212ce1021 13

A STUDY ON BEARING CAPACITY OF
SUBMERGED ROCK FOUNDATION
A Thesis submitted in partial fulfillment of the requirements

for the award of the Degree of
Master of Technology
in
Geotechnical Engineering
Civil Engineering Department
by
NIRUPAM BARUAH
212CE1021
DEPARTMENT OF CIVIL ENGINEERING

NATIONAL INSTITUTE OF TECHNOLOGY
ROURKELA - 769008
MAY 2014
A STUDY ON BEARING CAPACITY OF
SUBMERGED ROCK FOUNDATION
A Thesis submitted in partial fulfillment of the requirements

for the award of the Degree of
Master of Technology
in
Geotechnical Engineering
Civil Engineering Department
by
NIRUPAM BARUAH
212CE1021
Under the guidance of

Dr. N. ROY
DEPARTMENT OF CIVIL ENGINEERING

NATIONAL INSTITUTE OF TECHNOLOGY
ROURKELA-769008
MAY 2014
Department of Civil Engineering
National Institute of Technology
Rourkela – 769008
Odisha, India
www.nitrkl.ac.in
CERTIFICATE
This is to certify that the thesis entitled “A study on Bearing Capacity of Submerged Rock
Foundation” submitted by Mr. Nirupam Baruah (Roll No. 212CE1021) in partial fulfilment
of the requirements for the award of Master of Technology Degree in Civil Engineering with
specialization in Geo-Technical Engineering at National Institute of Technology, Rourkela is
an authentic work carried out by him under my supervision and guidance.
To the best of my knowledge, the matter embodied in the thesis has not been submitted to any
other University / Institute for the award of any Degree or Diploma.
Date: Dr. N. Roy
Place: Professor and Head

National Institute of Technology
Rourkela - 769008
ACKNOWLEDGEMENT
First of all I would like to express my heartfelt gratitude to my project supervisor Dr.
N. Roy, Professor and Head, Department of civil Engineering for his able guidance,
encouragement, support, suggestions and the essential facilities provided to successfully
complete my research work. I truly appreciate and value his esteemed guidance and
encouragement from the beginning to the end of this thesis, his knowledge and company at the
time of crisis will be remembered lifelong.
I sincerely thank to our Director Prof. S. K. Sarangi and all the authorities of the
institute for providing nice academic environment and other facilities in the NIT campus. I am
also thankful to all the faculty members of the Civil Engineering Department, especially Geo-
Technical Engineering specialization group who have directly or indirectly helped me during
the project work. I would like to extend my gratefulness to the Dept. of Civil Engineering, NIT
ROURKELA, for giving me the opportunity to execute this project, which is an integral part
of the curriculum in M. Tech program at the National Institute of Technology, Rourkela.
Submitting this thesis would have been a Herculean job, without the constant help,
encouragement, support and suggestions from my friends and seniors. Last but not the least I
would like to thank my parents, who taught me the value of hard work by their own example.
Date: Nirupam Baruah

Place: Roll No: 212CE1021
M. Tech (Geo-Technical Engineering)
NIT ROURKELA
Odisha – 769008
CONTENTS
CONTENTS PAGE NO.
Abstract I
List of Figures II
List of Tables IV
Nomenclature V
MATLAB variables VIII
Chapter 1 INTRODUCTION 1
1. Introduction 2
1.1 Origin of Project 2
1.2 Present study components 2
1.3 Lower Bound Theorem 3
1.4 Upper Bound Theorem 4
Chapter 2 LITERATURE REVIEW 5
2. Literature review 6
Chapter 3 THEORY 8
3. Theory 9
3.1 Introduction 9
3.2 Determination of Bearing Capacity 9
3.2.1 Assumptions 9
3.2.2 Failure mechanism 10
3.2.2.1 Two-sided mechanism 10
3.2.2.2 One-sided mechanism 11
3.2.3 Calculations 11
3.3 Algorithmic analysis 14
3.3.1 Flowchart 14
3.3.2 MATLAB Programming 15
3.4 Reliability Analysis 15
3.4.1 Importance 15
3.4.2 Definition 15
3.4.3 Overview 16
3.4.4 Reliability Index 16
3.4.5 Steps in Reliability Analysis 17
3.4.6 Error Propagation 18
3.4.7 Solution Techniques 18
3.4.7.1 The First Order Second Moment (FOSM) Method 18
3.4.7.2 The Second Order Second Moment (SOSM) Method 19
3.4.7.3 The Point Estimate Method 19
3.4.7.4 The Hasofer–Lind Method 19
3.4.7.5 Monte Carlo method 19
3.4.8 Recommendation 20
3.4.9 Statistical Terms 20
3.4.9.1 Population 20
3.4.9.2 Sample 20
3.4.9.3 Mean 20
3.4.9.4 Median 21
3.4.9.5 Mode 21
3.4.9.6 Range 21
3.4.9.7 Interquartile Range 21
3.4.9.8 Variance 21
3.4.9.9 Standard Deviation 21
3.4.9.10 Normal Distribution 22
3.4.9.11 Skew 22
3.4.9.12 Kurtosis 23
3.4.9.13 Standard Error 23
3.4.10 Microsoft Excel Spreadsheet 23
Chapter 4 ANALYSIS & DISCUSSION 24
4. Analysis and Discussion 25
4.1 Algorithmic Analysis 25
4.1.1 Flowchart 25
4.1.2 Algorithmic Model 25
4.1.3 Implementation Details 32
4.1.4 Execution Steps 36
4.2 Effect of submergence 47
4.2.1 Analysis process 47
4.2.2 Graphical representation 48
4.2.3 Observation 48
4.3 Comparison between mechanisms 49
4.3.1 Stochastic model creation 49
4.3.2 Graphical representation 51
4.3.3 Observation 52
4.4 Reliability Analysis 52
4.4.1 Monte Carlo Simulation 52
4.4.1.1 Effect of Sample Size 55
4.4.1.1.1 Analysis Process 55
4.4.1.1.2 Graphical representation 56
4.4.1.1.3 Observation 58
4.4.1.2 Determination of Reliability Index 58
4.4.1.3 Determination of Probability of failure 60
4.4.1.4 Result Summary 61
Chapter 5 CONCLUSION & FUTURE SCOPE 63
5. Conclusion and Future Scope 64
5.1 Conclusions 64
5.2 Future Scopes 64

Appendix – I 66
Appendix – II 67
Appendix – III 70
Appendix – IV 72
References 74
ABSTRACT
For more urbanization, engineering structures are not limited to underground tunnels or
metro stations, engineers are constructing underground multistoried building even underground
city. Therefore accurate assessment of rock strength is necessary for the rational design of
underground structures. Bearing capacity is one of the most important factor for assessment of
rock strength. Various studies on determination of bearing capacity for intact as well as jointed
rock mass have been performed. Concisely, various studies have been performed for
determination of bearing capacity of rock mass under dry condition while for submerged
condition, it is outnumbered. In this study, determination of submerged bearing capacity of
rock mass is attempted. Most cases, the rock substratum is considered to be homogeneous,
intact, but the practical scenario does not recommend so. In-situ rock mass variability renders
the deterministic analysis to be inefficient and hence there arises the necessity for probabilistic
or reliability analyses to model the uncertainties. Rather than calculating a deterministic factor
of safety, a reliability based analysis is more appropriate for geotechnical design. Present study
comprises of different aspects regarding ultimate bearing capacity of rock foundations for both
dry and submerged condition. These are, development of algorithmic modelling for
determination of ultimate bearing capacity for both dry and submerged condition, analysis of
behavior of ultimate bearing capacity due to the effect of submergence, comparative study
between theorem mechanisms and reliability analysis through Monte-Carlo simulation for
various parameters of rock mass. The data generated from the mentioned aspects are
represented in graphical form and histogram bar charts are developed from Monte Carlo
simulation.
I
LIST OF FIGURES
Figure No Title Page No
Fig. 3.1 Hodograph diagram of two sided mechanism 10

Fig. 3.2 Hodograph diagram of one sided mechanism 11
Fig. 3.3 Line diagram representing standard shape of normal distribution 22
Fig. 4.1 Flowchart of Algorithmic Model for determination of Ultimate
26
Bearing Capacity
Fig. 4.2 MATLAB Command Window showing comment lines of
33
ModelBCR
Fig. 4.3 MATLAB Command Window, asking input of foundation width 36
Fig. 4.4 MATLAB Command Window, asking input of surcharge magnitude 37
Fig. 4.5 MATLAB Command Window, asking input of orientation angle 37
Fig. 4.6 MATLAB Command Window, asking input of cohesion of intact
38
rock
Fig. 4.7 MATLAB Command Window, asking input of cohesion of jointed
38
rock
Fig. 4.8 MATLAB Command Window, asking input of frictional angle of
39
intact rock
Fig. 4.9 MATLAB Command Window, asking input of frictional angle of
39
jointed rock
Fig. 4.10 MATLAB Command Window, asking input of Bulk Unit Weight 40
Fig. 4.11 MATLAB Command Window, asking input of Ground water
40
condition
Fig. 4.12 MATLAB Command Window, inserting input as dry condition in
41
Ground water condition
Fig. 4.13 MATLAB Command Window, showing output with its inputs for a
41
simple problem of rock foundation without submergence
Fig. 4.14 MATLAB workspace showing variables for the stated problem (Dry
42
Condition)
Fig. 4.15 MATLAB Command Window, asking input of saturated unit weight 43
Fig. 4.16 MATLAB Command Window, asking input of depth of water 44
II
Fig. 4.17 MATLAB Command Window, showing output with its inputs for a
44
simple problem of rock foundation with submergence.
Fig. 4.18 MATLAB workspace showing variables for the stated problem
46
(Submerged Condition)
Fig. 4.19 Behavior of Ultimate Bearing Capacity with Submergence. 48
Fig. 4.20 Comparison graph of failure mechanism (Two sided & One Sided) 51
Fig. 4.21 Excel spreadsheet formulated for Monte Carlo Simulation. 54
Fig. 4.22 Histogram of 100 samples 56
Fig. 4.23 Scaled Histogram of 100 samples 56
Fig. 4.32 Reliability index with variation of sample size 59
Fig. 4.33 Probability of Failure with variation of sample size 61
III
LIST OF TABLES
Table No Title Page No
Table 4.1 Data set for analysis of effect of submergence 47

Table 4.2 Data set used for comparison between one side and two sided failure
50
mechanism
Table 4.3 Data set used for Monte Carlo Simulation 56
Table 4.4 Data set used for Reliability Index (β) determination 58
Table 4.5 Data set used for Probability of Failure (Pf) determination 60
Table 4.6 Various Results of Monte Carlo Simulation 62
IV
NOMENCLATURE
α: Orientation angle.
β: Reliability Index.
γ: Total unit weight of the rock mass.
γsat : Saturated unit weight of rock mass.
γw : Unit weight of water.
γ‫ ׳‬: Effective unit weight.
θ: Angle of velocity discontinuity line (CD) with horizontal.
μM : Mean values of M (Margin of Safety).
μQ : Mean values of Q (Load).
μR : Mean values of R (Resistance).
ξn : Function of α, θ, ϕi & ϕj.
ρRQ : Correlation Coefficient between R and Q (Load).
σF : Standard Deviations of F (Factor of Safety).
σM : Standard Deviations of M (Margin of Safety).
σQ : Standard Deviations of Q (Load).
σR : Standard Deviations of R (Resistance).
σM2 : Variances of M (Margin of Safety).
σQ2 : Variances of Q (Load).
σR2 : Variances of R (Resistance).
ϕi : Angle of friction of intact rock mass.
φj : Angle of friction of jointed rock mass.
V
B: Footing Width.
ci : Cohesion of intact rock mass.
cj : Cohesion of jointed rock mass.
CI95 : A 95% confidence interval.
CI99 : A 99% confidence interval.
dw : Depth of water from foundation base.
E[F] : Expected values of F.
F: Factor of Safety.
fn : Function of ξ & ϕj for two sided mechanism.
gn : Function of ξ & ϕj for one sided mechanism.
hC : Height from ground level to point C in mechanism.
hD : Height from ground level to point D in mechanism.
LCL95: Lower confidence limit at 95% confidence
M: Margin of Safety.
no : Number of joint set.
Nci : Bearing Capacity Coefficient.
Ncj : Bearing Capacity Coefficient.
Nq : Bearing Capacity Coefficient.
Nγ : Bearing Capacity Coefficient.
Nγsub : Bearing Capacity Coefficient.
Nγw : Additional factor depends on groundwater depth.
Pf : Probability of Failure.
q: Surcharge.
qu : Ultimate bearing capacity for dry condition.
VI
quw : Ultimate bearing capacity for submerged condition.
Q: Load.
R: Resistance.
Si : Spacing of ith joint set.
sx : The standard error.
t95 : The t value for a two-tailed test, alpha level of .05 with a given degrees of freedom.
t99 : The t value for a two-tailed test, alpha level of .01 with a given degrees of freedom.
UCL95: Upper confidence limit at 95% confidence.
̅
X: The sample mean.
xn : nth Co-ordinate Point.
Xn : nth Variable.
VII
MATLAB VARIABLES
alpha : Orientation angle (α).
B: Footing Width (B).
ci : Cohesion of Intact Rock (ci).
cj : Cohesion of Jointed Rock (cj).
dw : Depth of water from foundation base (dw).
f1 : Function of ξ & ϕj for two sided mechanism (fn).
g1 : Function of ξ & ϕj for one sided mechanism (gn).
GWC : Ground water condition, either ‘DRY’ or ‘SUBMERGED’.
hC : Height from ground level to point C in mechanism.
hD : Height from ground level to point D in mechanism.
mechanism : Mechanism to be used either one sided ‘OS’ or two sided ‘TS’.
Nci : Bearing capacity coefficient (Nci).
Ncj : Bearing capacity coefficient (Ncj).
Nq : Bearing capacity coefficient (Nq).
Nr : Bearing capacity coefficient (Nγ).
Nrsub: Bearing Capacity Coefficient (Nγsub).
Nrw : Additional factor depends on groundwater depth (Nγw).
VIII
phii : Angle of friction of Intact Rock (ϕi).
phij : Angle of friction of Jointed Rock (ϕj).
q: Surcharge (q).
qu : Ultimate bearing capacity for both dry and submerged condition (qu & quw).
r: Bulk unit weight (γ).
reff : Effective unit weight (γ‫)׳‬.
rsat : Saturated unit weight (γsat).
rw : Unit weight of water (γw ).
theta : Angle of velocity discontinuity line (CD) with horizontal (θ).
xi1 : Function of α, θ, ϕi & ϕj (ξn).
IX
INTRODUCTION
1. INTRODUCTION
1.1 ORIGIN OF PROJECT
Except from some soft-rock, most rock masses are excellent or in other words, hard
enough for foundation material. However, for very huge construction with unusually high loads
and special requirements like 100-storey skyscrapers, large bridge, arch-dam, nuclear power
plants, underground city and others, it is very essential to study the strength of rock mass.
Therefore, in those cases, it is essential to justify and determine the ultimate bearing capacity,
and also the allowable load for shallow foundations. In other cases, when the loads are large
and the rock mass is weak or highly fragmented, a more rigorous method to establish the
ultimate bearing capacity of the rocks is needed.
Sutcliffe et al. (2004), Yang and Yin (2005), Merifield et al. (2006), and Saada et al.
(2008) have performed various studies on rock foundations. In these studies, effect of
groundwater and joint spacing on the bearing capacity had not been investigated. Thereafter,
the upper bound theorem of limit analysis was used by Imani et al. (2012) to investigate the
bearing capacity of submerged jointed rock foundations.
Imani et al. (2012) performed splendid studies on upper bound solution for the bearing
capacity of submerged jointed rock foundations. These studies include various relationships
between the different bearing capacity parameters such as relationship between ultimate
bearing capacity for dry rock and ultimate bearing capacity for submerged rock for different
ground water depth, width of foundations along with cohesion and frictional angle of both
intact and jointed rock. He suggested that upper bound theorem should be used for relatively
simple structures.
1.2 PRESENT STUDY COMPONENTS
Since Imani et al. (2012) is the first published document for determining the bearing
capacity for submerged rock by upper bound solution, further studies should be carried out.
Most cases, the rock substratum is considered to be homogeneous, intact, but the
practical scenario does not recommend so. In-situ rock mass variability renders the
deterministic analysis to be inefficient and hence there arises the necessity for probabilistic or
reliability analyses to model the uncertainties. Rather than calculating a deterministic factor of
Page | 2
INTRODUCTION
safety, a reliability based analysis is more appropriate for geotechnical design. Present study
comprises of the following:
Firstly, An Algorithmic modelling has been developed as a step by step procedure for
calculation of bearing capacity. Mentioned algorithmic program is developed with the help of
MATLAB language. The program consists of computational commands in a sequence and the
computational commands are created in MATLAB script files. When the script file is executed,
computational commands in the order as they are enlisted are also executed and by putting
different input parameters e.g. footing width, orientation angle, rock mass strength properties,
physical properties, water table location and surcharge, output parameter i.e. ultimate bearing
capacity of rock mass can be evaluated.
Secondly, a study on behavior of ultimate bearing capacity due to the effect of

submergence is attempted. Graphical representation is carried out to observe variation of
ultimate bearing capacity with reduction of water table.
Thirdly, a comparative study between theorem mechanisms of upper bound theorem

for determination of submerged bearing capacity of rock mass has been investigated. Graphical
representation is carried out to observe the mentioned comparative study.
At last, a reliability based study through Monte-Carlo simulation is attempted. A

stochastic model with random input parameters of various strength parameters and different
ground water locations is created in Microsoft Excel Spreadsheet to identify the performance
of the correlations of upper bound theorem for determination of submerged bearing capacity
of rock mass. The data generated from the simulation is represented as histogram bar charts.
Different statistical descriptors like mean, median, standard deviation and many others are
generated for ten thousand samples. Furthermore, probability distribution function and
confidence interval were also evaluated.
This study will help researcher to investigate issues related to bearing capacity of rock
mass with submerged condition.
1.3 LOWER BOUND THEOREM
The lower bound theorem states that the collapse load obtained from any statically
admissible stress field will underestimate the true collapse load. A statically admissible stress
field is one which firstly satisfies the equations of equilibrium, secondly satisfies the stress
Page | 3
INTRODUCTION
boundary conditions and finally does not violate the yield criterion. Two basic assumptions
must be made about the material in order to apply the lower bound theorem.
1. The material is assumed to be perfectly plastic, that is, the material exhibits ideal
plasticity with an associated flow rule without strain hardening or softening. The level
of ductility exhibited by a jointed rock mass is dependent on the imposed stress state.
2. It is assumed that all deformations or changes in the geometry of the rock mass at the
limit load are small and therefore negligible.
It is recognized that the lower bound theorem has been applied less frequently than the
upper bound theorem as it is easier to construct a kinematically admissible failure mechanism
than it is to construct a statically admissible stress field. Furthermore, the lower bound theorem
is often difficult to apply to problems involving complex loading and geometry, particularly if
it is necessary to construct the stress fields manually.
1.4 UPPER BOUND THEOREM
The upper bound theorem states that the rate of work done by actual forces is less than
or equal to the rate of energy dissipation in any kinematically admissible velocity fields. A
kinematically admissible velocity field is compatible with the velocities at the boundary of the
geometrical masses. The solution obtained from upper bound theorem, in the non-conservative
side, is not less than the actual solution by a kinematically admissible velocity field. To obtain
the better solution (lower upper bounds), work has to be done for as many trial kinematically
admissible velocity fields as possible. The lowest possible upper bound solution is sought with
an optimization scheme by trying various possible kinematically admissible failure
mechanisms.
The upper bound theorem is a powerful tool for stability analysis and has been
widely used in many areas of geotechnical design. However, for practical problems
which involve nonhomogeneous material, complicated loadings and complex geometries,
this theorem quite difficult to apply. An upper bound to the collapse loads may obtain by
postulating a collapse mechanism, and computing the ratio of the plastic dissipation associated
with this mechanism to the work done by the applied loads.
Page | 4
LITERATURE REVIEW
2. LITERATURE REVIEW
Metropolis and Ulam (1949) stated the ‘Monte Carlo method’ with several statements
mentioned below:
1. Monte Carlo method is a statistical approach to the study of differential equations, or

more generally, of integro-differential equations that occur in various branches of the
natural sciences.
2. The mathematical description of Monte Carlo method is the study of a flow which
consists of a mixture of deterministic and stochastic processes.
Chen and Drucker (1969) stated that slip-line solutions give neither rigorous lower bound nor
rigorous upper bound solutions on the actual collapse load.
Sutcliffe et al. (2004) concluded the followings:
1. Inclusion of single week joint, reduction in strength is significantly affected by both the
strength of the joint relative to the properties of the intact rock material and to the
orientation of the joint set.
2. Inclusion of two joint sets, the strength of the joints as well as the joint orientation
significantly affects the result, although in this case the angle between the two joint sets
also plays an important role.
3. The inclusion of a third joint set vertically oriented results in a further loss in ultimate
bearing capacity of up to 40% as compared to the results for a rock mass with two joint
sets only. Parameters similar to those in the two joint case were again found to be
critical.
Yang and Yin (2005) established a theorem on the ultimate bearing capacity of a strip footing
with the modified failure criterion using the generalized tangential technique. Assumptions
were made such as the strip footing is long enough for consideration of plain strain problem,
rigid triangular blocks of symmetrical translation failure mechanism, homogenous and
isotropic rock masses are idealized as a perfectly plastic material, and follow the associated
flow rule and lastly, the failure of the rock masses is governed by a modified HB failure
criterion. Using the technique, an MC linear failure criterion, which is tangent to the actual HB
failure criterion, is proposed to calculate the rate of external work and internal energy
dissipation. Suggested equations in this theorem executes with a maximum difference of less
Page | 6
LITERATURE REVIEW
than 0.5% for bearing capacity factors. It is found that the surcharge load and self-weight have
effects on the ultimate bearing capacity, and that the contribution related to uniaxial
compressive strength can be separated from the ultimate bearing capacity while the
contributions related to surcharge load and unit weight cannot be separated from the ultimate
bearing capacity. It should be noted that lower bound solution is less than or equal to actual
solution.
Merifield et al. (2006) investigated bearing capacity of a surface strip footing resting on a rock
mass whose strength can be described by the generalized Hoek–Brown failure criterion and
drawn the following conclusions:
1. The effect of ignoring rock weight can lead to a very conservative estimate of the
ultimate bearing capacity. This is particularly the case for poorer quality rock types.
2. Estimating the ultimate bearing capacity of a rock mass using equivalent Mohr–
Coulomb parameters was found to significantly overestimate the bearing capacity.
3. Existing numerical solutions for weightless rock masses are generally conservative.
Imani et al. (2012) concluded that application of the upper bound theorem was used for
relatively simple configurations because of the non-homogeneity and discontinuous nature of
rock masses. Further observation stated that the Spacing Ratio, different mechanical properties
of the intact rock and the joint sets did not show a remarkable effect on the bearing capacities
obtained from the upper bound solution. Also it has been observed that the shape (i.e. either a
straight line or a log-spiral curve) of velocity discontinuity lines does not have any significant
effect on the submerged bearing capacity. Furthermore another conclusion is that the bearing
capacity of dry and submerged rocks indicate that submergence of the rock below the footing
base reduces the contribution of the rock weight in the bearing capacity.
Page | 7
THEORY
3. THEORY
3.1 INTRODUCTION
From civil engineering aspects, foundation is commonly known as the lowest part of a
structure that transmits its weight to the underlying soil or rock. Foundations are generally
classified into two major categories – shallow foundation and deep foundation. Considering
splendid research studies of foundations, it can be concluded that if the depth of foundation is
less than or equal to 4 times the width of foundation then the foundation is shallow foundation
or otherwise if the depth of foundation is more than 4 times the width of foundation then the
foundation is deep foundation.
Most commonly bearing capacity is the load-carrying capacity of foundation soil or

rock which enables it to bear and transmit loads from a structure. Whereas ultimate bearing
capacity is the theoretical maximum pressure which a foundation can support without the
occurrence of shear failure. In other words, ultimate bearing capacity is the external load
applied per unit area of the foundation after which shear failure occurs.
Intact rock permeability is small and negligible in comparison to the permeability of

joints. However, in field, the whole rock mass below the water table is saturated after a specific
time lag and leads to buoyancy of the rock mass. For various combinations of mechanical
properties of intact rock and joint set, there is a depth beyond which the groundwater has no
effect on the bearing capacity. This depth is called the ‘critical depth’.
3.2 DETERIMINATION OF BEARING CAPACITY
3.2.1 ASSUMPTIONS
For the calculations of bearing capacity of shallow foundations on rocks for both dry
and submerged conditions following assumptions were made:
1. A rock mass containing two orthogonal tight joint sets was considered.
2. The concept of spacing ratio proposed by Serrano and Olalla (1996) is used to account
the joint spacing.
3. The mechanical properties of the intact rock and the joint sets are not affected by the
groundwater.
4. The Mohr-Coulomb failure criteria was used for both the intact rock and the joint sets.
Page | 9
THEORY
5. Orientation angles (α) equal to 15o, 30o and 45o were considered for one of the joint
sets.
6. A two-sided symmetrical failure mechanism is used for the case of α=45o and a one-
sided asymmetrical failure mechanism is used for α=15o and 30o.
7. Velocity discontinuity line at failure plane is considered to be a straight line and its
beginning and end points are located at the junction of the junction of the two joint sets,
not within intact rock block.
3.2.2 FAILURE MECHANISM
The mechanism by which the minimum bearing capacity can be determined is formally
known as the most appropriate failure mechanism. Imani et al. (2012) suggested two different
failure mechanism depending upon joint set condition –
1. Two-sided mechanism.
2. One-sided mechanism.
3.2.2.1 Two-sided Mechanism
Two-sided mechanism is symmetrical in shape. This mechanism is considered only

with centric and vertical footing in the foundations. A hodograph diagram of two sided failure
mechanism is shown below.
Fig. 3.1: Hodograph diagram of two sided mechanism
Page | 10
THEORY
3.2.2.2 One-sided Mechanism
One-sided mechanism is asymmetrical in shape. In presence of joint set, the failure

mechanism may be affected by the joint set and shape is converted to asymmetrical. A
hodograph diagram of one sided failure mechanism is shown below.
Fig. 3.2: Hodograph diagram of one sided mechanism
3.2.3 CALCULATIONS
Two different sets of equation were proposed for each of the mechanisms. Each set
contains different equations for dry condition as well as submerged condition for determining
submerged bearing capacity of rock mass. For ease of calculations, a rock mass containing
orthogonal tight joint sets were considered. By knowing footing width (B), orientation angle
(α), rock mass strength properties (ci, cj, ϕi, ϕj), physical properties (γ, γsat), water table location
(dw), surcharge (q) one can find out the ultimate bearing capacity of rock mass by the following
equations:
For dry condition: qu = ci Nci + cj Ncj + q Nq + 0.5 γ B Nγ … eq. 1
For submerged condition: qu = ci Nci + cj Ncj + q Nq + 0.5 γ B Nγsub ………eq. 2
The angle of velocity displacement line (CD) with the horizontal direction is given by:
n0 S1
θ = tan-1 ( )–α ……………………………………………... eq. 3
B cosα
Assumptions made for calculations of ultimate bearing capacity are as follows –
Page | 11
THEORY
ξ1 = α + θ – ϕi – ϕj ……………………………………………....eq. 4
ξ2 = α + θ – ϕi + ϕj ………………………………………………eq. 5
ξ3 = α + ϕj ………………………………………………eq. 6
ξ4 = α + θ ………………………………………………eq. 7
ξ5 = ϕi – θ ………………………………………………eq. 8
ξ6 = α – ϕj ………………………………………………eq. 9
Now putting fn is the function of ξ & ϕj for two sided mechanism -
f1 = sinξ6 – sinξ5 cosξ1 ………………………………………………eq. 10
f2 = – sinξ6 sinξ5 + cosξ1 ………………………………………………eq. 11
f3 = sin2ϕj cosξ2 + sinξ1 ………………………………………………eq. 12
f4 = sin2ϕj + cosξ2 sinξ1 ………………………………………………eq. 13
Again putting gn as the function of ξ & ϕj for one sided mechanism -
g1 = cosξ1 sinξ2 – sin2ϕj ………………………………………………eq. 14
g2 = cosξ1 – sinξ2 sin2ϕj ………………………………………………eq. 15
g3 = sin2ϕj cosξ2 + sinξ1 ………………………………………………eq. 16
g4 = sin2ϕj + cosξ2 sinξ1 ………………………………………………eq. 17
Bearing capacity coefficients can be find out from these following equations –
For two-sided mechanism
2 cosϕj cosα
Ncj =
f1
[cos2ξ5 + ff23 x tanξ4 ( sin2ξ2 + tanα
f4
)] ………eq. 18
f2 2 cosϕi cosα
Nci = x ……………………………………....eq. 19
f1 cosξ4
f2 f4 2 sinξ3 tanξ4
Nq = x ……………………………………....eq. 20
f1 f3 tanα
Nγ = cosα [ 2f1f2 x cosα tanξ4 x ( sinξ5 + ff43 x sinξtanα

3 tanξ4
) – sinα ] ………eq. 21
Page | 12
THEORY
For one-sided mechanism
cosϕj cosα 2
Ncj =
cosξ3
[ tanα + cosg1 ξ2 + gg1g23 x tanξ4 ( sin2ξ2 + tanα
g4
)]………eq. 22
g2 cosϕi cosα
Nci = x ……………………………………....eq. 23
g1 cosξ3 cosξ4
g2 g4 tanξ3 tanξ4
Nq = x ……………………………………....eq. 24
g1 g3 tanα
Nγ = cosα [ gg21 x cosα tanξ4

cosξ3
g
x ( sinξ5 + 4 x
g3
sinξ3 tanξ4
tanα
) – sinα ]………eq. 25
Submerged bearing capacity coefficient (Nγsub) for both the mechanisms can be find out from
the following equation:
γ′ dw γ′
Nγsub =
γ
Nγ +
B
( 1 - γ ) Nγw ……………………………………....eq. 26
It had been considered that the water table is located at a depth dw from the foundation base.
Location of the water table can vary to any depth of rock mass. Therefore, two different heights
had been established to distinguish two different boundaries for the water table and an
additional factor (Nγw) had been established that depends on the groundwater depth. Referring
to fig. 3.1 and fig.3.2 water table location varies from foundation base level to initial point
(point C) of velocity discontinuity line CD (hC) for one condition while for another varies up
to last point (point D) of velocity discontinuity line CD (hD). It has been observed that beyond
the point D, water table has no effect on bearing capacity and ultimate bearing capacity will be
same as in the dry condition. These two boundary heights can be determined by the following
equations -
hC = B sinα cosα ……………………………………………....eq. 27
hD = B cos2α tan(α + θ) ………………………………………………eq. 28
For different condition of water table depth with boundary heights additional factor (Nγw) can
be evaluated from the following equations:
For two sided mechanism
a) If 0 ≤ dw ≤ hC
Page | 13
THEORY
dw 2 2 f2 2 f2 f4 4 f2 f4 sinξ3 tanξ4
Nγw= ( 1+ sinξ5 – sinξ3 ) + –2 …eq. 29
B sin2α f 1 f f 1 3 f f tanα1 3
b) If hC ≤ dw ≤ hD
4 f2 cosξ4 sinξ5 f4 sinξ3 2 dw f2 1 cosξ4 sinξ5

Nγw =
f1
tanξ4 cosα (
sinθ
+
f3 sinα
)– B f1 cosα
( sinθ
+
f4 sinξ3 B
f3 sinα
)+
dw
cosα [ 2f1f2 cosα tanξ4 sinξ5 ( 1 – cosαsinθ
sinξ4
) – sinα ]
……….. eq. 30
For one sided mechanism
a) if 0 ≤ dw ≤ hC
dw 2 g2 sinξ5 g2 g4 2 g2 g4 tanξ3 tanξ4

Nγw= ( 1+ – tanξ3 ) + –2 ……eq. 31
B sin2α g 1 cosξ3 g1 g3 g1 g3 tanα
b) if hC ≤ dw ≤ hD
2 g2 sinξ4 cosα sinξ5 g4 sinξ3 dw g2 1

Nγw =
g1 cosξ3
( sinθ
+
g3 sinα cosξ4
)– B g1 cosα cosξ3
cosξ4 sinξ5 g4 sinξ3 B g cosα tanξ4 sinξ5

( sinθ
+
g3 sinα
) +
d
cosα [
g
2
cosξ3
( 1- cosαsinθ
sinξ4
) – sinα]
w 1
…… eq. 32
It should be noted that if dw > hD, groundwater has no effect on bearing capacity and therefore,
Nγsub will be equal to Nγ for both mechanisms.
3.3 ALGORITHMIC ANALYSIS
An algorithm is an effective method expressed as a finite list of well-defined

instructions for calculating a function. Starting from an initial state and initial input, the
instructions describe a computation that, when executed, proceeds through a finite number of
well-defined successive states, eventually producing output and terminating at a final ending
state. Most commonly algorithm is known as a step by step procedure for calculations or
solving a problem.
3.3.1 FLOWCHART
A flowchart is a type of diagram that represents an algorithm, workflow or process,

showing the steps as boxes of various kinds, and their order by connecting them with arrows.
Page | 14
THEORY
Commonly flowchart is a diagrammatic representation of a solution to any problem. Flowcharts

are used in analyzing, designing, documenting or managing a process or program in various
fields. There are many different types of flowcharts, and each type has its own repertoire of
boxes and notational conventions.
3.3.2 MATLAB Programming
MATLAB is a high-level language and interactive environment for numerical

computation, visualization, and programming developed by MathWorks Inc., Natick,
Massachusetts, USA. MATLAB is a fourth generation programming language which is
powerful and popular language for technical computing and most prominently easy to use.
MATLAB can be used for mathematical computations, algorithm development, modeling,
matrix manipulations, plotting of functions and data, creation of user interfaces, interfacing
with programs written in other languages, including C, C++, Java, and FORTRAN and many
others.
Most commonly, a program is a sequence of computational commands. MATLAB

programs are generally written in MATLAB script file. A script file is a list of MATLAB
commands, called a program that is saved in a file. When the script file is executed, MATLAB
executes the commands in the order in which they are listed, and in which all the variables are
defined within the script file. MATLAB provides several tools that can be used to control the
flow of a program. Conditional statements and the switch structure (Appendix III) make it
possible to skip commands or to execute specific groups of commands in different situations.
3.4 RELIABILITY ANALYSIS
3.4.1 IMPORTANCE
In-situ rock mass variability renders the deterministic analysis to be inefficient; and,
hence there arises the necessity for probabilistic or reliability analyses to model the
uncertainties. Rather than calculating a deterministic factor of safety, a reliability based
analysis is more appropriate for geotechnical design. Such analysis indicates the performance
and reliability of a geotechnical problem, and can be used for risk-based decision making.
3.4.2 DEFINITION
The reliability of an engineering system can be defined as the confidence on its ability
to fulfill its design purpose for some time-period. The theory of probability provides the
Page | 15
THEORY
fundamental basis to measure this ability. The reliability of a structure can be viewed as the
probability of its satisfactory performance, according to some performance functions, for a
specific service and subjected to extreme conditions within a stated time-period. In estimating
this probability, system uncertainties are modeled using random variables with mean values,
variances, and probability distribution functions.
3.4.3 OVERVIEW
Generally, reliability analysis deals with the relation between the loads carried by a
system and its ability to carry those loads. Both the loads and the resistance may be uncertain,
so the result of their interaction is also uncertain. Loads and resistance may include not only
forces and stresses, but also seepage, settlement, and any other phenomena that might become
design considerations. The values of both load and resistance are uncertain, so these variables
have mean or expected values, variances, and co-variances, as well as other statistical
descriptors.
To initiate a reliability analysis, random fields of soil or rock mass properties are
commonly generated to derive the required statistical parameters, e.g. mean and standard
deviation. A method of reliability analysis is then selected for determining the probability of
failure and the reliability index. Some commonly used techniques are the Monte Carlo
simulation, First Order methods and Point Estimate method.
3.4.4 RELIABILITY INDEX
Reliability index is defined as the ratio of mean value of margin of safety to variance
of margin of safety. Reliability index depends on correlation coefficient, as reliability index
varies with the condition of correlation coefficient. Calculations of the reliability index are
more difficult when it is expressed in terms of the factor of safety, because factor of safety is
the ratio of two uncertain quantities while Margin of safety is their difference.
The margin of safety, M, is the difference between the resistance and the load:
Mathematically - M=R–Q eq. 33
From the elementary definitions of mean and variance, it follows that regardless of the
probability distributions of Resistance and Load, the mean value of Margin of Safety is
µM = µR − µQ eq. 34
Page | 16
THEORY
and the variance of Margin of Safety is
σM2 = σR2 + σQ2 – 2ρRQσRσQ eq. 35

Then, a reliability index, β, is defined as
μM
β= eq. 36
σM
μ𝑅 − μ𝑄
Therefore, β= eq. 37
2
√σ2𝑅 + σ𝑄 −2 ρ𝑅𝑄 σ𝑅 σ𝑄
Which expresses the distance of the mean margin of safety from its critical value (M = 0) in
units of standard deviation. If the load and resistance are uncorrelated, the correlation
coefficient is zero, and
μ𝑅 − μ𝑄
β= eq. 38
√σ2𝑅 + σ2𝑄
eq. 37 and eq. 38 shows variation of reliability index with the condition of correlation
coefficient.
Factor of safety, F, is defined as
R
F= eq. 39
Q
Failure occurs when F=1, and a reliability index is defined by
E [ F ]−1
β= eq. 40
σF
For very small values of reliability index the probability of failure is actually slightly
larger for the Normal distribution than for the others.
3.4.5 STEPS IN RELIABILITY ANALYSIS
To explain what is involved in the various approaches to reliability calculations, it is

useful to set out the steps explicitly. The goal of the analysis is to estimate the probability of
failure, with the understanding that ‘failure’ may involve any unacceptable performance. The
steps are:
i. Establish an analytical mode: There must be some way to compute the margin of safety,
factor of safety, or other measure of performance. It can be simple or it can be an elaborate
Page | 17
THEORY
computational procedure. There may be error, uncertainty, or bias in the analytical model,
which can be accounted for in the reliability analysis.
ii. Estimate statistical descriptions of the parameters: The parameters include not only
the properties of the geotechnical materials but also the loads and geometry. Usually, the
parameters are described by their means, variances, and co-variances, but other
information such as spatial correlation parameters or skewness may be included as well.
iii. Calculate statistical moments of the performance function: Usually this means
calculating the mean and variance of the performance function.
iv. Calculate the reliability index: Often this involves the simple application of reliability
index equations. Sometimes, as in the Hasofer–Lind method, the computational
procedure combines this step with the previous one.
v. Compute the probability of failure: If the performance function has a well-defined
probabilistic description, such as the Normal distribution, this is a simple calculation. In
many cases the distribution is not known or the intersection of the performance function
with the probabilistic description of the parameters is not simple. In these cases the
calculation of the probability of failure is likely to involve further approximations.
3.4.6 ERROR PROPAGATION
The evaluation of the uncertainty in the computed value of the margin of safety and the
resulting reliability index is a special case of error propagation. Basically, uncertainties in the
values of parameters propagate through the rest of the calculation and affect the final result.
Generally, each of the computational steps involves some error and uncertainty on its own, and
uncertainties in the original measurements of object properties will affect the numbers
calculated at each subsequent step.
3.4.7 SOLUTION TECHNIQUES
In more typical cases, the analyst must usually employ a technique that yields an
approximation to the true value of the reliability index and the probability of failure. Several
methods are available, each having advantages and disadvantages.
3.4.7.1 The First Order Second Moment (FOSM) Method
This method uses the first terms of a Taylor series expansion of the performance
function to estimate the expected value and variance of the performance function. It is called a
second moment method because the variance is a form of the second moment and is the highest
Page | 18
THEORY
order statistical result used in the analysis. If the number of uncertain variables is N, this method
requires either evaluating N partial derivatives of the performance function or performing a
numerical approximation using evaluations at 2N+1 points.
3.4.7.2 The Second Order Second Moment (SOSM) Method
This technique uses the terms in the Taylor series up to the second order. The
computational difficult is greater, and the improvement in accuracy is not always worth the
extra computational effort. The Second Order Second Moment methods have not found wide
use in geotechnical applications.
3.4.7.3 The Point Estimate Method
Rosenblueth (1975) proposed a simple and elegant method of obtaining the moments
of the performance function by evaluating the performance function at a set of specifically
chosen discrete points. One of the disadvantages of the original method is that it requires that
the performance function be evaluated 2N times, and this can become a very large number when
the number of uncertain parameters is large. Recent modifications reduce the number of
evaluations to the order of 2N, but introduce their own complications.
3.4.7.4 The Hasofer–Lind Method
Hasofer and Lind (1974) proposed an improvement on the First Order Second Moment
method based on a geometric interpretation of the reliability index as a measure of the distance
in dimensionless space between the peak of the multivariate distribution of the uncertain
parameters and a function defining the failure condition. This method usually requires iteration
in addition to the evaluations at 2N points. The acronym First Order Reliability method often
refers to this method in particular.
3.4.7.5 Monte Carlo Method
Essentially, Monte Carlo method is a statistical approach to the study of differential

equations, or more generally, of integro-differential equations that occur in various branches
of the natural sciences. In this approach the analyst creates a large number of sets of randomly
generated values for the uncertain parameters and computes the performance function for each
set. The statistics of the resulting set of values of the function can be computed and Reliability
Index is calculated directly. The method has the advantage of conceptual simplicity, but it can
require a large set of values of the performance function to obtain adequate accuracy.
Page | 19
THEORY
Computational operations can be divided roughly into two classes. Firstly, production
of random values with their frequency distribution equal to those which govern the change of
each parameter. Secondly, calculation of the values of those parameters which are
deterministic, i.e. obtained algebraically from the others. Monte Carlo simulation is categorized
as a sampling method because the inputs are randomly generated from probability distributions.
The data generated from the simulation can be represented as histograms or converted to error
bars, reliability predictions, tolerance zones, and confidence intervals.
One interesting feature of the method is that it allows one to obtain the values of certain
given operations on functions obeying a differential equations, without point to point
knowledge of the functions which are solutions of the equations. One more advantage is that
we avoid dealing with multiple integrations or multiplications of the probability matrices, but
instead sample single chains of events.
3.4.8 RECOMMENDATION
Before carrying out a reliability analysis it is very important to understand that most
practical methods of reliability analysis involve approximations, even if one or more steps are
exact. It is very obvious that different methods will give different answers. Therefore, it is often
a good idea to compare results from two or more approaches to gain an appreciation of the
errors involved in the computational procedures.
3.4.9 STATISTICAL TERMS
Statistical terms used in reliability analysis only are described below:
3.4.9.1 Population
The group from which data are collected or a sample is selected. The population
encompasses the entire group for which the data are alleged to apply.
3.4.9.2 Sample
An individual or group, selected from a population, from whom or which data are
collected.
3.4.9.3 Mean
The mean is simply the arithmetic average of a distribution of scores and it provides a
single, simple number that gives a rough summary of the distribution. It is important to
Page | 20
THEORY
remember that although the mean provides a useful piece of information, it does not tell
anything about how spread out the scores are (i.e., variance) or how many scores in the
distribution are close to the mean.
3.4.9.4 Median
The median is the score in the distribution that marks the 50th percentile. That is, 50%
of the scores in the distribution fall above the median and 50% fall below it. Median is often
used when distribution scores are dived into two equal groups (a median split). The median is
also a useful statistic to examine when the scores in a distribution are skewed or when there
are a few extreme scores at the high end or the low end of the distribution.
3.4.9.5 Mode
The mode is the least used of the measures of central tendency because it provides the
least amount of information. The mode simply indicates which score in the distribution occurs
most often, or has the highest frequency.
3.4.9.6 Range
The range is simply the difference between the largest score (the maximum value) and
the smallest score (the minimum value) of a distribution.
3.4.9.7 Interquartile Range
Unlike the range, the interquartile range is the difference between the score that marks
the 75th percentile (the third quartile) and the score that marks the 25th percentile (the first
quartile). If the scores in a distribution were arranged in order from largest to smallest and then
divided into groups of equal size, the interquartile range would contain the scores in the two
middle quartiles.
3.4.9.8 Variance
The variance provides a statistical average of the amount of dispersion in a distribution

of scores. In general, variance is used more as a step in the calculation of other statistics (e.g.,
analysis of variance) than as a stand-alone statistic.
3.4.9.9 Standard Deviation
The word ‘deviation’, in this case, refers to the difference between an individual score
in a distribution and the average score for the distribution. So if the average score for a
Page | 21
THEORY
distribution is 10, and an individual score is 12, then the deviation is 2. The other word
‘standard’ means typical, or average. So a standard deviation is the typical, or average,
deviation between individual scores in a distribution and the mean for the distribution.
3.4.9.10 Normal Distribution
The normal distribution is known in statistics as a theoretical distribution. There are a

number of statistics that begin with the assumption that scores are normally distributed. When
this assumption is violated (i.e., when the scores in a distribution are not normally distributed),
there can be dire consequences.
A more familiar name for the normal distribution is the ‘bell curve’, because a normal
distribution forms the shape of a bell. Fig. 3.3, represented as a normal distribution curve, which
has three fundamental characteristics. First, it is symmetrical, meaning that the upper half and
the lower half of the distribution are mirror images of each other. Second, the mean, median,
and mode are all in the same place, in the center of the distribution (i.e., the top of the bell
curve). Because of this second feature, the normal distribution is highest in the middle. Finally,
the normal distribution is asymptotic, meaning that the upper and lower tails of the distribution
never actually touch the baseline, also known as the x-axis.
Fig. 3.3: Line diagram representing standard shape of normal distribution.
3.4.9.11 Skew
When a sample of scores is not normally distributed (i.e., not the bell shape), there are
a variety of shapes it can assume. If there are a few scores creating an elongated tail at the
higher end of the distribution, it is said to be positively skewed. If the tail is pulled out toward
Page | 22
THEORY
the lower end of the distribution, the shape is called negatively skewed. So a positively skewed
distribution will have a higher mean than median, and a negatively skewed distribution will
have a smaller mean than median.
3.4.9.12 Kurtosis
Kurtosis refers to the shape of the distribution in terms of height, or flatness. When a
distribution has a peak that is higher than that found in a normal, bell-shaped distribution, it is
called ‘leptokurtic’. When a distribution is flatter than a normal distribution, it is called
‘platykurtic’. A leptokurtic distribution will have a greater percentage of scores closer to the
mean and fewer in the upper and lower tails of the distribution, whereas a platykurtic
distribution will have more scores at the ends and fewer in the middle than will a normal
distribution.
3.4.9.13 Standard Error
Technically, a standard error is, in effect, the standard deviation of the sampling
distribution of some statistic (e.g., the mean, the difference between two means, the correlation
coefficient, and many others.). Or in other words, standard error is the denominator in the
formulas used to calculate many inferential statistics. This is because the standard error is the
measure of how much random variation we would expect from samples of equal size drawn
from the same population.
3.4.10 MICROSOFT EXCEL SPREADSHEET
Microsoft Excel spreadsheet is a spreadsheet application developed by Microsoft,

Albuquerque, New Mexico, U.S. It features calculation, graphing tools, pivot tables, and a
macro programming language called Visual Basic for Applications. Microsoft Excel has the
basic features of all spreadsheets using a grid of cells arranged in numbered rows and letter-
named columns to organize data manipulations like arithmetic operations. It is very useful for
statistical, engineering and financial needs. In addition, it can display data as line graphs,
histograms and charts, and with a very limited three-dimensional graphical display.
Page | 23
ANALYSIS and DISCUSSION
4. ANALYSIS AND DISCUSSION
4.1 ALGORITHMIC ANALYSIS
4.1.1 FLOWCHART
A flowchart or in a broad sense a diagrammatic representation of the complete

algorithm for determination of bearing capacity has been developed. The flowchart represents
workflow or process of the algorithm in terms of different boxes in order and by connecting
them with arrows. With the help of this flowchart one can understand the complete process of
algorithm i.e. from input (footing width, orientation angle, rock mass strength properties,
physical properties, water table location, and surcharge) to output (ultimate bearing capacity)
and how the mechanisms are selected or how the water table condition is selected. Fig. 4.1
represents the flowchart of the algorithmic model for determination of ultimate bearing
capacity of rock foundation for both dry and submerged condition. It can be easily noticed that
after division of mechanism condition, workflow for each mechanism is quite same, but for
each one, theorem conditions are different and mathematical equations are also different. The
symbols used in the flowchart are described in the Appendix I.
4.1.2 ALGORITHMIC MODEL
For determination of ultimate bearing capacity of rock mass for both dry and submerged
condition, an algorithmic model has been developed in MATLAB script. Commands and
programs implemented in this model are mentioned in Appendix II and Appendix III
respectively. For any case, with known footing width (B), orientation angle (α), rock mass
strength properties (ci, cj, ϕi, ϕj), physical properties (γ, γsat), water table location (dw), surcharge
(q), ultimate bearing capacity (qu) of rock mass for both dry and submerged condition can be
determined with the help of this model in MATLAB software.
It is necessary to mention that some command lines are not short enough to write in a
single line. Therefore the incomplete portion is written in the next line or if it is not completed
in that line also then another line is added and so on. Line numbers are addressed before each
line to understand more precisely. But in MATLAB script file, line numbers will be addressed
automatically and if each single command line is converted into many lines without ellipsis
(…) then the program will run with error. So it is better to write a single command line in one
line as a line in MATLAB script file may contain 4096 characters.
Page | 25
START
Input Parameters
(B, q, α, ci, cj, ϕi, ϕj, γ)
Ground Water Condition=? SUBMERGED Input Parameters

DRY or SUBMERGED? (γsat, dw)
DRY
no
If α = 450? OS Mechanism
yes
TS Mechanism
Calculation of θ, ξ1, ξ2, ξ3, ξ4, ξ5, ξ6
yes Foundation Width can be

If B<=0? neither zero nor negative
no
yes Unit Weight can be neither

If γ <=0? zero nor negative
no
yes Orientation Angle can be

If α <=0? neither zero nor negative
no
TS Mechanism Mechanism Condition OS Mechanism

Whether TS or OS?
If 0<α-θ+ϕi,+ϕj<90? If 0<α+θ-ϕi,-ϕj<90?
yes
yes
no
no
Calculation of g1, g2,
Do not satisfy the Calculation of f1, f2, Do not satisfy the
g3, g4, Nci, Ncj, Nq, Nγ
constrain condition f3, f4, Nci, Ncj, Nq, Nγ constrain condition
DRY Ground Water Condition=? Ground Water Condition=? DRY

DRY or SUBMERGED? DRY or SUBMERGED?
SUBMERGED SUBMERGED
Ultimate Bearing Calculation of γeff, hC, hD Calculation of γeff, hC, hD Ultimate Bearing
Capacity for Capacity for
DRY condition DRY condition
yes no no yes
If dw<=hD? If dw<=hD?
yes Nγsub = Nγ Nγsub = Nγ yes

If hC<dw≤hD? If hC<dw≤hD?
no
no
Ground Water Table will

Calculation Calculation of Nγw not have any effect on Calculation of Nγw Calculation
of Nγw Ultimate Bearing Capacity of Nγw
Ultimate Bearing Calculation of Nγsub

Calculation of Nγsub Capacity for
SUBMERGED condition
END
Fig. 4.1: Flowchart of Algorithmic Model for determination of Ultimate Bearing Capacity
Page | 26
The algorithmic model for determination of ultimate bearing capacity of rock mass
developed in MATLAB script is given below –
1. %Algorithmic model for determination of Ultimate Bearing Capacity of Rock

Foundation.
%Model INPUT parameters are Foundation Width, Surcharge, Orientation Angle, Rock
2.
Mass Strength Properties, Physical Properties, Water Table Location.
3. %Model OUTPUT parameter is Ultimate Bearing Capacity of Rock Foundation.
4. clc
5. clear all
6. format compact
disp ('Welcome to the Algorithmic model for determination of Ultimate Bearing
7.
Capacity of Rock Foundation')
8. disp ('Please enter the magnitudes of the following INPUT Parameters one by one')
9. % INPUT PARAMETERS
10. B=input('Width of Foundation (in m)=');
11. q=input('Surcharge magnitude (in kPa)=');
12. alpha = input ('Orientation Angle (in degree)=');
13. ci=input('Cohesion of Intact rock (in kPa)=');
14. cj=input('Cohesion of Jointed rock (in kPa)=');
15. phii = input ('Angle of Friction of Intact rock (in degree)=');
16. phij = input ('Angle of Friction of Jointed rock (in degree)=');
17. r=input('Bulk Unit Weight (in kN/m^3)=');
18. disp('Type only the word in Capital letters')
19. GWC=input('Ground Water condition,(DRY or SUBMERGED)=','s');
20. switch GWC
21. case 'SUBMERGED'
22. rsat=input ('Saturated Unit Weight of Submerged Rock (in kN/m^3)=');
23. disp('Water Table Location should be Measured from Ground Level');
24. dw=input ('Depth of Water (in m)=');
25. end
26. if alpha == 45;
27. mechanism = 'TS';
28. else mechanism = 'OS';
Page | 27
29. end
30. clc
31. % CALCULATION PART
32. n0 = 100;
33. s1 = 0.039;
34. theta = atand(n0*s1/(B*cosd(alpha)))-alpha;
35. xi1=alpha+theta-phii-phij;
36. xi2=alpha+theta-phii+phij;
37. xi3=alpha+phij;
38. xi4=alpha+theta;
39. xi5=phii-theta;
40. xi6=alpha-phij;
41. if B<=0;
disp('Foundation Width can be neither zero nor negative. Please try again with
42.
correction of Foundation Width.')
43. elseif r<=0;
disp('Unit Weight can be neither zero nor negative. Please try again with correction of
44.
Unit Weight.')
45. elseif alpha<=0;
disp('Orientation Angle can be neither zero nor negative. Please try again with
46.
correction of Orientation Angle.')
47. elseif B>0&&r>0&&alpha>0;
48. switch mechanism
49. case 'TS'
50. if 0>(alpha-theta+phii+phij)||(alpha-theta+phii+phij)>90;
51. disp ('Do not satisfy the constrain condition.')
52. elseif 0<(alpha-theta+phii+phij)&&(alpha-theta+phii+phij)<90;
53. f1=sind(xi6)-sind(xi5)*cosd(xi1);
54. f2=-sind(xi6)*sind(xi5)+cosd(xi1);
55. f3=sind(2*phij)*cosd(xi2)+sind(xi1);
56. f4=sind(2*phij)+cosd(xi2)*sind(xi1);
Ncj=2*cosd(phij)*cosd(alpha)/f1*((cosd(xi5))^2 +
57.
f2/f3*tand(xi4)*((sind(xi2))^2 + f4/tand(alpha)));
Page | 28
58. Nci=f2/f1*2*cosd(phii)*cosd(alpha)/cosd(xi4);
59. Nq=f2*f4/(f1*f3)*2*sind(xi3)*(tand(xi4))/tand(alpha);
Nr=cosd(alpha)*(2*f2/f1*cosd(alpha)*tand(xi4)*(sind(xi5) +
60.
f4/f3*sind(xi3)*tand(xi4)/tand(alpha)) - sind(alpha));
61. switch GWC
62. case 'DRY'
63. qu=cj*Ncj+ci*Nci+q*Nq+0.5*r*B*Nr;
64. qu=qu/1000;
fprintf ('Ultimate Bearing Capacity of Rock Foundation is %5.3f MPa for
\n %3.1f m Foundation Width, %4.1f kPa Surcharge, %4.1f degree
Orientation Angle, \n %5.1f kPa Cohesion and %5.2f degree Angle of
65.
Friction of Intact Rock, \n %4.1f kPa Cohesion and %5.2f degree Angle
of Friction of Jointed Rock and \n %5.2f kN/m^3 Bulk Unit weight. \n',
qu,B,q,alpha,ci,phii,cj,phij,r)
67. rw=9.807;
68. reff=rsat-rw;
69. hC=B*sind(alpha)*cosd(alpha);
70. hD=B*(cosd(alpha))^2*tand(alpha+theta);
71. if dw<=hD;
72. if dw<=hC;
Nrw=2*dw/(B*sind(2*alpha))*(1+2*f2/f1*sind(xi5) - 2*f2*f4/(f1
73. *f3)*sind(xi3)) +4*f2*f4/(f1*f3)*sind(xi3)*tand(xi4)/tand(alpha) -
2;
74. elseif hC<dw&&dw<=hD;
Nrw=4*f2/f1*tand(xi4)*cosd(alpha)*(cosd(xi4)*sind(xi5)/sind(thet
a) + f4*sind(xi3)/(f3*sind(alpha))) - 2*dw*f2/(B*f1*cosd(alpha))
75. *(cosd(xi4)*sind(xi5)/sind(theta) + f4*sind(xi3)/(f3*sind(alpha))) +
B*cosd(alpha)/dw*(2*f2/f1*cosd(alpha)*tand(xi4)*sind(xi5)*(1-
cosd(alpha)*sind(xi4)/sind(theta)) - sind(alpha));
76. end
77. Nrsub=reff*Nr/r+dw/B*(1-reff/r)*Nrw;
78. elseif dw>hD;
Page | 29
79. Nrsub=Nr;
fprintf ('Note: Ground Water Table will not have any effect on
80.
Ultimate Bearing Capacity. \n')
81. end
82. qu=cj*Ncj+ci*Nci+q*Nq+0.5*r*B*Nrsub;
83. qu=qu/1000;
84. Friction of Intact Rock, \n %4.1f kPa Cohesion and %5.2f degree Angle
of Friction of Jointed Rock,\n %5.2f kN/m^3 Bulk Unit weight, %5.2f
kN/m^3 Saturated Unit weight, %3.1f m Water Table from
GL.\n',qu,B,q,alpha,ci,phii,cj,phij,r,rsat,dw)
85. end
86. end
87. case 'OS'
88. if 0>(alpha+theta-phii-phij) || (alpha+theta-phii-phij)>90;
89. disp ('Do not satisfy the constrain condition.')
90. elseif 0<(alpha+theta-phii-phij)&&(alpha+theta-phii-phij)<90;
91. g1=cosd(xi1)*sind(xi2)-sind(2*phij);
92. g2=cosd(xi1)-sind(xi2)*sind(2*phij);
93. g3=sind(2*phij)*cosd(xi2)+sind(xi1);
94. g4=sind(2*phij)+cosd(xi2)*sind(xi1);
Ncj=cosd(phij)*cosd(alpha)/cosd(xi3)*(tand(alpha) + (cosd(xi2))^2/g1 +
95.
g2/(g1*g3) *tand(xi4)*((sind(xi2))^2 + g4/tand(alpha)));
96. Nci=g2/g1*cosd(phii)*cosd(alpha)/(cosd(xi3)*cosd(xi4));
97. Nq=(g2*g4)/(g1*g3)*tand(xi3)*tand(xi4)/tand(alpha);
Nr=cosd(alpha)*(g2/g1*cosd(alpha) *tand(xi4)/cosd(xi3) *(sind(xi5) + g4/g3
98.
*sind(xi3) *tand(xi4)/tand(alpha)) - sind(alpha));
99. switch GWC
100. case 'DRY'
101. qu=cj*Ncj+ci*Nci+q*Nq+0.5*r*B*Nr;
102. qu=qu/1000;
Page | 30

103.
Friction of Intact Rock, \n %4.1f kPa Cohesion and %5.2f degree Angle
of Friction of Jointed Rock and \n %5.2f kN/m^3 Bulk Unit weight. \n',
qu,B,q,alpha,ci,phii,cj,phij,r)
105. rw=9.807;
106. reff=rsat-rw;
107. hC=B*sind(alpha)*cosd(alpha);
108. hD=B*(cosd(alpha))^2*tand(alpha+theta);
109. if dw<=hD;
110. if dw<=hC;
Nrw=2*dw/(B*sind(2*alpha))*(1+g2*sind(xi5)/(g1* cosd(xi3)) -
111. g2*g4/(g1*g3)*tand(xi3)) + 2*g2*g4/(g1*g3) *tand(xi3)
*tand(xi4)/tand(alpha) - 2;
112. elseif hC<dw&&dw<=hD;
Nrw=2*g2/g1*sind(xi4)*cosd(alpha)/cosd(xi3)
*(sind(xi5)/sind(theta) + g4*sind(xi3)/(g3*sind(alpha)*cosd(xi4))) -
dw*g2/(B*g1*cosd(alpha)*cosd(xi3))*(cosd(xi4)
113. *sind(xi5)/sind(theta) + g4*sind(xi3)/(g3*sind(alpha))) +
B*cosd(alpha)/dw *(g2/g1*cosd(alpha)*tand(xi4)
*sind(xi5)/cosd(xi3)* (1-cosd(alpha)*sind(xi4)/sind(theta))-
sind(alpha));
114. end
115. Nrsub=reff*Nr/r+dw/B*(1-reff/r)*Nrw;
116. elseif dw>hD;
117. Nrsub=Nr;
fprintf ('Note: Ground Water Table will not have any effect on
118.
Ultimate Bearing Capacity. \n')
119. end
120. qu=cj*Ncj+ci*Nci+q*Nq+0.5*r*B*Nrsub;
121. qu=qu/1000;
Page | 31

122. Friction of Intact Rock, \n %4.1f kPa Cohesion and %5.2f degree Angle
of Friction of Jointed Rock,\n %5.2f kN/m^3 Bulk Unit weight, %5.2f
kN/m^3 Saturated Unit weight, %3.1f m Water Table from GL.\n',
qu,B,q,alpha,ci,phii,cj,phij,r,rsat,dw)
123. end
124. end
125. end
126. end
4.1.3 IMPLEMENTATION DETAILS
It is already mentioned that by using the above algorithmic MATLAB program, with
known footing width (B), orientation angle (α), rock mass strength properties (ci, cj, ϕi, ϕj),
physical properties (γ, γsat), water table location (dw), surcharge (q) ultimate bearing capacity
of rock mass can be determined in MATLAB software. As the algorithmic model is composed
of MATLAB language, it is quite hard to understand it without knowing MATLAB language.
Therefore the complete process of the model execution or the statements used in model are
discussed below:
Step 1: First three lines are the comment lines that describe about the model, i.e. model purpose,
Model input parameters and Model output parameter. These comment lines are not
mandatory as they never execute, but it is often necessary to provide information about
the algorithm to the user. These comment lines can be viewed in command window
with the help of ‘help’ command. Suppose the model is saved as ‘ModelBCR’ (BCR
stands for Bearing Capacity of Rock), then whenever we execute ‘help ModelBCR’
command, those three lines will be displayed. Fig. 4.2 shows a view of displaying
comment lines of ModelBCR in MATLAB software. Next three lines, i.e. statement
lines from 4th to 6th formats the command window for display appearance. For a clear
and safe run or execution, display should be blank and there should not be any pre-
assigned values to any variable. Statement lines 7th and 8th provide basic information
for further execution of model to the user.
Page | 32
Fig. 4.2: MATLAB Command Window showing comment lines of ModelBCR
Step 2: Statement lines from 10th to 17th are the statements for input parameter. In this step, the
input parameters B, q, α, ci, cj, ϕi, ϕj and γ will be asked to insert for further operations.
Step 3: Statement lines from 18th to 25th specify the Ground Water Condition. It will be asked
to specify the ground water condition, whether it is ‘DRY’ or ‘SUBMERGED’. This
step contains a simple conditional program such as if it is specified the condition as
‘DRY’, statement lines from 20th to 25th will be skipped. On the other hand, for
‘SUBMERGED’ condition statement lines from 20th to 25th will be executed and the
input parameters γsat, dw will be asked to insert.
Step 4: Statement lines from 26th to 29th contains a simple program with conditional statement.
Program collects information as mentioned in step 2, more precisely, value of α from
user input and specifies the mechanism condition whether ‘two sided’ or ‘one sided’
for further execution.
Step 5: Statement lines from 32nd to 40th comprises of theorem equations to calculate θ, ξ1, ξ2,
ξ3, ξ4, ξ5, ξ6.
Step 6: Statement line 41st collects information as mentioned in step 2 that B is equal to zero
or negative or positive. If B is either negative or zero then a message will be displayed
“Foundation Width can be neither zero nor negative. Please try again with correction
of Foundation Width.” and further operations will be skipped to the end. For positive
values of B, execution of 42nd line will be skipped and will be proceed to the next step.
Step 7: Statement line 43rd collects information as mentioned in step 2 that γ is equal to zero or
negative or positive. If γ is either negative or zero then a message will be displayed
“Unit Weight can be neither zero nor negative. Please try again with correction of Unit
Weight.” and further operations will be skipped to the end. For positive values of γ,
execution of 44th line will be skipped and will be proceed to the next step.
Page | 33
Step 8: Statement line 45th collects information as mentioned in step 2 that α is equal to zero or
negative or positive. If α is either negative or zero then the program will display a
message will be displayed “Orientation Angle can be neither zero nor negative. Please
try again with correction of Orientation Angle.” and further operations will be skipped
to the end. For positive values of α, execution of 46th line will be skipped and will be
proceed to the next step.
Step 9: Statement lines from 48th to 126th will be executed only when the conditions in step 6,
step 7 and step 8 are satisfied i.e. for positive values of B, γ and α. These lines contain
the main calculation part for ultimate bearing capacity determination. There are several
nested program within this step. As the execution proceeds with positive values of B, γ
and α, 48th line statement will collect information about mechanism condition as
mentioned in step 4 and thereafter it is decided, the commands under decided
mechanism will be executed i.e. statement lines either from 49th to 86th or from 87th to
125th. It should be noted that statement lines either from 49th to 86th or from 87th to
125th, algorithm structure is same, but theorem equations or conditions are different.
Statement lines from 49th to 86th comprises of theorem equations or conditions for Two-
sided mechanism. On the other hand, statement lines from 87th to 125th comprises of
theorem equations or conditions for One-sided mechanism. Suppose α = 450, which will
support two-sided mechanism and therefore statement lines from 49th to 86th will
execute.
Statement line 50th will verify the constrain condition of two sided theorem with input
data. If the constrain condition is satisfied then 51st line will be skipped and further
operations of calculating f1, f2, f3, f4, Nci, Ncj, Nq, Nγ will be proceed, otherwise the
computation will be ended with a display message “Do not satisfy the constrain
condition”.
After calculation of f1, f2, f3, f4, Nci, Ncj, Nq, Nγ further computations will be carried out
by statement lines from 61st to 85th under nested program for ground water condition.
If user mentioned ‘DRY’ condition in Step 3 then statement lines from 63rd to 65th will
be executed and then ultimate bearing capacity will be calculated and further operations
will be terminated.
But if user mentioned ‘SUBMERGED’ condition in Step 3 then statement lines from
63rd to 65th will be skipped and statement lines from 67th to 84th will be executed. It
Page | 34
should be noted that in most calculations, unit weight of water is generally taken as
either 9.81 kN/m3 or 10kN/m3, but for more perfection in result of ultimate bearing
capacity, unit weight of water is taken as 9.807kN/m3. In ‘SUBMERGED’ condition,
firstly, γeff, hC, hD will be calculated by statement lines from 67th to 70th and then proceed
to nested program for different location of ground water.
Statement line 71st will decide whether the water table is above the critical point or
below the critical point. It is obvious that if location of water table is below the critical
point then ground water will have no effect on bearing capacity. If the conditional
statement in statement line 71st confirms that water table is below the critical point, i.e.
dw>hD, execution of statement lines from 72nd to 77th will be skipped and operations
will be carried out as specified in statement lines 79th and 80th, which will add a display
message ‘Note: Ground Water Table will not have any effect on Ultimate Bearing
Capacity.’. On the other hand, if water table is above or just at critical point i.e. dw≤hD
e.g. in case of currently running problem, statement lines from 72nd to 77th will be
executed and 79th and 80th line will be skipped. As the model proceeds to line 72nd,
nested program of calculation of Nγw for different ground water condition will execute.
If the conditional statement in 72nd line is satisfied then 73rd line will be executed and
execution of 75th line will be skipped. On the other hand, if the conditional statement
in 72nd line is not satisfied then execution of 73rd line will be skipped and with
compulsion, model will satisfy the condition in 74th and 75th statement lines will be
executed. After calculation of Nγw for either condition between two mentioned
conditions, 77th line statement will be executed and thereafter Nγsub will be calculated.
After calculation of Nγsub for ‘SUBMERGED’ condition, ultimate bearing capacity will
be calculated by executing the command lines 82nd and 83rd. Statement line 84th will
display the final output i.e. ultimate bearing capacity of rock foundation with the
specified input parameters.
As it is already mentioned in 1st paragraph of step 9, this chapter, same subsection, that
algorithm structure is same for statement lines from 49th to 86th or for statement lines
from 87th to 125th. Therefore when the orientation angle is other than 450 or α≠450, it
will follow one-sided mechanism and therefore statement lines from 87th to 125th will
execute. These statement lines will follow the same procedure as followed by statement
lines from 49th to 86th, but each theorem calculations and conditions are a bit different,
Page | 35
such as constrain condition is different, calculations of g1, g2, g3, g4 will be carried out
instead of f1, f2, f3, f4.
4.1.4 EXECUTION STEPS
Before execution, first of all, the complete algorithm must be saved in a MATLAB
script file. This task can be easily done by choosing ‘Save As’ option from ‘EDITOR’ tab in
MATLAB or simply and commonly by pressing ‘Ctrl+S’ key, inbuilt shortcut key. It should
be ensured that the file name with which the file is saved must satisfy the MATLAB conditions
(must begin with a letter, can include digits and underscore but no spaces, and up to 63
characters long and others) for writing a file name. It is very essential that user must check
whether the file is located in ‘Current Folder’ or not. If not, then either the MATLAB file
directory have to change to the location where the file is saved or the file have to be moved to
directory location. After completing the above steps user may execute the complete model by
choosing ‘Run’ option from ‘Editor’ tab or by typing the file name in command window or by
simply pressing ‘F5’ key, inbuilt shortcut key. As the model run user must proceed the
following steps:
Fig. 4.3: MATLAB Command Window, asking input of foundation width.
Step 1: In command window three lines will appear. First line describes the model purpose,
second line notifies to put the input parameters and third line asks to insert input of
‘Width of Foundation (in m)’. Fig. 4.3 shows a view of ‘MATLAB Command Window’
of first step execution. Suppose width of foundation is 1m, as stated in 2nd line, only
magnitude have to be entered, therefore input will be 1.
Page | 36
Fig. 4.4: MATLAB Command Window, asking input of surcharge magnitude.
Step 2: After entering the input value of foundation width, next line appears asking to insert
input of ‘Surcharge magnitude’ (in kPa). Fig. 4.4 shows a view of ‘MATLAB
Command Window’ of second step execution. Suppose surcharge magnitude is 20kPa,
then input will be 20.
Fig. 4.5: MATLAB Command Window, asking input of orientation angle.
Step 3: After entering the input value of surcharge magnitude, next line appears asking to insert
input of ‘Orientation Angle' (in degree). Fig. 4.4 shows a view of ‘MATLAB Command
Window’ of third step execution. Suppose orientation angle is 450, then input will be
45.
Page | 37
Fig. 4.6: MATLAB Command Window, asking input of cohesion of intact rock.
Step 4: After entering the input value of orientation angle, next line appears asking to insert
input of ‘Cohesion of Intact rock’ (in kPa). Fig. 4.6 shows a view of ‘MATLAB
Command Window’ of fourth step execution. Suppose cohesion of intact rock is 5MPa,
as the input is asked to enter in kPa unit therefore input will be 5000.
Fig. 4.7: MATLAB Command Window, asking input of cohesion of jointed rock.
Step 5: After entering the input value of cohesion of intact rock, next line appears asking to
insert input of ‘Cohesion of Jointed Rock’ (in kPa). Fig. 4.7 shows a view of ‘MATLAB
Command Window’ of fifth step execution. Suppose cohesion of jointed rock is 50kPa,
then input will be 50.
Page | 38
Fig. 4.8: MATLAB Command Window, asking input of frictional angle of intact rock
Step 6: After entering the input value of cohesion of jointed rock, next line appears asking to
insert input of ‘Angle of Friction of Intact Rock’ (in degree). Fig. 4.8 shows a view of
‘MATLAB Command Window’ of sixth step execution. Suppose frictional angle of
intact rock is 350, then input will be 35.
Fig. 4.9: MATLAB Command Window, asking input of frictional angle of jointed rock
Step 7: After entering the input value of angle of friction of intact rock, next line appears asking
to insert input of ‘Angle of Friction of Jointed Rock’ (in degree). Fig. 4.9 shows a view
of ‘MATLAB Command Window’ of seventh step execution. Suppose frictional angle
of jointed rock is 350, then input will be 35.
Page | 39
Fig. 4.10: MATLAB Command Window, asking input of Bulk Unit Weight
Step 8: After entering the input value of angle of friction of jointed rock, next line appears
asking to insert input of ‘Bulk Unit Weight’ (in kN/m^3). Fig. 4.10 shows a view of
‘MATLAB Command Window’ of eighth step execution. Suppose bulk unit weight of
rock mass is 27kN/m3, then input will be 27.
Fig. 4.11: MATLAB Command Window, asking input of Ground water condition
Step 9a: After entering the input value of bulk unit weight, next two lines appear asking to
insert input of ‘Ground Water Condition’ (DRY or SUBMERGED) with a guide
message which indicates that input have to type in capital letters. Fig. 4.11 shows a
view of ‘MATLAB Command Window’ of ninth step execution. Suppose there is no
water table i.e. ground water condition is dry, then input will be DRY. Fig. 4.12 shows
a view of ‘MATLAB Command Window’ at the verge of completion of ninth step.
Page | 40
Fig. 4.12: MATLAB Command Window, inserting input as dry condition in Ground
water condition
Step 10a: After entering the input value of ground water condition, output will be displayed.
First line will show the ultimate bearing capacity of rock foundation, and the next
lines will show the input parameters for which bearing capacity is determined. Fig.
4.13 shows a view of ‘MATLAB Command Window’ displaying output for dry
condition.
Fig. 4.13: MATLAB Command Window, showing output with its inputs for a simple
problem of rock foundation without submergence.
Therefore, for the case of 1 m foundation width, 20 kPa surcharge, 450 orientation
angle, 5 MPa cohesion of intact rock, 50 kPa cohesion of jointed rock, 350 angle of friction for
both intact and jointed rock, 27 kN/m3 bulk unit weight, the model for determination of ultimate
bearing capacity yields ultimate bearing capacity 210.833 MPa. The same result obtained by
Imani et al. (2012).
Page | 41
Variables addressed in the process of execution were created and stored in workspace.
MATLAB workspace provides complete information about the variables i.e. variable name,
value, size, minimum and maximum values, statistical descriptors e.g. mean, median, mode,
standard deviation and many others. Fig. 4.14 shows a view of MATLAB workspace after
completion of execution of Algorithmic Model for dry condition.
Fig. 4.14: MATLAB workspace showing variables for the stated problem
(Dry Condition)
Page | 42
In many practical cases, presence of ground water may found. For submerged condition,
the execution process is slightly different. First eight steps are same as dry condition, but from
ninth step, execution process will be different. Therefore, ninth and tenth step are discussed in
two different manners which are distinguished by 9a and 9b or 10a and 10b. Steps 9a and 10a
are for dry condition while steps 9b and 10b are for submerged condition. Suppose input
parameters up to eight step are same, but from ninth step inputs are different due to submerged
condition. Therefore, execution from ninth step for submerged condition is discussed below:
Step 9b: After entering the input value of bulk unit weight, next two lines will be appeared
asking to insert input of ‘Ground Water Condition’ (DRY or SUBMERGED) with a guide
message which indicates that input have to type in capital letters. Fig. 4.15 shows a view of
‘MATLAB Command Window’ after ninth step execution. Suppose for currently running
problem, there is availability of water table i.e. ground water condition is submerged, in that
case, input will be SUBMERGED.
Fig. 4.15: MATLAB Command Window, asking input of saturated unit weight.
Step 10b: After entering the input of ground water condition as submerged, next line appears
asking to insert input of ‘Saturated Unit Weight of Submerged Rock’ (in kN/m^3).
Fig. 4.15 shows a view of ‘MATLAB Command Window’ of tenth (b) step
execution. Suppose saturated unit weight is 30kN/m3, then input will be 30.
Page | 43
Fig. 4.16: MATLAB Command Window, asking input of depth of water.
Step 11: After entering the input of saturated unit weight, next two lines will be appeared asking
to insert input of ‘Depth of water’ (in m) with a guide message which indicates that
water table location should have to be measured from ground level. Fig. 4.16 shows a
view of ‘MATLAB Command Window’ of eleventh step execution. Suppose water
table is at depth of 0.4m from ground level, then input will be 0.4.
Fig. 4.17: MATLAB Command Window, showing output with its inputs for a simple
problem of rock foundation with submergence.
Step 12: After entering the input value of water depth, output will be displayed. First line will
show the ultimate bearing capacity of rock foundation, and the next lines will show
the input parameters for which bearing capacity is determined. Fig. 4.17 shows a view
of ‘MATLAB Command Window’ displaying output for submerged condition.
Page | 44
Therefore, for the case of 1 m foundation width, 20 kPa surcharge, 45 0 orientation

angle, 5 MPa cohesion of intact rock, 50 kPa cohesion of jointed rock, 350 angle of friction for
both intact and jointed rock, 27 kN/m3 bulk unit weight, 30 kN/m3 saturated unit weight and
ground water at 40 cm from ground level, the model for determination of ultimate bearing
capacity evaluates ultimate bearing capacity as 209.580 MPa.
As like dry condition, for submerged condition also variables addressed in the process
of execution were created and stored in workspace. But this time, ground water condition is
replaced by submerged condition, and variables associated with submerged condition is newly
created and stored in MATLAB workspace. Fig. 4.18 shows a view of undocked MATLAB
workspace window after completion of execution of Algorithmic Model for submerged
condition.
It has been observed that bearing capacity for dry condition is higher than the bearing
capacity for submerged condition. Therefore it can concluded that due to the effect of
submergence, bearing capacity is slightly reduced.
Page | 45
Fig. 4.18: MATLAB workspace showing variables for the stated problem
(Submerged Condition)
Page | 46
4.2 EFFECT OF SUBMERGENCE
It is obvious that in presence of water, ultimate bearing capacity will be reduced. But it
is quite difficult to say the behavior of reduction i.e. reduction is rapidly growing or gradually
growing or it is of uneven nature. Therefore, a concise study on the effect of submergence is
carried out to observe the behavior of reduction of ultimate bearing capacity in presence of
water. It should be noted that, in this study, water depth is measured from ground level.
4.2.1 ANALYSIS PROCESS
With the help of the algorithmic model presented before, ultimate bearing capacity for
different ground water depth can be evaluated. The same set of data analyzed before are used
with variation range of ground water depth from 0.0 m to 4.0 m for analysis of effect of
submergence in ultimate bearing capacity. Table 4.1 shows the data set for analysis of effect
of submergence.
Table 4.1: Data set for analysis of effect of submergence
VALUE
PARAMETER
Minimum Maximum
Width of Foundation, B (m) 1
Surcharge Magnitude, q (kPa) 20
Cohesion of Intact Rock, ci (kPa) 5000
Cohesion of Jointed Rock, cj (kPa) 50
Angle of Friction of Intact Rock, φi (degree) 35
Angle of Friction of Jointed Rock, φj (degree) 35
Orientation Angle, α (degree) 45
Bulk Unit Weight, γ (kN/m3) 27
Saturated Unit Weight, γsat (kN/m3) 30
Water Table from ground, dw (m) 0.0 4.0
For better analysis, large number of values have been generated and for that, variation
scale has been taken with very little difference, therefore, 0.01 m or 1 cm variation scale of
ground water depth has been taken into account. Large number of result set has been evaluated,
i.e. 401 sets of result comprising of ultimate bearing capacity for each different ground water
depth. Graphical representation may be carried out in MATLAB software. But for better
visualization, graphical representation is carried out in Microsoft excel spreadsheet. Data
Page | 47
generated in MATLAB workspace is sorted in a tabulation form and then exported to excel
spreadsheet.
4.2.2 GRAPHICAL REPRESENTATION
Two sets of column comprising of ground water depth and ultimate bearing capacity
are represented graphically in the form of scatter graph with ground water depth as horizontal
axis and ultimate bearing capacity as vertical axis. Additionally, critical point for each bearing
capacity is located as critical line which intersect the ultimate bearing capacity curve. Value of
intersecting point is shown nearby vertical axis, pointed by a line created and extended from
the point of intersection. A trend line is then created to represent the nearest shape of the curve
showing its polynomial equation along with R-squared value. Plot area is filled in gradient
form to observe the presence of water table and gradual reduction of the same. Fig. 4.19
represents the behavior of ultimate bearing capacity with gradual reduction of water table.
Fig. 4.19: Behavior of Ultimate Bearing Capacity with Submergence.
4.2.3 OBSERVATION
From the graphical representation, it is clearly observed that ultimate bearing capacity
changes for the same set of parameters i.e. for same foundation width, surcharge magnitude,
Page | 48
orientation angle, strength and physical properties of rock mass, where water table depth is
unstable. From fig. 19 it is observed that the ultimate bearing capacity is gradually increasing
with the gradual reduction of water table depth. Technically, increment curvature is not purely
linear, firstly at initial state, i.e. at high level of water it shows linear behavior with 450 gradient,
after which it reaches transition state i.e. water table is about to vanish, it shows nonlinear
behavior in form of exponentiation and finally at perfectly dry state, it shows a completely
linear straight line with no gradient from which it can be concluded that further reduction of
water table depth will not have any effect on ultimate bearing capacity.
Therefore, it can be concluded that for reduction of water table depth, keeping other
factors constant, ultimate bearing capacity of rock mass will be constantly increased up to the
critical point and after that water table will not have any effect on ultimate bearing capacity.
4.3 COMPARISON BETWEEN MECHANISMS
For determination of ultimate bearing capacity, Imani et al. (2012) suggested two
different failure mechanism depending upon joint sets, which were already discussed in chapter
3, subsection ‘Failure Mechanism’. Condition for choosing each mechanism mainly depends
upon the orientation angle of joint sets. Assumption made for failure mechanism clearly
indicates that if orientation angle is 450, failure mechanism will be ‘Two-Sided Failure
Mechanism’ otherwise it will be considered as ‘One-Sided Failure Mechanism’. Suggested
theorem by Imani et al. (2012) for determination of ultimate bearing capacity for submerged
rock foundation comprises of 32 equations i.e. from eq. 1 to eq. 32 in chapter 3, subsection
‘Calculations’, out of which 12 equations are common to both failure mechanisms i.e. two
sided failure mechanism and one sided failure mechanism, remaining 20 equations are divided
into two categories i.e. 10 equations are for each mechanism. Considering all those equations
all at a time, makes the computational approach very time consuming and tedious. Hence, an
attempt has been made to study regarding the comparison between one sided failure mechanism
and two sided failure mechanism. It should be noted that the comparison is done by keeping
orientation angle constant at 450, whereas the other parameters involved in determination of
bearing capacity varies individually.
4.3.1 STOCHASTIC MODEL CREATION
For comparison of failure mechanisms, ultimate bearing capacity is calculated for both
one sided and two sided mechanism for orientation angle, 450. For determination of ultimate
Page | 49
bearing capacity, 10 parameters are involved, wherein, only orientation angle is constant i.e.
450. For an ease of calculation, foundation width (B) and surcharge magnitude (q) have been
assumed constant, such as 1 m foundation width and 20 kPa surcharge. For deterministic
analysis, i.e. out of 7 variable parameters, considering 6 as constant and remaining one varies
and repeating the same, computational approach and observation of comparison will create
confusion. Therefore, rather than considering a deterministic analysis, a stochastic approach is
carried out, where all these parameters vary simultaneously with random values within a
specified range. Hence, for randomly varying parameter, coefficient of variation is taken as 5%
for strength parameters, 2% for physical properties, and 100% for water table depth into the
nominal values used by Imani et al. (2012). Data set used for this approach is enlisted in table
4.2.
Table 4.2: Data set used for comparison between one side & two sided failure mechanisms
VALUE
PARAMETER
Minimum Maximum
Orientation Angle, α (degree) 45
Width of Foundation, B (m) 1
Surcharge Magnitude, q (kPa) 20
Cohesion of Intact Rock, ci (kPa) 4750 5250
Cohesion of Jointed Rock, cj (kPa) 47.50 52.50
Angle of Friction of Intact Rock, φi (degree) 33.25 36.75
Angle of Friction of Jointed Rock, φj (degree) 33.25 36.75
Bulk Unit Weight, γ (kN/m3) 26.46 27.54
Saturated Unit Weight, γ (kN/m3) 29.40 30.60
Water Table from ground, dw (m) 0 4.00
For stochastic approach, a stochastic model has been developed in Microsoft excel
spreadsheet. Stochastic model is so called due to the fact of random value generation.
Thereafter, theorem equations are formulated in spreadsheet with the help of pre-assigned
formulae of excel. Then 5000 rows have been developed such that each row formulated with
the same set of equations. It should be noted that every single trail or run will give different set
of outputs but graphical representation will be same. Presented study is based on only one trail,
but it has been observed that in this case, each trail will evaluate the same. Calculated ultimate
bearing capacity varies from 146.83 MPa, lower bound bearing capacity to 346.08 MPa, upper
Page | 50
bound bearing capacity. But for present case, stochastic model evaluates bearing capacity
ranges from 150.41 MPa to 338.10 MPa, within the range of lower to upper bound bearing
capacity. Excel formulae used for this approach are enlisted in Appendix-IV.
4.3.2 GRAPHICAL REPRESENTATION
Two sets of array (column) comprising of ultimate bearing capacity for one sided
mechanism and ultimate bearing capacity for two sided mechanism are represented graphically
in the form of scatted graph with ultimate bearing capacity for two sided failure mechanism as
horizontal axis and ultimate bearing capacity for one sided failure mechanism as vertical axis.
A trend line is then created to represent the closest shape of plotted graph showing its equation
along with R-squared value. Fig. 4.20 represents the comparison graph of ultimate bearing
capacity between one sided and two sided failure mechanism.
Fig. 4.20: Comparison graph of failure mechanism (Two sided & One Sided)
Page | 51
4.3.3 OBSERVATION
From the graphical representation, it is clearly observed that the ultimate bearing
capacity for one sided failure mechanism and two sided failure mechanism are exactly same
for uneven variation of theorem parameters keeping orientation angle constant at 450.
Technically, presented graph shows a purely linear behavior with 450 gradient which implies
that plotting points have the same axis coordinate. Therefore, theorem equations for two sided
failure mechanism can be eliminated for computational approach. In other words, out of 32
theorem equations, 10 equations can be eliminated i.e. eq. 10, 11, 12, 13, 18, 19, 20, 21, 29 and
30 (in chapter 3, subsection ‘Calculations’) and ultimate bearing capacity can be determined
from rest 22 theorem equations. Additionally, the theorem assumption 6 mentioned in chapter
3, subsection ‘Assumptions’ can be eliminated, as for all condition of orientation angle,
ultimate bearing capacity can be determined considering one sided failure mechanism.
Therefore, it can be concluded that the ultimate bearing capacity for submerged rock
foundation can be determined by considering only one sided failure mechanism independently.
4.4 RELIABILITY ANALYSIS
In most of the cases, rock substratum is assumed to be homogeneous, intact for

simplicity in analysis. But in practical cases, rock substratum generally occurs with non-
homogeneous nature with joint sets. Therefore, in-situ rock mass variability provides that
deterministic analysis to be inefficient. Reliability analysis can be used to observe the
performance and reliability of rock mechanics problem and also can be used for risk based
decision making. Hence, a probabilistic or reliability analysis is needed to examine the
uncertainties of rock mechanics problem. Therefore, an attempt has been made to model the
uncertainties by reliability analysis. Theorem of determination of ultimate bearing capacity for
submerged rock foundation consists of several complex equations, for which statistical
calculations and reliability index determination are hard enough to compute. Among all the
reliability method, Monte Carlo method is selected due to minimize computation effort and
most prominently statistical calculations and reliability index can be directly computed in
Monte Carlo method.
4.4.1 MONTE CARLO SIMULATION
Statistical calculations and reliability index determination by Monte Carlo method has
been carried out in Microsoft excel spreadsheet. Excel formulae used for this simulation are
Page | 52
enlisted in Appendix IV. Step by step procedure for this analysis developed in excel
spreadsheet is described below:
Step 1: First of all a model capable of calculating deterministic output, lower and upper bound
output along with stochastic output is created. Stochastic formulation has been
developed similarly as mentioned in sub-section ‘Stochastic model creation’ in this
chapter. Theorem equations are formulated for deterministic calculations as well as
stochastic calculations to compare the outputs i.e. ultimate bearing capacities. Then
theorem parameters foundation width and surcharge magnitude has been kept constant
for all sets of calculations. Deterministic calculation set is carried out with nominal
values used by Imani et al. (2012). Lower and Upper bound calculation sets are
developed with minimum and maximum values for each parameter by formulating
different coefficient of variation to each parameter. Thereafter with the help of random
function pre-assigned in excel, stochastic calculation set is carried out with normally
distributed random values ranging from minimum to maximum as formulated for lower
and upper bound calculation respectively. It should be noted that every single run or
trail will generate random input values within the specified range and evaluates output
within the range of lower and upper bound outputs. Additionally, constrain condition
and groundwater condition are also formulated to provide the information about
whether the calculated ultimate bearing capacity satisfies the constrain condition or not
and it is under dry or submerged condition. Cells formulated for deterministic approach
are filled with green color, whereas for stochastic approach cells are filled with blue
color for comparative look. A view of excel spreadsheet containing stochastic model
for Monte Carlo Simulation and dataset used for the same is represented in fig. 4.21.
Step 2: Now, theorem equations are formulated again in one single row in another five
spreadsheet and then 100, 500, 1000, 5000 and 10000 arrays have been developed
accordingly such that each array formulated with the same set of equations. That means
each row will calculate different ultimate bearing capacity for randomly generated
values for each parameter. It is already mentioned that re-running the simulation will
generate different set of output as inputs. After completing the simulation, reliability
index and probability of failure calculations are carried out. Reliability index values
and probability of failure values for 25 trails of 100, 500, 1000, 5000 and 10000 samples
are stored and then plotted in graph along with mean and medians of each 25 trails.
Thereafter, variation of reliability index and probability of failure with foundation
Page | 53
width and different coefficient of variation are carried out. Graphical representation and
results obtained in this step are enlisted in the following subsections in this chapter.
Fig. 4.21: Excel spreadsheet formulated for Monte Carlo Simulation.
Step 3: From output column, i.e. column for ultimate bearing capacity, minimum and maximum
bearing capacity are evaluated. For creating a histogram chart, an array of bins is
developed within a range of minimum to maximum bearing capacity with 40 evenly
spaced numbers. Next, an array of count is developed with the help of frequency
function. This array represents the number of counts for each corresponding bin. Now,
an array for scaled histogram is created to represent probability distribution such that
area under the curve is equal to 1. Lastly, another array is added to represent cumulative
probability with respect to bins.
Histogram chart provides a visual representation, for which it is very essential to plot
histogram chart throughout statistical analysis. To create a histogram chart, first of all
a ‘column chart’ has been created with the arrays of bins and counts. Then selected the
chart area which will active two new tabs in Menu bar under ‘Chart Tools’, then clicked
‘Select Data’ nested under ‘Design’ tab, due to that, a new window ‘Select Data Source’
Page | 54
will be popped up. Now ‘Legend Entries’ are edited by putting count array in ‘Series
1’ and cumulative probability array in ‘Series 2’ and ‘Horizontal Axis’ is edited by
putting bins array values. Again selected the chart area and clicked ‘Change Chart
Type’ nested under ‘Design’ tab, due to that, a new window ‘Change Chart Type’ will
be popped up, then clicked ‘Combo’ tab and then selected ‘Clustered Column – Line
on Secondary Axis’. With this, the critical part of plotting Histogram chart is ended.
Now the columns are filled with ‘violet’ color and cumulative probability line is filled
with ‘green’ color and accordingly chart area and axis labels are also modified for better
visualization. Furthermore to compare histogram with probability distribution, a scaled
histogram is developed with bins as horizontal axis and frequency as vertical axis.
Step 4: From the output array of ultimate bearing capacity, various statistical functions are
calculated e.g. for central tendency - mean and median; for spread - standard deviation,
minimum, maximum, quartiles and ranges; for shape - skewness, kurtsis. Statistical
formulae used in this step are already discussed in subsection ‘Statistical terms’ in
chapter 3. Then percentiles for 90% interval and 95% interval are also formulated.
Results obtained in statistical analysis are enlisted in the next subsection ‘Result
Summary’ in this chapter.
Step 5: For calculating confidence interval, firstly, standard error is calculated and then upper
and lower confidence limits are calculated with normal distribution function. Normal
distribution functions determines inverse of the standard normal cumulative
distribution. The distribution has a mean of zero and a standard deviation of one.
Various analysis starting with variation of counts and cumulative probability with bins
due to different sample size, variation of reliability index and probability of failure with sample
size, statistical analysis e.g. central tendency, spread, shape and many others are carried out.
4.4.1.1 Effect of Sample Size
Regarding Monte Carlo simulation, it is very important to study the variation behavior
of output i.e. ultimate bearing capacity due to the sample size. Therefore a study for variation
behavior with increment of sample size is carried out.
4.4.1.1.1 Analysis Process
As mentioned in step 3, subsection ‘Monte Carlo Simulation’ different arrays

comprising of bins, counts, frequency, cumulative probability are developed. Thereafter
Page | 55
histogram charts along with scaled histograms with 40 equally spaced bins are developed for
different sample size. 100, 500, 1000, 5000 and 10000 samples are used for this analysis.
Dataset used for analysis for Monte Carlo Simulation is given in table 4.3.
Table 4.3: Data set used for Monte Carlo Simulation

Coefficient Deter- Lower Upper
Parameters of ministic Bound Bound
variation values values values
Width of Foundation, (m) 0% 1
Surcharge magnitude (kPa) 0% 20
Cohesion of Intact Rock (kPa) 5% 5000 4750 5250
Cohesion of Jointed Rock (kPa) 5% 50 47.5 52.5
Frictional angle of Intact Rock (degree) 5% 35 33.25 36.75
Frictional angle of Jointed Rock (degree) 5% 35 33.25 36.75
Orientation Angle (degree) 5% 45 42.75 47.75
Bulk Unit Weight (kN/m3) 2% 27 26.46 27.54
Saturated Unit Weight (kN/m3) 2% 30 29.4 30.6
Water Table from ground (m) 100% 2 0 4
4.4.1.1.2: Graphical representation

Histogram charts along with scaled histograms developed for 100, 500, 1000, 5000 and
10000 samples size are represented from fig. 22 to fig. 31.
Fig. 4.22: Histogram of 100 samples Fig. 4.23: Scaled Histogram of 100 samples
Page | 56
Page | 57
4.4.1.1.3 Observation
From fig. 22 to fig. 31, it is clearly observed that small sample size gives totally uneven
distributions for counts, frequency as well as cumulative probability. As the sample size
increases distributions for counts, frequency and cumulative probability change from fairly
disturbed nature to evenly distributed nature. Fig. 28 to fig. 31 represents a fair variation of
counts and probability distributions but fig. 30 and fig. 31 represented for 10000 samples shows
quite better distribution behavior than fig. 28 and fig. 29.
Therefore it is concluded that a higher sample size should be used for Monte Carlo
Simulation such as 5000 or 10000 if possible. And hence 10000 samples are used for reliability
index determination, probability of failure determination along with statistical analysis.
4.4.1.2 Determination of Reliability Index
Reliability index, β can be directly computed from 10000 sample size although it is
important to study the variation behavior of reliability index with variation of sample size.
Therefore to study this behavior following study has been carried out.
Firstly reliability index is calculated for 100, 500, 1000, 5000 and 10000 samples.
Thereafter, reliability index values obtained from 25 trails of 100, 500, 1000, 5000 and 10000
samples are stored. From stored data, minimum and maximum values of 25 trails for each
sample size are calculated and then variation range is formulated to observe variation behavior
due to sample size. To observe central tendency behavior mean and median of 25 trails for each
sample size are also formulated. Table 4.4 presents the data set obtained from above
formulations.
Table 4.4: Data set used for Reliability Index (β) determination
Sample Minimum Maximum Variation Median Mean

Number (β) (β) Range (β) (β)
100 4.637 5.912 1.275 5.273 5.224
500 4.775 5.618 0.842 5.122 5.117
1000 4.937 5.317 0.380 5.140 5.130
5000 4.996 5.209 0.214 5.126 5.120
10000 5.064 5.181 0.117 5.130 5.131
Page | 58
4.4.1.2.2 Graphical Representation
A scatter graph is plotted with every values obtained from 25 trails for every sample
size as points and mean and median values are plotted as lines accordingly. To observe more
precisely, points are marked in blue color whereas lines for mean and median values are plotted
in red and green colors respectively.
Fig. 4.32: Reliability index with variation of sample size
From fig. 4.32 it is observed that reliability index, β varied within a high range for
sample size 100. Furthermore, central tendency behavior for sample size 100 is also poor as
mean and median points are distinctly different. As the sample size increases variation range
of reliability index decreases. Mean and median values are approximately same when sample
size are equal or greater than 500. Most important point is, in case of sample size 10000,
variation occurrence from mean or median is less than 0.1. Hence, reliability index is computed
for 10000 sample size. From table 4.4, it is observed that reliability index varies from 5.064 to
5.181 and mean and median values are 5.131 and 5.130 respectively. Although mean and
median both represents central tendency, but for present case, median is more valuable than
mean as variation distribution is not studied. Hence, reliability index is computed to be the
same with median value of 25 trails for 10000 sample size i.e. 5.130.
Page | 59
Therefore, it is concluded that variation of reliability index decreases with increase of

sample size and hence higher sample size e.g. 10000 is better enough to compute reliability
index. Another conclusion is - reliability index, β is determined as 5.130.
4.4.1.3 Determination of Probability of failure
Probability of failure, Pf can be directly computed from 10000 sample size although it
is important to study the variation behavior of probability of failure with variation of sample
size. Therefore to study this behavior following study has been carried out.
Firstly probability of failure is calculated for 100, 500, 1000, 5000 and 10000 samples.
Thereafter, probability of failure values obtained from 25 trails of 100, 500, 1000, 5000 and
10000 samples are stored. From stored data, minimum and maximum values of 25 trails for
each sample size are calculated and then variation range is formulated to observe variation
behavior due to sample size. To observe central tendency behavior mean and median of 25
trails for each sample size are also formulated. Table 4.5 presents the data set obtained from
above formulations.
Table 4.5: Data set used for Probability of Failure (Pf) determination
Sample Minimum Maximum Variation Median Mean

Number (Pf) (Pf) Range (Pf) (Pf)
100 0.028 0.044 0.017 0.035 0.036
500 0.031 0.042 0.011 0.037 0.037
1000 0.034 0.039 0.005 0.037 0.037
5000 0.036 0.039 0.003 0.037 0.037
10000 0.036 0.038 0.002 0.037 0.037
4.4.1.3.2 Graphical Representation
A scatter graph is plotted with every values obtained from 25 trails for every sample
size as points and mean and median values are plotted as lines accordingly. To observe more
precisely, points are marked in blue color whereas lines for mean and median values are plotted
in red and green colors respectively.
Page | 60
Fig. 4.33: Probability of Failure with variation of sample size
From fig. 4.33 it is observed that probability of failure varied within a high range for
sample size 100. Furthermore, central tendency behavior for sample size 100 is also poor as
mean and median points are distinctly different. As the sample size increases variation range
of probability of failure decreases. Mean and median values are approximately same when
sample size are equal or greater than 5000. In the case of 10000 sample size, variation
occurrence from mean or median is less than 0.001. Hence, probability of failure is computed
for 10000 sample size. From table 4.5, it is observed that probability of failure varies from
0.036 to 0.038 and mean and median values are 0.037 and 0.037 respectively. As the mean and
median values are exactly same, hence probability of failure is computed to be the same with
either values of mean and median of 25 trails for 10000 sample size i.e. 0.037.
Therefore, it is concluded that variation of probability of failure decreases with increase

of sample size and hence higher sample size e.g. 10000 is better enough to compute probability
of failure. Another conclusion is - probability of failure is determined as 0.037.
4.4.1.4 Result Summary
Statistical calculations and Reliability calculations through Monte Carlo simulations

for 10000 samples are enlisted in table 4.8. It is already mentioned that Monte Carlo Simulation
Page | 61
is developed in stochastic model, therefore every single simulation will generate different
results.
Table 4.6: Various Results of Monte Carlo Simulation
Analysis Parameters / Terms Value

Central Tendency Mean 217.835
(location) Median 210.758
Standard Deviation 41.92
Minimum 135.13
Maximum 415.36
Spread Range 280.23
Quartile 3 242.33
Quartile 1 186.73
Interquartile Range 55.61
Skewness 0.830
Shape
Kurtsis 0.676
Q (0.05) 161.337
90 % interval
Q (0.95) 297.235
Q (0.025) 155.205
95 % interval
Quantiles, Percentiles, Q (0.975) 317.340
Q (/ 2) 155.205
Alpha,  = 0.05
Q (1 - / 2) 317.340
Standard Error 0.419
Reliability Index 5.130
Reliability Analysis
Probability of failure 0.037
Page | 62
CONCLUSION & FUTURE SCOPE
5. CONCLUSION AND FUTURE SCOPE
5.1 CONCLUSIONS
On the basis of present study the following conclusions are drawn-
1. For reduction of water table depth, keeping other factors constant, ultimate bearing
capacity of rock mass will be constantly increased up to the critical point and after that
water table will not have any effect on ultimate bearing capacity. Highest ultimate bearing
capacity occurs in critical point from where further reduction of water table will not have
any effect on it. Lowest ultimate bearing capacity occurs when water table is located at
ground level.
2. Ultimate bearing capacities for theorem mechanism two-sided and one-sided are exactly
same for joint orientation angle 450. That is ultimate bearing capacity for submerged rock
foundation can be determined by considering only one sided failure mechanism
independently. Theorem assumption made for joint orientation angle for selection of
mechanism can be eliminated.
3. As the sample size increases distributions for counts, frequency and cumulative
probability change from fairly disturbed nature to stable nature. Therefore, it is
recommended to use higher sample size as possible e.g. 10000.
4. Variation of reliability index decreases with increase of sample size and hence higher
sample size e.g. 10000 is recommended to compute reliability index. For present case,
reliability index, β is determined as 5.130.
5. Variation of probability of failure decreases with increase of sample size and hence higher
sample size e.g. 10000 is recommended to compute probability of failure. For present
case, probability of failure is determined as 0.037.
5.2 FUTURE SCOPES
On the basis of the present work it is observed that in the area of ultimate bearing capacity
determination of rock mass following studies are essential in future –
1. Theorem for determination of ultimate bearing capacity of submerged rock whose

strength may be described by non-linear failure criteria to be developed.
Page | 64
CONCLUSION & FUTURE SCOPE
2. In the present study, reliability analysis is based on normally distributed variable.

Comparative study to be conducted with log-normal distributed or beta distributed
variables.
Page | 65
APPENDIX - I
FLOWCHART SYMBOLS
Flowchart symbols used in the stated flowchart (Fig. 4.1) only are given below:
Flowchart Symbol Name Description
Process An operation or action step.
Terminator A start or stop point in a process.
Decision A question or branch in the process
Indicates data inputs and outputs to and from a

Data (I/O)
process
Indicates the direction of flow for materials and/or
Flow Line
information.
Page | 66
APPENDIX - II
MATLAB COMMANDS
MATLAB characters, commands implemented in the presented algorithm program only are
given below:
Arithmetic Operators
Character Description
+ Addition
- Subtraction
* Scalar and array Multiplication
/ Right division
= Assignment operator
() Parentheses
Relational and Logical operators
<= Less than or equal
& Logic AND (if both are true, result is true)
Order of precedence of Arithmetic, Relational and Logical operators
Precedence Order Operation

1 Parentheses
2 Exponentiation
3 Logical NOT
4 Multiplication, Division
5 Addition, Subtraction
6 Rational Operators
7 Logical AND
N.B. – 1. For nested parentheses, inner one will execute first than the others.
2. For operations having same precedence, expression will execute from left to right.
Page | 67
MATLAB character
' Single quote (creates string)
; Semicolon (suppresses display of output)
, Comma (separates array subscript, commands)
… Ellipsis (continuation of line)
% Percent (suppresses display of line)
Display formats and Managing commands
Command Description
format compact Eliminates empty lines
clc Clears Command Window
clear Removes all variables from memory
Trigonometric functions
Command Description
cosd (x) Cosine of x (x in degrees)
sind (x) Sine of x (x in degrees)
tand (x) Tangent of x (x in degrees)
Input and Output commands
Command Description
disp Displays output directly
fprintf Displays output in a specified format
input Prompts for input to assign variable
\n Starts a new line
%x.y f Creation of fixed point notation after x spacing with y decimal
points.
Page | 68
Flow control commands
Command Description
case Conditionally execute specified commands
else Conditionally execute specified commands
elseif Conditionally execute specified commands
end Terminates conditional statements and loops
if Conditionally execute specified commands
switch Switches among several cases based on expression
Page | 69
APPENDIX - III
MATLAB PROGRAMS
MATLAB programs implemented in the stated algorithm program only are given below:
I. if-elseif-else-end structure
As the program executes, it reaches the if statement.
i. If the conditional expression is true, the program executes group 1 of commands between the
if and the elseif statements and then skips to the end.
ii. If the conditional expression in the if statement is false, the program skips to the elseif
statement. If the conditional expression in the elseif statement is true, the program executes
group 2 of commands between the elseif and the else and then skips to the end.
iii. If the conditional expression in the elseif statement is false, the program skips to the else
and executes group 3 of commands between the else and the end.
Fig.: Flow chart of if-elseif-end structure
Page | 70
II. switch-case statement
The first line is the switch command in the form of ‘switch switch expression’. The switch
command is followed by one or several case commands. Each has a value (can be a scalar or a
string) next to it (value1, value2, etc.) and an associated group of commands below it. After
the last case command there is an optional otherwise command followed by a group of
commands. The last line must be an end statement. After execution of program if there is more
than one match, only the first matching case is executed. If no match is found and the otherwise
statement (which is optional) is present, the group of commands between otherwise and end is
executed. If no match is found and the otherwise statement is not present, none of the command
groups is executed.
Fig.: Flow chart of switch-case statement
Page | 71
APPENDIX - IV
MICROSOFT EXCEL FORMULAE
Microsoft excel formulae used in the study of comparison between mechanisms and reliability
analysis only are given below:
Basic Math Functions

Function Description
+ Add numbers.
- Subtract numbers.
* Multiply numbers.
/ Divide numbers.
Math and Trigonometry Functions

Syntax Description
Returns the arctangent, or inverse tangent, of a number
ATAN(number)
(in radian).
COS(number) Returns the cosine of the given angle (in radian).
DEGREES(angle) Converts radians into degrees.
Returns e (=2.71828182845904) raised to the power of a
EXP(<number>)
given number.
SIN(number) Returns the sine of the given angle (in radian).
RADIANS(angle) Converts degrees to radians.
SQRT(number) Returns a positive square root.
Adds all the numbers specified as arguments. Each
SUM(number1,[number2],...]) argument can be a range, a cell reference, an array, a
constant, a formula, or the result from another function.
TAN(number) Returns the tangent of the given angle (in radian).
Logical Functions
Syntax Description
IF(logical_test,
Returns one value if a condition you specify evaluates to TRUE,
[value_if_true],
and another value if that condition evaluates to FALSE.
[value_if_false])
Page | 72
Statistical Functions
Syntax Description
Returns the average (arithmetic mean) of the
AVERAGE(number1,[number2], ...)
arguments.
Counts the number of cells that contain numbers,
COUNT(value1,[value2], ...)
and counts numbers within the list of arguments.
Calculates how often values occur within a range
FREQUENCY(data_array,bins_array) of values, and then returns a vertical array of
numbers.
MAX(number1,[number2] ,...) Returns the largest value in a set of values.
MEDIAN(number1,[number2], ...) Returns the median of the given numbers.
MIN(number1,[number2], ...) Returns the smallest number in a set of values.
Returns the kurtosis of a data set. Kurtosis
characterizes the relative peakedness or flatness
KURT(number1,[number2], ...)
of a distribution compared with the normal
distribution.
Returns the skewness of a distribution. Skewness
SKEW(number1,[number2], ...) characterizes the degree of asymmetry of a
distribution around its mean.
Compatibility Functions
Syntax Description
Returns the inverse of the standard normal cumulative
NORMSINV(probability) distribution. The distribution has a mean of zero and a
standard deviation of one.
PERCENTILE(array,k) Returns the k-th percentile of values in a range.
QUARTILE (array,quart) Returns the quartile of a data set.
STDEV(number1,[number2],...]) Estimates standard deviation based on a sample.
Page | 73
REFERENCES
Baecher G, Christian J (2003) ‘Reliability and Statistics in Geotechnical Engineering’, John

Wiley & Sons, Inc, New York, USA.
Chen W F, Drucker D C (1969) ‘Bearing capacity of concrete blocks or rock’. Journal of the
Engineering Mechanics Division. Proceedings of the American Society of Civil
Engineers 95(EM4). p. 955–79.
Das B M (2009) ‘Shallow Foundations Bearing Capacity and Settlement’, Second edition.
CRC Press, Taylor & Francis Group, USA.
Das B M (2011) ‘Principles of Foundation Engineering’, Seventh Edition. CENGAGE

Learning, USA.
Gilat A (2011) ‘MATLAB an Introduction with Applications’, Fourth edition, John Wiley &
Sons, Inc, New York, USA.
http://www.vertex42.com/ExcelArticles/mc/MonteCarloSimulation.html (Last checked:

21.05.2014).
http://en.wikipedia.org/wiki/Flowchart (Last checked: 21.05.2014)
Imani M, Fahimifar A, Sharifzadeh M (2012) ‘Upper Bound Solution for the Bearing Capacity
of Submerged Jointed Rock Foundations’. Rock Mech Rock Eng 45:639–646.
Kiureghian A D, Lin H Z, Hwang S J (1987) ‘Second-order reliability approximations.’

Journal of Engineering Mechanics, ASCE 113 (8): 1208–1225.
Matlab 8 (2012) Mathworks Inc., Natick, Massachusetts, USA.
Merifield RS, Lyamin AV and Sloan SW (2006) ‘Limit analysis solutions for the bearing
capacity of rock masses using the generalised Hoek–Brown criterion’. Int J Rock Mech
Min Sci 43:920–937.
Metropolis N, Ulam S (1949) ‘The Monte Carlo method’. Journal of the American Statistical
Association 43:335–341.
Microsoft Excel 15.0.4433.1506 (2013) Microsoft, Albuquerque, New Mexico, U.S.
Saada Z, Maghous S, Garnier D (2008) ‘Bearing capacity of shallow foundations on rocks

obeying a modified Hoek–Brown failure criterion’. Comput Geotech 35:144–154.
Page | 74
Serrano A, Olalla C (1996) ‘Allowable bearing capacity of rock foundations using a non-linear
failure criterium’. Int J Rock Mech Min Sci Geomech Abstr 33:327–345.
Sutcliffe D J, Yu H S, Sloan S W (2004) ‘Lower bound solutions for bearing capacity of jointed
rock’. Comput Geotech 31:23–36.
Urdan T C (2010), ‘Statistics in plain English’, third edition, Routledge, Taylor & Francis
group, New York, USA.
Venkatramaiah C (2006) ‘Geotechnical Engineering’, Revised third edition. New age

International Publishers, New Delhi, India.
Yang X-L, Yin J-H (2005) ‘Upper bound solution for ultimate bearing capacity with a modified
Hoek–Brown failure criterion’. Int J Rock Mech Min Sci 42:550–560.
Page | 75

212ce1021 13

Uploaded by

Copyright:

Available Formats

212ce1021 13

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

212ce1021 13

Uploaded by

Copyright:

Available Formats

A STUDY ON BEARING CAPACITY OF

SUBMERGED ROCK FOUNDATION

A Thesis submitted in partial fulfillment of the requirements

DEPARTMENT OF CIVIL ENGINEERING

A Thesis submitted in partial fulfillment of the requirements

Civil Engineering Department

Under the guidance of

DEPARTMENT OF CIVIL ENGINEERING

Date: Dr. N. Roy

Place: Professor and Head

Date: Nirupam Baruah

CONTENTS PAGE NO.

5. Conclusion and Future Scope 64

5.2 Future Scopes 64

Figure No Title Page No

Fig. 3.1 Hodograph diagram of two sided mechanism 10

Table No Title Page No

Table 4.1 Data set for analysis of effect of submergence 47

γ: Total unit weight of the rock mass.

γsat : Saturated unit weight of rock mass.

γw : Unit weight of water.

γ‫ ׳‬: Effective unit weight.

θ: Angle of velocity discontinuity line (CD) with horizontal.

μM : Mean values of M (Margin of Safety).

μQ : Mean values of Q (Load).

μR : Mean values of R (Resistance).

ξn : Function of α, θ, ϕi & ϕj.

ρRQ : Correlation Coefficient between R and Q (Load).

σF : Standard Deviations of F (Factor of Safety).

σM : Standard Deviations of M (Margin of Safety).

σQ : Standard Deviations of Q (Load).

σR : Standard Deviations of R (Resistance).

σM2 : Variances of M (Margin of Safety).

σQ2 : Variances of Q (Load).

σR2 : Variances of R (Resistance).

ϕi : Angle of friction of intact rock mass.

φj : Angle of friction of jointed rock mass.

ci : Cohesion of intact rock mass.

cj : Cohesion of jointed rock mass.

CI95 : A 95% confidence interval.

CI99 : A 99% confidence interval.

dw : Depth of water from foundation base.

E[F] : Expected values of F.

fn : Function of ξ & ϕj for two sided mechanism.

gn : Function of ξ & ϕj for one sided mechanism.

hC : Height from ground level to point C in mechanism.

hD : Height from ground level to point D in mechanism.

LCL95: Lower confidence limit at 95% confidence

no : Number of joint set.

Nci : Bearing Capacity Coefficient.

Ncj : Bearing Capacity Coefficient.

Nq : Bearing Capacity Coefficient.

Nγ : Bearing Capacity Coefficient.

Nγsub : Bearing Capacity Coefficient.

Nγw : Additional factor depends on groundwater depth.

qu : Ultimate bearing capacity for dry condition.

Si : Spacing of ith joint set.

sx : The standard error.

UCL95: Upper confidence limit at 95% confidence.