i

COMPARISON OF BACK ANALYSIS AND MEASURED LATERAL

DISPLACEMENT OF CANTILEVER DIAPHRAGM WALL

FAUDZIAH BTE SHUKOR

A project report submitted in partial fulfilment of the

requirements for the award of the degree of

Master of Engineering (Civil-Geotechnics)

Faculty of Civil Engineering

Universiti Teknologi Malaysia

MAY 2009

iii

DEDICATION

T O MY DEAR HUSBAND AND FAMILY

iv

ACKNOWLEDGEMENT

Formost, I wish to express my sincere appreciation to my supervisor, Professor

Dr.Aminaton bt. Marto. She has given me her full patient, advice, guidance, comments

and her valuable time in assisting me to complete the project. She has assisted me in my

difficulties of getting the work done. Without her patience and guide I feel I may not be

able to complete the project in the time schedule. Thank you again to my dear Professor

Dr.Aminaton.

I would like also to express my gratitude to Ir.Law Kim Hing who has provided

me the data for the case study. He has also assisted me in trouble-shooting during the

running of the analysis.

My sincere thanks also to my other colleagues, friends and lecturers who have

some how or rather been directly or indirectly helped me in the project.

Finally, my gratitude to my husband and children who have given me courage and

strength to complete the project.

v

ABSTRACT

This study is aimed to evaluate the performance of a cantilever diaphragm wall

earth retaining system, constructed by staged excavation. The actual performance

extracted from geotechnical instrument is compared with the numerical method. The

numerical method is using Soil Hardening model. The simulation of the computer

analysis has been carried out using different soil stiffness parameter which have been

correlated from 2000N, 2500N, 3000N and 3500N values (with N is the Standard

Penetration Test values). The results obtained had been used to compare with the actual

performance of the field data. From the results of the finite element analysis, the obtained

lateral displacement profile is reasonably in close agreement when compared to the

instrumentation profile. Thus the soil stiffness parameters ( refE50 , ref

oedE , ref

urE ) which have

been used to correlate based on 3500N values are suitable for the soil at the site of Kenny

Hill formation, analysed in this project. The instrumentation data and the analyses have

yielded very useful information for deep basement construction in terms of the selection

of the soil parameters.

vi

ABSTRAK

Kajian ini adalah bertujuan untuk menilai prestasi sistem tembok penahan tanah

diaphragm kantilever dalam pembinaan pengorekan tanah secara berperingkat. Prestasi

sebenar tembok yang di perolehi dari alatan instrumentasi geoteknikal akan di

bandingkan dengan cara numerikal. Cara numerikal tersebut adalah mengunakan model

Hardening soil model. Kekuatan tanah yang di perbandingkani dengan nilai 2000N, 2500

N, 3000N dan 3500N (dimana N adalah nilai dari Standard Penetration Test) disimulasi

dengan mengunakan computer. Keputusannya digunakan untuk membuat perbandingan

dengan prestasi sebenar yang diperolehi dari data tapak. Keputusan dari simulasi

komputer , menghasilkan pergerakan tembok adalah menepati pergerakkan jika

dibandingkan dengan profil instrumentasi. Dengan itu parameter kekuatan tanah

( refE50 , ref

oedE , ref

urE ) dimana perbandingkan 3500N adalah bersesuaian untuk tanah Kenny

Hill formation.. Data dari instrumentasi dan analisis telah menghasilkan maklumat

penting dalam pemilihan parameter tanah untuk kerja pembinaan basemen.

vii

TABLE OF CONTENTS

CHAPTER

TITLE PAGE

1

TITLE PAGE

DECLARATION

DEDICATION

ACKNOWLEDGEMENT

ABSTRACT

ABSTRAK

TABLE CONTENTS

LIST OF TABLES

LIST OF FIGURES

LIST OF SYMBOLS

INTRODUCTION

1.1 Background of the Problem

1.2 Statement of Problem

1.3 Objectives of the Study

1.4 Scope and Limitation

1.5 Significant of the Study

i

ii

iii

iv

v

vi

vii

x

xi

xiii

1

1

2

3

3

4

2 LITERATURE REVIEW 5

2.1 Retaining Wall Type

2.2 Excavation Effects

2.3 Geotechnical Instrumentation

2.4 Previous Research In Deep Excavation

2.5 Elastic Properties of Soil

2.6 Applications OF Finite Element Analysis

2.6.1 Plaxis

2.6.2 Hardening Soil Model

5

5

7

7

12

14

16

18

viii

2.7 Design Approach

2.7.1 Stability Analysis

2.7.2 Free Earth Support Method and

Fixed Earth Support Method

2.7.3 Stress and Deformation Analysis

2.7.4 Settlement Induced by the

Construction of Diaphragm Wall

2.7.5 Excavation depth

2.7.6 Analysis of ground surface settlement

Induced by excavation

2.7.6.1 Peck’s Method

2.8 Mobilisation of Earth Pressure

2.9 Engineering properties of retained soils

2.9.1 Angle of Internal Friction

2.9.2 Cohesion

2.9.3 Unit Weight of soil

2.9.4 Active Earth Pressure

2.9.5 Passive Earth Pressure

2.9.6 At-Rest Pressure

21

21

24

24

27

28

29

29

31

31

31

32

33

33

34

35

3 RESEARCH METHODOLOGY 36

3.1 Introduction

3.2 Technical Literature

3.3 Data Collection

3.4 Parametric Study

3.5 Analysis of Data

36

36

36

37

37

ix

4 CASE STUDY OF DEEP EXCAVATION IN

KUALA LUMPUR

39

4.1 Introduction

4.2 Project Description

4.3 Stratigraphy Profile

4.4 Field Instrumentation

4.5 Stage Construction

4.6 Finite Element Simulation

4.6.1 2-D Modeling

4.6.2 Soil Parameters and constitutive model

4.6.3 Parametric Study

39

39

40

42

44

47

47

47

50

5 ANALYSIS RESULTS AND DISCUSSION 52

5.1 Introduction

5.2 Influence Soil Stiffness ( refE50 , ref

oedE )

52

52

6 CONCLUSION 61

6.1 Introduction

6.2 Conclusion

6.3 Recommendations

61

61

62

REFERENCES

63

x

LIST OF TABLES

TABLE NO. TITLE PAGE

2.1

2.2

2.3

2.4

4.1

4.2

4.3

4.4

4.5

4.6

4.7

4.8

5.1

5.2

5.3

Typical Values in Terms of Top Movement Caused

from Rotation of Basement Wall at its Base

Empirical Equations for Es (Bowles, 1998)

Ranges of Es for Various Soils (Bowles, 1998)

A Correlation between the Angle of Internal Friction

and the Standard Penetration Test (Winterkom and

Fang)

Simulation of Construction Work

Soil Parameters for Hardening Soil Model

Typical Parameters for Hardening Soil Model

Stiffness Soil Parameters for Hardening Soil Model

(2000N)

Stiffness Soil Parameters for Hardening Soil Model

(2500N)

Stiffness Soil Parameters for Hardening Soil Model

(3000N)

Stiffness Soil Parameters for Hardening Soil model

(3500N)

Soil Stiffness Input in Soil Hardening Model

Comparison of the Back Analysis in the Numerical

Simulation with the Actual Performance for

Horizontal Displacement at the Final Stage, at the

Soil Surface.

Extreme Bending Moment and Shear Force Values

for the Different Soil Stiffness

Vertical Displacement Values at Final Stage for the

Different Soil Stiffness

6

13

14

32

44

48

48

49

49

49

49

50

52

60

60

xi

LIST OF FIGURES

FIGURE NO. TITLE

2.1

2.2

2.3

2.4

2.5

2.6

2.7

2.8

2.9

Deformed Mesh for Analysis of Sheet Pile Wall

(Bakker and Brinkgreve 1991)

Hyperbolic Stress-Strain Relations in Primary

Loading Standard Drained Triaxial Test.

(Brinkgreve et al,2004)

Definition of in Oedometer Test Results

(Brinkgreve et al,2004)

Overall Shear Failure Modes (a) Push-In and (b)

Basal Heave (Chang, 2006)

Free Earth Support Method (a) Deformation of

Rretaining Wall and (b) Distribution of Earth

Pressure (Chang, 2006)

Fixed Earth Support Method (a) Deformation of

Retaining Wall and (b) Distribution of Earth

Pressure (Chang, 2006)

Envelope of Ground Surface Induced by

Trenched Excavation (Clough and O’ Rourke,

1990)

Envelope of Ground Surface Settlement

Induced by Diaphragm Wall Construction (Ou

and Yang, 2000)

Relationship between Maximum Wall

Deflections and Excavation Depths (Ou

et,1993)

17

19

21

23

25

26

27

28

29

xii

2.10

2.11

2.12

2.13

3.1

4.1

4.2

4.3

4.4

5.1

5.2

5.3

5.4

5.5

5.6

5.7

Peck’s Method (1969) for Estimating Ground

Surface Settlement

Mohr Circle Diagram (Macnab,2002)

Rankine Diagram Active Pressure

(Macnab,2002)

Rankine Diagram Passive Pressure

(Macnab,2002)

Methodology Flow Chart

Diaphragm Wall Elevation and Llocation Plan

Geological Map of Malaysia

SPT-N Values Profile

Stages of Construction and Simulation on

Hardening Soil Model

Horizontal Wall Displacement versus Wall

Depth for 2000N

Horizontal Wall Displacement versus Wall

depth for 2500N

Horizontal Wall Displacement versus wall

depth for 3000 N

Horizontal Wall Displacement versus Wall

depth for 3500N

Comparison of Horizontal Wall Displacement

for 2000N, 2500N, 3000N and 3500N with

Measured Field Data,at the Final Excavation

Stage

Typical Profile for Computed Shear Force

Diagram for the Diaphragm Wall

Typical Profile for Computed Bending Moment

Diagram for the Diaphragm Wall

30

32

34

34

39

40

41

43

46

53

54

55

56

57

59

60

xiii

LIST OF SYMBOLS

A Cross sectional area

'c Effective cohesion

uc Undrained shear strength

d Distance from wall

E Young’s modulus

refE Young’s modulus for Mohr-Coulomb model

refE50 Secant stiffness in triaxial test

ref

oedE Tangent stiffness in oedometer test

ref

urE Unloading reloading stiffness

sE Young’s modulus of sand or clay

f Yield function

H Total depth of excavation

I Moment of inertial

pI Plasticity index

NC

oK oK value for normal consolidation

xk Permeability in horizontal direction

yk Permeability in vertical direction

L Width of excavation

m Power for stress level dependency stiffness

n Porosity

'P Mean effective stress

refp Reference stress for stiffness, p

ref = 100 kpa

xiv

q Deviator / Deviatoric stress

qc Tip resistance of cone penetration test

RL Reduced level

fR Failure ratio

erRint Interface reduction factor

SPT’N’, N Standard penetration test N value

u Pore water pressure

Lw Liquid limit

'φ Effective friction angle

uφ Undrained friction angle

ψ Dilatancy angle

τ Shear stress

σ 1 Effective stress

σ Total stress

'

3

'

,2

'

1 , σσσ Major, intermediate and minor principal stress, respectively

ν Poisson’s ratio

urv Poisson’s ratio for unloading reloading

aεε ,1 Axial strain

γ Soil unit weight

unsatγ Soil unit weight�

satγ Soil unit weight

1

CHAPTER 1

INTRODUCTION

1.1 Background of Problem

The most cost effective and practical method to support an excavation is to

slope back the sides of the excavation. The design requires only the soil properties of

the excavated area so as to determine the angle of reponse. The concept of the design

requires wider space for the proposed slope and the excavation may not be deep.

Another setback will be the excavated area is near to the adjacent building. The

subject of the construction for deep excavation become more complicated, when

there is no adequate space, slope cannot be accommodated and therefore an earth

retaining system is the solution. The three basic types of earth retaining systems are

cantilevered, braced or tied-back system.

The design of the retaining wall system can be from a simple empirical

method towards a more complicated complex computer analysis. Whatever the

method, the design aspects are the stresses, loads related to the wall system and the

effect of construction method. Other than those design methods, the designer past

experience is significant too. The designer must be equipped with the subject

literature, aided with the geotechnical journals and texts for him to apply an

appropriate solution from many options to excavation support problem.

However in the increasingly competitive environment where “value

engineering” is required followed by the reluctance of the client to invest in

“geotechnical cost”, the designer is required to produce a design with minimal cost.

Whatever the preferred solution from the many available solutions, the risks involved

have to be evaluated and any failures of the retaining wall must not happened. The

2

preferred solution must therefore produce safe and economical design taking all

aspects into consideration.

1.2 Statement of Problem

The occurrence of ground displacement due to excavation works in deep

excavation is based on many factors such as the stratigraphy, soil properties, lateral

earth pressure, the method of constructions, contractual matters, soil loads, water

table, seepage problem and workmanship. These factors need to be considered in the

design procedure to understand the ground response due to excavation. Related to the

deep excavation, other important aspect is the evaluation of the foundation for the

adjacent properties and the effects of the excavation on the serviceability of the

adjacent structures.

The subsoil stratigraphy is generally obtained by auger deep boring supported

with Standard Penetration Test result from borehole. With the data taken from the

multiple boreholes and then drawn in the geometry, can lead to understand the soil

stratigraphy. The soil properties such as friction angle, cohesive intercept, and

poisson’s ratio are required for any design method of earth retaining structure

whether using numerical or simplified method.

This data used in the design process is confirmed by the instrumentation

monitoring on the behavior of the earth retaining system. Both the predicted and

actual behavior of the retaining system must be the same during/and at end of the

construction period. The need to monitor the behavior of the earth retaining wall in

conjunction with the stage of excavation which is required to evaluate the actual

behavior is similar to the predicted or as designed. If otherwise immediate action is

necessary to resolve the problem cause by the unexpected behavior of the supported

soil. This may be due to the construction nearby causing the change in the water

level or wrong interpretation of soil data.

This study is aimed to evaluating the performance of a cantilever diaphragm

wall earth retaining system, constructed by staged excavation. The actual

3

performance extracted from geotechnical instrument will be compared with the

numerical method. The numerical method is using Soil Hardening model.

1.3 Objective of the Study

The objectives of the study are as follows:

i) To determine soil properties and the standard penetration test (SPT-N)

values of the soil layers at site location.

ii) To determine the correlation between the soil stiffness parameter with

the field SPT-N values by comparing the monitoring measurement

results with the values obtained from the finite element analysis result

using Hardening Soil model for the lateral displacement of cantilever

diaphragm wall.

iii) To determine the wall deflection using the previous researchers

correlation of soil stiffness and SPT-N values.

1.4 Scope and Limitation

This case study has been conducted on particular project, which is represent

by a development with a 7 m depth basement at a site somewhere in Kuala Lumpur.

The detail of the subsoil parameters, existing and proposed soil platform, the stages

of excavation, the retaining wall system and the on-site instrumentation data has been

used in the case study. The analysis has been carried out with numerical method

using Finite Element Analysis by the aid of computer program (Plaxis). The lateral

displacement results compare with the actual behaviour of the retaining wall system,

taken from the instrumentation data. However, the instrumentation data used was for

the horizontal deflection of the wall only. The limitation of using the Plaxis analysis

is limited to Soil Hardening Model using the available soil data obtained from the

available borehole results.

4

1.5 Significant of the Study

This study will be very useful to geotechnical engineers to be used as

reference for analyzing the stability of the retaining wall system with respect to the

lateral movement created during construction. The outcome of the study will show

that the design and construction method of the project will be used as guidance to

similar condition of other construction sites for prediction of horizontal movement.

The soil parameter for the individual soil type can be used as a guide to all designers

with similar condition. The design model is also significant whereby it can be used

for similar condition.

5

CHAPTER 2

LITERATURE REVIEW

2.1 Retaining Wall Type

Deep excavation earth retaining is divided into three systems namely

cantilever wall system, braced system and tied-back system. The designer need to

understand this system so as to evaluate the appropriate system to be adopted. The

choice of support system depends on the soil type, depth of excavation, availability

of space, budget, its advantage and disadvantage (Chang, 2006)

The cantilever wall system is most suitable at shallow depth where the lateral

earth pressure and water pressure is not a problem. Terzaghi and Peck (1967)

considered that excavation whose depth were less than 6 m are defined as shallow

excavations and those deeper than that as deep excavations. In related to the deeper

depth, a much reliable retaining wall system is needed, where it can be tied back or

strutted. In certain design, the construction adopts temporary tieback during the earth

excavation where it is released once the permanent lateral support is provided by the

structure. This type of construction is adopted in the top down construction

procedure.

2.2 Excavation Effects

There are many factors affecting the excavation works. The designer must

truly understand the factors both in theoretical and practical applications. It is a most

specialized work involving understanding of soil types and behaviour, soil forces and

6

subsurface profile. The parameters taken shall be most appropriate and to much

accuracy.

The earth retaining wall needs to be restrained at the top to avoid horizontal

soil yielding. Once the wall moves, it indicates the soil shear resistance occurs. The

studies done by Raj (1999) relate the top translation movement due to the rotational

response of the base of the wall with regard to the development of earth pressure.

Thus indicate that the lateral force is related to the active earth pressure. Typical

values in terms of top movement caused from rotation of basement wall at its base

are shown in Table 2.1.

Table 2.1: Typical Values in Terms of Top Movement Caused

from Rotation of Basement Wall at its Base (Raj,1999)

Soil and condition Amount of

translation

Cohensionless, dense

0.001 - 0.002H

Cohensionless, loose

0.002 - 0.004H

Cohesive, firm

0.01 - 0.02H

Cohesive, soft

0.02 - 0.05H

where H = height of basement wall in meter

7

2.3 Geotechnical Instrumentation

Geotechnical instrumentation is very significant in any deep excavation. It

monitors the response of soil and wall under excavation works at any stage. The

monitoring can be viewed carefully. The interpretation of the result is important to

the wall stability. If any problem occurred such as soil movement it can be

encountered and solved successfully during the construction period.

One of the commonly used instruments is the inclinometer. It measures the

deformation of the wall through the depth of the wall. It can be placed in the wall or

install in the subsoil immediately behind the wall. Usually it is extended 3 meter

below the wall toe to measure the movements. From the results obtained from the

inclinometer, the lateral movement profile can be used to validate the design and

make immediate measures for any mistakes or wrong assumptions done during the

initial design stage.

In wall monitoring, the instruments are used to monitor the deflection.

Brahana et al (2007) concluded that the lateral deflection for anchored walls ranged

from 0.1 % to 0.5% of the wall height. Their result of study, described that from the

instrumentation data, suggest that changes of support has no effect on the earth

pressure. Their findings were that the trapezoidal earth pressure is still applicable to

be used in both temporary and permanent structure.

2.4 Previous Research in Deep Excavation

Knowing the history of failures caused by excavation is of great importance.

The designer is able to understand the unpredicted nature of soil behavior and

properties under the condition of unloading and loading. Hence the designer can list

the do and don’ts and what criteria are of significant and which criteria have been

ignored unpurposely.

Poulos and Chen (1996, 1997) and Poulos (2002) involved in the case study

where an excavation had caused the nearby building sitting on pile failed. The wall

8

of the excavation was not supported causing soil-pile interaction failed due to

excavation induced ground movement, causing additional forces and moments to the

adjacent area.

Gue and Tan (2004) described a case study of a construction site

experiencing base heaving and the sheet pile wall system moved excessively causing

the pile integrity damaged. Through the back analysis, the 12 m long sheet pile

cannot retain the soil when the excavated depth exceeds 3.5 m level. Introduction of

props at the top of the retaining wall would have avoided the failure.

Another type of failure was the collapse of the 25 m deep excavation due to

the buckling of the bracing (Magnus et al 2005). The cause of failure was due to the

conceptual technical error when using the complex numerical model. Even though

observation method was established during the construction, its weakness was that

the performance of the excavation was not checked regularly.

�

The observation method was introduced by Peck (1969) to enhance the

development of geotechnical engineering with the approach of “learn as you go”. It

is a valuable application where in any excavation; expectation of uncertainty is due

to be met. The predicted value need to be evaluated during the excavation by the

actual performance.

Ikuta et al (1994) described a case study on the observation method, of a top

down construction method. The first few monitoring reading for prediction of ground

movement was taken to be used to improve in the predictions. This would optimize

the cost of the whole construction works resulted from optimizing the design due to

the availability of the instrumentation data. The numerical method has solved to

clarify the important parameter related to excavation. Therefore, the application of

these numerical models has led to increasingly useful information, knowledge and

experience. Thus allowing geotechnical design to be safe yet economical for any

construction works under high risk environment.

Liew et al (2008) described a case history involving the construction of a

two-storey basement. The scope of work was to evaluate the condition of a distressed

9

temporary shoring structure, find the probable causes and then propose options for

remedial. The original sequence of works was that the temporary shoring support

which was made up of 12 m long steel sheet piles were driven from RL 59.0 m, to

facilitate the installation of contiguous bored pile (CBP) which is 16 m long and 750

mm diameter After the installation of CBP, excavation proceeded from RL 54.6 m to

52.0 m. At this stage, ground stresses at the adjacent area were observed. There were

ground subsidence, tension cracks and the CBP showed some deviation. The cause of

deviation of the CBP wall was due to the over excavation of the pile cap excavation

works, without the support of the planned raking strut. To confirm this cause, the

finite element simulation was carried out. The results showed that analyses agreed

with the measured wall displacement. The removal of the passive berm and the over

excavation had reduced the lateral resistance of the sheet pile and mobilised the

strength of the retaining wall from serviceability state to ultimate limit state

condition. The deviated sheet pile wall induced additional lateral force to the CBP

wall and the high induced force had damaged the CBP pile, which cause effects to

the ground displacement. Other findings from this case study were as follows:-

i. The existing backfilling over the natural valley had created

perched ground water regime which lead to unfavourable

behavior of backfill.

ii. Filling over valley without proper compaction result in

unstable backfill.

iii. Desk-study of pre-development ground contours of the site was

crucial so as to indentify potential geotechnical problems.

Freiseder et al (1999) carried out hardening soil model to analyze the

deformation behavior of an excavation for a 6.3 m deep underground car park. The

numerical analysis was carried out prior to the construction so as to refine the

analysis during construction. The data from the early stages of construction obtained

from the insitu site measurements were utilized. The results from the analysis and

site measurement were compared. The numerical model became a valuable tool to

10

assess the ground distress at any stage of critical phase. The site is located at

Salzburg, Austria which was next to the main station and existing underground

railway line. The allowable displacement should not exceed 10 mm at the top of

diaphragm wall or the railway line. Therefore finite element analysis was carried out

using the obtained data from the initial stage of construction. The aim was to refine

the analysis further. In the analysis, the underground railway line was modelled to

account for the stress redistributions (the underground railway line was completed

two years before). The displacements were set to zero. The model was performed

using undrained condition and consolidation analysis. The consolidation analysis

allowed the dissipation of excess pore water pressure matching the actual state of

construction.

Tan (2007) in his paper described a case history of incorrect use of numerical

modeling of the excavation system that result in the under design of the diaphragm

wall. In the paper he described the importance of understanding the difference

between the effects of change in volume due to volumetric shear distortion, The

Mohr coulomb model shows no volume change due to shear The stress path response

to undrained loading or unloading following the constant mean effective stress (p’)

path. The soil model will fail at the point where the initial p’ value at the start of the

unloading sequence meets the Mohr Coulomb failure criterion. As for soft clays

(normally consolidated) that tend to shear under drained condition, the volume will

change and contract. For the undrained loading / unloading condition, the soil is

prevented from contracting thus develop large excess pore water pressure within the

soft soil, consequence is that the effective stress path of the soft clay in undrained

shearing is to curve back from the constant p’ line. This was the aspect of the soil

behavior that was not incorporated in the plaxis design of the project. The

underestimate value of the strength of the soft clays resulted in under-prediction of

maximum wall deflection by a factor of 2. The other effect of this error, had led to

undersize of the diaphragm wall and also inadequately of the penetration wall depth.

This caused redundancy in the design, whereby plastic hinge occurred between 9th

and 10th

level strut. After the removal of the upper soil, the plastic hinge was formed

and yielded the soil and triggered the total collapse of the excavation support system.

11

Liew et al (2007) performed a study to evaluate the effectiveness and

economical of the proposal design / solution for the construction of a 3-level

basement excavation (11m depth). The basement design was using semi top-down

method with permanent ring slab at B1 level. The wall thickness is 600mm thick and

the depth was 15m. The 325mm thick B1 slab was utilized as the temporary prop to

surrounding diaphragm wall until excavation works had completed. The ring slab

also resisted the lateral load from diaphragm wall. In the study, computer

programmed, plaxis was carried out for the analysis of deformation problems both in

soil and rock. The model adopted was hardening soil model using effective stress

analyses. The soil material is assigned as undrained condition. Consolidation analysis

was to simulate the actual soil behavior from undrained to drained condition due to

excess pore water pressure. The effective Young’s Modulus (E’) and the effecting

unloading / reloading Young’s Modulus (E’ur) are correlated based on 2000N and

6000N respectively (as recommended by Tan et al (2001)). As for the wall interface

elements, Rinter of 0.67 and 0.5 are commonly used for sandy and silty soil

respectively. According to Gaba, A. R. et al Rinter of 0.8 can be adopted for concrete

cast against soil (diaphragm wall) provided that the wall is not subjected to high

vertical load which can cause the wall to settle relative to the soil. The results of the

study found that the lateral wall displacement as initial cantilever stage of excavation

adopted soil stiffness 3000N which matched computer analysis result with the

measured wall lateral displacement. For further excavation stage, the soil stiffness

reduced gradually to 1850N at the last stage of excavation works. Both the measured

wall displacement and the analysis results matched well which showed that the

maximum wall displacement is about 2% of the maximum excavation depth (11 m).

The study concluded that the effective Young’s Modulus (E’) and the effecting

unloading / reloading Young’s Modulus (E’ur) which are correlated based on 2000N

and 6000N respectively are suitable to be adopted in Kenny Hill Formation residual

soil.

James et al (2008) in his paper gave an account of the commonly adopted

steel embedded walls and steel shores in Hong Kong with an illustration from three

case histories. One of his findings was that when using sectional properties of the

steel for retaining earth system, was more consistent and uniform than that of the cast

in-situ reinforced walls, because it would eliminate the uncertainty in bending

12

stiffness of the wall and provided a good basis for design assumptions such as soil

stiffness. The soil stiffness in Hong Kong was correlated to SPT-N values.

According to James et al (2006), back analysis of wall deflection at Lok Ma Chah

(LMC) spur line was conducted by Pan et al (2006) and the result of the analysis was

that the elastic moduli were found to be 1500N for fill, 4000N for coarse alluvium /

completely decomposed granite. Another similar study had been done at the Hong

Kong Polytechnic University (HKPU) site, in which one of the back analysis of the

wall deflection, showed that the correlations of SPT-N values for soil stiffness are

1500N for fill and marine deposits, and 4000N for completely decomposed granite.

However the author advised that the application of these factors in the design of

excavation must be carefully examined since there are other factors attributing to

settlement.

Yee et al (2008) described three case studies in Malaysia for the excavations

works in problematic soils such as marine clays and ex-mining soils overlying

limestone. In all the three cases; the earth retaining support systems were using deep

soil mixing technique. One of the sites was at Kuala Lumpur and the project was a 12

storey condominium with 2 levels of basement car park. A dry deep soil mixing

technique was used to support the stress sloped excavation along the 3 sides of sites.

The subsoil condition was that for the first 8m the soil comprised of loose silty sand

deposits and ex-mining soils. The underlain layer was the medium dense silty sand

layer overlying limestone bedrock. The water table was about 2 m from the ground

level. From the monitoring results, the wall deflection was less than 1% of the

excavation depth (without lateral props).

2.5 Elastic Properties of Soil

The stress – strain modulus sE is one of the elastic properties of soil. It can

be obtained from the slope (tangent or secant) of stress – strain curves from triaxial

tests. It can also be estimated from field tests. Triaxial tests, tend to improve the

value of sE , since the confining pressure “stiffens” the soil thus a larger initial

tangent modulus is obtained. Whilst unconfined compression tests tend to have a

13

conservative value of sE , i.e. the computed (usually initial tangent modulus) value is

small, thus �H being larger than the value obtained at in situ.

Due to high cost to obtain the data of sE value from the laboratory, the

standard penetration test (SPT) has been used frequently to obtain the stress-strain

modulus sE . According to Chang (2006), the equations in Table 2.2 give a number of

equations for the estimation of the sE values. Table 2.3 gives the ranges of sE for

various soil types.

Table 2.2: Empirical Equations for sE (Bowles, 1998)

Soil type SPT-N (kPa)

Sand sE = 500(N + 15)

(normally consolidated) sE = (15,000 ~ 22,000) In N

sE = (35,000 ~ 50,000) log N

Sand (saturated) sE = 250(N+15)

Sand (over consolidated) sE = 18,000+750N

Gravelly sand and gravel sE = 1,200(N + 6)

sE = 600(N + 15) N<15

sE = 600(N + 15)+2,000 N > 15

Clayey sand sE = 320(N + 15)

Silty sand

Soft Clay

sE = 300(N + 6)

Note

N - SPT-N value; I kPa = I kN/m2 = 0.1 t/ m

2

14

Table 2.3: Ranges of sE for Various Soils (Bowles, 1988)

Soil Type

sE

(Mpa)

Very Soft Clay

2 – 15

Soft Clay

5 - 25

Medium Stiff Clay

15 - 50

Stiff Clay

50 - 100

Sandy Clay

25 – 250

Silty Sand

5 - 20

Loose Sand

10 - 25

Dense Sand

50 - 81

Loose Gravel

50 - 150

Dense Gravel

100 - 200

Shale

150 - 5000

Silt

2 - 20

Notes:

The table only lists the possible range of sE .

sE of in situ soils is related to water conrent, density, stress history, etc

2.6 Applications of Finite Element Analysis

Applications of Finite Element started in the early 1970’s. Its applications

nowadays are tremendously used by geotechnical engineers to solve much complex

problems. Some of the applications are to predict ground movements, undertaking

parametric studies and to back-analyze past case-histories.

15

Palmer et al (1972) used finite element to evaluate the importance of different

parameters on the performance of a braced excavation at Oslo Subway Norway . A

unique technique by input many variables simulating various soil interaction and

behavior.. The soil deformation modulus was the most significant. The other

parameters were of nominal effect except for the wall and strut stiffness.

In the year 1977, Burland et al described the performance of the multi-

propped excavation at New Palace yard. The soil profile consists of sand and gravel

overlying stiff, fissured London clay. Their finding was that the cause of

approximately 50 % of the total ground movements both in vertical and horizontal

direction was due to the installation of the diaphragm wall and piles.

In the year 1983, Eisenstein et al performed the study on excavation in glacial

till, overconsolidated stiff, fissured silty-clay. The location was at Edmonton, Canada.

They ran finite element to analyze the lateral pressures, making many models on

stress-strain analysis-linear and non-linear elasticity using data derived from the

triaxial tests. Their observations were that the plane strain data and the stress path

using the triaxial data matched with the actual field performance of the

displacements. They described that the wall flexibility contribute to a reduction of

the lateral pressures and the ground movements.

Potts et al (1984) used finite element analysis to study the performance of a

propped retaining wall in deep excavation.. The study was on initial value of earth

pressures at rest. With a high value of Ko the prop force, bending moment were

substantial high thus exceeding the predicted value by the limit equilibrium

calculation. But with lower value of Ko, the bending moment and the prop forces

were smaller when compared with the value derived from the limit equilibrium

calculation which show a conservative value.

Parametric studies on the effect of soil/wall/prop stiffness and pre-excavation

earth pressures coefficient were investigated by Powrie et al (1991) on a deep

excavation of 9 m depth in stiff over consolidated boulder clay. The wall was

propped by a continuous slab. The analysis showed that the stiffness of the soil was

the major value that contributes to the deformations, not the wall flexural rigidity. As

16

for the bending moment, it was largely governed by the pre-excavation earth pressure

coefficient. The effect from the slab-wall connection was of lesser effect.

Ng and Lings (2000) performed simple models using a linear elastic-perfectly

plastic Mohr-Coulomb and a non-linear model Brick Model (Simpson, 1991) to

evaluate the effectiveness of these two models. The program simulated the

construction of a top down multi-propped construction in the stiff over consolidated

Gault clay at Lion Yard , Cambridge . The studies show that the Mohr-coulomb with

the “wished-in-place” wall was able to predict the deflection and the maximum

bending moment of the wall, in condition that the soil value was estimated perfectly.

The model could not estimate the ground deformation pattern. For the strut loads the

model estimated a very high value.

Law (2008) described the behaviour of a braced excavation at Kenny Hill

Formation. Finite element analyses were carried out to assess the effects of soil

stiffness parameters and the soil constitutive models using elastic-perfectly plastic

Mohr-Coulomb model and elasto-plastic Hardening Soil model implemented in

Plaxis. One of the findings was that the predicted wall deflections using the

correlation of 2500N was in good agreement with field measured data.

2.6.1 Plaxis

There are many computer aided programme in solving problem using finite

element numerical method. According to Burd (1999) the Plaxis programme started

at Delft University of Technology in early 1970’s when Peter Vermeer started to do a

programme of research on finite element analysis on the design and construction of

Eastern Scheldt Storm-Barrier in Netherland. He established the finite element code

called Elplast which was able to calculate the elastic-plastic plane using six-nodded

triangular elements, written in Fortran VV.

In the year 1982 Rene de Borst under the supervision of Pieter Vermeer ,

performed his master’s programme related topic on the analysis of cone penetration

test in clay. The study of axisymmetric led to the existence of Plaxis (Plasticity

17

Axisymmetry). The study was on a six-nodded triangles in the element. The setback

being the programme exhibit “locking” for incompressible model. This was later

improved by Sloan and Randolph (1982) highlighting the effect of “locking” was due

to the incompressibility constraints imposed to the nodal displacements. They had

made 15 – noded triangle thus increasing the number of nodes in the element. Their

findings benched the usage of 15-noded triangle as the simplest element for any

analysis in axisymmetric. This experiment had led De Borst and Vermeer to

implement the 15-noded triangle in Plaxis thus solving the problem of cone

penetrometer.

The development of Plaxis proceed with the problem to solve the soil-

structure interaction effects. This led to the study on beam element by Klaas Bakker

under the supervision of Pieter Vermeer. The outcome of the experiment using beam

element was applicable to flexible retaining wall and later application to the analysis

of flexible footings and rafts. Baker’s work formulated the implementation of 5-

noded beam element in Plaxis (Bakker et al (1990), Bakker et al (1991)). The 5-

noded beam element is compatible to the 15-noded triangular elements (has 5 nodes).

Baker’s work was novel for the invention of hybrid method introducing the

displacement of degree-of –freedom to the element behavior. The lack of degree of

freedom has made solution to reduce the number of variables thus simplified the

element . The early application of this is as shown in the Figure.2.1.

Figure 2.1: Deformed Mesh for Analysis of Sheet Pile Wall.

(Bakker and Brinkgreve 1991)

18

2.6.2 Hardening Soil Model

The Hardening Soil model is an advanced model used for simulating the

behavior of both stiff and soft soils (Schanz, 1998). Compared with an elastic –

perfectly model, the yield surface of the hardening plasticity model is not fixed in

principal stress space. It expands because of the plastic straining. The two types of

hardening are the shear hardening and compression hardening. In shear hardening,

the model is irreversible to strain due to primary deviatoric loading. As for

compression hardening, the model is irreversible to plastic strains due to primary

compression in oedometer loading and isotropic loading.

When soil are subjected to primary deviatoric loading, the soil shows a

decrease in its stiffness and develop an irreversible plastics strains. In a drained

triaxial test, the axial strain and deviatoric stress, can be expressed as a hyperbola.

This relationship between the axial strain and deviatoric stress, known as hyperbolic

model was initially formulated by Kondner (1963) and later by Chang et al (1970).

The model was then superseded by the Hardening Soil model. The Soil Hardening

model uses the theory of plasticity. It includes soil dilatancy, and the model has a

yield cap.

The distinct characteristic of the model are : plastic straining due to primary

deviatoric loading ( refE50 ), plastic straining due to primary compression ( ref

oedE ),

Elastic unloading / reloading (ref

urE ), failure according to Mohr-Coulomb model (c,φ

and ψ ), and stress dependent stiffness according to a power law (m).

The formulation of the Hardening Soil model is the hyperbolic relationship

between the vertical strain, 1ε , and the deviatoric stress q, in primary triaxial loading.

The drained triaxial test established yield curves, described by the following

Equations (2.1) and (2.2):

ai qqI

q

E

I

/1

−=− ε for qfq < (2.1)

19

f

iR

EE

−=

2

2 50 (2.2)

whereby aq is the asymptotic value for shear strength, iE is the initial

stiffness, qf is the ultimate deviatoric stress.

The relationship of these equations is plotted in Figure 2.2. The parameter

50E is the confining stress dependent stiffness modulus for primary loading and is

described by the Equation (2.3):

m

ref

ref

pc

cEE �

�

�

�

��

�

�

+

−=

φ

σφ

cot

cot'

3

5050 (2.3)

where refE50 is a reference stiffness modulus corresponding to the reference

confining pressureref

p .

aq

1− ε

Figure 2.2: Hyperbolic Stress-Strain Relations in Primary Loading for a

Standard Drained Triaxial Test (Brinkgreve et al, 2004).

Using plaxis, the ref

p is taken as 100 stress units as a default setting. The

actual stiffness depends on the minor principal stress, '

3r which is the confining

pressure in a triaxial test. '

3r is negative for compression. The value of stress

dependency is given by the power m. Simulation for soft clays, the power m is taken

E50

1 Eur

1

axial strain devia

toric s

tress I�

1 - �

3I

Failure line

fq

Asymptote

20

as equal to 1.0. Janbu (1963) reports values of m around 0.5 for Norwegian sands and

silts. Von Soos (1980) reports values ranges 0.5 < m < 1.0.

In Equation (2.1), the ultimate deviatoric stress fq , and the quantity aq are

defined as follows :

( )φ

φφ

snrcq f

−−=

1

sin2cot

1

3 ffa Rqq /= (2.4)

The minor principal stress, 1

3r , is usually negative. The ultimate deviatoric

stress fq is derived from the Mohr-Coulomb failure criterion. As value of deviatoric

stress, q is equal to ultimate deviatoric stress fq , the Mohr-Coulomb failure criterion

is satisfied, thus perfectly plastic yielding occurs. This is described in the Mohr-

Coulomb model. Failure ratio fR is the ratio between fq and aq , which is smaller

than 1. For plaxis, fR is equal 0.9 as the default setting.

The stress paths for unloading and reloading used stress dependent stiffness

modulus as follows .

m

ref

ref

ururpc

cEE �

�

�

�

��

�

�

+

−=

φ

σφ

cot

cot'

3 (2.5)

where ref

urE is the reference Young’s Modulus for unloading and reloading, with

respect to the reference pressure ref

p . In most cases it is appropriate to set ref

urE

equal to 3 refE50 ;this is the default setting used in Plaxis.

As 50E have ben defined by Equation (2.2) the oedometer stiffness has to be defined.

The equation as follows in Equation (2.6)

( )mpc

cref

oedoed refEE+

−=

φ

σφ

cot

cot'

1

(2.6)

21

When oedE is the target stiffness modulus as shown in Figure 2.3. ref

oedE is a tangent

stiffness at a vertical stress of refp=−

'

1σ .

1σ−

ref

oedE

ref

p 1

1ε−

Figure 2.3: Definition of ref

oedE in Oedometer Test Results

(Brinkgreve et al, 2004).

2.7 Design Approach

Chang (2006) expressed that failures or collapse of excavations are disastrous

and must be avoided. The failure may arise from the stress on the support system

exceeding the strength of the material, or from the shear stress in soils exceeding the

shear strength etc. Therefore in excavation design, various analyses have to be

performed as follows.

i) Stability Analysis

ii) Stress and Deformation Analysis

2.7.1 Stability Analysis

The method of analyzing whether the soils at the excavation are able to bear

the stress generated by excavation are called stability analysis. This includes the

shear failure analysis, sand boiling analysis and upheaval analysis. When the shear

stress at a point exceeds or equals to the shear strength of the soil at the point, the

22

point is in failure limit state. If all these failure points are connected a failure surface

is formed, which will cause collapse or failure to the excavation. This kind of failure

is called overall shear failure. The failure modes can be in form of push in or based

heave as shown in Figure 2.4. Push-in is caused by earth pressures and both the sides

of retaining wall are reaching the limiting state. The soil moved towards the

excavated zone until reaching the full-zone failure.

This analysis views the earth retaining wall as a free body, where equilibrium

between the external and internal forces is achieved. When push-in occurs, there are

different extents of movements of the embedded part of retaining wall. This varies

the earth pressure on the wall. To analyze this earth pressure, fixed earth support

method and the free earth support method are adopted.

The basal heave is formed from the weight of the soil outside the excavation

exceeding the bearing capacity of the soil below the excavation bottom. The soil

move causing the excavation bottom to heave so much that cause the excavation to

collapse (Figure 2.4).

Therefore when analyzing the basal heave, several possible heave failure

surfaces are analyzed, but the smallest factor of safety is the most probable critical

failure surface. When basal heave occurs, the soil surrounding the bottom of

excavation heaves. If the soil is a soft clay, the earth pressure on both sides of the

wall will reached the limiting state.

23

a) Wall Bottom “Kick-Out”

b) Failure Surface

Figure 2.4: Overall Shear Failure Modes (a) Push-In and (b) Basal Heave

(Chang, 2006)

24

2.7.2 Free Earth Support Method and Fixed Earth Support Method

For the push-in failure, the two methods to analyse these failures are free

earth support method and fixed earth support method. The free earth method assumes

that the embedment depth of the earth retaining wall is allowed to move under the

influence of lateral earth pressure is in the limiting state and the profile of the earth

pressure is as shown in Figure 2.5. As shown in Figure 2.6 the fixed earth support

method assumes that the embedment length of the retaining wall fixed at a point

below the excavation surface. Rotation occurs at this fixed point. Therefore when the

lateral earth pressure on the retaining wall is in the limiting state, the earth pressure

around the fixed point on the two sides of the wall, may not reach the active or

passive pressure, as shown in Figure 2.6. For a cantilever wall, the analysis method

of free earth support is not applicable because in the free earth support method there

is not fixed point at the embedded length of the wall. Therefore the passive and

active forces on the wall are not in equilibrium. As for the strutted wall, the free earth

support method can be applied. The strutted forces on the wall together with the

passive and active forces are in equilibrium. If applying the strutted wall using the

fixed earth support method, the wall depth will be too deep and not economical.

2.7.3 Stress and Deformation Analysis

The stress and deformation due to excavation, occured from either

unbalanced forces or construction defects. The greater the unbalanced forces, the

larger the movement of the soils. Therefore stress analysis is required for the design

of structural component. The deformation analysis is to analyze the wall deflection

and soil movement due to the excavation works so that adjacent buildings are not

affected.

Both the stress and deformation analysis methods for excavation apply

simplified and numerical method. As for the simplified method, it is in the form of

monitoring results of excavation from many case histories in which the results are

sort out into stress and deformation characteristic of wall and soils. These

characteristics are significant to assist in understanding the actual behavior of

excavation. The field measurement results also represent the results of every

25

important element on deformation. Thus, this can help in predicting for similar

excavation projects, in terms of soils condition and construction methods and designs.

a) Deformation of Retaining Wall

b) Distribution of Earth Pressure

Figure 2.5: Free Earth Support Method (a) Deformation of Retaining Wall and (b)

Distribution of Earth Pressure (Chang, 2006).

26

a) Deformation of Retaining Wall

b) Distribution of Earth Pressure

Figure 2.6: Fixed Earth Support Method (a) Deformation of retaining wall and (b)

Distribution of Earth Pressure (Chang, 2006)

27

.

2.7.4 Settlement Induced by the Construction of Diaphragm Wall

Clough and O’Rourke (1990) showed that the ratio of the maximum

settlement (due to the construction of diaphragm walls) to the depth of excavation is

0.15% according to field monitoring results as shown in Figure 2.7. Therefore it is

important to monitor soil settlement within the diaphragm wall so as to protect

adjacent properties.

Ou and Yang (2000) studied the settlement results from a construction site

using diaphragm wall at Taipei Rapid Transit System. The study showed that the

maximum settlement induced by a single panel is 0.05 Ht % ( Ht is the depth of the

excavation) and for several panels the settlement was 0.07 Ht % (Figure 2.8). The

maximum amount of total settlement was 0.13 Ht % and it occurred 0.3 Ht from the

wall. This value of total maximum settlement was less than that of Clough and

O’Rourke’s (1990) envelope (0.15 Ht %). The settlement was not significant at 1.5 -

2 Ht from the wall.

Figure 2.7: Envelope of Ground Surface Settlements Induced by Trenched

Excavation (Clough and O’Rourke, 1990)

28

Figure 2.8: Envelope of Ground Surface Settlement Induced by the Diaphragm

Wall Construction (Ou and Yang, 2000)

2.7.5 Excavation Depth

Ou et al (1993) studied the relationship between the deformation of

excavation and its depth (see Figure 2.9) many case histories showed that that the

deformation of the wall increases with the increased of the depth of excavation. As

for soft clay the deformation of the wall is greater than in sand. The maximum

deformation (d hm )can be calculated as in Equation (2.7)

d hm =(0.2-0.5%) Hc ( 2.7)

Where cH = excavation depth

The upperbound of the d hm value is applied for soft clay and lower bound is for

sand clay.

29

Figure 2.9: Relationship between Maximum Wall Deflections and

Excavation Depths (Ou et al,1993)

2.7.6 Analysis of Ground Surface Settlement Induced by Excavation

To predict the ground surface settlement there are four empirical formulas for

the solution as follows (Chang, 2006):

i) Peck’s method

ii) Bowle’s method

iii) Clough and O’Rourke’s method

iv) Ou and Hsieh’s method

2.7.6.1 Peck’s Method

Peck (1969b) proposed a method to predict excavation – induced ground

surrface settlement based on field measurement. The case studies were at Chicago

and Oslo sites, where he established curves relationship between the ground surface

settlement (dv) and distance from the wall (d) (Figure 2.10).

In this method the soil is divided into three categories which is related to the

soil characteristics as follows:-

30

Type I - sand and soft to still clay

Type II - very soft to soft clay

1. Limited depth of clay below the excavation bottom

2. Significant depth of clay below the excavation bottom but cbb NN <

Type III very soft to soft clay to a significant depth below the excavation

bottom and cbb NN ≥ .

Where bN is the stability number of soil ,is defined as γ ue SH / , where γ is

the unit at of soil, eH is the excavation depth, uS is the undrained shear strength of

soil, and cbN is the critical stability number against basal heave.

Peck’s method had results of case histories before 1969. The wall employed

was mostly steel sheet piles or soldier piles. Therefore the application of these

relation curves is not applicable to all excavation, but at present it is still being used

by engineers to ground surface settlement.

Figure 2.10: Peck’s Method (1969) for Estimating Ground Surface Settlement

31

2.8 Mobilisation of Earth Pressure

In the design of embedded retaining wall, the distribution and intensity of

mobilised earth pressure is important. It will affect the overall stability of the

retaining wall, the bending moments and the prop / anchor forces.

For a cohensionless soil, c’ = 0, and the limiting effective stress σ h’ is

defined as follows:

σ h’ = Ka . σ v’ or Kp . σ v’ (2.8)

σ v’ = � . (z-u) (2.9)

Where z is the depth below ground surface, u is the pore water pressure, � is

the bulk unit weight, Ka and Kp are the active and passive earth pressure coefficients.

2.9 Engineering Properties of Retained Soils

It is necessary to understand the basic concepts of the engineering properties

of soil retained by the earth retaining wall (Macnab, 2002). The properties are

derived both by measurement, some developed and other are derived. However they

form the input data for empirical and analytical methods.

2.9.1 Angle of Internal Friction

The angle of internal friction defines the increase in shear strength of the soil

with increasing confining pressure. It is derived by plotting a series of triaxial tests as

Mohr-Circles diagram. The asymptote of several Mohr-Circles is called the Mohr-

Coulomb envelope (see Figure 2.11). The angle of internal friction is the slope of the

Mohr-Coulomb envelope. It is most obvious in cohesionless soil especially sands and

gravels. For soft cohesive soil such as clay it will approach to zero. The angle of

friction can be obtained from cone penetration tests, or laboratory test derived from

undisturbed sample taken from the field. A correlation between the angle of internal

friction and the standard penetration test is as shown in Table 2.4

32

Figure 2.11: Mohr Circle Diagram (Macnab, 2002).

Table 2.4: A Correlation Between the Angle of Internal Friction and

the Standard Penetration Test ( Winterkom and Fang)

2.9.2 Cohesion

Cohesion is a property exhibited in fine grained soils such as clays and silts.

It is formed by the atomic attractive force tend to create shear strength in the

material with the absence of confining pressure. In the Mohr-Coulomb envelop, the

intercept at the shear strength parameter. It is pronounced in cohesive soil. In soft

clays the ø is almost negligible.

33

Sometimes cohensionless soil will stand vertically when cut. It exhibit the

characteristic of cohesive soil. Actually the sands and gravels are bounded together

by capillary attraction forces because of the presence of moisture content. The

cohesion are weak and this phenomenon is called the apparent cohesion. This type

of cohesion can also be due to the result of particle cementation caused by the

mineralogy or thixotropic action. The thixotropic action is due to the previous high

stress history of the soil. However the apparent cohesion will permit a vertical face

of an excavated cohensionless soil for a short period. The cohesion disappears once

the exposed soil dries out.

2.9.3 Unit Weight of Soil

It is the special volume of soil and expressed as weight per unit volume.

2.9.4 Active Earth Pressure.

Rankine in his theory of earth pressure is defined as a wedge of soil which

would move if not restrained. When a face was cut on soil, a wedge which was

defined by an angle measured from the vertical axis 45° - ø/2 from the toe of the

wall was caused by gravity trying to more upward and downward outlined in

Figure 2.12.

The gravitational force was counteracted by the shear stresses along the

line AB. The unbalance force is the function of the soil weight. Therefore the

coefficient of active earth pressure is derived as in Equation (2.10)

Ka = tan ² ( 45° - ø/2) ( 2.10 )

34

Figure 2.12: Rankine Diagram-Active Pressure (Macnab, 2002)

2.9.5 Passive earth pressure.

Rankine’s theory described that if a force was applied to a face of soil to

make it back into its original position, the force equilibrium would be the

summation of the gravitational load of the failure wedge plus the shear stress along

the line CD ( Figure 2.13). Therefore the coefficient of passive earth pressure is

defined as in Equation (2.11)

Kp= tan ² ( 45° + ø/2) (2.11)

Figure 2.13: Rankine Diagram-Passive Pressure (Macnab, 2002).

35

2.9.6 At-Rest Pressure

The at-rest pressure is function of the weight of the soil. The at rest

condition is when the excavated face of the soil is in place without the use of the

shear strength of the soil. The coefficient of at-rest earth pressure is designated as Ko

is derived approximately by Equation (2.12).

Ko = 1 – sin ø (2.12)

36

CHAPTER 3

RESEARCH METHODOLOGY

3.1 Introduction

The study consists of 4 phases. Firstly was to develop a case study

considering what the aims and objectives. Following that a literature reviews of the

said case study. Secondly, data collection for the said case-study has been carried out.

Thirdly, parametric study has been carried out to validate the analysis with the actual

performance at site. Finally, a conclusion has been for the study. Figure 3.1 shows

the flow chart of the study.

3.2 Technical Literature

The technical literature has been captured from geotechnical journals,

conference papers, previous thesis and relevant text books related to the study. The

literature review described in Chapter 2, highlight the factors to be considered in

deep excavations where the soil-structure interaction plays an important element to

be considered, other than the soil parameter itself. The literature also described the

causes of failure from past case-history, which will be a guide to present and future

designer pertaining to design of retaining wall.

3.3 Data Collection

The data was collected from a professional designer who was involved in the

project, for the construction of a Basement at Jalan Kia Peng. The data collected

were as follows:

37

i) Subsurface soil profile from borehole report, SPT-N value of different

stratigraphy soil layers, ground water profile and laboratory results.

ii) Monitoring data from instrumentation which comprised of inclinometer

readings for measuring the lateral displacement.

iii) Other details such as location map, layout of basement, diaphragm wall

details and layout of instrumentation.

�

3.4 Parametric Study

�

Parametric study has been carried out to correlate the soil-stiffness

parameters with SPT-N values using the Hardening soil model. By back-analysis

these parameters, the horizontal displacement of the wall has been obtained. The

results are compared with the actual displacement at the site. �

�

�

3.5 Analysis of Data

�

The SPT-N value obtained from the soil investigation report has been

analyzed to obtain the average SPT-N value in each soil layers. The SPT-N value is

used to correlate the soil stiffness i.e. at 2000N values, 2500N values, 3000N values

and 3500N (with N is the Standard Penetration Test values) values, in the Hardening

Soil model.

�

From the four models, the results of the lateral horizontal displacement has

been analysed and compared with the instrumentation result. The validation is to

observe which of the four numerical models give good agreement with the actual

performance of the wall. This choosen model gives the soil stiffness ( refE50 �

ref

oedE ��

ref

urE ) that can be adopted to same soil condition. Finally, a conclusion is made from

the whole study of the performance of the earth retaining wall with respect to the

back analysis of this case-study using Hardening Soil model.

38

Figure 3.1: Methodology Flow Chart

METHODOLOGY

Literature Review

• Journal , conference &

papers

• Thesis

Proposal Writing & Presentation

• Introduction

• Objectives of the study

• Literature review

• Research methodology

Data Collection

• Soil investigation report

• Field instrumentation data

• Site layout plan and detail drawing

Parametric Study

• Back analysis the data obtained to evaluate

the input soil parameters E values are in

good agreement with the actual results.

• Input the different E values of soils to the

model.

Study of Field Measurements

Data

• Data obtained is evaluated

Comparison Study

• Evaluate of numerical result and field data.

• Comparison of the E values of the different soil layer.

Conclusion

• Comments and recommendations

Completion of Study and Preparation for Presentation &

Submission of Report

39

CHAPTER 4

CASE STUDY OF DEEP EXCAVATION

4.1 Introduction

The case study is the construction of a cantilever diaphragm wall as the

retaining wall system for the basement. Data obtained from the field borehole test i.e.

the SPT-N values are evaluated and used to correlate with the parameter for soil

stiffness values, 2000N values, 2500N values 3000N values and 3500N values.

These values are used in Hardening Soil model to obtain the values of refE50 , ref

oedE

and ref

urE . The refE50 is equal to ref

oedE and the values ref

urE is equal refE 503 . These

inputs are used in the modeling and the results of lateral displacement are compared

with the actual data from the site. The comparison some how are evaluated whether

good agreement between the modeling and actual values, after evaluating the soil

stiffness values.

4.2 Project Description

The case study is the construction of 30 storey building with 2 storey parking

area. The site is located at Kuala Lumpur in a Kenny Hill Formation. The

excavation work has 8 stages of construction. The retaining wall system is a

cantilever reinforced diaphragm with wall thickness of 600 mm and the total depth

being 17 m. The final embedded length is 10 m. The layout of the site is as shown in

Figure 4.1.

40

7 m excavated area

10 m

diaphragm wall

Figure 4.1: Diaphragm Wall Elavation and Location Plan

4.3 Stratigraphy Profile

The geological map of Kuala Lumpur in Figure 4.2 shows that the site is on

the Kenny Hill Formation. The Kenny Hill formation is from the Permian to

Carboniferous in age. It consists of a series of horizontal interbedded shales, phyllite

Excavation area

KLCC

Inclinometer I 3

Diaphragm wall

LOCATION PLAN

PPPPLANPLAN

41

and quartzite, whereby it has experience the process of extensive weathering giving

the product of thin residual soil followed with a very thick weathered rock layer. In

between, quartz intrusions have made their way but in small thickness (Tan et al,

2007).

From the soil investigation report indicates that the top 6 m consist of

alluvial deposits of very loose to loose silty sand with SPT-N values of 8.

Underneath the layer is the medium dense silty sand with 3 meter thickness, having

SPT-N values of 15. The layer underneath the medium dense silty sand (6 m

thickness) is the dense silty sand of SPT-N values of 60. Underneath the layer is the

hard-strata of Kenny Hill formation with SPT-N value of 100. This hard strata is of

clayey silt material. The soil SPT-N values are as shown in Figure 4.3.

Figure 4.2: Geological Map of Malaysia

42

4.4 Field Instrumentation

Extensive instrumentation equipments were installed at the site. For this study,

the data obtained from inclinometers I-3 is used to verify the results obtained from

back-analysis by the numerical modeling.

The inclinometer I-3 was installed to monitor the lateral displacements during

and after construction. The inclinometer was located behind the wall.

43

Figure 4.3: SPT-N Values Profile

44

4.5 Construction Stage

The stages of construction in the excavation works are as shown in Table 4.1

These sequences of work are simulated in the back analysis. The cantilever wall is

17m length with 10 m embedded into the hard stratum.

Table 4.1: Simulation of Construction Work

Stage Depth (m) of Excavation Work Description

2 1.5 Excavate 1.5 m depth

3 1.5 Installation of 17 m wall

4 2.5 Excavate 1.0 m depth

5 3.5 Excavate 1.0 m depth

6 5.5 Excavate 2.0 m depth

7 6.5 Excavate 1.0 m depth

8 7.5 Excavate 1.0 m depth

9 8.5 Excavate 1.1 m depth

Stage 2

Excavation works to 1.5 m depth

Stage3

Installation of 17 m diaphragm wall

45

Stage 4

Excavation works to 2.5 m depth

Stage 5

Excavation works to 3.5 m depth

Stage 6

Excavation works to 5.5 m depth

46

Stage 7

Excavation works to 6.5 m depth

Stage 8

Excavation works to 7.5 m depth

Stage 9

Excavation works to 8.5 m depth

Figure 4.4: Stages of Construction and Simulation on Hardening Soil Model

47

4.6 Finite Element Simulation

Numerical modeling using the computer-aided program ‘PLAXIS’ was used

to simulate the sequence of stage excavation. The back analysis was performed to

verify the behavior of the wall in terms of horizontal displacement.

4.6.1 2D Modeling

The model was a 2-D analysis under plain strain conditions (Brinkgreve,

2002) utilizing 15 node elements. The model boundary conditions were fixed by

standard fixities whereby the side boundary, the x-direction is fixed and the y-

direction is free to move vertically. The bottom boundary was fixed from movement

in x and y direction. As for the initial stresses, it used gravity loading.

The model, as shown in Figure 4.4, was built as per site construction

sequence, with the removal of elements the same as the stages done on site. At initial

stage, the excavation was to a depth of 1.5 m below ground level followed by the

installation of the 600 mm thick diaphragm wall. This wall was activated as a beam

element in the model. Next the stages of excavation were made by removal of the

excavated element in the model. The processes of removal of the elements were

repeated to a depth of 7 m from the top of the wall (8.5 m below the existing ground

level.

4.6.2 Soil Parameters and Constitutive Model

The soil profile shows loose weathered soil before reaching the hard stratum.

The choice of model is the hardening soil-model. The soil properties were taken from

the laboratory test results. The SPT-N values for each layer of soil were taken as

average value after plotting them as shown in figure 4.3. The data for the soil

parameters are shown in Table 4.2 to Table 4.7.

48

Table 4.2: Soil Parameters for Hardening Soil Model

Symbol Unit S1 S2 S3 S4

SPT’N’ Blows/300mm 8 15 60 100

'c kPa 3 5 8 12

'φ [ ]0 28 30 33 35

urv [ ]− 0.2 0.2 0.2 0.2

Note: S1 soil at 1st layer

S2 soil at 2nd

layer

S3 soil at 3rd

layer

S4 soil at 4th

layer

Table 4.3: Typical Parameters for Hardening Soil Model

Symbol Unit S1 S2 S3 S4

ψ [ ]0 0

0 0 0

xk m/s 1 x 10-6

1 x 10-6

1 x 10-7

1 x 10-7

yk m/s 1 x 10-6

1 x 10-6

1 x 10-7

1 x 10-7

m [ ]− 0.5 0.5 0.7 0.7

refp kPa 100 100 100 100

NC

nK [ ]− 0.5 0.5 0.455 0.426

fR [ ]− 0.9 0.9 0.9 0.9

erRint [ ]− 0.8 0.8 0.8 0.8

49

Table 4.4: Stiffness Soil Parameters for Hardening Soil Model (2000N)

Symbol Unit S1 S2 S3 S4

refE50 kPa 1.60 x 10

4 3 x 10

4 1.2 x 10

5 2 x 10

5

ref

oedE kPa 1.60 x 104

3 x 104 1.2 x 10

5 2 x 10

5

ref

urE kPa 4.8 x 104

9 x 104 3.6 x 10

5 6 x 10

5

Table 4.5: Stiffness Soil Parameters for Hardening Soil Model (2500N )

Symbol Unit S1 S2 S3 S4

refE50 kPa 2.1 x 0

4 3.75 x10

4 1.5 x 0

5 2.5 x 10

5

ref

oedE kPa 2.1 x104

3.75 x104 1.5 x 0

5 2.5 x 10

5

ref

urE kPa 6.3 x 04

1.125x105 4.5 x10

5 7.5 x 10

5

Table 4.6: Stiffness Soil parameters for Hardening Soil Model (3000N)

Symbol Unit S1 S2 S3 S4

refE50 kPa 2.4 x 10

4 4.5 x 10

4 1.8 x 10

5 3 x 10

5

ref

oedE kPa 2.4 x 104

4.5 x 104 1.8 x 10

5 3 x 10

5

ref

urE kPa 7.2 x 104

1.35 x 105 5.4 x 10

5 9 x 10

5

Table 4.7: Stiffness Soil Parameters for Hardening Soil Model (3500N)

Symbol Unit S1 S2 S3 S4

refE50 kPa 2.8 x 10

4 5.25 x 10

4 2.1 x 10

5 3.5 x 10

5

ref

oedE kPa 2.8 x 104

5.25x 104 2.1 x 10

5 3.5 x 10

5

ref

urE kPa 8.4 x 104

1.57 x 105 6.3 x 10

5 10.5 x 10

5

50

The values for the Young’s Modulus were obtained by using a correlation of

the blow count from the SPT-N values as shown in Table 4.4 to Table 4.7. The

Young’s Modulus have been taken with the correlation based on 2000N values,

2500N values, 3000N values and 3500N values respectively.

The wall interface elements erRint of 0.8 was adopted as the wall (concrete cast

in soil) was not subjected to vertical load which will cause the wall to settle relative

to the soil. erRint is the ratio of crtitical state angle of shearing Ø’crit over the angle of

shearing at peak, Ø’peak.

4,6.3 Parametric Study

After a basic model has been created, parametric study is commenced by

simulating the back analysis with the same value for certain input parameters and

changing the value for certain input parameters.

In the back analysis, the soil stiffness has a considerable influence on the

lateral soil movement. The soil stiffness has to be modified in order to reasonably

match the measured ground movement. The soil stiffness adopted in the back

analysis using hardening soil model is tabulated in Table 4.8

Table 4.8: Soil Stiffness Input in Soil Hardening Model

Analysis

No

refE50

(Kpa)

ref

oedE

(Kpa)

refref

ur EE 503=

(Kpa)

1

2

3

4

2000N

2500N

3000N

3500N

2000N

2500N

3000N

3500N

6000N

7500N

9000N

10500N

Note : SPT-N value from soil investigation report.

51

In the simulation, the results from the correlation between the soil stiffness

( refE50 , ref

oedE , ref

urE ) with the SPT-N values are evaluated for its effect on the wall

horizontal displacement. These results are then compared with the actual lateral

deflection results obtained from the field monitoring measurement.

52

CHAPTER 5

ANALYSIS OF RESULTS AND DISCUSSION

5.1 Introduction

The back analysis results presented in this chapter included the lateral

displacement of the soil due to excavation.

5.2 Influence of Soil Stiffness (ref

E50 ,ref

oedE )

As described in the earlier chapter, the influence of soil stiffness ( refE50 , ref

oedE )

is very significant in the lateral movement of the wall. This has been simulated in the

parametric study by taking the values of the individual SPT-N values of the soil layer

multiplying by the factors 2000N, 2500N, 3000N and 3500N respectively.

Table 5.1 : Comparison of the Back Analysis in the Numerical Simulation with the

Actual Performance for Horizontal Displacement at the Final Stage, at

the Soil Surface

Excavation

Stages

Displacement results from different soil stiffness parameters

(m)

Instrument

Data

No I-3(m)

2000 N 2500 N 3000N 3500N

5

6

7

8

9

5.33 x 10-3

14.65 x 10-3

22.03 x 10-3

31.82 x 10-3

44.05 x 10-3

4.96 x 10-3

13.14 x 10-3

19.54 x 10-3

28.64 x 10-3

39.57 x 10-3

4.83 x 10-3

12.62 x 10-3

18.53 x 10-3

27.57 x 10-3

37.23 x 10-3

4.62 x 10-3

12.06 x 10-3

18.12 x 10-3

25.62 x 10-3

35.59 x 10-3

*

*

*

*

40 x 10-3

Note: * no data available

53

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

6.0

6.5

7.0

7.5

8.0

8.5

9.0

9.5

10.0

10.5

11.0

11.5

12.0

12.5

13.0

13.5

14.0

14.5

15.0

15.5

16.0

16.5

17.0

0.000 0.005 0.010 0.015 0.020 0.025 0.030 0.035 0.040 0.045 0.050

Horizontal Wall Displacement (m)

Wa

ll D

ep

th (

m)

stage3

stage 4

stage 5

stage 6

stage 7

stage 8

stage 9

Final Excavation Level

Figure. 5.1: Horizontal Wall Displacement versus Wall Depth

for 2000N

54

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

6.0

6.5

7.0

7.5

8.0

8.5

9.0

9.5

10.0

10.5

11.0

11.5

12.0

12.5

13.0

13.5

14.0

14.5

15.0

15.5

16.0

16.5

17.0

0.000 0.005 0.010 0.015 0.020 0.025 0.030 0.035 0.040

Horizontal Wall Displacement (m)

Wall D

epth

(m

)stage 3

stage 4

stage 5

stage 6

stage 7

stage 8

stage 9

Figure 5.2: Horizontal Wall Displacement versus Wall Depth

for 2500N

55

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

6.0

6.5

7.0

7.5

8.0

8.5

9.0

9.5

10.0

10.5

11.0

11.5

12.0

12.5

13.0

13.5

14.0

14.5

15.0

15.5

16.0

16.5

17.0

0.000 0.005 0.010 0.015 0.020 0.025 0.030 0.035 0.040

Horizontal Wall Displacement (m)

Wall D

epth

(m

)

stage 3

stage 4

stage 5

stage 6

stage 7

stage 8

stage 9

Final Excavation Level

Figure 5.3: Horizontal Wall Displacement versus Wall Depth

for 3000N

56

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

6.0

6.5

7.0

7.5

8.0

8.5

9.0

9.5

10.0

10.5

11.0

11.5

12.0

12.5

13.0

13.5

14.0

14.5

15.0

15.5

16.0

16.5

17.0

0.000 0.005 0.010 0.015 0.020 0.025 0.030 0.035 0.040

Horizontal Wall Displacement (m)

Wa

ll D

ep

th (

m)

stage3

stage 4

stage 5

stage 6

stage 7

stage 8

stage 9

Final Excavation Level

Figure 5.4: Horizontal Wall Displacement versus Wall Depth

for 3500N

57

0

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05

Horizontal displacement (m)W

all

Dep

th (

m)

2000N

2500N

3000N

measured

3500N

Figure 5.5: Comparison of Horizontal Wall Displacement for

2000N, 2500N, 3000N and 3500N with measured field data,

at the final excavation stage

Table 5.1 compares the horizontal displacement at the surface, of the

measured data obtained from geotechnical instrument with data obtained from the

finite element analysis. The comparisons start from the fifth to the final stage of

excavation. The values of horizontal displacement decreases as the soil stiffness

increases. The measured displacement matches the displacement for soil stiffness at

2500N, but actually the profile for 3500N matches the profile for the measured data.

58

The maximum displacement at the surface for the measured data at the end of the

excavation works shows a value of 40 mm. Therefore the maximum translation of

the wall with the height of the wall (7 m) is 0.57 %. For the theoretical value

adopting the 3500N soil stiffness, whereby the maximum horizontal displacement is

35.59 mm, the maximum translation is 0.51%. These values of maximum translation

of the wall both theoretically and actual are categorized under cohesive and firm soils.

The lateral displacement results extracted from the analysis of the wall using

soil stiffness parameters ranges from 2000N, 2500N, 3000N and 3500N are shown in

Table 5.1 and Figure 5.1 to Figure 5.5. By discussing the four concepts of

deformation, analysis based on different soil stiffness (Figure 5.1 to Figure 5.5) show

that the increase in the soil stiffness will result in the decrease of the wall horizontal

displacement. Referring to Figure 5.5 at which the theoretical calculated wall

displacement based on the four different soil stiffness parameter has been plotted

against wall depth and compared to the actual wall lateral movement obtained from

the geotechnical instrument. The profile for 3500N matches with the actual lateral

deflection at the excavation depth at 0.5 m to 7.5 m. The measured values for the

lateral displacement at the excavation depth of 7.5 m onwards, are smaller than the

analysis values. These may possibly indicate that the soil is under higher stiffness

than the adopted stiffness value. At the top surface that is from the ground surface to

the excavation depth of 0.5 m, the curve for the lateral displacement obtained from

the measured data deviate significantly. These may possibly indicate some

disturbance.

Thus the value for the horizontal displacement for soil stiffness 3500N is in

good agreement with the field measurement. This shows that the design and

construction method of the project will be appropriate to similar condition of other

construction sites for prediction of horizontal movement. The soil stiffness parameter

for the individual soil type can be used as a guide to all designers with similar

condition.

59

94.00 96.00 98.00 100.00 102.00 104.00 106.00 108.00 110.00 112.00 114.00 116.00 118.00

20.00

22.00

24.00

26.00

28.00

30.00

32.00

34.00

Shear forcesExtreme shear force -190.24 kN/m

Figure 5.6: Typical profile for computed shear force diagram for the

diaphragm wall

92.00 94.00 96.00 98.00 100.00 102.00 104.00 106.00 108.00 110.00 112.00 114.00 116.00

20.00

22.00

24.00

26.00

28.00

30.00

32.00

34.00

Bending momentExtreme bending moment -792.30 kNm/m

Figure 5.7: Typical profile for computed bending moment diagram for the

diaphragm wall

60

Table 5.2: Extreme bending moment and shear force values for the different soil

stiffness

Soil Stiffness Extreme bending moment

kNm/m

Extreme shear force

kN/m

2000N 792.30 190.24

3000N 738.94 182.49

3500N 728.76 180.96

Table 5.3: Vertical displacement values at final excavation stage for the different

soil stiffness

Soil Stiffness Vertical displacement

(1 x 10-3

m )

2000N 3.95

2500N 3.10

3000N 2.36

3500N 2.25

Figure 5.6 shows that the typical profile for shear force diagram and Figure

5.7 shows the typical profile for bending moment diagram for the diaphragm wall.

Table 5.2 shows the values for the shear force and the bending moment diagram for

the different soil stiffness. Table 5.3 shows the values for the vertical displacement

for the different soil stiffness. From the tables and figures it shows that the values for

the bending moment, shear force and displacement decreases as the soil stiffness

increases. This is a normal phenomenon because the soil strength influences many

aspects of the excavation works. The computed shear and the bending moment

diagram can then be adopted for the design of the wall.

61

CHAPTER 6

CONCLUSION

6.1 Introduction

The simulations of the computer analysis using different soil stiffness

parameter have been correlated from 2000N, 2500N, 3000N and 3500N values. The

results are to compare with the actual performance of the field data.

6.2 Conclusion

Based on the results of the analysis, the following conclusions can be made:

i) From the results of the finite element analysis the obtained lateral

displacement profile shows reasonably close agreement with the

lateral movement obtain from the instrumentation. The soil stiffness

parameters ( refE50 , ref

oedE , ref

urE ) which have been used to correlate based

on 3500N values are suitable for the soil condition.

ii) The maximum obtained translation of the wall system with height of

7 m is 0.57% for the actual wall movement and 0.51% for the

theoretical calculated wall displacement adopting soil stiffness of

3500N. These values of maximum translation of the wall both

theoretically and actual are categorized under cohesive and firm soils

iii) The instrumentation data and the analyses have yielded very useful

information for deep basement construction in terms of the selection

of the soil parameters.

62

iv) The Hardening Soil model which has been adopted in the analysis is

appropriate for Kenny Hill residual soils.

6.3 Recommendations

The result of the study has basically modeled an excavation sequence of work

with soil stiffness parameters matching the actual site performance. The model can

be further improved by varying erRint . This parameter is also significant to simulate

the soil-structure interaction behavior so as to give a more accurate result of wall

movement related to the soil behind it.

63

REFERENCES

Chang Yu Ou (2006) Deep Excavation, Theory and Practice. Taylor and

Francis/Balkema.

Raj Jalla (1999) ) Basement Wall Design for Geotechnical Aspects. Journal of

Architectural Engineering. Pp. 89-91.

Daniel C.Brahana, William M Tanner, E.I.T., and Larry D. Mullins (2007).

Deflection and earth pressure measurements of an anchored concrete shoring

wall. Seventh International Symposium on Field Measurements in Geomechanics.

Karki R (2006) Effects of Deep Excavations on Circular Tunnels in Fine-Grained

Soils. A thesis submitted to the Department of Civil and Geological Engineering ,

University of Saskatchewan, Saskatoon, S. K. ,Canada

Liew S.S (2008). Case Studies of Support of Open Excavation and Distressed

Retaining Wall in Malaysia. Seminar on Excavation and Retaining Wall.

Freiseder M.G. and Schweiger H.F. (1999). Performance of the hardening soil model

for a deep excavation. Beyond 2000 in computational geotechnics- 10 years of

plaxis international @ 1999 Balkema, Rottendam.Pp 263-270

Tan S. A .(2007) Advances and Impact of Numerical Modeling on Excavation Works

in Singapore. Development, Advancement and Achievement of Geotechnical

Engineering in Southeast Asia, 40th

Anniversary Commemorative of the

Southeast Asian Goetechnical Society. Pp 233-242

Liew S.S and Gan S.J. (2007) Back Analyses and Performance of Semi Top-Down

Basement Excavation in Sandy Alluvial Deposits. 16 th

South-East Asian

Geotechnical Conference. Pp. 833-837

James W.C.Sze and Albert N.L.Ho (2008). An Account of Deep Excavation Using

Steelwork in Hong Kong. Seminar on Excavation and Retaining wall (2008).

64

Yee Y.W. and Chua C.G.(2008). Case Studies: Deep Soil Mixing in Excavation.

Seminar on Excavation and Retaining Wall (2008).

Bowles J.E. (1998) Foundation. Analysis and Design. Fourth Edition. Mc Graw-Hill

International Editions.

Law K.H. (2008). Performance of Multi-propped Deep Excavation in Kenny Hill

Formation. A thesis for Universiti Teknologi Malaysia. Unpublished.

Burd H.J.(1999) The History of Plaxis .Beyond 2000 in Computational Geotechnics-

10 years of Plaxis International. A.A. Balkema/Rotterdam/Brookfield

Brinkgrave R.B.J. (2005) . Plaxis 2D. Tutorial Manual Version 9.0

Macnab A (2002). Earth Rretention Systems Handbook. M.C. Graw Hill Handbook.