study 1.

Q(φ, c) (kN/m2) Q2(φ, c)

Q(+,+)=148.158 21951

Q(-,-)=36.872 1359.6

Q(-,+)=54.975 3022.3

Q(+,-)=99.371 9874.7

Using point estimation method for two variables to find the weights 8&E, _), the
mean is calculated as follows listed in Appendix B.
Forming the characteristics of the safety margin and according to Harr (2002), the
coefficient of correlation will be ρ (Q, D) = 0.75: From equation A1, we have the
reliability index:

http://www.iaeme.com/IJCIET/index.asp 70 editor@iaeme.com
 Reliability Based Analysis of Bearing Capacity of Footings on Soft Soil Strengthened by
Stone Columns

C−D ̅ ̅
β=
σ2 C + σ2 D − 2Pσ C . σ D

66.510
β= =1.607
& )
&49.89)2 +&12.47)2 −2 0.75 &49.89)&12.47)

From probability tables, ψ (β) = 0.4460

From Equation 13:
g
e&f) = − ψ&β) = j. jklm
h
Probability of failure =5.39 %
Table 4 shows a summary of the reliability calculations for bearing capacity of
different cases of floating stone columns considering different values of factor of
safety.
It can be noticed that the probability of failure values obtained from reliability
analysis based on Equation 14 derived in this study are always higher than those
obtained depending on the conventional equations which are always conservative.

2.5. Reliability Estimation for End Bearing Stone Columns

The following equation was obtained using the SPSS program form previous studies
data with a determination coefficient (R2) = 0.871 to estimate the bearing capacity of
end bearing stone column group for a range of undrained shear strength between (6-
25) with a diameter of 50 mm and L/D = 8:
= NT U.[X- Y U.--n \Y U.UZ- *]YB.ZWW (15)
In order to validate equation 15, Figure 2 is drawn to make a comparison between
the measured bearing capacity and predicted values by this equation.

Figure 2: Comparison Between Measured and Predicted Values of Bearing Capacity of End
Bearing Stone Columns.

http://www.iaeme.com/IJCIET/index.asp 71 editor@iaeme.com
 Mohammed Y. Fattah, Mohammed A. Al-Neami and Ahmed S. H. Al-Suhaily

Equation 15 will be used in a step by step reliability analysis of one of the

experiments carried out in this work as illustrated in the following section.

Table 4: Summary of Reliability Calculations for Bearing Capacity of Different Cases of

Floating Stone Columns Considering Different Factors of Safety.

2.5.1. Case Study 2

For a Single end bearing stone column group with footing of 64.6 mm diameter with
as = 0.6 having the parameters listed in Table 5.

Table 5: Parameters of Stone Column and Bed Soil used in Reliability Estimation.

Values form Standard

Parameter X+ X-
Experimental Work Deviation

Angle of friction of
40 5 45 35
stone column, φ (o)

Cohesion of the soil 18 6.3 18.34 9.66

bed, cu (kN/m2)

http://www.iaeme.com/IJCIET/index.asp 72 editor@iaeme.com
 Reliability Based Analysis of Bearing Capacity of Footings on Soft Soil Strengthened by
Stone Columns

The bearing capacity is a function of two independent variables, therefore the

bearing capacity will be calculated 22=4 times
The respective values of Q(φ, c) are calculated in (kN) as listed in Table 6.

Table 6: Bearing Capacity Values Q(φ, c) for Case Study 2.

φ, c) (kN/m2)
Q(φ Q2(φ
φ, c)

Q(+,+)=255.37 65212.805

Q(-,-)=63.27 4003.698

Q(-,+)=108.832 11844.405

Q(+,-)=148.47 22043.522

Using point estimation method for two variables to find the weights8&E_), the
mean will be calculated as listed in Appendix B.
Forming the characteristics of the safety margin and according to Harr (2002) the
coefficient of correlation will be ρ (Q, D) = 0.75,: From equation A1, we have the
reliability index:

C−D ̅ ̅
β=
σ2 C + σ2 D − 2Pσ C . σ D
114.891
β= =1.588
& )
&87.229)2 +&21.807)2 −2 0.75 &87.229)&21.807)

From probability tables ψ (β) = 0.444

From equation 13:
g
o&f) = − ψ&β) = j. jkpg
h
Probability of failure =5.61 %
Table 7 shows a summary of the reliability calculations for different cases of end
bearing stone column considering different values of factor of safety.
It can be noticed that the probability of failure values obtained from reliability
analysis based on equation 15 derived in this study are always higher than those
obtained depending on the conventional equations which are always conservative.
Reliability calculations provide a means of evaluating the combined effects of
uncertainties, and a means of distinguishing between conditions where uncertainties
are particularly high or low. Moreover, reliability analysis provides a framework for
establishing appropriate factors of safety and other design targets and leads to a better
appreciation of the relative importance of uncertainties in different parameters as was
stated by Christian and Baecher (2001).
The higher values of factors of safety associated with allowable bearing capacity
obtained by probabilistic approach clarifies clearly the importance of uncertainty
studies in foundation engineering and strongly demands the need to include
probabilistic framework in geotechnical engineering design.

http://www.iaeme.com/IJCIET/index.asp 73 editor@iaeme.com
 Mohammed Y. Fattah, Mohammed A. Al-Neami and Ahmed S. H. Al-Suhaily

Table 7: Summary of Reliability Calculations of Bearing Capacity of Different Cases of End

Bearing Stone Columns Considering Different Values of Factor of Safety.

3. RELIABILITY RELATIONS
Reliability estimation for all cases have been done using EXECL and presented in the
form of histograms of calculation. The results are shown in Figures 3 to 6.
Based on the results presented in Figures 3 to 6, it can be seen that the reliability
index increases with the increase of factor of safety for both stone columns and soil
replacement, while the probability of failure decreases with the increase of factor of
safety. For a given factor of safety, the probability of failure is approximately the
same whether the stone column was singular or within a group.
The reliability and the probability of failure depend majorly on the approach used
in the estimation of bearing capacity. A number of equations were derived in the
literature to estimate the bearing capacity but the degree of conservation for each one
is different from one to another.
The reliability analysis helps in choosing the most proper equation for the
estimation of bearing capacity which provides adequate factor of safety with a
sufficient degree of economy.

http://www.iaeme.com/IJCIET/index.asp 74 editor@iaeme.com
 Reliability Based Analysis of Bearing Capacity of Footings on Soft Soil Strengthened by
Stone Columns

Figure 3: Reliability Index for Different Cases of Floating Stone Column Groups
Considering Different Values of Factor of Safety.

Figure 4: Probability of Failure of Different Cases of Floating Stone Column Groups

Considering Different Values of Factor of Safety.

http://www.iaeme.com/IJCIET/index.asp
w.iaeme.com/IJCIET/index.asp 75 editor@iaeme.com
 Mohammed Y.. Fattah, Mohammed A. Al-Neami
Al and Ahmed S. H. Al-Suhaily
Al

Figure 5: Reliability Index of Different Cases off End Bearing Stone Column Considering
Consider
Different Values of Factor of Safety.

Figure 6: Probability off Failure of Different Cases off End Bearing Stone Column Groups
Group
Considering Different Values of Factor of Safety.
In some cases analyzed in this study, the probability of failure was found to be
less than 1% depending
epending on the value of reliability index obtained from reliability
tables which are always greater than 2.2 based on standard normal distribution. This
may have to be reanalyzed depending on more real values of probability of failure
using different typess of distribution.

http://www.iaeme.com/IJCIET/index.asp 76 editor@iaeme.com
 Reliability Based Analysis of Bearing Capacity of Footings on Soft Soil Strengthened by
Stone Columns

4. CONCLUSIONS
• The reliability index increases with the increase of factor of safety for both stone
columns and soil replacement, while the probability of failure decreases with the
increase of factor of safety.
• For a given factor of safety, the probability of failure is approximately the same
whether the stone column was singular or within a group.
• The reliability and the probability of failure depend mainly on the approach used in
the estimation of bearing capacity, a number of equations were derived in the
literature to estimate the bearing capacity but the degree of conservation for each one
is different from one to another.
• In some cases analyzed in this study, the probability of failure was found to be less
than 1% depending on the value of reliability index obtained from reliability tables
which are always greater than 2.2 based on standard normal distribution.
• The probability of failure values obtained from reliability analysis based on the
equations derived in this study for estimation of the bearing capacity of floating and
end bearing stone columns are always higher than those obtained depending on the
conventional equations which are always conservative.

REFERENCES
[1] Abbawi Z. W., Evaluation of Improvement Techniques for Ballasted Railway
Track Model Resting On Soft Clay, Ph.D. Thesis, Building and Construction
Engineering Department, University of Technology, Baghdad, Iraq, 2010.

[2] Al-Baiaty S. E., Comparison Between The Efficiency of Ordinary and Geogrid
Encased Stone Columns, M.Sc. Thesis, Building and Construction Engineering
Department, University of Technology, Baghdad, Iraq, 2012.

[3] Al-Suhaili A. S. H., Experimental Study on Uncertainties and Reliability of Soil

Properties Improved by Stone Columns", M.Sc. thesis, Building and Construction
Engineering Department, University of Technology, Baghdad, Iraq, 2014.

[4] Al-Waily M. J., Stress Concentration Ratio of Model Stone Columns Improved
by Additives, Ph.D. Thesis, Building and Construction Engineering Department,
University of Technology, Baghdad, Iraq, 2008.

[5] Barksdale R. D. and Bachus R. C., Design and Construction of Stone Columns,
Volume I”, Prepared for the Federal Highway Administration, U.S. Department
of Transportation. P. 239, 1983.

[6] Bouassida M., De Buhan P. and Dormicux L., Bearing Capacity of a Foundation
Resting on Soil Reinforced by a Group of Columns,' Geotechnique, 1995, Vol.
45, No. 1, pp. 25-34.

[7] Bowels J.E., Foundation Analysis and Design, 5th Edition , McGraw-Hill Book
Company, P. 1175, 1996.

[8] Christian, J.T. and Baecher, (2001), "Factor of Safety and Reliability in
Geotechnical Engineering – Discussion” Journal of Geotechnical Engineering,
ASCE, 127(8) , pp. 700-703.

[9] Dasaka S.M., Rajaparthy S.R. and Sivakumar Babu G.L., Reliability Analysis of
Allowable Pressure of Strip Footing in Spatially Varying Cohesionless Soil,
Proceedings Mini-symposium on Reliability Evaluation in Geotechnics, 9th

http://www.iaeme.com/IJCIET/index.asp 77 editor@iaeme.com
 Mohammed Y. Fattah, Mohammed A. Al-Neami and Ahmed S. H. Al-Suhaily

International Conference on Structural Safety and Reliability (ICOSSAR), Rome,

Italy, 2005, pp. 909-915.

[10] Ditlevsen O., Uncertainty Modeling, McGraw-Hill, New York, 1981.

[11] Duncan J. M. and Honorary, Factors of Safety and Reliability in Geotechnical

Engineering, Journal of Geotechnical and Geoenvironmental Engineering, ASCE,
2000, 126(4), pp. 307-316.

[12] Ellingwood B., Galambos T. V., MacGregor J. G., and Cornell C. A.,
Development of a Probability Based Load Criterion for American National
Standard A58, Nat. Bur. Stand. Spec. Publ. 577, Washington, D.C, 1980.

[13] Etezad, M., Hanna, A. M. Ayadat, T., Bearing Capacity of a Group of Stone
Columns in Soft Soil, International Journal of Geomechanics, 15(2), 04014043,
2015.

[14] Fattah M. Y., Reliability-Based Design Procedure of Axially Loaded Piles,

Journal of Engineering, 2010, No.1, Vol. 16, pp. 4462-4477.

[15] Fattah, M. Y., Shlash, K. T., Al-Waily, M. J., Experimental Evaluation of Stress
Concentration Ratio of Model Stone Columns Strengthened by Additives,
International Journal of Physical Modelling in Geotechnics, ICE, Vol. 13, No.
(3), pp. 79–98, http://dx.doi.org/10.1680/ijpmg.12.00006, 2013.

[16] Harr M. E., Reliability-based Design in Civil Engineering, Henry M. Shaw

Lecture, Dept. of Civil Engineering, North Carolina State University, Raleigh,
N.C., 1984.

[17] Harr M. E., Accounting for Variability (Reliability), Chapter 16 in “The Civil
Engineering Handbook, edited by W. F. Chen and J. Y. R. Liew, Second Edition,
CRC Press, 2002.

[18] Haugen E. B., Probabilistic Approaches to Design, John Wiley & Sons, New
York, 1968.

[19] Honjo Y., Hara T., and Kieu Le T. C., Level III Reliability Based Design of
Examples Set by ETC10, Proceeding of the 2nd International Workshop on the
Evaluation of Eurocode 7, EUcentre. Pavia, Italy, 2010.

[20] Kameswara S., Somayazulu J. R. and Rama K. S., Granular Trenches and Stone
Columns as Ground Improvement Techniques, Third International Conference on
Case Histories in Geotechnical Engineering, St. Louis, Missouri, 1983, paper No
737.

[21] Kulhawy F. H., On the Evaluation of Soil Properties, ASCE Geotech. Spec.
Publ., 1992, No. 31, 95–115.

[22] Phoon K. K. and Kulhawy F. H., Evaluation of Model Uncertainties for

Reliability-Based Foundation Design, Proc. 9th International Conference on
Applications of Statistics and Probability in Civil Engineering, San Francisco (2),
2004, pp. 1351- 1356.

[23] Rahil F.H., Improvement of Soft Clay Underneath a Railway Track Model Using
Stone Columns Technique, Ph.D. Thesis, University of Technology, Iraq, 2007.

[24] Esraa A. Mandhour, Saad N. Al-Saadi and Saad F. Ibrahim, “Study of the
Efficiency of Stone Columns in Soft Clay: Considering the Effect of Clay

http://www.iaeme.com/IJCIET/index.asp 78 editor@iaeme.com
 Reliability Based Analysis of Bearing Capacity of Footings on Soft Soil Strengthened by
Stone Columns

Minerals in Soil", International Journal of Civil Engineering & Technology

(IJCIET), 5(9), 2014, pp. 241–251.

[25] Ranjan C. and Rao A. S. R, "Basic and Applied Soil Mechanics, Wiley Eastern
Limited, 640 pp, 1993.

[26] Rethati. L., Probabilistic Solutions in Geotechnics, Elsevier Amsterdam, 451 pp,
1988.

[27] K.V. Maheshwari, Dr. A.K. Desai and Dr. C.H. Solanki, "Bearing Capacity of
Fiber Reinforced Soil", International Journal of Civil Engineering & Technology
(IJCIET), 4(1), 2013, pp. 159–164.

APPENDIX A
The number of standard deviations that the mean value of the safety margin is beyond
S = 0, is called the reliability index, β; that is:
I
q
β= (A1)
σI q
The reliability index is seen to be the reciprocal of the coefficient of variation of
the safety margin, or:
B
β= (A2)
rst&q)
Application to their definitions produces the following identities (a, b, and c are
constants), (Ditlevesen, 1981):
u + v7 + A = + vu 7 + u A (A3)

6 + v7 + A = v - 6 7 + -
6 A + 2v. . @6 7, A] (A4)
@6 7, A]≤ ' 7 . ' A (A5)

6 + v7 + A = v - 6 7 + -
6 A + 2v. . ' 7 . ' A . 8
From Eq. A4, we have:
u K = u N − u O =N̅ − O
w (A6)
Equation 17 produces:
' - K = ' - N + ' - O − 28' N . ' O (A7)
Hence,
L ̅ xM
w
β= (A8)
?y L z?y M x-{? L .? M

APPENDIX B
B.1 Calculations for Case Study 1
u | = |I = ∑ |&E_)8&E_) = 88.680 kN/m2
while the bearing capacity form the experimental value for the same case is 82.15
kN/m2.
u| -
= |I = ∑ | - &E_)8&E_)= 10353.6

http://www.iaeme.com/IJCIET/index.asp 79 editor@iaeme.com
 Mohammed Y. Fattah, Mohammed A. Al-Neami and Ahmed S. H. Al-Suhaily

the variance is :
} Q = •€Qh • − &• Q )h = 2489.43
and Equation (6) for the standard deviation gives:

σQ = 6 | =49.894
for the coefficient of variation, Equation (9) requires:
σQ 49.894
CoV Q = ∗ 100 % = ∗ 100 = 56.263 %
EQ 88.680
For factor of safety = 4, the demand will be from Equation (10):
w
w=
D = 22.170
„.…
The standard deviation of the demand will be equal to:
σ D = u&O) ∗ N@†&O) = 22.17 ∗ 0.5626 = 12.47
to find the safety margin :
K = N̅ − O
w = 88.68 − 22.170 = 66.510 kN/m-

B.2 Calculations for Case Study 2

u | = |I = ∑ |&E_)8&E_) = 153.187 kN/m2
while the bearing capacity form the experimental results for the same case is 153
kN/m2
u| -
= |I = ∑ | - &E_)8&E_)= 31075.4
We have for variance:
} Q = •€Qh • − &• Q )h = 7609
the standard deviation is:

σQ = 6 | =87.229
for the coefficient of variation equation 9 requires:
σQ 87.229
CoV Q = ∗ 100 % = ∗ 100 = 56.943 %
EQ 153.187
For factor of safety = 4, the demand will be from equation 10
w
w=
D = 38.296
„.…
The standard deviation of the demand will be equal to:
σ D = u&O) ∗ N@†&O) = 38.296 ∗ 0.569 = 21.807
to find the safety margin :
K = N̅ − O
w = 153.187 − 38.296 = 114.891 kN/m-

http://www.iaeme.com/IJCIET/index.asp 80 editor@iaeme.com