Fundamentals of Foundation Engineering (2023)

Fundamentals of Foundation Engineering
This book aims to introduce the principle and design of various foundations, covering shallow
foundations, mat foundations, earth retaining structures, excavations, pile foundations, and slope
stability. Since the analysis and design of a foundation are based on the soil properties under
short-term (undrained) or long-term (drained) conditions, the assessment of soil properties from
the geotechnical site investigation and the concept of drained or undrained soil properties are
discussed in the first two chapters. Foundation elements transfer various load combinations
from the superstructure to the underlying soils or rocks. The load transfer mechanisms, vertical
stress or earth pressure distributions, and failure modes of each foundation type are clearly
explained in this book. After understanding the soil responses subjected to the loadings from the
foundation, the design methods, required factors of safety, and improvement measures for each
foundation type are elaborated.
This book presents both theoretical explication and practical applications for readers to
easily comprehend the theoretical background, design methods, and practical applications
and considerations. Each chapter provides relevant exercise examples and a problem set for
self-practice. The analysis methods introduced in the book can be applied in actual analysis
and design as they contain the most up-to-date knowledge of foundation design. This book
is suitable for teachers and students to use in foundation engineering courses and engineers
who are engaged in foundation design to create a technically sound, construction-feasible, and
economical design of the foundation system.
Fundamentals of Foundation
Engineering
Chang-Yu Ou, Kuo-Hsin Yang, Fuchen

Teng, Jiunn-Shyang Chiou, Chih-Wei Lu,
An-Jui Li, Jianye Ching, Jui-Tang Liao
Designed cover image: The Taipei 101, once the tallest building in the world, is
located in Taipei City. Below the surface are thick soft clays with SPT-N values
between 2 and 10, and below the clays are andesite formations. The foundation
must carry the load of the 101-story tower above ground and the 6-story
podium. In this case, pile foundations are used to transfer the load to the rock
formation. The foundation and excavation is designed by Sino Geotechnology,
Inc., Taiwan.
First published 2024
by CRC Press/Balkema
4 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN
e-mail: [email protected]
www.routledge.com—www.taylorandfrancis.com
CRC Press/Balkema is an imprint of the Taylor & Francis Group, an informa business
© 2024 Chang-Yu Ou, Kuo-Hsin Yang, Fuchen Teng, Jiunn-Shyang Chiou, Chih-
Wei Lu, An-Jui Li, Jianye Ching, Jui-Tang Liao.
The right of Chang-Yu Ou, Kuo-Hsin Yang, Fuchen Teng, Jiunn-Shyang Chiou,
Chih-Wei Lu, An-Jui Li, Jianye Ching, Jui-Tang Liao to be identified as authors
of this work has been asserted in accordance with sections 77 and 78 of the
Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this book may be reprinted or reproduced or
utilised in any form or by any electronic, mechanical, or other means, now
known or hereafter invented, including photocopying and recording, or in any
information storage or retrieval system, without permission in writing from the
publishers.
Although all care is taken to ensure integrity and the quality of this publication
and the information herein, no responsibility is assumed by the publishers nor
the author for any damage to the property or persons as a result of operation
or use of this publication and/or the information contained herein.
ISBN: 978-1-032-39498-5 (hbk)

ISBN: 978-1-032-39497-8 (pbk)
ISBN: 978-1-003-35001-9 (ebk)
DOI: 10.1201/9781003350019
Typeset in Times New Roman
by Apex CoVantage, LLC
Contents
Preface xi
About the authors xiii
1 Geotechnical site investigation 1

1.1 Introduction 1
1.2 Plan for subsurface exploration 5
1.2.1 Investigation methods 5
1.2.2 Site investigation layout 5
1.2.3 Depth of subsurface exploration 7
1.3 Borehole exploration 7
1.3.1 Wash boring method 9
1.3.2 Rotary drilling method 9
1.4 Soil sampling methods 11
1.4.1 Thin-walled tube sampler 12
1.4.2 Split-spoon sampler 13
1.4.3 Piston sampler 14
1.4.4 Core sampler 14
1.5 Sampling disturbance 16
1.6 Laboratory test 17
1.7 Field test 18
1.7.1 Standard penetration test (SPT) 19
1.7.2 Cone penetration test (CPT) 21
1.7.3 Vane shear test (VST) 24
1.7.4 Pressuremeter test (PMT) 26
1.7.5 Dilatometer test (DMT) 29
1.7.6 Plate loading test (PLT) 31
1.8 Groundwater investigation 32
1.8.1 Groundwater table and pressure measurement 32
1.8.2 Field permeability test 34
1.9 Geophysical survey 37
1.9.1 Electrical and electromagnetic survey 37
1.9.2 Seismic survey 38
vi Contents
1.10 Reports and interpretation 41

1.10.1 Abstract of investigation plan 41
1.10.2 Basic information collection 42
1.10.2.1 Location 42
1.10.2.2 Regional geology 42
1.10.3 Borings and tests 42
1.10.3.1 Layout and depth of borings 43
1.10.3.2 Field test 43
1.10.3.3 Laboratory tests 43
1.10.4 Investigation results 43
1.10.5 Foundation Analysis 46
1.10.5.1 Settlement 47
1.10.5.2 Soil liquefaction potential assessment 47
1.11 Summary and general comments 48
2 Principles of foundation design 51

2.1 Introduction 51
2.2 Geotechnical analysis method 52
2.2.1 Drained behavior and undrained behavior 52
2.2.2 Effective stress analysis and total stress analysis 53
2.2.3 Stress paths 56
2.2.4 Short-term analysis and long-term analysis 58
2.3 Estimation of design soil parameters 61
2.3.1 Soil strength 61
2.3.2 Soil modulus 68
2.3.3 Soil consolidation 75
2.3.4 At-rest lateral earth pressure coefficient 75
2.4 Type of loads and design loads 78
2.5 Foundation design methods 78
2.5.1 Working stress design (WSD) method 78
2.5.2 Ultimate strength design (USD) method 79
2.5.3 Limit state design (LSD) method 80
2.6 Characteristic soil parameters 81
2.6.1 Concept of the characteristic soil parameter 83
2.6.2 Determination of the characteristic soil parameter 84
3 Shallow foundations 90
3.1 Introduction 90
3.2 Types of shallow foundations 91
3.3 Components of shallow foundation design 91
3.4 Vertical bearing capacity 91
Contents vii
3.4.1 Ultimate bearing capacity 91

3.4.2 Bearing failure modes 92
3.4.3 Terzaghi’s bearing capacity theory 94
3.4.4 General bearing capacity equation 99
3.4.5 Factor of safety and allowable bearing capacity 103
3.4.6 Modification of bearing capacity equations for
the groundwater table 104
3.5 Eccentrically loaded foundations 106
3.5.1 Bearing pressure distributions of eccentrically loaded
foundations 106
3.5.2 Ultimate bearing capacity of eccentrically loaded
foundations 108
3.6 Foundation settlement 111
3.6.1 Immediate settlement 111
3.6.2 Consolidation settlement 116
3.6.3 Allowable bearing capacity considering allowable settlement
(example of saturated clay) 118
3.6.4 Tolerable settlement of buildings 119
3.7 Applications of plate loading tests 122
3.7.1 Estimation of qu 122
3.7.2 Estimation of settlement 122
3.8 Mat foundations 123
3.8.1 Bearing capacity of mat foundations 123
3.8.2 Compensated foundations 123
3.8.3 Stress analysis of mat foundations 125
3.8.4 Piled mat foundations 128
4 Lateral earth pressure 134

4.1 Introduction 134
4.2 Lateral earth pressure at rest 135
4.3 Rankine’s earth pressure theory 140
4.3.1 Active earth pressure 140
4.3.2 Passive earth pressure 142
4.3.3 Earth pressure with sloping ground 144
4.4 Coulomb earth pressure theory 152
4.4.1 Active earth pressure 152
4.4.2 Passive earth pressure 154
4.5 Displacement and earth pressure 157
4.6 Caquot and Kerisel’s solution 158
4.7 Earth pressure due to surcharge 164
4.8 Earth pressure due to earthquakes 166
viii Contents
5 Earth retaining structures 175

5.2 Types of earth retaining structures 176
5.2.1 Classification 176
5.2.2 Wall selection 184
5.3 Gravity and semi-gravity walls 189
5.3.1 Typical dimensions and drainage design 189
5.3.2 Failure modes 191
5.3.3 Force components for stability analyses 192
5.3.4 Evaluation for overturning 194
5.3.5 Evaluation for sliding 195
5.3.6 Evaluation for bearing capacity 197
5.4 Reinforced walls 203
5.4.1 Typical dimensions and drainage design 203
5.4.3 Force components for stability analyses 207
5.4.4 Evaluation for reinforcement breakage 210
5.4.5 Evaluation for reinforcement pullout 211
5.4.6 Evaluation for facing connection 212
5.5 Nongravity walls 217
5.5.1 Types and applications 217
5.5.3 Earth pressure distribution 219
5.5.4 Stability analyses for cantilevered walls 221
5.5.5 Stability analyses for supported walls 223
6 Excavation 241
6.2 Excavation methods 242
6.2.1 Braced excavation 242
6.2.2 Top-down construction 244
6.3 Stability analysis 245
6.3.1 Base shear failure 245
6.3.2 Upheaval failure 260
6.3.3 Sand boiling 261
6.4 Stress analysis 264
6.4.1 Struts 265
6.4.2 Wales 269
6.4.3 Retaining walls 270
6.5 Deformation analysis 275
6.5.1 Wall deflection 276
6.5.2 Ground settlement 277
Contents ix
6.6 Control of excavation-induced movement 279

6.7 Monitoring system 284
6.7.1 Monitoring items 284
6.7.2 Arrangement of the monitoring system 287
7 Pile foundations 295

7.2 Classification of piles 295
7.2.1 Functions of piles 295
7.2.2 Materials 298
7.2.3 Method of installation 300
7.3 Installation of piles and its impact on soils 301
7.3.1 Driven pile 301
7.3.2 Effect of pile driving in clays 303
7.3.3 Effect of pile driving in sands 306
7.3.4 Bored and cast-in-place pile 307
7.3.5 Effect of installing the bored pile 310
7.4 Axial load capacity of piles 311
7.4.1 Load transfer mechanism 311
7.4.2 End-bearing resistance 312
7.4.3 Frictional resistance 315
7.4.4 Design with SPT data 320
7.4.5 Uplift capacity 321
7.4.6 t-z curve 322
7.5 Lateral loaded piles 329
7.5.1 Classification of lateral piles 330
7.5.2 Lateral load capacity for rigid piles 331
7.5.3 Lateral load capacity for nonrigid piles 337
7.6 Pile load test 341
7.6.1 Configuration of pile load test 341
7.6.2 Interpretation of test results 342
7.6.3 Correlation of pile load test results with other in situ tests 345
7.7 Settlement analysis of single piles 345
7.8 Negative skin friction 347
7.9 Group pile effect 349
8 Slope stability 356

8.2 Failure modes 356
8.3 Slope stability analyses 359
8.3.1 Factor of safety in the limit equilibrium method 359
8.3.2 Basic limit equilibrium methods 360
x Contents
8.3.3 Infinite slope 361

8.3.4 Finite slope 364
8.4 Slope stability charts 372
8.5 Stabilization methods 375
Solutions 383
Index 419
Preface
The most important consideration in the design of a structure such as a building, bridge, retain-
ing wall, or geotechnical structure is a stable foundation. An unstable foundation puts the safety
of the entire structure at risk. The analysis and design of a foundation are closely related to
the basic properties of soils, especially their drained and undrained behaviors. The method for
analysis and design introduced in this book focuses particularly on the drained or undrained
behavior of the target soil. Chapter 2 elucidates the concept and analysis methods for soils under
drained and undrained conditions. The concept of drained or undrained properties associated
with necessary soil parameters and analysis methods is also applied to the analysis and design
methods presented in each chapter.
Since a large number of studies related to the analysis and design methods for foundation
have been developed, this book is unable to cover them all. Therefore, for teaching purposes,
only basic principles of analysis and design are provided in this book, followed by an introduc-
tion to their application. If students learn the principles of analysis and design from this book,
then in future practical design work, they can perform a sound analysis or design considering
relevant design codes.
Generally, the method for foundation design includes the allowable stress method, strength
design method, and method considering various limit states. Chapter 2 explains the basic con-
cepts of these design methods. Although the allowable stress method is a mainstream method
in foundation design, the strength design method has been adopted to design upper structures.
Obviously, it is not reasonable to adopt different design methods for the upper structure and
foundation, especially considering the greater uncertainties involved in the soil parameters than
in the parameters of structures. Consequently, an increasing number of country building codes
adopt more rigorous strength design methods and even more advanced limiting state design
methods. The limiting strength design method considers the design at various limiting states, for
example, considering ultimate strength, deformation, or earthquakes.
To make it easy for students to understand the basic principles of basic analysis and design,
each chapter introduces the basic principles of basic design from the perspective of the allow-
able stress method. If readers can understand the concepts of the strength design method and
limit state design method, they will be able to understand the implications better behind the
design codes when they must use them in actual design work.
The topics for foundation design cover a wide range, and relevant studies are becoming
increasingly advanced. It is not easy for a single author to be an expert on every topic. There-
fore, eight distinguished scholars from different fields contribute to the contents of this book.
The book chapters and the contributing authors are arranged as follows. Chapter 1 is about
geotechnical site investigation, written by Dr. Chih-Wei Lu and Dr. Jui-Tang Liao. Chapter 2
xii Preface
is about principles of foundation design, written by Dr. Chang-Yu Ou, Dr. Jianye Ching, and
Dr. Jiunn-Shyang Chiou. Chapter 3 is about shallow foundations, written by Dr. Jiunn-Shyang
Chiou. Chapter 4 is about lateral earth pressure, written by Dr. Chang-Yu Ou. Chapter 5 is about
earth retaining structures, written by Dr. Kuo-Hsin Yang. Chapter 6 is about excavation, writ-
ten by Dr. Chang-Yu Ou. Chapter 7 is about pile foundations, written by Dr. Fuchen Teng. And
Chapter 8 is about slope stability, written by Dr. An-Jui Li. Each author is an expert in the field
corresponding to the book chapter he wrote. All authors have tried their best to make the content
of this book accessible, accurate, and current. All chapters in this book have been reviewed and
approved by all authors.
About the authors
Chang-Yu Ou is a chair professor of the Department of Civil and Construction Engineering

at National Taiwan University of Science and Technology (NTUST), Taipei, Taiwan. He
received his doctoral degree from Stanford University, USA, in 1987. His areas of interest
are deep excavations, soil behavior, soft ground tunneling, and ground improvement. He has
published more than 200 journal and conference papers. He also has published two mono-
graph books regarding deep excavations in English by Taylor & Francis in 2006 and 2022,
respectively, and three deep excavation books in Chinese.
Kuo-Hsin Yang is a professor in the Geotechnical Engineering Program of the Department of
Civil Engineering at the National Taiwan University (NTU). He completed his PhD at the
University of Texas at Austin. His research interests involve slope stability and earth retain-
ing structures, application of geosynthetics, geotechnical engineering modeling, and geo-
disaster engineering. He has many years’ experience in research and practice on the analysis,
design, and case study of slopes and earth retaining structures using both numerical modeling
and physical tests.
Fuchen Teng is an associate professor of the Department of Civil and Construction Engineer-
ing at National Taiwan University of Science and Technology (NTUST), Taipei, Taiwan.
He received his doctoral degree from NTUST in 2011. Before joining the Department of
Civil and Construction Engineering, he worked as a post-doctoral fellow and engineer at
Northwestern University and Sinotech Engineering Consultants during 2013–2016. He is a
member of the ATC6 of the International Society of Soil Mechanics and Geotechnical Engi-
neering. His areas of interest are ground improvement, soil stress-strain-strength behaviors,
soil–structure interaction problems, and geological disposal of rad-wastes.
Jiunn-Shyang Chiou is an associate professor of the Department of Civil Engineering at NTU.
He received his doctoral degree from NTU in 2001. He was a research fellow of the National
Center for Research on Earthquake Engineering, Taiwan. He was a geotechnical engineer
of Sinotech Engineering Consultants Ltd. His areas of interest are geotechnical earthquake
engineering, foundation engineering, seismic design of structure foundations, and soil–
structure interaction analysis. He has published more than 60 journal and conference papers.
Chih-Wei Lu is a professor at NTUST. He completed his Ph. at Kyoto University, Japan. Dr.
Lu worked as a geotechnical engineer in Moh and Associates Inc., experiencing an engineer-
ing practice in foundation design, geotechnical investigation, and soil liquefaction potential
assessment. His research interests involve soil dynamics and soil liquefaction, effective stress
analysis on interaction of superstructure–foundation–soil system, and innovative design for
geotechnical structures.
xiv About the authors
An-Jui Li received his PhD degree from the University of Western Australia in 2009. Before
joining NTUST in 2017 as an associate professor, he worked as a lecturer at Central Queens-
land University and Deakin University in Australia for 7+ years. An-Jui teaches geotechnical
engineering units. His primary area of research includes rock mechanics, excavation, and
slope stability. Currently, he has published 50+ journal and conference articles.
Jianye Ching is a distinguished professor in the Department of Civil Engineering at NTU.
He obtained his PhD in 2002 in the University of California at Berkeley. His main research
interests are geotechnical reliability analysis and reliability-based design, basic uncertainties
in soil properties, random fields and spatial variability, reliability-based geotechnical design
codes, and probabilistic site characterization. He is the author or co-author of more than 100
publications in international journals.
Jui-Tang Liao received his PhD from NTUST. He established Land Engineering Consultant
Co. Ltd. in 1988. He possesses a 35-year experience in engineering and also was an adjunct
associate professor of the Department of Civil and Construction Engineering at NTUST. His
specialties are geological investigation, geotechnical monitoring, and landslide mitigation.
Chapter 1
Geotechnical site investigation
1.1 Introduction
Geotechnical engineering problems vary due to different project types and geological condi-
tions. Thus, purposes and characteristics of construction should be recognized sufficiently to
develop a complete site investigation plan for performing the optimum engineering solution
and thereby assist in solving potential geotechnical problems before, during, and after the con-
struction work. A function of foundation that stabilizes structures is to transfer superstructure
loads to the soil or rock or to stabilize temporary excavations. It is known that not only is risk
of foundation construction higher for subsurface structures than surface structures during the
construction period but also that foundation design expenditures could be considerably reduced
because of an integral site investigation performed for the construction work cautiously.
The scope of foundation engineering covers generally building, underground excavation, road
or railway, dam, slope engineering, etc. The most significant components of a building founda-
tion are permanent substructures and temporary stabilization structures for excavations. There
are diverse temporary stabilization methods for excavations, such as internal bracings, retaining
walls, and ground anchors. After the temporary excavation face is stabilized, a permanent sub-
structure can be constructed, such as a mat foundation, pile foundation, or caisson foundation.
Based on the stable subsurface structure, the subsequent superstructure is constructed afterward.
However, since most foundation projects are underground projects, they are not visibly
exposed. Given underground geological conditions or groundwater conditions, along with
their variable distribution characteristics, to a geotechnical engineer, planning and designing
are essential. The geological survey should first be executed to investigate the geological and
groundwater conditions of different sites in each project and should be used to thoroughly con-
sider the potential problems and possible damage modes that may be encountered in subsequent
construction and even after completion to proactively respond to them.
Providing the best solution for engineering design is the most important task for geotechni-
cal engineers. Factors such as the economy, safety, and feasibility of construction are taken into
account for obtaining the best solution. When determining the solution, a cost-optimal plan
is required for engineering project, but safety should always be prioritized. Safety of differ-
ent conditions, including safety during construction and long-term safety under heavy rain or
earthquake conditions after construction, is required to be analyzed. There are a few examples
of appropriate site investigation not being performed for the geotechnical design, thus causing
damages, shown in this chapter to remind the readers of the importance of site investigation.
Figure 1.1 shows a problem when conducting basement excavation work in soft clay, which
contains a very soft clay layer classified as CL (low-plasticity clay) or CH (high-plasticity clay)
DOI: 10.1201/9781003350019-1
2 Geotechnical site investigation
Figure 1.1 Tilt in the adjacent building due to an inadequate support system during excavation.
by the unified soil classification system, and the standard penetration test SPT-N value is less
than 4. A braced retaining system was adopted. However, an excavation failure occurred, which
caused a serious tilt in adjacent buildings during excavation. The cause of the failure was related
to the insufficient understanding of the soil properties below the surface, resulting in inade-
quate design of braced retaining system. To avoid such disasters, better site investigation must
be carried out first to determine the major design parameters, such as groundwater condition,
Geotechnical site investigation 3
compressibility, and strength of soils. Then, a reasonable stability analysis should be performed,
and a safe braced retaining system should be designed.
Figure 1.2 shows several three-story houses built on loose sandy ground. The soil profile
is classified as SM (silty sand) and SP (poorly graded sand), and the SPT-N value is smaller
than 10. The houses were founded safely until several years after construction. The foundation
of the shallow-founded houses experienced severe soil liquefaction in an earthquake, and the
occurrence of liquefaction-induced lateral spread due to loss of bearing capacity in the adjacent
retaining wall caused permanent inclination of the houses. Soil liquefaction is easily triggered
Figure 1.2 Tilt in building due to liquefaction of loose silty sand (N <10).
in loose sandy soils during seismic motion, in which the effective stress of the soils reduces con-
siderably due to excess pore water pressure. It told us a lesson that a more accurate site inves-
tigation should be carried out to investigate the SPT-N of the soils and groundwater depth, and
laboratory tests of general physical property of soils should be conducted to consider construc-
tion on soil liquefiable sites. To ensure reasonable safety against soil liquefaction, construction
work in seismic hazard zones has proved to be critical.
Figure 1.3 is a case of a hillside community in northern Taiwan. The anchored retaining wall
experienced failure during Typhoon Winnie in Taiwan in August 1997. The site is located on a
dip slope with interbedded sandstone and shale. It is observed that when the downslope force
increased because the rising groundwater became larger than the resistance force, slope sliding
occurred, resulting in instantaneous failure of the anchored retaining wall. Furthermore, the
Figure 1.3 Dip slope slides at Lincoln community (hillside residential community).
sliding rock mass of the dip slope quickly rushed into the houses and led to the damage and tilt
of the house, resulting in 28 persons’ deaths. The failure of the anchored retaining wall is attrib-
uted to the insufficient understanding of geological and groundwater conditions.
The discontinuity type and groundwater conditions of the slope can significantly affect slope
stability. Therefore, this failed case showed that when designing a retaining wall of man-made
slopes, detailed information about the rock type, discontinuity type, strength, and groundwater
level in the site investigation is needed. Then, a reliable stability analysis of the slope and retain-
ing wall can be conducted to ensure slope safety under unfavorable conditions, such as heavy
rain or earthquakes, to prevent serious landslides.
1.2 Plan for subsurface exploration

The design of the building foundation and basement excavation requires detailed background
information for each site, including geological conditions, groundwater conditions, strength of
geotechnical material, and so on. Given the unique geological conditions of each site, as well as
the fact that the characteristics and scale of different engineering projects are not the same, such
as the building height, excavation depth, or condition of the adjacent house, each case poses a
different engineering problem, which needs to be resolved individually in a subsurface explora-
tion plan, readers can refer to relevant reference (Geotechnical Engineering Office, 2017).
1.2.1 Investigation methods

Table 1.1 shows a comprehensive list of investigation methods employed for field investigation,
including data collection, surface geology surveys, subsurface exploration (soil boring and geo-
physical prospecting), groundwater investigation, in situ and laboratory tests, and monitoring.
Generally, the data collection, surface geological survey, and soil boring in subsurface explora-
tion are often executed in a civil engineering project.
Data collection includes collecting aerial photo, topography, and geological information.
Subsurface geological surveys include investigating the landforms, geology, and topography of
surfaces in the field. Both are used to preliminarily develop a geological model of the study area
for filtering potential problems and planning more detailed investigations.
Table 1.1 also lists commonly applied investigation methods in engineering practice. The
timing for the method application is entirely dependent on the scale, complexity, and cost of the
construction project. A detailed field investigation is carried out, from which reliable geological
conditions at the construction site can be captured. Readers can fere to the relevant references
(Welman and Head, 1983).
1.2.2 Site investigation layout

To evaluate the investigation area, the first step is to prepare a site investigation plan. For exam-
ple, when executing a construction project, one of the major commonly employed methods is
soil boring. Several factors should be considered:
1. Investigation area
The investigation area generally covers the entire building site. However, if the area is sur-
rounded by neighboring buildings, a wider area of the buildings should be investigated for
potential damage from construction. In this case, the investigation area will be widened
6
Table 1.1 Engineering types and investigation methods
Investigation method Surface Geophysical Borehole drilling and sampling test Groundwater investigation Field Labo-
survey survey test ratory
test
Engineering types Data Recon- Seis- Electrical Drill- Standard Cone Vane Borehole Lateral Obser- Field Pump- Water Plate
collec- nais- mic and elec- ing and penetra- penetra- shear tel- load vation perme- ing logging loading
tion sance survey tromagnetic sampling tion test tion test test eviewer test well and ability test test test
survey piezometer test
Building Shallow foundation ○ – – – ○ ○ ○ – – – ○ – – – ○ ○

Pile foundation ○ – – – ○ ○ △ – △ △ ○ – – – – ○
Foundation ○ – – – ○ ○ △ ○ – – ○ ○ ○ – – ○
excavation
Underground Buried structure ○ ○ – – ○ ○ △ △ – △ △ – △ – – ○
structure Large-scale ○ ○ ○ ○ ○ – – – – ○ △ ○ – △ – ○
underground
structure
Road or Cut ○ ○ ○ ○ ○ ○ – – – – △ – – – ○ ○
railway Fill ○ ○ △ – ○ ○ ○ – – – △ – – – ○ ○
Tunnel ○ ○ ○ ○ ○ ○ – – △ – ○ ○ ○ △ – ○
Bridge ○ ○ – – ○ ○ – – – ○ △ △ – – ○ ○
Dam Dam foundation ○ ○ ○ ○ ○ △ – – – ○ △ – – ○ – ○
Dam ○ ○ – – ○ – – – – – – – – – – ○
Slope ○ ○ ○ ○ ○ ○ – – △ △ – ○ ○ △ ○ –
Legend: ○, often used; △, sometimes used; –, rarely

to include the entire influential construction area. Additionally, the foundation type of the
neighboring buildings, the geological condition, and the use and material of the pipeline
should be included in the investigation.
2. The number of soil borings
The number of soil borings should be determined to obtain a complete picture of the sub-
surface soil condition in the project. Generally, at least two soil borings are required for an
investigation site and three for a slope area. According to the author’s experience, one addi-
tional soil boring per 600 m2 is deployed, and the total number of soil borings is adjustable
when the site is over 6,000 m2. Notably, appropriate additional soil borings are required for
the site of the variable ground profile.
3. Others
Vital buildings, such as buildings in public places, and the safety of these buildings are par-
ticularly important and therefore require a more detailed site investigation. If the geological
condition of the site is simple and filled with uniform gravel and rock layers, then the inves-
tigation may only need to fit to a minimum requirement. On the other hand, if the geological
condition is complex, the investigation should be executed in more detail.
1.2.3 Depth of subsurface exploration

In an investigation plan, determination of subsurface depth is required. The principle of the
determination is that the investigated depth should be able to provide sufficient information for
examining the safety of the construction work and of the complete structure. In engineering
practice, the minimum boring depth aims at reaching where the net increase in vertical effective
stress Δ - ' under a foundation with depth, as shown in Figure 1.4, appearing limited influence,
0
can be determined by the following rules:
1. Determine the net increase in effective stress Δ - ' under a foundation with depth.
0
2. Estimate the variation of the vertical effective stress - ' with depth.
v
3. Determine the depth (D = D1) at which the effective stress increase Δ - ' is equal to (1/10) q
0
as q equals estimated net stress on the foundation.

4. Determine the depth (D = D2) at which (Δ - ' / - v' ) = 0.05.
0
5. Determine the depth (D = D3) which is the distance from the lower face of the foundation to
bedrock, as shown in Figure 1.4.
6. Choose the smaller of the three depths (D1, D2, and D3) for determining minimum depth of
boring.
7. Explore minimum 3 m below the bearing layer for pile foundation, as shown in Figure 1.4.
1.3 Borehole exploration

Borehole drilling is one of the most commonly used methods to carry out geological investi-
gations and has three purposes: profile determination, soil sampling, and monitoring system
installation. Usually, a few boring holes after being soil-sampled can assist to determine the soil
profile in the investigated area by an experienced geotechnical engineer. The soil boring work,
including boring machines, rods, and samplers, aims to explore different geological materials,
such as soils and rocks. The soil boring methods often employed on plains sites are the wash
boring method combined with the split tube or thin-walled tube method. On slope sites, the
rotary method is used to sample the soils. Moreover, in engineering practice, the wash boring
8
Figure 1.4 Preliminary determination of borehole depth.
method is often combined with the field test method of the standard penetration method (SPT)
using a split-spoon sampler, introduced in the following sections. For clarifying the function of
the penetration methods and the samplers to the readers, this chapter firstly introduces the bore
hole exploration in this section and samplers in a later section, readers can also refer to the refer-
ences (Hvorslev, 1949).
1.3.1 Wash boring method

A fishtail drill bit (see Figure 1.5) in the wash boring method is commonly employed for pen-
etrating soil washing and boring until it reaches the scheduled sampling borehole depth. A split-
spoon sampler or thin-walled tube is then used to sample the soil for different purposes that will
be introduced in the following sections. A schematic picture of wash boring operation in Fig-
ure 1.6 shows that wash boring uses a hollow drilled rod with a sharp chisel, of which water is
pressurized in the drill rod, raised, dropped, and rotated and controlled from the ground. The soil
water slurry is forced up to the ground surface and is circulated into the drill rod. A horizontal
water jet is usually used to avoid direct downward washing that disturbs the sample when being
applied with sampling methods. The sampling interval is usually 1 m or 1.5 m but should be less
than 2 m. Denser intervals are necessarily for soil profiles with significant changes.
1.3.2 Rotary drilling method

Rotary drilling is the use of a sharp rotating drill bit to penetrate through the soil or rock layers in
the ground by cutting or crushing with an exerted continuous downward pressure that changes
with the soil formation, the hardness of the rocks, and the depth. The face discharge of the drill
bit, as shown in Figure 1.7, needs to be maintained sharp in rotating and penetrating process. It
becomes one of the major costs of this work; therefore, the selection of the drill bit type with the
corresponding formation requires wisdom and experience of the contractors and workers. Also,
for keeping the hole clear of cutting and reclamation, and the drill bit lubricated and less heated,
a circulation of fluid is usually applied in this drilling work. A schematic picture of a rotary drill-
ing method can be referred in Figure 1.8
Figure 1.5 Fishtail drill bit.

Figure 1.6 Schematic of wash boring operation.
Figure 1.7 Diamond drill bits of the drill bit.

The same system to wash boring in Figure1.6
Coupling
Drive pipe
Drive shoe
Flush-type casing
Diamond casing shoe

Drill rod coupling
Drill rod
Core barrel
Reamer
Diamond bit
Figure 1.8 Typical rig with wireline coring equipment.
1.4 Soil sampling methods

For obtaining soils in the ground for different design parameters, the corresponding sampling
methods are required. Generally speaking, for the properties of soil strength and compressibil-
ity, an undisturbed soil sample is preferred; for physical soil properties, a disturbed soil sample
which is less costly is commonly used, so that three types of sampling methods and their distur-
bance to the soils are introduced herein.
1.4.1 Thin-walled tube sampler

The thin-walled tube sampler, made of mostly stainless steel, is commonly used to obtain rela-
tively undisturbed samples of cohesive soils for strength, compressibility, and consolidation
behavior. The sampler with a sharpened opening is commonly used, as shown in Figure 1.9,
with a 76.2 mm outside diameter and a 72.9 mm inside diameter, and typically is 700 mm to
900 mm long. The pushing of the sampler should be gentle to keep the soils undisturbed and
the samplers in a regular shape. Wax is used for sealing moisture, and caps are used to close
up the tubes. The thin-walled tube should be delivered to the laboratory in very limited vibra-
tory motion and preserved in a freezer for testing. ASTM D1587/D1587M-15 (2015) can be
referred to herein.
Figure 1.9 Thin-walled tube sampler: (a) illustration, (b) picture, (c) close-up picture.
1.4.2 Split-spoon sampler

Because thin-walled tube samplers have difficulty in penetrating hard soils, such as gravelly
soils, sandy soil, and cemented soils, the thick-walled tube pushed into soils with hammer drops
is often used to sample disturbed soils for general property tests. In the standard penetration test,
the sampler usually has a 35 mm inside diameter and a 15 cm long thick-walled open-drive split
barrel sampler (see Figure 1.10). It consists of a driving shoe, 560 mm split tube, and coupling
part connected to a drill rod. There are 8 small tubes in the split tube, of which representative
ones are shipped to the laboratory for analysis, observation, and test. The application of the
split-spoon sampling method in the standard penetration test is discussed in Section 1.6. Readers
can refer to the reference (ASTM D3550-01, 2001).
Figure 1.10 Split-spoon sampler: (a) illustration, (b) assembly photo, (c) disassembly photo.
1.4.3 Piston sampler

When sampling soils that are very sensitive or that consist of a high fines content, a piston
sampler that consists of a thin-walled tube fitted with a piston can be used, in which the piston
is attached to the length of the rod passing through the sampler head and running inside the
hollow-boring rod. During the sampling process, the sampler is lowered into the borehole with a
piston located at the lower end of the tube. The tube and piston are locked together with a clamp-
ing device at the top of the rods, as shown in Figure 1.11. Hence, the piston is able to prevent
water or loose soils from entering the tube when sampling as well as retain the soil in the tube
when being withdrawn with a vacuum. There are several types of piston samplers with different
dimensions of the tubes; however, the operation principles are similar (Osterberg, 1952). There-
fore, in Figure 1.11, the dimensions of the instruments are omitted.
1.4.4 Core sampler

The core sampler is usually used for sampling stiff to hard clays, silts, and sands with some
cementation, as well as soft rock. The sampler is not suitable for gravelly soils and loose,
cohesionless soils. The frequently employed rotary samplers are triple-tube core barrels with a
removable thin-walled tube, known as a liner, inside an outer tube provided with a cutting bit,
as shown in Figure 1.12. The rotary sampler has a diameter of 50.8 mm, namely, NX, which is
often used for both plains and slopes, and 63.5 mm, namely, HQ, which is often used for slopes,
with a length of 61 cm. The outer tube with the cutting bit is rotated and pushed into the soil to
Figure 1.11 Sampling pr ocess by piston sampler.

Source: Redrawn from Osterberg (1952).
Figure 1.12 Triple-tube core barrel.
Table 1.2 Common borehole sizes
Core barrel type Hole diameter Core diameter Core sample

(Sample preservation with labeled core box)
NX 75.8 mm 50.8 mm
HQ 96.3 mm 63.5 mm
the required depth, and the sample enters the liner. The inner tube, that is, the liner, provided
with a smooth cutting shoe, remains stationary, and the sample cut by the rotating outer tube
slides into the liner. The rock core sample is thus collected in the liner. The rock core is reserved
in the core box labeled with the project name, date, and sample depth for further geological
investigation, as shown in Table 1.2.
Table 1.3 RQD classification index
RQD (%) Rock mass quality
<25% Very poor

25–50% Poor
50–75% Fair
75–90% Good
90–100% Excellent
After boring and sampling, a method known as rock quality designation (RQD), which uses
the core sample to interpret the geological status of the area to measure the quality of a rock core
taken from a borehole, is usually conducted. RQD signifies the degree of jointing or fracture in
a rock mass measured as a percentage, where an RQD of 75% or more indicates good-quality
hard rock and less than 50% indicates low-quality weathered rocks, as shown in Table 1.3. It
should be reminded to the readers that RQD is defined as the percentage of intact drill core
pieces longer than 10 cm recovered during a total core run.
1.5 Sampling disturbance

When pushing samplers introduced in Section 1.4 into the ground to withdraw soil, distur-
bances to the soil samples are unavoidable. When using the thin-walled tube method, inside
wall friction is one of the principal causes of disturbance of the sample (Hvorslev, 1949). The
thin-walled tube method is one way to reduce or eliminate wall friction between the soil and
the sampler by providing inside clearance by making the diameter of the cutting edge slightly
smaller than the inside diameter of the sampler tube. The area ratio can be expressed as:
d 02 - di2
Ar = (1.1)
di2
Where d 0 = outside diameter of the sampler tube.
di = inside diameter of the sampler tube.

Ar = area ratio (if Ar is smaller than or equal to 10%, the samples are considered undisturbed).
An example is shown to calculate the area ratio in Figure 1.12:
In the case of the thin-walled tube method, the outside diameter is 76.2 mm, and the inside
diameter is 72.9 mm. According to eq. 1.1, Ar = 9% < 10%. (Undisturbed sample)
In the case of the split barrel method, the outside diameter is 51 mm, and the inside diameter
is 35 mm. According to eq. 1.1, Ar = 112% >> 10%. (Disturbed sample)
Notably, although a thin-walled tube is able to withdraw undisturbed soil samples, the trans-
portation of the samples to the laboratory is also a key factor to keep the samples undisturbed.
Table 1.4 lists the types of laboratory tests applicable to disturbed and undisturbed soil
samples.
Table 1.4 Test for disturbed samples and undisturbed samples
Disturbed samples Undisturbed samples

(Clayey soil and fine-grained soils)
Sieve analysis Shear strength

Liquid limit and plastic limit Compressibility
Density Permeability
Soil classification –
Table 1.5 Common laboratory tests for soils
Laboratory tests Soil property
General physical property tests
Water (moisture) content test ω, γ sat , γ d

Specific gravity test Gs
Atterberg limits test LL, PL, PI, SL
Soil classification
Sieves and hydrometer analysis Soil particle size distribution curve
Soil classification
Relative density test γ d max , γ d min , Dr
Permeability tests
Constant head permeability test k (for coarse-grained soil)

Falling head permeability test k (for fine-grained soil)
Triaxial permeability test k (for fine-grained soil)
Compressibility tests
Oedometer test - c', OCR, C c , C ur
Shear strength tests
Direct shear test c′, 0 '

Triaxial compression test
Triaxial consolidated drained (CD) c′, 0 '
Triaxial consolidated undrained (CU) c′, 0 ' (with pore water pressure measurements)
Triaxial unconsolidated undrained (UU) su
Triaxial unconfined compression test (UC) q u, s u
1.6 Laboratory test

A laboratory test is essential to explore the mechanical and physical properties of soils in
the field, and engineers are then able to provide a correct estimate of the engineering per-
formance of the subsurface soil. Thus, practitioners should make good use of laboratory
tests. Table 1.5 summarizes the laboratory tests and their applicability. Test procedures that
are key to ensuring test quality should follow the corresponding specifications, which are
beyond the scope of the book. The relevant reference (Casagrande and Fadum, 1940) can
be referred to reader.
Symbols used in Table 1.5
ω (%) water content γ d max (kN/m 3)

ma ximum dr y unit c′ (kN/m 2)effective
weight cohesion
γ sat (kN/m 3) saturated unit γ d min (kN/m 3) minimum dry unit 0' effective
weight weight friction angle
γ d (kN/m 3) dry unit weight Dr (%) relative density q u (kN/m 2) unconfined
compression
strength
Gs specific gravity k (m/s) permeability s u (kN/m 2) undrained shear
strength
LL liquid limit - c' (kN/m 2) preconsolidation
stress
PL plastic limit OCR overconsolidation
ratio
PI plasticity Index C c compression index
SL shrinkage limit C ur unload–reload
index
Table 1.6 Common laboratory tests for rocks
Laboratory tests Applicability
Rock physical properties tests Obtains unit weight, moisture content ( γ m )

Point load strength test Provides an index of uniaxial compressive
strength (q u)
Unconfined compressive strength of intact Uniaxial compressive strength (q u)
rock core
Direct shear test for weak rock c′ , 0 '
1.7 Field test

Field tests refer to all the in situ tests being executed during soil boring exploration. It includes
the standard penetration test (SPT), cone penetration test (CPT), vane shear test, pressuremeter
test, flat dilatometer test, and plate loading test. SPT is capable of withdrawing soil samples for
laboratory tests, while CPT is capable of measuring continuous soil properties during penetra-
tion. In situ field tests have inherent advantages that are able to avoid disturbances of soil sam-
pling. As mentioned in Section 1.5, undisturbed soil samples can be obtained through cautious
sampling procedures; however, great attention must be paid to sample transportation and the test
procedure in the laboratory. Much valuable information on soil behaviors can be provided, such
as undrained and drained soil strength, stress path, and stress‒strain responses of soils in the
laboratory test. Therefore, both field tests and laboratory tests are required to explore the precise
geological information of the site.
Although a field test can measure the properties of in situ soils, it is usually unable to derive
the soil parameters directly that are required in a foundation design, such as the friction angle,
undrained shear strength, Young’s modulus, and so on. Therefore, some transformation equa-
tions or empirical formulas are necessary. Section 2.3 provides several useful transformation
equations that have been published in literatures for readers’ reference.
1.7.1 Standard penetration test (SPT)

The equipment is shown in Figure 1.14 and comprises a split tube with a driving head and either
a heavy-duty cutting shoe or a solid cone point. The head of the tube is threaded for connection
(via a series of drill rods) to a hammer. The device is driven into the ground at the base of the
borehole with a free-falling 63.5 kg trip hammer dropping to a distance of 760 mm. The hammer
and rods should be vertical, and therefore, the borehole should be plumb.
During the test, a count is made for the number of blows required to drive it though the initial
150 mm and a subsequent 300 mm test zone. The initial 150 mm is considered to be in distrib-
uted material. The recording of the blow count of the test zone is becoming widely accepted and
is strongly recommended. The device is driven into the ground at the base of the borehole with a
free-falling 63.5 kg trip hammer dropping a distance of 760 mm. The soils pushed into the split
tube introduced in Figure 1.10 during the penetration are for general physical property soil test.
Tables 1.7 and 1.8 are a general relationship between SPT-N value and the strength of clay or
the relative density of sand.
Figure 1.13 Calculation of the area ratio Ar .

Figure 1.14 SPT in operation.
Table 1.7 Relationship of the consistency of clay, SPT-N value, and unconfined compressive
strength qu >.
SPT-N value (blows) Consistency Unconfined compressive

strength q u (kPa)
<2 Very soft <25

2~4 Soft 25~50
4~8 Medium 50~100
8~15 Stiff 100~200
15~30 Very stiff 200~400
>30 Hard >400
Source: Terzaghi et al. (1996).
Table 1.8 Relative density of sands according to SPT-N results
SPT-N value (blows) Relative density
0~4 Very loose

4~10 Loose
10~30 Medium
30~50 Dense
>50 Very dense
Source: Terzaghi et al. (1996).
Due to variability in local practice and equipment, the energy efficiency of the drop hammer onto
the drop rods varies from region to region. Moreover, the overburden stress also affects the SPT-N
value significantly. The SPT-N values obtained should be corrected. The relevant correction and
transformation equation for the parameters that are required for design can be found in Section 2.3.
1.7.2 Cone penetration test (CPT)

The cone penetration test (CPT) is also a commonly used field investigation method in engineer-
ing practice. A steel cone penetrates the ground for soil testing similar to the SPT, and data on
tip resistance, side friction, and pore pressure during steady penetration is acquired for profiling
continuous soil layers along the depth and estimating the soil strength, soil classification, and
pile bearing capacity.
The configuration of the steel cone in the CPT, with an apex angle of 60° and cone bottom
cross-sectional area of 10 cm2, as shown in Figure 1.15, has been standardized in recent years
Figure 1.15 Illustration of CPT.

(David, 2018). Several empirical or transformation equations and figures to correlate design
parameters have been developed, as discussed in Section 2.3.
Figure 1.16 demonstrates the work procedures of the CPT, in which the truck weight is
employed to provide a reaction. Most commercially available CPT rigs operate electronic fric-
tion cone and piezocone penetrometers, whose testing procedures are outlined briefly in this
book. These devices produce a computerized log of tip and sleeve resistance, the ratio between
the two, induced pore pressure just behind the cone tip, pore pressure ratio (change in pore
pressure divided by measured pressure with a penetration speed of 2 cm/s), and lithologic inter-
pretation of each 2 cm interval, which are continuously logged and printed out. A guideline for
evaluating soil types can be summarized: sands are identified when qc > 5 MPa and u2 ≈ uo, and
Figure 1.16 CPT in operation: (a) setting the CPT truck; (b) forcing the cone into the soil at
equal rate (20 mm/s).
the presence of intact clays is prevalent when qc < 5 MPa and u2 > uo (hydro). The magnitude
of pore water pressures can be helpful to indicate intact clays such as soft (u2 ≈ 2·u0), firm (u2 ≈
4·u0), stiff (u2 ≈ 8·u0), and hard (u2 ≈ 20·uo). Fissured overconsolidated clays tend to have nega-
tive u2 values such that u2 < 0. As meaning of all the symbols can be listed as followings:
Tip resistance, qc, is theoretically defined as the ratio of penetration force.

Sleeve friction, fs, is measured by tension load cells embedded in the sleeve as the probe is
advanced through the soil.
The hydrostatic pressure, u0.
The pore pressure, u2, is measured during penetration.
Compared with the SPT, the advantage of the CPT is that it is capable of acquiring continu-
ous data on soils (see Figure 1.17), and the disadvantage is not being capable of sampling soils.
In engineering practice in many countries, the SPT is still the main method employed in soil
investigations, for two main reasons: SPT data are more abundantly collected, and the SPT labor
cost is relatively lower than that of CPT trucks in many countries. In Figure 1.17, clayey soils
are indicated by low tip resistance and high pore pressure, where qc < 5 MPa and u2 > uo, such
as the depths of 2–6 m and 10–17 m; on the other hand, high tip resistance and low pore pres-
sure (qc > 5 MPa and u2 ≈ uo) indicate sandy soils or sand mixtures, with the depths of 8–10 m.
Figure 1.17 Typical CPT results.

1.7.3 Vane shear test (VST)

The vane shear test is a method for determining the undrained shear strength of soils, especially
soft clays, in the field, while undisturbed soil sampling by the thin tube method is not easy to
perform.
The vane shear test (VST) consists of a four-bladed vane, which is typically approximately 55
mm or 65 mm in diameter and 110 mm or 130 mm in height, with a height-to-diameter ratio of
2, depending on the test purpose, as shown in Figure 1.18. A rectangular bladed vane or tapered
Figure 1.18 Variable types of four-bladed vane: (a) rectangular vane (b) to (d) tapered vane.
bladed vane with longer height is used in soft clay, while a tapered bladed vane with shorter
height is used in harder clay. The vanes of the instrument are placed into a borehole using a
suitably stiff rod until it reaches the desired depth. Then, the vane is gradually lowered into the
undisturbed soils until the top of the vane is 50 cm into the soils. The vane is rotated using torque
applied at the handle at a rate of 0.1º per second, until the soils reach the failure state. A picture
of vane shear test performed in field is seen in Figure 1.19.
Eq. 1.2 assumes that the undrained shear strength (su,VST) of the soil is constant on the cylin-
drical sheared surface and at the top and bottom faces of the sheared cylinder, as shown in
Figure 1.20. The torque applied ( M max ) equals the sum of the resisting torque at the sides and
Figure 1.19 Performing a vane shear test: (a) illustration of vane shear test; (b) applying torque
by twisting the crank at constant speed and recording the rotation angle θ and
torque until the soils fail. Then, the max torque Mmax can be determined.
Figure 1.20 Definition of the parameters of the undrained shear strength formula.
that at the top and bottom. Thus, the shear strength of a given soil sample is calculated with the
following formula.
M max
su, VST = (1.2)
{ - D h } { - D3 }
2
| |+| |
{ 2 } { 6 }
Where:
su,VST = undrained shear obtained by VST (kN/m2).
M max = max torque applied on the rod (kN-m).
D = diameter of the vane (m).
h = height of the vane (m).
The undrained shear strength obtained from a vane shear test depends on the rate of applica-
tion of torque, which is, in turn, related to the plasticity of soils. Therefore, the obtained su,VST
must be corrected. The relevant correction can be found in Section 2.3.
1.7.4 Pressuremeter test (PMT)

A pressuremeter, also named the borehole lateral load in situ test, is a device performed by
applying pressure to the sidewalls of a borehole and recording the corresponding deformation.
It was designed by Menard (1956) for determining the stress‒strain relations of in situ soil at a
certain depth in a borehole. By using these stress‒strain relations, one can determine the elastic
modulus, preconsolidated stress, friction angle, undrained strength, soil pressure of the remain-
ing soil, etc.
Pressuremeters enter the ground by pushing into a pre-boring hole at a depth at which the
probe should be placed. Then, increments of pressure by gas or water are applied to the inside of
the membrane, forcing it to press against the material and loading a cylindrical two-dimensional
state of stress. The probe of the pressuremeter, which is the major testing unit, consists of three
inflatable cells, one of which is above the other, as shown in Figure 1.21. The middle cell is the
measuring membrane cell, which is filled with water during the test and has a diameter of 58 or
74 mm and a height of 203 or 184 mm. The other two cells, which are at the top and bottom of
the mid cell, are guard cells that protect the main cell from the end effects caused by the finite
Figure 1.21 (a) PMT in operation, and (b) illustration of the pressuremeter test.
length of the cable (so-called end effects) to reduce testing errors. Relevant information can be
referred to Baguelin et al. (1978) and Clarke (1994).
A self-boring pressuremeter test, namely, the SBPMT, is developed to overcome testing dif-
ficulties when experiencing frequent borehole collapse in soft soils or expansions in expansive
soils before pressuremeter placement.
Figure 1.22 presents a typical pressuremeter curve to estimate Young’s modulus, preconsoli-
dation stress, internal friction angle, undrained shear strength, static earth pressure, etc. that can
be found in Menard (1956). The formula used for determining Young’s modulus is shown as an
example in eq. 1.3.
EP = 2 + ( K D / 1.5)0.47 - 0.6 (1.3)
or
EP = 2(1+y )(V0 + Vm )(-P / -V )
Where:
EP = pressuremeter modulus.
KD = horizontal stress index (KD = (P0- u0)/- v'0).
γ = Poisson’s ratio (generally taken as 0.3 for pressuremeter applications).
V0+ ΔV = volume of probe.
ΔV = volume increase in the straight-line portion of the test curve.
ΔP = pressure increase corresponding to ΔV volume increase.
Figure 1.22 Typical pressuremeter test result.

Source: Redrawn from Rashed et al. (2012).
1.7.5 Dilatometer test (DMT)

The flat dilatometer test (DMT) was developed in Italy by Silvano Marchetti (1980) and is a
push-in type in situ test that is quick, simple, economical, and highly reproducible for explor-
ing ground profile, soil property, and modulus values by applied displacement and measured
pressure. The device consists of an arrangement for lateral expansion after reaching the desired
depth. This test uses a stainless steel blade with an 18° wedge, 240 mm length, 95 mm width,
and 15 mm thickness. The blade has a 60 mm diameter circular steel membrane mounted flush
on one side.
The blade in Figure 1.23(a) is connected to a control unit on the ground surface by a pneumatic-
electrical tube (transmitting gas pressure and electrical continuity) running through the insertion
rods. A gas tank, connected to the control unit by a pneumatic cable, supplies the gas pressure
required to expand the membrane. The control unit is equipped with a pressure regulator, pres-
sure gage, audiovisual signal, and vent valves.
First, pushing the blade into the ground against the reaction fame. Upon reaching the desired
depth, the gas pressure was applied through a nylon tube connected to the surface assembly with
control units. The lift-off pressure P0 required to move the center of the membrane in line with
the blade was measured approximately 15 sec after the blade reached the desired depth. Fig-
ure 1.23(b) shows that the pressure P1 is the pressure required to move the center of the mem-
brane by 1.1 mm into the soil and is recorded within 15 to 30 sec after the last reading. Upon
reaching P1, the membrane is quickly deflated, and the blade is pushed to the next test depth.
When DMT is performed inside a borehole, the diameter of the borehole should be kept minimal
(100–120 mm) to avoid buckling of the drill rod. The test is suitable for clays, sand, and silts.
Figure 1.23 Structure of (a) the sensor, (b) mechanism of operation.

Source: Redrawn from Marchetti et al. (2001).
Although not suitable for gravels, the blade is strong enough to pass through a gravel layer of
0.5 m thickness. By using field record and data reduction P0 (corrected first reading) and P1 (cor-
rected second reading), one can determine the coefficient of earth pressure at rest, as shown in
the following (others can be found in Silvano Marchetti (1980)). Figure 1.24 presents a typical
dilatometer test result with parameter KD; furthermore, K0, DMT is able to be obtained by eq. 1.4.
The coefficient of earth pressure in situ can be determined by the formula:
K0,DMT = (KD/1.5)0.47-0.6 For ID <1.2 (1.4)
Where:
K0,DMT = coefficient of earth pressure at rest, determined by the dilatometer test.
KD = horizontal stress index (KD = (P0- u0)/- ' ).
v0
- ' = the initial effective stress and u0 is the initial pore water pressure.
v0
P0 = the reading of the dilatometer test.

ID = the material index (ID = (P1- P0)/(P0- u0)).
Figure 1.24 Typical dilatometer test result.

Source: Redrawn from Marchetti et al. (2015).
1.7.6 Plate loading test (PLT)

The plate load test (PLT) is designed to determine the vertical deformation and strength char-
acteristics of soil by assessing the force with time when a rigid or flexible plate is made to act
on the soil vertically. Then, the vertical plate load test can be used to evaluate the ultimate bear-
ing capacity, shear strength, deformation parameters, and vertical subgrade reaction of the soil
beneath the plate. If the plate is designed to act horizontally, the horizontal subgrade reaction of
soils can be determined. Testing can be executed at the ground surface, in pits, or in trenches.
The isotropic elastic continuum of the soils is hypothesized, and the stress and deformation
information of the soil mass are approached in the plate load test.
The coefficient of the vertical subgrade reaction coefficient kv is defined as the ratio of the
uniform loading pressure exerted q to the plate displacement s, which is kv = q /s . Importantly,
since elastic theory is used to derive kv, kv is valid only under small levels of deformation, varies
with the depth of the soils, and is affected by the size of the plate.
Figure 1.25 shows an example of a vertical plate loading test in operation. When the PLT
is used to evaluate the kv in the gravel layer, it is necessary to adjust the size of the plate to
Figure 1.25 Plate vertical loading test in operation.

Cohesive soil
Figure 1.26 Typical plate loading test result.

Source: Redrawn from Anjan (2019).
approach the real value of kv for the foundation design. Attention is also paid on the typical
performance of plate load test on different soils in Figure 1.26 for finding the best test result in
the plate load test.
The results from a plate load test can be used to estimate the bearing capacity and settlement of
a foundation. Section 3.7 introduces the application of the plate load test to the foundation design.
1.8 Groundwater investigation

Groundwater, the major disaster-causing factor in geotechnical engineering, undoubtedly affects
the stability of construction; therefore, it is a critical work factor in field investigations. Due
to geo-structural inhomogeneity, discontinuities, or fissures, groundwater seepage is complex
because of the characteristics of multiple aquifers (aquifers, confined aquifers, confining layers,
subsurface perched water, etc.) that result in difficulties in field investigation. For a better under-
standing of the profile of groundwater variation, integrated investigation methods are necessary.
This section introduces frequently employed methods, including groundwater table observation
and water pressure monitoring and in situ permeability testing.
1.8.1 Groundwater table and pressure measurement

The groundwater table changes with the season and has a relevant relationship with rainfall. The
groundwater table depth, water pressure profile, relationship between groundwater table and
rainfall in the time domain, extreme conditions due to the groundwater table rise, groundwater
influence on slope stability, and path of groundwater seepage are all critical in a field investiga-
tion of groundwater in geotechnical engineering.
The following paragraphs introduce two frequently employed methods in engineering prac-
tice: observation wells used for observing the groundwater table and piezometers used for moni-
toring water pressure in sandy or aquifer layers at different depths.
Observation wells are often constructed with a 50 mm diameter PVC opening with holes and
screened with a filter pack or geotextile in completed boreholes, and the depth of groundwater
table is measured in the PVC, as shown in Figure 1.27. Generally, the groundwater table meas-
ured in the PVC of observation wells represents a mixed groundwater table along its penetration
depth; therefore, observation wells are more often employed in a single aquifer.
Open standpipe piezometers are typically installed with a solid casing down to the depth of
interest and a slotted or screened casing within the zone where water pressure is being meas-
ured. Above and below the depth of interest, the water is sealed into the borehole with clay or
bentonite to prevent the groundwater supply at other depths, as shown in Figure 1.28.
For monitoring groundwater pressure in the piezometer as well as the groundwater table in
the observation well in modern days, data can be obtained manually as well as automatically,
Figure 1.27 Illustration and photo of the water observation well and the pipe (measuring the
water level of a single aquifer).
Figure 1.28 Illustration and photo of the piezometer well and the piezometer (measuring the
water level of any aquifer).
and pressure gauges (transducer) can be vibrating-wire, pneumatic, or strain-gauge in opera-

tion, converting pressure into an electrical signal. These piezometers are cabled to the surface,
where they can be read by data loggers or portable readout units, allowing for faster or more
frequent reading. This capability can be a critical component in a warning system. Importantly,
however, a manual reading of the groundwater table is necessary for confirming the accuracy of
the installed automatic instrument.
Figure 1.28 presents an illustration of the piezometer wells and a picture of the piezometer.
The bentonites in the borehole in Figure 1.28 were used for sealing water from other soil layers.
Bentonites swell and form a semisolid, gel-like mass after mixing with water because of their
special property, which can fill voids and limit water movement to create barriers between the
soil and water.
1.8.2 Field permeability test

The soil permeability, also termed hydraulic conductivity, of soils in the field can be measured
using several methods that include constant and falling head laboratory tests on undisturbed
soil specimens sampled from the field in principle; however, in reality, the permeability of fine
soils in laboratory testing is determined through triaxial test indirectly. Alternatively, perme-
ability may be measured in the field by field constant head or falling head tests, or pumping
tests. Also, the permeability may also be empirically obtained through the grain size distribu-
tion in engineering practice for a preliminary design purpose. A good knowledge of soil per-
meability is needed for estimating the quantity of seepage and designing drainage systems in
construction site planning.
1. Permeameter test
The permeameter test measures soil permeability at the bottom of a borehole, as shown in Fig-
ure 1.29. The test measures the hydraulic conductivity by temporarily filling water into PVC in
the borehole to create an unbalanced pressure head between water inside PVC and in soils at the
bottom of the borehole with a constant head for sand, as in Figure 1.29(a), and gravel and a fall-
ing head for silt and clay, as in Figure 1.29(b). Based on the relationship of time and filling water
in PVC, field permeability can be measured. The equations for the permeability of the constant
head and falling head of the in-field test are similar to those of laboratory tests in soil mechanics
books, but with horizontal seepage. In Figure 1.30, hc shows the constant head, h0 shows the
initial head of time t0, and h1 shows the initial head of time t1. Figure 1.30 also illustrates that
the tests in boreholes that only involve a relatively small volume of soil or rock around the test
section provide “small-scale” values of permeability in the field. If the soil/rock is heterogene-
ous or has significant fabric, such tests may not be representative of the mass permeability of
the strata. Large-scale tests (such as pumping tests) may provide better results. Readers can also
refer to the reference (Hvorslev, 1951).
2. Pumping test
A large-scale pumping test contains pumping wells and observation wells, in which a pumping
well with a diameter larger than 30 cm and an observation well with a diameter larger than 15
cm are usually installed. The principle of a pumping test in Figure 1.31 is that if water from the
well is pumped and the discharge q is recorded, the drawdown and in-observation wells h1 and
h2 are measured, and the hydraulic conductivity (k) and other parameters can be estimated based
on the measured drawdown curve and pumping discharge in the soil mechanics textbook.
Figure 1.29 Illustration of borehole permeability tests: (a) constant head test, (b) falling head
test.
Figure 1.30 Borehole constant head test in operation.
Figure 1.31 Illustration of the pumping test for the unconfined problem.
Generally, the observation period of a large-scale pumping test is longer than 24 hr. The large-
scale pumping test result can be used for a workability evaluation of underground drainage work,
such as horizontal drains, vertical pumping wells, or catchment wells. However, readers should
keep in mind that overpumping at the site may cause adjacent buildings to tilt and settle, which is
avoidable if the knowledge in the section is understood. Herein, a pumping test only in an uncon-
fined layer is introduced, and the test in the confined layer may be interesting to some readers in
relation to depressurization of confined aquifers, as referred to in the soil mechanics textbook.
1.9 Geophysical survey

Geophysical surveys are implementations of geophysical methods, which include electrical and
electromagnetic surveys, seismic surveys, gravity surveys, magnetic methods, and radioactiv-
ity methods, to determine the geological, structural, physical, and mechanical characteristics of
the foundation soil with airborne, marine, and land geophysical surveys and borehole surveys.
These methods are used to explore the layout, thickness, and properties of individual layers
below the terrain surface on which the construction of a specific structure is planned. Nonde-
structive testing methods have the merits of being quick, having lower cost, and offering con-
tinuous data logging and more comprehensive exploration of the site (Tokimatsu and Uchida,
1990; Griffiths and King, 1981; Gunn et al., 2015). This section will focus on electrical and
electromagnetic survey and seismic survey methods.
1.9.1 Electrical and electromagnetic survey

Electrical resistivity, which measures the resistivity profile of layers, has been used widely in
prospecting for groundwater, bedrock features, faults, buried objects based on conductivity dif-
ferences of various cements, minerals, porosities, water contents, etc. of geological materials.
2D electrical resistivity tomography (ERT) can be used for mapping the distributions of the
presence of groundwater and geological structures according to the resistivity of materials.
Electrode array configurations are aimed at gathering data used to estimate lateral and vertical
variations in ground resistivity values. An example of the results of implementing 2D electrical
resistivity tomography (2D ERT) is shown in Figure 1.32. The ground resistivity detection line
Figure 1.32 Electrical and electromagnetic survey operation.

Figure 1.33 An example of a 2D ERT survey result.
is approximately 50 m long, and the detection depth range is approximately 20 m. The ground
resistivity image profile detection data is presented to obtain the distribution of apparent resis-
tivity (unit: Ohm-m) (see Figure 1.33). After reading the ground resistivity profile image, the
survey area can be approximately divided into two layers. Layer 1: 0~3 m below the ground
surface, with a higher resistivity distribution of approximately 10~80 ohm-m. The underground
material has a larger particle or a higher void ratio, resulting in a higher resistivity. Layer 2: 3–20
m below the ground surface, generally showing a relatively low electrical resistivity of approxi-
mately 5~12 ohm-m. The particle size of the underground material is finer or the water content
is higher, resulting in lower resistivity. Through ground resistivity detection, a wide range of
survey data can be used to access the soil layer distribution. Importantly, because electrical
properties are indirectly determined in this nondestructive testing method, the composition of
the stratum, the size of the particles, and the water content and salinity of the stratum, which
are factors affecting the resistivity responses, still require the core samples through boring to
connect the resistivity and physical properties of those formation materials. The ERT provides
more accurate investigation results when calibrated with other direct geological investigation
methods, such as the SPT and CPT. Finally, readers should keep in mind that rainfall records
must be involved in ERT projects because the resistivity of shallow layers is easily affected by
rainfall and droughts.
1.9.2 Seismic survey

There are two major types of seismic surveys, refraction and reflection, depending on the spe-
cific type of waves being utilized. Active seismic waves are generated by using an impact source,
such as a sledgehammer. The generated waves transmit differently according to densities, frac-
tures, porosities, etc. in different geological media. The refraction and reflection recorded by an
array of sensors or geophones are a topic of interest. After the measured wave velocity used for
representing different geological materials is back-analyzed, the geological conditions of the
area of interest can be obtained according to Table 1.9.
Figure 1.34 shows an example of the seismic refraction method implementation. In this case,
the refraction seismic survey line is approximately 150 m long, and the detection depth range
is approximately 50 m. The seismic wave is recorded at each measuring point on the survey
line, the seismic wave velocity is calculated, and then the velocity layered profile distribution
is explored. The refraction seismic survey shows that the ground can be approximately divided
into five layers with the corresponding velocity wave. Layer 1: approximately 0~3 m below the
Table 1.9 Approximate range of P wave velocities Vp for some common geological materials
Medium Velocity (km/s)
Min Max
Air 0.31 0.36

Weathered soil horizon 0.10 0.50
Gravel, dry sand 0.10 0.60
Loam 0.30 0.90
Wet sand 0.20 1.80
Clay 1.20 2.50
Water 1.43 1.59
Friable sandstone 1.50 2.50
Dense sandstone 1.80 4.00
Chalk 1.80 3.50
Limestone 2.50 6.00
Marl 2.00 3.50
Gypsum 4.50 6.50
Ice 3.10 4.20
Granite 4.00 5.70
Metamorphosed rock 4.50 6.80
Source: Bettina Albers (2011).
Figure 1.34 A seismic refraction survey result.
ground surface, and the seismic wave velocity is 0.4~0.6 km/sec. This surface soil layer is soft
or porous due to the high energy loss during seismic wave transmission, so the wave veloc-
ity transmission is slow. Layer 2: approximately 1~11 m below the surface, and the seismic
wave velocity is approximately 1.1 km/sec. This layer has a large-grained rock composition,
or the formation composition is relatively compact; thus, the energy loss in the propagation
is relatively low, and the wave velocity is fast. Layers 3 and 4: approximately 20–40 m below
the ground surface, the seismic wave velocity is approximately 2.0 km/sec. It is a collapsed
layer or a rock layer composed of many cracks or filled with muddy materials. The energy loss
is relatively high during seismic wave propagation, so the wave velocity is slower than layer
5. However, this property requires further confirmation by geological survey core sampling.
Layer 5: approximately 20~40 m below the ground surface, and the seismic wave velocity is
approximately 3.50 km/sec. This layer is a rock formation. The rock mass is relatively intact,
with few cracks and mud inclusions. The energy loss of wave transmission is low, and the wave
velocity is therefore transmitted quickly. On the other hand, a reflection method, as shown in
Figure 1.35, is also usually used for determining soil profile for shallow depth than seismic
refraction method, of which V1 and V2, the shear velocities of the first and the second layer, are
analyzed based on signals recorded by surface geophone for identifying the soil profile.
Essentially, soil boring methods can provide direct investigation results but are limited to
explore a small scale of the geological materials in boreholes drilled in the area of interest.
When they are used alongside geophysical surveys, a two- or three-dimensional geological
Figure 1.35 Seismic survey operation.

structure can be developed and provide overall information on subsurface geological condi-
tions, especially useful for complex ground profile exploration.
1.10 Reports and interpretation

A site investigation report generally consists of two parts: a factual description of the investiga-
tion results and an engineering analysis with relevant recommendations.
A reliable report should precisely describe the features of the project site, the work per-
formed, and the results obtained.
• The scope and purpose of the project and its importance.

• The basic conditions of the site, such as topographical and hydraulic features.
A concise description and summary of the amounts, procedures, and results of various field
and laboratory tests should be prepared. Tables containing bore logs and other field and labora-
tory test results and a simplified geological profile should be presented.
Analyses of allowable bearing capacity, settlement, liquefaction potential, and/or recom-
mended foundation type should also be provided in the report. Recommendations for further
investigation should be reported specifically if there are still unclarified subjects due to the
complexity of the geological conditions at the project site. Table 1.10 shows a list of items in
geotechnical investigation report.
A building foundation construction, which is located in Taipei City, north Taiwan, is intro-
duced here as an example to facilitate discussion on which foundation and excavation plan in
that project are introduced. To integrate the aforementioned knowledge, the investigation works
used to explore the geological conditions are introduced in what follows.
1.10.1 Abstract of investigation plan

The site plan of a building construction introduced in this chapter is shown in Figure 1.36. The
excavation area is approximately 3,200 m2, the length is approximately 60 to 105 m, the width
is approximately 40 m, and the shape is similar to a trapezoid. This building is expected to have
18 stories above the ground surface and 5 stories underground. The design depth of excavation
Table 1.10 List of items in a geotechnical investigation report
General description Analysis Other
1. The purpose and importance 1. Analyze and discuss the test 1. Recommendations,
of the project. results. such as dewatering,
2. Topographical features and 2. Determine soil profiles and ground improvement,
hydrological conditions of the geotechnical and hydraulic construction
site. parameters. procedures, and
3. A brief description of the 3. Determine allowable bearing further investigation.
various field and laboratory pressures, pile loads, etc.
tests carried out. 4. Evaluate soil liquefaction
4. Boring log/CPT. potential.
5. Geophysical surveying. 5. Analyze excavation- induced
6. Table listing the testing results. influences.
Figure 1.36 Site location plan and layout of borings.

Source: Redrawn from Ou et al. (1998).
is approximately 19.7 m below the ground surface. To plan and design an appropriate founda-
tion, a series of investigations is performed to understand the geological conditions and obtain
the soil properties (Liao, 1996).
1.10.2 Basic information collection
1.10.2.1 Location
Figure 1.36 shows that there are existing roads and 4-story and 12-story buildings around the plan
site. Therefore, the safe protection and reduction of construction-induced vibration and noise for
adjacent buildings and residents are important during the design and construction phases.
1.10.2.2 Regional geology

This plan site is located in the Taipei Basin, formed by a thick alluvium formation, known as the
Sungshan Formation, which lies above the Chingmei Gravel Formation. The Sungshan Forma-
tion has six alternating sand and clay layers. Sungshan layers II, IV, and VI are silty clay (CL),
and layers I, III, and V are silty sand (SM). The 25 m thick Sungshan layer IV, classified as soft
to medium clay, is the main layer that affects excavation behavior in the study.
1.10.3 Borings and tests

The basic information collected indicates that the plan site will be situated on the clay depos-
its. Clay usually has lower shear strength and poor bearing capacity, and the settlement of
consolidation must also be taken into consideration. The subsurface exploration items in this
project site include geological borehole drillings and laboratory and field tests.
1.10.3.1 Layout and depth of borings

• Amount of geological borehole drilling: the base area of this project site is approximately
3,200 m2, and the principle of borehole configuration is at least 1 borehole for each 600 m2.
Therefore, 3,200 m2/600 m2 = 5.3 boreholes, and 6 boreholes are required and averagely
distributed in the plan site, as shown in Figure 1.36.
• Depth of geological borehole drilling: the excavation depth is 19.7 m, so the design of the
borehole drilling depth must be deeper than the excavation depth to a hard stratum.
• Sampler: A thin-walled tube sampler is adopted to obtain undisturbed soil for laboratory test-
ing in this project.
1.10.3.2 Field test

Field tests in this plan site investigation consist of a standard penetration test (SPT) and cone
penetration test (CPT). Moreover, the vane shear test (VST) is used to directly determine the
undrained shear strength of clay to ensure the quality of the test data. The shear strength of clay
by VST will be very useful to obtain data for engineering design.
1.10.3.3 Laboratory tests

Laboratory soil Atterberg limit tests, triaxial permeability tests, oedometer tests, and shear
strength tests are performed to determine the physical and mechanical soil properties in this
project. Various shear strength tests, including the direct shear test, triaxial unconsolidated und-
rained test (UU), and triaxial consolidated undrained test (CU), are conducted to obtain neces-
sary parameters that will be adopted in the design of the foundation.
1.10.4 Investigation results

After geological borehole drilling, laboratory and field testing are conducted in this site, a part of a
boring log is obtained, as shown in Figure 1.37 as an example. According to the logs of Boreholes
A-1, A-2, and A-3, designated as Section A-A in Figure 1.36, a soil profile can be determined, as
Figure 1.38. Note that if inhomogeneous or non-uniform soil layers are justified, the soil profiles at
other sections should be determined, for example, the section along boreholes A-4, A-2, and A-6.
The excavation depth for this excavation project is 19.7 m; the layer consists of silty clay (CL
classified in USCS) influencing the behavior of this excavation with a SPT-N value of the layer
2 to 8 blows, which is assessed as soft to medium stiff soil. A distribution of the SPT-N values
along the depth is shown in Figure 1.39.
Because this plan site consists mainly of clay, the main points of the foundation design will
be considered bearing capacity, consolidation settlement, and heave; therefore, a detailed analy-
sis needs to be carried out in the report. For obtaining a more detailed examination on the soil
strength of the clay soil underlying the foundation, undisturbed soil is obtained for UU test in
the laboratory and VS test is applied in the field. According to the shear strength test results, su is
approximately 63 kPa at a depth of 19.7 m, as shown in Figure 1.40. Finally, a simplified subsoil
profile judged by an experienced geotechnical engineer with the recommended parameters in
Table 1.11 is to use for engineering design in the report.
Borehole Log
Job title: Taipei survey Location: Kuan Fu N. RD. Borehole: A-3 Sheet 1 of 3
Depth: 52.5 m Elevation: 6.07 m G.W.L: 5.58 m Date: 1998/10/30~11/01
Boring data Testing
Particle size Water Unit Specific Void Liquid Plasticity
Depth Sample SPT-N Log Soil USCS analysis(%) content weight gravity ratio limit index
(m) no. blows description
G S M C (%) (kN/m3) Gs e LL PI
0.7
1 Sand and
S-1 4 gravel CL 0 11 50 39 34.5 17.85 2.67 0.201 50.5 23.7
2
3 S-2 2 CL 30 18.44
Gray-black
4 T-1 silty clay 0 1 60 39 40.8 17.66 2.68 0.214 29 8.5
S-3 1 CL 39.5 17.17
5
6.10 SM 0 3 68 29 42.2 17.56 2.68 0.217 30 8.8

6 S-4 2
7
Sand mixed 25.6 18.93
S-5 3 SM
sandy silt
8
9 9.0
10 S-6 2 CL 30.9 18.54
11 S-7 3 CL 27.8 19.62
12 T-2 CL-ML 0 19 71 10 29.5 18.74 2.71 0.187 24.6 5.3

S-8 2 CL 0 5 70 25 29.3 18.93 2.74 0.187 23.6 7.4
13
14 S-9 3 CL 36.8 18.25

Gray-black
15 silty clay
S-10 3 CL 0 1 48 51 41.1 17.85 2.74 0.217 40.9 18.5
16
17 S-11 3 CL 38.5 18.05
18
T-3 0 0 52 48 40.8 17.66 2.7 0.215 38.3 15.6
19 S-12 4 CL 34.8 17.95
20
Figure 1.37 A part of one of the boring logs.

Figure 1.38 Geological profile of Figure 1.36 A-A'.

Source: Redrawn from Liao (1996).
Figure 1.39 Distribution of the SPT-N values along the depth.

Figure 1.40 Undrained shear strength by variable tests versus depth.

Table 1.11 Simplified classification of strata and test results
Layer Depth USCS SPT-N (kN/m 3) c′ (kPa) 0' ω (%) LL PI s u (kPa)

(m) value (deg.)
1 0~5.6 CL 2~4 18.25 – 30 33~38 33~36 13~16 –

2 5.6~8 SM 4~11 18.93 0 31 21~25 – – –
3 8~33 CL 2~8 18.15 0 29 32~40 29~39 9~23 40~100
4 33~35 SM 22~24 19.62 0 31 25~26 – – –
5 35~37.5 CL 9~11 19.13 – 29 28~32 37~41 15~19 150
6 37.5~46 ML or SM 14~37 19.62 0 32 23~24 – – –
7 >46 GP >100 – – – 1 – – –
Note: γ t = unit weight; ω = water content; LL = liquid limit; PI = plasticity Index.
1.10.5 Foundation Analysis

For geotechnical reports, geotechnical engineers usually evaluate the foundation analysis
according to the geotechnical investigations introduced earlier, which may include bearing
capacity, settlement, and soil liquefaction potential assessments (Construction and planning
agency ministry of the interior, R.O.C., 2000). The calculation method of bearing capacity and
settlement can be found in Chapter 3.
Procedures for determining ultimate soil bearing capacity and structure loading are exhibited
as follows:
• Ultimate bearing capacity qu: The soil bearing capacity is calculated based on the general
bearing capacity equation (Section 3.2). Because in this example, the bearing layer is located
in clay with low permeability, the undrained behavior is considered here, that is, c = su and
φ = 0. According to the shear strength test result shown in Figure 1.40, the su at a depth of
19.7 m is approximately 63.3 kPa. The foundation width B and length L are 40 m and 100 m,
respectively. The depth of foundation Df is 19.7 m. Based on these parameters, the ultimate
bearing capacity qu is calculated to be 779.8 kPa.
• Structure loading qs (a combined load of foundation and building): It is preliminarily esti-
mated that the dead and live load for each story is 14 to 16 kPa. The building contains 23
stories and a foundation; therefore, the structure loading is approximately 360 kPa. An accu-
rate calculation for the structure result will be conducted in the detail design stage, and an
approximate estimation in the site investigation report is usually carried out as follows:
Table 1.12 An approximate estimation of load
Story Total stories Dead load (kPa) Live load (kPa) Total (kPa)
2~18F 17 12 2 17 × (12 + 2) = 238

1F 1 12 5 1 × (12 + 5) = 17
B1~B5 5 11 5 5 × (11 + 5) = 80
Mat foundation 1 25 0 1 × 25 = 25
Sum (kPa) 360
1.10.5.1 Settlement
Because the foundation will be constructed on the clay layer, the consolidation settlement is
necessarily taken into consideration:
• Preconsolidation stress - c': The laboratory testing results show that pc′ = 215 kPa, and the in
situ effective overburden pressure = 193 kPa.
• Overconsolidation ratio OCR: OCR = 1.11 > 1, which indicates overconsolidated clay.
• Compression index Cc: According to Section 2.3.3, an empirical correlation Cc ≈ PI 74 for
undisturbed clay, we therefore can obtain Cc = 0.22 as PI = 16.
• Swelling index Cur: An empirical correlation for Cur ≈ PI 370 is helpful here so that Cs =
0.043 can be determined.
• Consolidation settlement calculation: Hence, the clay layer is in an overconsolidation state
and the stress increment of the clay layer is very small because the weight of the designed
structure is close to the excavated soil; hence, the consolidation settlement can be ignored in
this project.
1.10.5.2 Soil liquefaction potential assessment

Soil liquefaction occurs mainly in saturated sand or fine-grained soils that can cause the
failure of buildings. The foundation of this building is to be situated on the clay layer, so the
soil liquefaction risk of the foundation is evaluated to be very low. There are many soil liq-
uefaction potential assessment methods for readers’ further references: SPT-N result based
(Seed and Idriss, 1971; Tokimatsu and Yoshimi, 1983; Seed et al., 1985, JRA, 1996), CPT-
qc result based (Shibata and Teparaksa, 1988), and seismic test result based (Tokimatsu and
Uchida, 1990).
1.11 Summary and general comments

Foundation engineering faces different geological hazards, such as a large settlement, bear-
ing capacity failure, earthquake-induced liquefaction, soil piping, ground heave, and landslides
during the construction or operation phase, which are related to the geological conditions and
material properties. Therefore, the purposes of geotechnical site investigation are to survey
subsurface conditions and to determine potential geological hazards by various investigation
methods and technologies. Furthermore, geotechnical site investigation is also fundamental and
essential to obtain correct parameters, including physical, hydraulic, and mechanical proper-
ties, to facilitate an appropriate design to avoid disasters and enhance the safety and economic
performance of construction.
Each investigation method or technique has a distinct purpose, advantage, disadvantage, and
applicability; moreover, no single method or technique can simultaneously obtain the various
information of the ground for design and analysis, so in good engineering practice, a series of
geotechnical site investigation methods is often applied during the investigation process. In
addition, many investigation methods have been developed and continue to progress world-
wide. This chapter lists some common and fundamental methods, such as borehole drilling
and sampling, laboratory tests, standard penetration tests, groundwater investigations, electrical
resistivity tomography, and seismic surveys. This section briefly integrates previous methods
into a case study to provide readers with basic knowledge of geotechnical site investigation and
its applications.
On the other hand, each investigation method or technique entails specific background
knowledge (e.g., geology, geophysics). When those methods are being applied, practition-
ers sometimes face difficult cases with special and complex geological conditions, and it is
not easy for a single individual to know all kinds of knowledge and different investigation
techniques, so geotechnical site investigation work requires teamwork and group discussion
to integrate the partial data obtained from various investigation methods and opinions from
the team to build a reasonable and accurate geological model of the site for supporting design
at the next stage.
Accurate and reliable investigation results are also based on reasonable investigation pro-
ject planning and appropriate implementation, but determining the investigation plan and
contents is often constrained by a limited timeframe and budget and the environmental condi-
tions. The adoption of the most efficient and feasible method is always a challenge for devel-
opment units and geotechnical engineers. Therefore, geotechnical engineers must first clearly
understand the engineering project and the required parameters for design and analysis and
based on thorough consideration and propose an appropriate investigation project. Moreover,
investigation work is not a one-off task. A supplementary investigation can be conducted
if there are still some doubts or unresolved questions. After the reports are carefully and
comprehensively compiled, they can provide useful and important information to facilitate a
viable engineering design.
Problems
1.1 Please indicate the purpose of each test.
Soil physical property tests
1. Triaxial unconsolidated undrained test (UU)

2. Triaxial consolidated undrained test (CU)
3. Oedometer test
4. Direct shear test
5. Permeability test
6. Dynamic triaxial test
1.2 Please describe the methods of downhole P-S logging and crosshole velocity logging,
with an introduction to their application purpose and procedures for determining the soil
parameters.
1.3 Please derive eq. 1.2 for determining the undrained shear strength with a vane shear test.
1.4 What is the difference between the blow count and SPT-N value of the standard penetra-
tion test (SPT)?
1.5 If soil permeability at depths of 5 m and 20 m in the boring log of Table 1.11 are needed,
what types of test method would be used in the field and the laboratory?
1.6 Table P1.6 provides field standard penetration numbers’ record data. Determine the N
value at each depth.
Table P1.6
Depth (m) N1 N2 N3 N -value
1.5 2 3 2
3.0 2 2 4
4.5 1 3 3
6.0 2 3 4
7.5 3 4 4
1.7 A dilatometer test was conducted in a clay deposit. The saturated unit weight of the soil is
19 kN/m2, and the groundwater level is 2 m below the ground surface. Now, we have the
P0 = 250 kN/m2 and P1 = 300 kN/m2 at a depth of 5 m. Determine the horizontal stress
index (KD) and the coefficient of earth pressure (K0,DMT).
1.8 Compare the area ratios of the split-spoon sampler (outer diameter d0 = 50.8 mm and
inner diameter di = 34.9 mm) used in the standard penetration test and the thin-walled tube
(outer diameter d0 = 76.2 mm with wall thickness t = 1.55 mm).
References
Anjan, P. (2019), Geotechnical Investigations and Improvement of Ground Conditions, Duxford, UK and
Cambridge, MA: Woodhead Publishing.
ASTM D1587/D1587M-15 (2015), Standard Practice for Thin-Walled Tube Sampling of Fine-Grained
Soils for Geotechnical Purposes, ASTM International, West Conshohocken, PA.
ASTM D3550-01 (2001), Standard Practice for Thick Wall, Ring-Lined, Split Barrel, Drive Sampling of
Soils, ASTM International, West Conshohocken, PA.
Baguelin, F., Jézéquel, J.F. and Shields, D.H. (1978), The Pressuremeter and Foundation Engineering,
Clausthall: Trans Tech Publications.
Casagrande, A. and Fadum, R.E. (1940), Notes on Soil Testing for Engineering Purposes, Cambridge, MA:
Harvard University Press.
Clarke, B.G. (1994), Pressuremeters in Geotechnical Design, London: CRC Press.
Construction and planning agency ministry of the interior, R.O.C. (2000), Design Specifications of Foun-
dation and Structure for Buildings (In Chinese), Taiwan: R.O.C.
David, S. (2018), Cone Penetration Test Design Guide for State Geotechnical Engineers, Minnesota
Department of Transportation, Research Services & Library, 395 John Ireland Boulevard, MS 330, St.
Paul, Minnesota 55155–1899.
Geotechnical Engineering Office (2017), Guide to Site Investigation, FHWA NHI-16–072, Washington,
DC, USA.
Griffiths, D.H. and King, R.F. (1981), Applied Geophysics for Geologists and Engineers: The Elements of
Geophysical Prospecting, 2nd edition, Oxford: Pergamon Press.
Gunn, D.A., Chambers, J.E., Uhlemann, S., Wilkinson, P.B., Meldrum, P.I., Dijkstra, T.A., Haslam, E.,
Kirkham, M., Wragg, J., Holyoake, S., Hughes, P.N., Hen-Jones, R. and Glendinning, S. (2015), Mois-
ture monitoring in clay embankments using electrical resistivity tomography. Construction and Building
Materials, Vol. 92, pp. 82–94.
Hvorslev, M.J. (1949), Subsurface Exploration and Sampling of Soils for Civil Engineering Purposes, U.S.
Army Corps of Engineers, Waterways Experiment Station, Vicksburg, Miss.
Hvorslev, M.J. (1951), Time Lag and Soil Permeability in Ground-Water Observations, Bull. No. 36,
Waterways Exper. Sta. Corps of Engrs, U.S. Army, Vicksburg, Mississippi, pp. 1–50.
JRA (Japanese Road Association) (1996), Specification for Highway Bridges, Part V, Japan: Seismic
Design, Japan.
Liao, J.T. (1996), Performances of a Top Down Deep Excavation, doctoral thesis (In Chinese), Taiwan.
Marchetti, S. (1980). In situ tests by flat dilatometer. Journal of the geotechnical engineering division,
106(3), pp. 299–321.
Marchetti, S., Marchetti, D., Monaco, P. (2015). Flat dilatometer (DMT). Applications and recent develop-
ments. In Proceedings of the Indian Geotechnical Conference, Pune, India, pp. 16–19.
Marchetti, S., Monaco, P., Totani, G. and Calabrese, M. (2001), The Flat Dilatometer Test (DMT) in Soil
Investigations, ISSMGE Committee TC16, Italy: University of L’Aquila.
Osterberg, J.O. (1952), New piston type soil sampler. Engineering News-Record, Vol. 148, No. 17,
pp. 77–78.
Ou, C.Y., Liao, J.T. and Lin, H.D. (1998), Performance of diaphragm wall constructed using top-down
method. Journal of Geotechnical and Geoenvironmental Engineering, Vol. 124, No. 9, pp. 798–808.
Rashed, A., Bazaz, J. B., and Alavi, A. H. (2012). Nonlinear modeling of soil deformation modulus
through LGP-based interpretation of pressuremeter test results. Engineering Applications of Artificial
Intelligence, 25(7), pp. 1437–1449.
Seed, H.B., Yokimatsu, K., Harder, L.F. and Chung, R.M. (1985), Influence of SPT procedures in soil
liquefaction resistance evaluation. Journal of Geotechnical Engineering, ASCE, Vol. 111, No. 12,
pp. 1425–1445.
Shibata, T. and Teparaksa, W. (1988), Evaluation of liquefaction potentials of soil using cone penetration
tests. Soils and Foundations, Vol. 28, No. 2, pp. 49–60.
Terzaghi, K., Peck, R.B. and Mesri, G. (1996), Soil Mechanics in Engineering Practice, 3rd edition, New
York: John Wiley & Sons, Inc.
Tokimatsu, K. and Uchida, A. (1990, June), Correlation between liquefaction resistance and share wave
velocity, soils and foundations. Japanese Society of Soil Mechanics and Foundation Engineering, Vol.
30, No. 2, pp. 33–42.
Tokimatsu, K. and Yoshimi, Y. (1983), Empirical correlation of soil liquefaction based on SPT N-value and
fines content. Soils and Foundations, Vol. 23, No. 4, pp. 56–74.
Chapter 2
Principles of foundation design
2.1 Introduction
Foundation design is based on the properties of subsurface soil and loading on a foundation.
Soil properties are highly related to the soil under drained or undrained conditions. This chapter
will first explain the undrained and drained behavior of soil. Once we understand the drained
and undrained behavior of soil, we can select appropriate soil parameters with drained or und-
rained analysis for target soils based on the effective stress or total stress analysis, which is also
covered in this chapter.
Soil parameters, such as the cohesion and friction angle, Young’s modulus, Poisson’s
ratio, compression index, unloading/reloading index, and coefficient of earth pressure at rest,
are necessary for foundation design. In principle, these soil parameters should be obtained
from related soil tests. However, some parameters can also be estimated by correlation with
other soil properties, such as unit weight, density, N value from the standard penetration
test (SPT), cone tip resistance from the cone penetration test (CPT), etc. This chapter pro-
vides reliable methods and correlations for obtaining design soil parameters for cohesive and
cohesionless soils.
Three design methods are used for foundation design: working stress design, ultimate
strength design, and limit state design methods. The first two methods are conventional. Work-
ing stress design involves designing a foundation system to sustain the design load (design
stress) without exceeding its allowable load (allowable stress). The uncertainties of external
load and soil resistance are lumped and addressed by a single factor of safety. Since the con-
cept of working stress design is straightforward and simple, it is introduced to design various
foundations in this book. Moreover, since both the external load and soil properties are subject
to different degrees of uncertainties or variabilities, it is more reasonable to assign different
factors to load and soil parameters. This design method is called ultimate strength design and
has been adopted in some foundation design codes. This book also introduces the concept of
ultimate strength design.
In fact, a foundation should be designed to avoid not only failure, that is, the ultimate limit
state, but also excessive settlement, that is, the serviceability limit state. The design method that
addresses both limits is often called limit state design. Recently, some geotechnical structural/
foundation codes have adopted this method for design. The concept of limit state design will be
introduced in this book. Finally, the determination of soil parameters that considers uncertain-
ties and variabilities in soil properties is elaborated in the last section.
DOI: 10.1201/9781003350019-2
52 Principles of foundation design
2.2 Geotechnical analysis method
2.2.1 Drained behavior and undrained behavior

When a soil is stressed, the ease of pore water flow out will significantly affect soil deformation
behavior. Figure 2.1 shows a schematic diagram of saturated coarse-grained soil under normal
pressure, '- . The spring represents the coarse-grained soil, which will deform as it is stressed.
Therefore, the stress in the spring represents the effective stress in the soil. The initial pore water
pressure is in the hydrostatic condition, that is, uw = y w z . The initial effective stress and total
stress are denoted by - v'0 and σ v0 , respectively. As a pressure -- acts on a piston, causing
the piston to move down, the water would flow out rapidly and the volume thus decreases. The
spring is therefore stressed. Under such a condition, the excess pore water pressure is nil and
the pore water pressure remains the same as the hydrostatic pressure (Figure 2.1a). The applied
pressure -- directly transmits to the soil, causing the effective stress in the soil to be increased
to - v'0 + -- (Figure 2.1b). The total pressure increases to - v0 + -- (Figure 2.1c). Such defor-
mation behavior of the soil is called drained behavior.
Figure 2.2 shows a schematic diagram of saturated fine-grained soil, that is, clay, under nor-
mal pressure, -- . Similarly, γ w z , - v'0 , and σ v0 represent the initial pore water pressure, effec-
tive stress, and initial total stress, respectively. Under these conditions, when a pressure --
acts on a piston, the pore water flows out very slowly. Therefore, at the initial stage of loading,
the pressure -- is borne entirely by the pore water, and the spring is unstressed. The stress in
the spring or the effective stress in the soil remains unchanged (Figure 2.2b). This condition is
called undrained behavior or short-term behavior. Pore water pressure is thus the sum of hydro-
static water pressure and excess pore water pressure, while excess pore water pressure is --
(Figure 2.2a). If there is a shear stress-induced pore water pressure, then the excess pore water
pressure will be generated by normal pressure ( -- ) as well as the shear stress.
Figure 2.1 Schematic diagram of saturated coarse-grained soil under normal pressure: (a) pore
water pressure, (b) effective stress, and (c) total stress.
Principles of foundation design 53
Figure 2.2 Schematic diagram of saturated fine-grained soil under normal pressure: (a) pore
water pressure, (b) effective stress, and (c) total stress.
However, as shown in Figure 2.2, the pore water will eventually flow out. The piston eventu-
ally moves down, and the volume still changes. Excess pore water decreases or dissipates, and
effective stress increases accordingly, but the total stress remains unchanged. When the excess
pore water pressure is nil, the stress in the spring is equal to - v'0 + -- . This condition is also
called drained behavior or long-term behavior.
Materials with drained behavior are called drained material. Conversely, materials with und-
rained behavior are called undrained material. When conducting geotechnical analysis, one
should first determine whether the target soil belongs to drained materials or undrained materi-
als based on the soil classification test results of the target soil and the description of the site
investigation and perform drained analysis or undrained analysis accordingly.
2.2.2 Effective stress analysis and total stress analysis

The effective stress analysis method and total stress analysis method are two types of analysis
methods in geotechnical engineering.
The effective stress analysis method treats the soil as a two-phase material, that is, soil parti-
cles and pore water. When an external force acts on the soil, the stress and strain experienced by
the soil depend on the contact stress between soil particles. The concept of effective stress repre-
sents the contact stress in soil mechanics. Due to the large void ratio of the soil, when the volume
of the soil is changed by force, the pore water will flow out of (or in) the soil unit, so there is no
excess pore water pressure (see Figure 2.1b). Basically, when soil particles are immersed in pore
water, the shear strength or deformation of the soil is related only to the effective stress, not to the
total stress. Therefore, the effective parameters, such as effective strength parameters (c′, 0 ' ),
effective unit weight ( y ' ), effective Poisson’s ratio ( u ' ), and effective Young’s modulus ( E ′ ),
Figure 2.3 Types of analysis: (a) effective stress analysis and (b) total stress analysis.
should be adopted, and the groundwater level should be set in the analysis (Figure 2.3a). Effec-
tive stress analysis is applicable to coarse-grained soils with large void ratios, such as sand and
gravel soil, or fine-grained soil under long-term conditions.
When fine-grained soil, such as clay, is stressed, due to very small clay voids, the pore
water cannot be drained in a short period of time, and excessive pore water pressure is gener-
ated (see Figure 2.2a). Under the undrained condition, the amount of excess pore water pres-
sure affects the effective stress of the soil, which, in turn, affects the stress‒strain behavior
of the soil. Therefore, total stress analysis is also called undrained analysis. This analysis
method regards clay and pore water as a one-phase material or a soil‒water mixture, with no
pore water existing in the material (Figure 2.4). When the material is stressed, no excessive
pore water pressure is generated because no pore water exists in this material. Therefore,
even if the clay is below the groundwater level, the groundwater level should not be set in the
analysis; otherwise, additional pore water pressure will be generated, that is, the groundwater
level should be set outside the range of influence. The groundwater level is usually assumed
at the bottom of the soil deposit or with no groundwater level. The total stress or undrained
parameters, such as total or undrained strength parameters ( c , φ ), saturated unit weight
(below groundwater level) ( γ sat ), total or undrained Poisson’s ratio ( µu ), and total or und-
rained Young’s modulus ( Eu ), should be used in the analysis (Figure 2.3b). The total stress
analysis method is applicable to the undrained behavior of fine-grained soils with a small void
ratio, such as clay.
Figure 2.4 Concept of the total stress analysis for undrained material: (a) soil in water and
(b) modeling as a soil-water mixture.
Table 2.1 Application of the effective stress analysis and total stress analysis
Method of Parameters Groundwater Pore water Application

analysis level pressure
Effective c′, ϕ′, γ′ (1)

, E′, μ′ Setting the Existent Granular soil, such as sand,
stress actual silt, or gravel
analysis location Fine soil, such as clay, drained
under long-term conditions
Total stress c, ϕ, γ sat(2), E u, μ u Assumed Nonexistent Fine soil, such as clay, under
analysis nonexistent undrained conditions
Note:
(1): for the hydrostatic condition only, where γ′ = g sat − γ w.
(2): for the soil below the groundwater level.
As shown in Figure 2.2, as the total pressure for saturated clay increases (or decreases), the
increased (or decreased) pressure acts on the pore water, causing the excess pore water pres-
sure to increase (or decrease), so the effective stress increment is zero. Based on the principle
of effective stress, if the effective stress is not changed, the undrained shear strength will not
change accordingly. This phenomenon is equivalent to the stress state of saturated clay in the
triaxial unconsolidated undrained (UU) test. Therefore, the strength parameter can be repre-
sented by ϕ = 0, c, where c is the cohesion. Notably, c is also called undrained shear strength,
normally denoted su.
Table 2.1 summarizes the parameters used and the setting of the groundwater level in the effec-
tive stress and total stress analysis. The application of the analyses is also summarized in this table.
2.2.3 Stress paths

Figure 2.5a shows the Mohr’s circle of a triaxial test specimen subject to horizontal pressure (σh)
and vertical pressure (σv). The normal stress and shear stress at the top of Mohr’s circle are equal
to (σv + σh)/2 and (σv − σh)/2, respectively. According to the theory of Mohr’s circle, these nor-
mal stress and shear stress are the stresses acting on a plane inclined at an angle of 45° with the
horizontal (Figure 2.5b). Let p and q represent the normal and shear stresses at the top of Mohr’s
circle or on a 45° plane with the horizontal, which can be expressed by the following equations:
-v +-h
p= (2.1)
2
-v --h
q= (2.2)
2
Where:
σv = vertical stress.
σh = horizontal stress.
Assuming that the initial pore water pressure is equal to 0 in the triaxial soil specimen after
consolidation, the pore water pressure in the test specimen is equal to the excess pore water pres-
sure (ue) that is generated during undrained shear. The effective Mohr’s circle can be obtained
by shifting the total Mohr’s circle to the left by an amount of ue, as shown in Figure 2.6. The
effective normal (p′) and shear stresses (q) at the top of the effective Mohr’s circle or a 45° plane
with the horizontal can be derived as follows:
- v' + - h' (- v - ue ) + (- h - ue ) - v + - h
p' = = = - ue (2.3)
2 2 2
- ' - - h' - v - - h
q' = q = v = (2.4)
2 2
Figure 2.5 Mohr’s circle for a soil specimen: (a) the stress state at the top of Mohr’s circle and
(b) the stresses on a 45° plane with the horizontal.
Figure 2.6 Total and effective Mohr’s circles.
Figure 2.7 Stress path: (a) a series of Mohr’s circles in the undrained test and (b) total and
effective stress paths.
Connecting values of p and q on every Mohr’s circle during shear on the p-q diagram results
in the path of shearing, called the total stress path (TSP), as shown in Figure 2.7. Similarly,
connecting the values of p ′ and q on each effective Mohr’s circle, we obtain a path called the
effective stress path (ESP) (Figure 2.7). In the triaxial consolidated drained (CD) test, the total
Mohr’s circles and the TSP are equal to the effective Mohr’s circles because there is no excess
pore water pressure generation in the triaxial CD test, q = q ' , and p = p '. Therefore, the ESP
and TSP would be the same curve. In the triaxial consolidated undrained (CU) test, if ue is posi-
tive, the ESP shifts to the left of the TSP by an amount of ue (Figure 2.7b).
As mentioned, the TSP represents the total normal and shear stresses on a plane inclined at an
angle of 45° with the horizontal, while the ESP represents the effective normal and shear stresses
on the plane. Therefore, the TSP and ESP can represent the changes in the normal stress, shear
stress, and excess pore water pressure on the plane. By observing these changes, we are able to
understand the shearing behavior of the soil specimen. We will use this concept to explain short-
term strength and long-term strength in the next section. For more explanation of the stress path
and application in geotechnical problems, readers can refer to the relevant references (Holtz
et al., 2011).
Kf
Figure 2.8 Relation between the K f line and Mohr–Coulomb failure line.
The Mohr–Coulomb failure line, which is a line tangent to Mohr’s circles, can be transformed
into the p-q diagram, which is called the K f line. The following geometric relationship exists
between the K f line and the Mohr–Coulomb failure line (Figure 2.8):
sin 0 = tan n (2.5)

a
c= (2.6)
cos 0
Where a and η are the intercept on the q-axis and the angle of the K f line with respect to the
horizontal, respectively.
2.2.4 Short-term analysis and long-term analysis

Figure 2.9 shows the typical stress‒strain behavior of dense and loose sand. The stress acting
on loose sand increases up to the ultimate strength while its volume decreases, that is, the soil
compresses. On the other hand, the stress acting on dense sand first increases up to the peak
strength and then decreases with increasing strain to the ultimate strength. In turn, the soil first
compresses and then dilates. With the same type of sand, the values of the ultimate strength for
the dense and loose states are nevertheless close. The reason is that the strain at the ultimate
strength is rather large, and the soil has been disturbed significantly. Under such conditions, soil
strength relates largely to the characteristics of intergranular friction (i.e., roughness of the soil
particles).
Normally consolidated (or lightly overconsolidated) clay in the drained test shows stress‒
strain behavior similar to that of loose sand. Because drained shearing tends to cause saturated
normally consolidated (or lightly overconsolidated) clay to contract, positive pore water pres-
sure is produced in the undrained test, where the volume of the clay remains unchanged. A
representative stress‒strain behavior along with the excess pore water pressure generated for
normally consolidated (or lightly overconsolidated) clay during undrained shear is shown in
Figure 2.10. The effective stress path (ESP) is therefore shifted to the left of the total stress path
(TSP) by an amount of ue, as shown in Figure 2.11.
Since the excess pore water pressure in the drained test is zero, curve ab in Figure 2.11 also
represents the ESP and TSP. With the q value with regard to point b larger than that with regard
Figure 2.9 Typical stress-strain behavior and volume change in sand and clay under drained
shear conditions.
Figure 2.10 A typical stress-strain behavior and excess pore water pressure strain for clay
under undrained shear conditions.
Figure 2.11 Possible effective and total stress paths for normally or lightly overconsoli-
dated clay.
Figure 2.12 Possible effective and total stress paths for highly overconsolidated clay.
to point c, the undrained shear strength, τc or su, of normally consolidated (or lightly overcon-
solidated) clay is smaller than the drained shear strength, τb, of clay with the same consolidation
pressure. That is, if positive pore water pressure is generated in an undrained test, the shear
strength under undrained conditions will be smaller than that under drained conditions under the
same consolidation pressure. The undrained shear strength should then be adopted for design.
The undrained shear strength refers to the condition that the excess pore water pressure is not
dissipated in a short time. Therefore, it is also called the short-term strength.
The behavior of highly overconsolidated clay is similar to that of dense sand. As shown in
Figure 2.9, drained shearing will cause the saturated highly overconsolidated clay to compress
first and then dilate, and therefore, undrained shearing will produce first positive and then nega-
tive pore water pressure (Figure 2.10). Figure 2.12 shows the TSP, as represented by curve adc,
and its corresponding ESP, as represented by curve ad′b, for saturated clay sheared under the
undrained condition. By definition, the undrained shear strength (su) is the shear strength (τb)
corresponding to point b, which is also called the short-term strength. In the long-term condition,
excess pore water pressure would dissipate completely, and its TSP and ESP would be the same,
as denoted by curve ad. The drained shear strength corresponding to point d is also called the
long-term strength. The q value with regard to point b is larger than that with regard to point d.
Consequently, a highly overconsolidated clay normally has a larger undrained shear strength, τb
or su, than the drained shear strength (long-term), τd, under the same consolidation pressure. In
other words, if the undrained test produces negative pore water pressure (or the volume dilates
in a drained test) at the ultimate state, the undrained shear strength will be larger than the drained
shear strength. Therefore, the drained shear strength for a highly overconsolidated clay should
be adopted for design.
2.3 Estimation of design soil parameters

Soils can be classified into two categories: cohesionless and cohesive. Cohesionless soils include
gravels, sands, and silts. Cohesive soils include clays and organic soils.
Cohesive soils usually exhibit undrained behaviors during loading; hence, undrained (total
stress) properties, such as the undrained shear strength (su), undrained Young’s modulus (Eu),
and undrained Poisson’s ratio (μu), are important design parameters for foundation design. The
undrained friction angle (ϕu) is zero whenever su is adopted to characterize the strength of an
undrained soil. The consolidation deformation of cohesive soils is also an important foundation
design problem, but it is a drained problem. The compression index (Cc) and unloading–reload-
ing index (Cur) are the relevant parameters for consolidation deformation.
Cohesionless soils usually exhibit drained behaviors during loading; hence, drained (effec-
tive stress) properties, such as the effective friction angle (ϕ′), effective Young’s modulus (E′),
and effective Poisson’s ratio (μ′), are important design parameters for foundation design. The
effective cohesion (c′) for cohesionless soils is zero. Cohesionless soils also show consolidation
deformation, but the deformation is considerably less than that for cohesive soils.
The at-rest lateral earth pressure coefficient (K0) is also an important design soil parameter for
both cohesive and cohesionless soils. Because K0 is defined as the ratio of the in situ effective
horizontal stress to the in situ effective vertical stress, K0 is always effective by definition (even
for cohesive soils).
2.3.1 Soil strength

The shear strength of soil can be expressed as the Mohr‒Coulomb criterion:
t f = c + - tan (0 ) (2.7)
where:
τf = shear strength.
c = cohesion.
σ = normal stress.
φ = friction angle.
However, eq. 2.7 is rarely adopted in practice. In practical applications, eq. 2.7 is usually applied
in two special forms. For the effective stress (drained) analysis, the following form is usually
adopted for cohesionless soils:
t f = - ' tan (0 ' ) (2.8)

where:
- ' = effective stress.

0 ' = effective friction angle.
For the total stress (undrained) analysis, the following form is usually adopted for cohesive soils:
t f = su (2.9)
where:
su = undrained shear strength.
Cohesionless soils
For cohesionless soils, the main shear strength parameter is the effective friction angle (ϕ′).
Depending on the soil density, the stress‒strain curve of the cohesionless soil may exhibit dif-
ferent behaviors. For a dense sand, the (peak) ϕ′ value is high and is developed under small
strain (Figure 2.9). For very loose sand, ϕ′ is relatively small and is developed under
large strain (Figure 2.9). Laboratory tests, such as triaxial compression tests, can be adopted
to determine ϕ′ for undisturbed samples of cohesionless soils. However, it is challenging to
extract such undisturbed samples for cohesionless soils. In practice, ϕ′ for a cohesionless soil
is typically estimated by correlation with other soil properties, such as unit weight, density,
N value for the standard penetration test (SPT), cone tip resistance for the cone penetration
test (CPT), etc.
Table 2.2 shows typical values of ϕ′ for cohesionless soils with different densities (loose vs.
dense). The effective friction angle of a cohesionless soil can also be estimated by its dry unit
weight (γd) and relative density (Dr), as shown in Figure 2.13.
It is common to estimate ϕ′ based on the in situ SPT N value, which is the number of
blow counts for the standard sampler to advance 30 cm in its penetration depth. Two types
of corrections for SPT N are necessary for better correlation to ϕ′: (a) a correction for the
energy efficiency and (b) a correction for the overburden stress. Due to variability in local
practice and equipment, the energy efficiency of the drop hammer onto the drop rods varies
by region. Skempton (1986) reviewed SPT data from Japan, China, the United States, and
Table 2.2 Typical values of the effective friction angles for cohesionless soils
Soil material Effective friction angle ϕ′ (degrees)
Loose Dense
Sand, round grains, uniform 27.5 34

Sand, angular grains, well-graded 33 45
Sandy gravel 35 50
Silty sand 27 to 33 30 to 34
Inorganic silt 27 to 30 30 to 35
Source: Terzaghi and Peck (1967); reproduced from Kulhawy and Mayne (1990).
Dr
Figure 2.13 Effective friction angle versus dry unit weight and relative density.
Source: Redrawn from NAVFAC (1982).
the United Kingdom and suggested the following corrections for the SPT N value (Kulhawy
and Mayne, 1990):
N 60 = CER CB CS CR N (2.10)
where:
N60 = SPT N value corrected to a standard energy ratio of 60%.

CER, CB, CS, CR = correction factors for the energy ratio, borehole diameter, sampling method,
and rod length, respectively.
The values of CER, CB, CS, and CR are given in Table 2.3.
In addition to the energy efficiency, the overburden stress also significantly affects the SPT
N value. To obtain a better correlation to ϕ′, it is necessary to apply the following correction
for N60:
( N1 )60 = CN N 60 (2.11)
where:
(N1)60 = N60 value corrected to a reference overburden stress of one atmosphere pressure.
CN = correction factor for overburden stress.
Table 2.3 Values for different SPT N correction factors
Factor Equipment variables Correction
Term Value
Energy ratio ( Japan) Donut hammer with Tombi release C ER 1.3

( Japan) Donut hammer with 2-turn rope release 1.1
(China) Pilcon hammer 1.0
(China) Donut hammer 0.9
(US) Safe hammer 0.9
(US) Donut hammer 0.75
(UK) Various hammers with trip release 1.0
(UK) Various hammers with 2-turn rope release 0.8
Borehole diameter 65–115 mm CB 1.0
150 mm 1.05
200 mm 1.15
Sampling method Standard sampler CS 1.0
Sampler without liner 1.2
Rod length >10 m CR 1.0
6–10 m 0.95
4–6 m 0.85
3–4 m 0.75
Source: Skempton (1986); Kulhawy and Mayne (1990).
The following overburden correction factor proposed by Liao and Whitman (1986) is
well-known:
CN = ( pa - v'0 ) (2.12)
0.5
where:
σ′v0 = in situ overburden (vertical effective) stress.

pa = one atmosphere pressure = 101.3 kN/m2.
The overburden correction factor calculated by eq. 2.12 is unbounded near the ground surface,
where Σ′v0 is close to zero. Idriss and Boulanger (2008) recommended an upper bound of 1.7 for
CN. The effective friction angle ϕ′ for a cohesionless soil can be estimated based on (N1)60 by the
following transformation equation (Hatanaka and Uchida, 1996):
0 ' = 15.4 x ( N1 )60 + 20o (2.13)
This equation was developed based on ϕ' data of expensive undisturbed sand samples obtained
by ground freezing. The coefficient multiplied by (N1)60 was not 15.4 in the original formula pro-
posed by Hatanaka and Uchida (1996) but was revised to 15.4 by Mayne et al. (2001) to correct
the SPT energy ratio in Japan. The equation in eq. 2.13 is plotted in Figure 2.14. Additional data
N N
Figure 2.14 Correlation between ϕ′ and (N 1) 60 for cohesionless soils.

Source: Ching et al. (2017).
for undisturbed sand/gravel samples extracted by ground freezing or block sampling are also
plotted in Figure 2.14. The transformation equation proposed by Chen (2004) seems to provide
a better fit to these additional data:
0 ' = 9.2 x log [[( N1 )60 ]] + 27.5o (2.14)
It is also possible to estimate ϕ ′ based on the cone tip resistance (qc) for a piezocone penetra-
tion test (CPTu). Similar to SPT N, some corrections for qc are necessary: (a) a correction for
the pore pressure acting on unequal areas of the cone geometry and (b) a correction for the
overburden stress. The cone tip resistance corrected for the pore pressure acting on unequal cone
areas is denoted by qt:
qt = qc + (1- a ) ubt (2.15)
where:
qt = cone tip resistance corrected for the pore pressure acting on unequal cone areas.
a = net area ratio due to unequal areas of the cone geometry (typically, a = 0.7–0.85)
ubt = pore water pressure measured behind the cone tip.
In addition to the correction for unequal cone areas, an overburden correction is necessary:
Qtn = CN x ( qt pa ) (2.16)
where:
Qtn = qt/pa value corrected to a reference overburden stress of pa = 101.3 kN/m2.

CN = correction factor for overburden stress in eq. 2.12.
The effective friction angle ϕ′ for a cohesionless soil can be estimated based on CPTu (Kulhawy
and Mayne, 1990):
0 ' = 11.0 x log ( Qtn ) + 17.6o (2.17)
This transformation equation is plotted in Figure 2.15. The data of reconstituted sands that are
used to develop this equation are shown in the figure (Qtn is based on calibration chamber tests).
Cohesive soils
For cohesive soils (clays), the main shear strength parameter for foundation design is the und-
rained shear strength (su). However, su is not a fundamental soil property. It is the response of a
clay during undrained loading. It depends on the initial condition, effective stress, stress history,
Qtn
Figure 2.15 Correlation between ϕ′ and Q tn for cohesionless soils.

Source: Kulhawy and Mayne (1990).
mode of loading, rate of loading, etc. For a remolded clay (e.g., reconstituted clay prepared in
the laboratory), the undrained shear strength can be estimated by the liquidity index (Locat and
Demers, 1988):
sure pa = 0.0144 x LI -2.44 (2.18)
where:
sure = undrained shear strength of a remolded clay.

LI = liquidity index.
Figure 2.16 illustrates this equation together with data for remolded clays.
For an intact (in situ) clay, the su value is affected by the effective stress, stress history, load-
ing mode, etc. For the loading mode of embankment failures, Bjerrum (1972), Jamiolkowski
et al. (1985), and Mesri and Huvaj (2007) proposed the following transformation equations to
estimate su based on the vane shear test (VST), overconsolidation ratio, and CPTu:
su = u B x su ,VST Bjerrum (1972) (2.19)

su - v'0 = 0.23 x OCR 0.8
Jamiolkowski et al. (1985) (2.20)
su = ( qt - - v0 ) 16 Mesri and Huvaj (2007) (2.21)
sure pa
Figure 2.16 Correlation between s ure/p a and LI for remolded clays.

Data source: Ching and Phoon (2014).
where:
su,VST = undrained shear strength by VST.

μB = VST correction factor proposed by Bjerrum (1972).
OCR = overconsolidation ratio.
σv0 = total overburden stress.
Figure 2.17 shows how the VST correction factor (μB) changes with the plasticity index (PI).
The data shown in the figure are back-calculated by real failure cases of embankments, foot-
ings, and excavations. Figures 2.18 and 2.19 illustrate the transformation equations proposed by
Jamiolkowski et al. (1985) and Mesri and Huvaj (2007), respectively, together with some data
from in situ clays.
Although the correlation is relatively weak, the su value for in situ clay can also be estimated
by the SPT N value (Kulhawy and Mayne, 1990):
su pa = 0.06 x N 60 (2.22)
Figure 2.20 illustrates this transformation equation together with data from in situ clays.
2.3.2 Soil modulus

For cohesionless soils, the main parameters for immediate deformation are the effective (or
drained) Young’s modulus (E′) and effective Poisson’s ratio (μ′). For cohesive soils, they are
the undrained Young’s modulus (Eu) and undrained Poisson’s ratio (μu). The Young’s modulus
of a soil varies over a wide range, depending on the initial condition, effective stress, loading
mode, etc. Due to its dependence on effective stress, the Young’s modulus is not a constant. As
illustrated in Figure 2.21, for a particular stress‒strain curve, the modulus can be defined as
Figure 2.17 VST correction factor (μ B) versus PI.

Source: Bjerrum (1972) and Ladd et al. (1977).
v0
su
Figure 2.18 Correlation between s u/σ′ v0 and OCR for in situ clays.
u
s
q
t v0
Figure 2.19 Correlation between s u and (q t–σ v0) for in situ clays.
su pa
N N
Figure 2.20 Correlation between s u/p a and N 60 for in situ clays.

Figure 2.21 Different definitions for the soil modulus.

the initial modulus (Ei), tangent modulus at a certain stress level (Et), and secant modulus at a
certain stress level (Es). In the following, the notation E is used as the generic Young’s modulus,
whereas the notations Ei, Et, and Es are used in special cases. For instance, Eu denotes a generic
undrained modulus, whereas Eus denotes the secant undrained modulus. For the secant modulus,
it is common to adopt a stress level of one-half of the peak strength, as illustrated in Figure 2.21.
Poisson’s ratio
The effective Poisson’s ratio (μ′) varies over a relatively narrow range. Table 2.4 shows typical
values of μ′ for various soils. A clay can also behave in a drained manner under long-term load-
ing. The typical value of μ′ for clays is also shown in the table. The undrained Poisson’s ratio
(μu) is equal to 0.5 because undrained soils are incompressible.
Cohesionless soils
The typical range for the effective Young’s modulus (E′) of a cohesionless soil is shown in
Table 2.5. E′ can also be estimated by correlation with the SPT N value. Many equations for E′
versus SPT N have been proposed, and Kulhawy and Mayne (1990) summarized them by the
following transformation equations:
E ' pa = 5 x N 60 sand with fines

E ' pa = 10 x N 60 clean nornally consoliddated sand (2.23)
E ' pa = 15 x N 60 clean overconsolidated sand
The correlations of these equations are usually relatively weak.

The pressuremeter test (PMT) provides a direct measurement of the in situ horizontal modu-
lus (denoted by EPMT). Figure 2.22 illustrates the correlation between EPMT and SPT N for cohe-
sionless soils. The data points from Ohya et al. (1982) in the figure are Japanese cases, and
Table 2.4 Typical ranges of the effective Poisson’s ratio (μ′)
Soil Range of μ′
Clay 0.2 to 0.4

Dense sand 0.3 to 0.4
Loose sand 0.1 to 0.3
Source:Kulhawy and Mayne (1990).
Table 2.5 Typical ranges of E′/p a
Density Range of E′/p a

Loose 100 to 200
Medium 200 to 500
Dense 500 to 1000
Source: Poulos (1975); Kulhawy and Mayne (1990).

EPMT pa
N N
Figure 2.22 Correlation between E PMT/p a and N 60 for cohesionless soils

Data source: Ohya et al. (1982) and Ching (2020).
Kulhawy and Mayne (1990) proposed the equation EPMT/pa = 9.08 × N0.66 to fit them. Because
Japanese SPT has a higher energy ratio, the coefficient of 9.08 is modified to convert to N60:
EPMT pa = 7.64 x N 600.66 (2.24)
In practice, eq. 2.24 can be adopted to estimate E′ because E′ ≈ EPMT (Kulhawy and Mayne, 1990).
The oedometer test provides a measurement of the tangent effective constrained modulus
(M′t), the ratio of the consolidation stress increment to the vertical strain increment. This M′t
value is often correlated to the cone tip resistance for CPTu. Mayne (2007) proposed the follow-
ing transformation equation:
M t' = 5 x ( qt - - v0 ) (2.25)
Figure 2.23 illustrates the correlation between (qt–σv0) and M′t together with some data for cohe-
sive and cohesionless soils. Once M′t is estimated by eq. 2.25, the tangent effective Young’s
modulus (E′t) can be computed:
(1 + u ' ) (1 - 2u ' )
Et' = M t' x (2.26)
1- u '
Cohesive soils
The typical range for the undrained Young’s modulus (Eu) of cohesive soils is shown in Table 2.6.
Eu is commonly estimated by undrained shear strength (su). Ladd and Edgers (1972) conducted
a series of undrained direct simple shear (DSS) tests for clays. Based on the Eus/su, OCR, and PI
Mt
qt v0
Figure 2.23 Correlation between (q t–σ v0) and M′ t.

Data source: Ching (2020).
Table 2.6 Typical ranges of E u/p a
Consistency Range of E u/p a
Soft 15 to 40
Medium 40 to 80
Stiff 80 to 200
Source: Kulhawy and Mayne (1990).
data of these DSS tests, Duncan and Buchignani (1976) proposed the Eus/su–OCR relationship
in Figure 2.24. The DSS data from Ladd and Edgers (1972) are shown in the figure, along with
DSS data from other studies.
The PMT provides a direct measurement of the in situ horizontal modulus (EPMT). Figure 2.25
illustrates the correlation between EPMT and SPT N for cohesive soils. Again, based on the Japa-
nese cases from Ohya et al. (1982), Kulhawy and Mayne (1990) proposed EPMT/pa = 19.3 × N0.63.
To convert to N60, the coefficient of 19.3 is modified:
EPMT pa = 16.36 x N 600.63 (2.27)
Eq. 2.27 can be adopted to estimate Eu because Eu ≈ EPMT for clays (Kulhawy and Mayne, 1990).
Eus su
us u
s
E
Figure 2.24 Correlation between E us/s u and OCR.

Data source: Ladd and Edgers (1972) and Ching (2020).
EPMT Pa
N N
Figure 2.25 Correlation between E PMT/p a and N 60 for clays.

Data source: Ohya et al. (1982) and Ching (2020).
2.3.3 Soil consolidation

The consolidation deformation of cohesive soils is an important foundation design problem, and
the compression index (Cc) and unload–reload index (Cur) are the two main design parameters.
Their definitions are shown in Figure 2.26. Kulhawy and Mayne (1990) proposed the following
transformation equations to estimate Cc and Cur based on PI:
Cc ≈ PI 74 (2.28)
Cur ≈ PI 370 (2.29)
Figure 2.27 illustrates these equations with data for cohesive soils. Note that eq. 2.28 signifi-
cantly underestimates the Cc values for clays with LI > 1 (sensitive clays).
2.3.4 At-rest lateral earth pressure coefficient

The at-rest lateral earth pressure coefficient (K0) is defined as the ratio of the in situ horizontal
effective stress to the in situ vertical effective stress. Mayne and Kulhawy (1982) proposed the
following equation to estimate K0 by OCR for both cohesive and cohesionless soils:
K 0 = (1 - sin 0 ' ) OCR sin 0 ' (2.30)
Figure 2.28 illustrates this equation for selected ϕ′ values, together with comparison data. Mar-
chetti (1980) proposed the following equation to estimate K0 by the horizontal stress index (KD)
for the dilatometer test (DMT):
K 0 = ( K D 1.5 ) - 0.6 (2.31)

0.47
Figure 2.29 illustrates this equation along with comparison data.
Figure 2.26 Definitions of C c and C ur.

Cc
C
ur
Figure 2.27 Correlation between (C c, C ur) and PI.

Data source: Ching et al. (2022).
K0
Figure 2.28 Correlation between K 0 and OCR.

K0
KD
Figure 2.29 Correlation between K 0 and K D.

2.4 Type of loads and design loads

Loads supported by a foundation need to be estimated in foundation design. The sources of
loads generally include the following:
1. Dead loads: loads due to the weight of the structure (i.e., the load associated with all weights
permanently attached to the structure).
2. Live loads: the weight of people, traffic surcharge, and objects that are on the structure but
are not permanently attached to it.
3. Environmental loads: the loads caused by environmental factors, such as water and earth
pressures, wind, snow, wave action, and seismic loads.
Loads adopted in design are called design loads. Design loads are calculated according to the
design methods specified in applicable codes by considering load combinations. Load combina-
tions typically include dead loads, live loads, and environmental loads.
2.5 Foundation design methods

There are three design methods adopted in, but not limited to, foundation design: working stress
design, ultimate strength design, and limit state design. The first two methods are conventional
and force-based. Design criteria are used to evaluate whether the design load exceeds the load
capacity of a foundation system (or component). In contrast, the limit state design method is
nonconventional and is both “force-based” and “displacement-based” (generally called “per-
formance-based”). In addition to checking for the load capacity, this method evaluates whether
the displacement of a foundation system (or component) under the design load exceeds its dis-
placement capacity. The design concepts for the three methods are introduced in the following
sections.
2.5.1 Working stress design (WSD) method

The concept of the working stress design method is straightforward. It is to design a system or
component to sustain the design load (or design stress) without exceeding its allowable load
(or allowable stress). Therefore, this method is also called the allowable stress design (ASD)
method. The design criterion is expressed as follows:
Qd ≤ Qa (2.32)
where Qd is the design load (or working load or service load) and Qa is the allowable load.
In this design method, a factor of safety is commonly adopted to compute the allowable load
Qa, which is expressed as:
Qu
Qa = (2.33)
FS
where Qu is the ultimate capacity and FS is the factor of safety. A large FS implies a small allow-
able load. The magnitude of FS varies with the importance of structures, design load conditions,
target failure probability, consequence of failure, uncertainty in design load and capacity, etc.
Different values are adopted in the long-term load condition and short-term load conditions.
A larger FS is adopted for the long-term condition than for the short-term condition. For founda-
tion structures, such as shallow foundations under vertical loading, an FS of 3 is commonly used
in long-term conditions, under which the elastic response of the foundation system is expected.
Because the concept of this method is straightforward and simple, it is introduced for designing
various foundations in this book.
2.5.2 Ultimate strength design (USD) method

Contrary to the WSD method, which generally assumes elastic structural behavior, the ultimate
strength design (USD) method considers the ultimate strength of structures to resist the largest
loads anticipated during their design lives. The design criterion of this method is commonly
expressed as follows:
E ( LF ) Q
i ni < ( RF ) Rn (2.34)
where LFi and Qni are the load factor and nominal load, respectively, for load type i. Different i
indices represent different load sources, such as the dead load and live load. RF and Rn are the
resistance factor and nominal resistance, respectively.
In eq. 2.34, the left-hand-side term of the inequality represents the design load, which is
a combination of nominal loads; the latter are multiplied by load factors that are typically
greater than 1. The design load should be less than or equal to the design resistance. As
defined by the right-hand-side term of the inequality in eq. 2.34, the design resistance is the
nominal resistance multiplied by a resistance factor, which is typically less than 1. The word
“nominal” means that both the loads and resistances are determined according to certain
guidelines.
Because the load and resistance factors are used to modify the nominal loads and resistance,
respectively, the ultimate strength design method is also called the load and resistance fac-
tor design (LRFD) method. Different values of the load and resistance factors are adopted to
account for uncertainties in load and resistance, the target probability of failure, and the conse-
quence of failure. The LRFD method was initially applied to structural design practice for rein-
forced concrete structures in the United States. It has been widely applied in the design of bridge
foundations in North America (AASHTO, 2012; CSA, 2014). For example, in the AASHTO
(American Association of State Highway Transportation Officials) bridge design specifications,
load factors of 1.25–1.5 (or 0.9 when the effect of load is favorable) for dead loads and of
1.35–1.75 for live loads are adopted for the ultimate state. The resistance factors depend on
foundation types, soil types, analysis methods, and determination methods of soil parameters.
The suggested resistance factors for the bearing capacity evaluation of shallow foundations for
the ultimate state are given in Table 2.7.
Notably, the design load for the WSD method is a combination of unfactored loads, that
is, the load factor LFi in eq. 2.34 is not adopted. Uncertainties in loads and resistances are
lumped in FS. However, for the USD method, the design load is a combination of factored loads.
Uncertainties in loads and resistances are reflected in load and resistance factors, respectively.
Because both the USD and WSD methods are force-based, additional checking for foundation
settlement and displacement is needed.
Table 2.7 Resistance factors for shallow foundations for the ultimate state suggested in
AASHTO bridge design specifications
Method/soil/condition Resistance factor
Bearing Theoretical method (Munfakh et al., 2001), in clay 0.50

resistance Theoretical method (Munfakh et al., 2001), in sand, 0.50
using CPT
Theoretical method (Munfakh et al., 2001), in sand, 0.45
using SPT
Semiempirical method (Meyerhof, 1957), all soils 0.45
Footings on rock 0.45
Plate load test 0.55
Sliding resistance Precast concrete placed on sand 0.9
(base) Cast-in-place concrete on sand 0.8
Precast or cast-in-place concrete on clay 0.85
Soil on soil 0.9
Sliding resistance Passive earth pressure 0.5
(lateral)
Source: AASHTO (2012).
2.5.3 Limit state design (LSD) method

The limit state design (LSD) method introduces a more complete view to design a system con-
sidering different limit states. A limit state is a set of critical conditions to avoid. Each state shall
have a required performance. Two basic limit states, serviceability and ultimate limit states, are
commonly used for design. For the serviceability limit state, a system fails to perform its intended
function due to excessive settlement or displacement. For the ultimate limit state, a system is no
longer safe due to overall instability, bearing failure, failure of foundation components, etc.
The AASHTO bridge design specifications have implemented the LSD method. In addition
to the ultimate limit state, the format of the LRFD method eq. 2.34 is applied to other limit
states. In the AASHTO bridge design specifications, different load factors are specified for
different limit states. For the serviceability limit state, the load factors are generally set to 1,
and the deformations and displacements of structures and their components are checked to
ensure they are below the tolerable limits. For the ultimate limit state, the aforementioned
USD method is adopted to check the strength and stability of structures and their components.
The Eurocode 7 (Geotechnical Design, EC7) (CEN, 2004) has also implemented the LSD
method. A distinct feature of EC7 is the use of the characteristic soil parameter, defined as a
cautious estimate of the value affecting the occurrence of the limit state. More discussions of the
characteristic soil parameters are given in Section 2.6. The characteristic value of a soil param-
eter is usually more conservative than the nominal value. For instance, given soil shear strength
data, the nominal shear strength may be calculated as the average value of the test data (called
the sample mean). However, the characteristic shear strength is usually less than the sample
mean. The sample mean is not a cautious estimate of the value affecting the limit state (the true
mean) because there is an approximately 50% probability that the sample mean underestimates
the true mean. A cautious estimate requires a smaller probability (e.g., 5%) of underestimating
the true mean, so the characteristic shear strength is usually less than the sample mean. In EC7,
a conservative design is achieved on two levels. At the first level, the use of the characteristic
soil parameter leads to a conservative design. At the second level, the characteristic soil param-
eters are further divided by partial (safety) factors for additional conservatism. The first level of
conservatism concerns soil parameter variabilities and uncertainties, whereas the second level
concerns the required safety. Different from the AASHTO bridge design specifications, EC7
does not adopt load and resistance factors in eq. 2.34 but adopts partial factors. Soil strengths
(such as friction angle and undrained shear strength) are divided by partial factors, whereas
loads are multiplied by partial factors:
E (Y ) Q
Q i ki < R( Xk Y X ) (2.35)
where Qki is the characteristic load for load type i; (γQ)i is its partial factor; R is the resistance,
which depends on soil parameter X; Xk is the characteristic soil parameter; and γX is its partial
factor. The partial factors (γQ)i and γX play a similar role as the traditional safety factor, but each
load or soil parameter now has its own partial safety factor. For the serviceability limit state,
partial factors are not adopted (set to one). For the ultimate limit state, the characteristic loads
are multiplied by their partial factors, whereas the characteristic soil parameters are divided by
their partial factors.
Recently, the performance-based design (PBD) method was proposed based on the con-
cept of the limit state design (LSD) method by considering the types of actions and impor-
tance of structures. The design procedure is commonly expressed in terms of a matrix (often
called the performance matrix), as displayed in Figure 2.30. In this matrix, the structural
performances (limit states) and the magnitude of external actions are taken as two axes.
The required performance for the magnitude of each external action is indicated in the
matrix depending on the importance of the structure (Honjo, 2003). The magnitude of exter-
nal actions is divided into three levels based on the occurrence frequency and impact of
the actions. Three limit states, serviceability, reparability, and ultimate, are considered to
account for different levels of structural damage. According to this matrix, for example, an
important structure is required to be in the serviceability limit state under high frequency
and low impact and medium frequency and medium impact of loading and in the reparability
limit state under low frequency and high impact loading. The railway foundation design code
of Japan (Railway Technical Research Institute, RTRI, 2012) has shifted toward PBD. In the
code, both force and displacement are checked to ensure they meet the requirements. To per-
form PBD, a complete capacity curve of a foundation system is needed, as presented in Fig-
ure 2.31. The capacity curve displays the relationship between the foundation load capacity
and displacement. The limit states in the code consider general and earthquake conditions.
The general condition includes the serviceability and safety limit states. The foundation is
not allowed to exceed the yield point of the foundation system (Δy, Qy). The long-term and
short-term allowable bearing capacities are set to 1/3 and 2/3 of the yield bearing capacity
of the foundation (Qy), respectively. The earthquake condition includes the reparability limit
state for checking the residual displacement and the safety limit state. The foundation can be
allowed to exceed the yield point of the foundation system, but the reparability and safety of
the foundation must be considered. Table 2.8 presents an example of detailed performance
requirements for shallow foundations under different limit states provided in the code. In this
table, the performance requirements for limit states under general and earthquake conditions
are completely specified.
2.6 Characteristic soil parameters

Eurocode 7 (EC7) requires the determination of characteristic soil parameters. This section dis-
cusses the concept of the characteristic soil parameter as well as its determination.
Figure 2.30 Performance matrix.

Source: Following Honjo (2003).
Figure 2.31 Capacity curve of a foundation system.

Table 2.8 Example of performance requirements for shallow foundations
Required Performance item Performance requirements

performance
Safety limit General Ground failure Vertical load ≤ allowable

bearing capacity (f r = 0.67)
Horizontal stability Horizontal load ≤ allowable
lateral capacity (f r = 0.67)
Overturning Moment ≤ allowable moment
capacity (f r = 0.67)
Earthquake Ground failure Plastic zone at base ≤ B/4
Horizontal stability Horizontal load ≤ allowable
Overturning Max. rotation ≤ 30/1,000 rad
Serviceability Long-term Vertical Vertical load ≤ allowable
limit performance bearing capacity (f r = 0.33)
Horizontal Horizontal load ≤ allowable
Rotational Eccentricity ≤ B/6
Short-term Vertical Vertical load ≤ allowable
performance bearing capacity (f r = 0.5)
Rotational Eccentricity ≤ B/4
Reparability Earthquake level I Vertical Vertical load ≤ allowable
limit (residual bearing capacity (f r = 0.67)
displacement) Horizontal Horizontal load ≤ allowable
Rotational Moment ≤ allowable moment
capacity (f r = 0.67)
Earthquake level II Vertical Plastic zone at base ≤ B/6
Rotational Max. rotation ≤ 20/1,000 rad
Source: RTRI (2012).
2.6.1 Concept of the characteristic soil parameter

Eurocode 7 defines the characteristic soil parameter as “a cautious estimate of the value affect-
ing the occurrence of the limit state” (EC7 Clause 2.4.5.2(2)). There are two main aspects in the
definition of characteristic values: (a) a cautious estimate and (b) the value affecting the occur-
rence of the limit state. These two aspects are elaborated next.
For the first aspect, a cautious estimate is needed because soil parameters show variabili-
ties and uncertainties. For instance, soils are inherently variable in space, so there are spa-
tial variabilities. Various laboratory and field tests are also subject to measurement errors. The
transformation equations introduced in Section 2.3 are not exact, so there are transformation
uncertainties. There is only a finite amount of site investigation data, and the sample mean of the
finite data is different from the true mean, so there are statistical uncertainties.
For the second aspect, the value affecting the occurrence of the limit state is related to the
failure mechanism. EC7 provides some guidelines on the value affecting the occurrence of
the limit state. For instance, EC7 Clause 2.4.5.2(4) lists factors to consider when selecting the
characteristic value, which include, among other factors, “the extent of the zone of ground gov-
erning the behavior of the geotechnical structure at the limit state being considered.” In the lit-
erature, the zone governing the limit state is sometimes called the “influence zone.” EC7 Clause
2.4.5.2(7) further states that “the value of the governing parameter is often the mean of a range
of values covering a large surface or volume of the ground.” Consequently, the value affecting
the occurrence of the limit state is often the mean value over the influence zone.
For instance, the slip circle for the base shear failure of an excavation in clay is shown in
Figure 6.10. The design soil parameter relevant to this ultimate limit state is the undrained shear
strength (su). When the characteristic su is determined, the value affecting the occurrence of the
ultimate limit state may be taken as the mean su value over the influence zone defined by slip
arcs a-g-f-b in the figure.
2.6.2 Determination of the characteristic soil parameter

The characteristic soil parameter can be determined based on statistical methods, as EC7 Clause
2.4.5.2(11) specifies: “If statistical methods are used, the characteristic value should be derived
such that the calculated probability of a worse value governing the occurrence of the limit state
under consideration is not greater than 5%.” In other words, the probability that the character-
istic soil parameter underestimates the true mean over the influence zone should not be greater
than 5%. If the probability distribution of a soil parameter is lognormal, the characteristic soil
parameter can be determined as the 5% fractile of the lognormal distribution:
(
-1.645x ln 1+COVm2 _)
Xk = mxe 1+ COVm2 (2.36)
where Xk is the characteristic soil parameter, m is the sample mean over the influence zone,
COVm is the coefficient of variation of this sample mean, and ‒1.645 is the 5% fractile of the
standard normal random variable (namely, the probability that the standard normal random vari-
able is less than ‒1.645 is 5%). The COVm quantifies all variabilities and uncertainties, including
the spatial variability, measurement error, transformation uncertainty, and statistical uncertainty
(Phoon and Kulhawy, 1999). For the slip circle example earlier, m in eq. 2.36 can be taken as the
sample mean of the su data over the influence zone (slip arc a-g-f-b in Figure 6.10). For COVm in
eq. 2.36, Table 2.9 shows the range of COVm for the mean shear strength (su or ϕ′) over the influ-
ence zone with a size (depth) of 5 m. If the shear strength data are based on laboratory triaxial
Table 2.9 Range of COV m for soil shear strength parameters (s u and ϕ′)
Soil parameter Test Soil type COVm (%) Transformation equation
su Direct (lab UC) Clay 10–40 –

su Direct (lab UU) Clay 7–25 –
su Direct (lab CIUC) Clay 10–30 –
su VST Clay 15–50 Eq. 2.19
su qt Clay 30–40 Eq. 2.21
su SPT N Clay 40–55 Eq. 2.22
ϕ′ Direct (lab CD) Clay, sand 6–20 –
ϕ′ qt Sand 10 Eq. 2.17
ϕ′ SPT N Sand 11* Eq. 2.13
* Also based on the transformation uncertainty COV summarized in Table 1 of Ching et al. (2017).
Source: Phoon and Kulhawy (1999), assuming the size of the influence zone = 5 m.
Table 2.10 Range of COV x for other soil parameters (E and K 0)
Soil parameter Test Soil type COVm(%) Transformation equation
E PMT Direct (PMT) Sand 15–70 –

ED Direct (DMT) Sand 10–70 –
E PMT SPT N Clay 85–95 Eq. 2.27
K0 Direct (PMT) Clay 15–45 –
K0 Direct (PMT) Sand 20–55 –
K0 KD Clay 35–50 Eq. 2.31
Source: Phoon and Kulhawy (1999), assuming the size of the influence zone = 5 m.
tests (direct tests), su or ϕ′ data are directly obtained. Transformation equations, such as those in
Section 2.3, are not necessary, so COVm is typically smaller because there is no transformation
uncertainty. If the shear strength is estimated by a transformation equation, COVm is typically
larger. Table 2.9 shows the results for the shear strength. Table 2.10 further shows the range
of COVm for the modulus and at-rest lateral earth pressure coefficient (E and K0). The increase
in shear strength and modulus usually has a positive effect on resistance or capacity. Their
characteristic values can be determined by eq. 2.36. However, the increase in K0, for example,
sometimes has an adverse effect on resistance or capacity. In this case, eq. 2.36 should not be
adopted. Instead, the following equation should be adopted:
(
1.645x ln 1+ COVm2 )
Xk = mxe 1+ COVm2 (2.37)
where 1.645 rather than ‒1.645 is adopted in eq. 2.36.
Example 2.1
Consider the base shear failure (an ultimate limit state) of an excavation in clay. The influence
zone is defined by the slip arc a-g-f-b in Figure 6.10. Suppose the influence zone extends from
depths of 20 m to 25 m. The site investigation data include the undrained shear strength data
from the unconsolidated undrained test (UU) and SPT N data shown in Figures EX 2.1a and
su N s N
u
Figure EX 2.1 Site investigation data: (a) s u data from UU, (b) SPT N data, and (c) s u data trans-
formed from SPT N.
EX 2.1b, respectively. Determine the characteristic undrained shear strength for the excavation
by considering two scenarios: (a) only the undrained shear strength (su) data are available, and
(b) only the SPT N data are available.
Solution
Let the characteristic undrained shear strength be denoted by su,k.
(a)
There are 7 su data points from depths of 20 m to 25 m (22.6, 32.9, 37.8, 40.2, 28.4, 31.9, and
25.0 kPa). The sample mean (m) of the su data is:
m = (22.6 + 32.9 + 37.8 + 40.2 + 28.4 + 31.9 + 25.0)/7 = 31.2 kPa
The influence zone has a size of 25 m – 20 m = 5 m, consistent with that assumed in Table 2.9,
so the table is applicable. The COVm checked from the table (soil parameter = su; test = lab
UU) is COVm = 7–25%. An intermediate value of COVm = 15% is adopted. Inserting the m and
COVm values into eq. 2.36 yields su,k = 24.1 kPa.
(b)
First, the SPT N data are transformed to su by eq. 2.22 (Figure EX 2.1c). There are three trans-
formed su data from depths of 20 m to 25 m (24.4, 24.4, 30.5 kPa). The sample mean of the
transformed su data is:
m = (24.4 + 24.4 + 30.5)/3 = 26.4 kPa
The COVm checked from Table 2.9 (soil parameter = su; test = SPT N) is COVm = 40–55%. An
intermediate value of COVm = 45% is adopted. Inserting the m and COVm values into eq. 2.36
yields su,k = 11.9 kPa.
For this particular example, the characteristic value su,k obtained by SPT N is relatively low
because of the significant uncertainty (COVm = 45%). This significant uncertainty is mainly due
to the uncertainty of the transformation equation (i.e., eq. 2.22). The characteristic value su,k
obtained by lab UU does not require such a transformation equation.

This chapter introduces various phenomena of drained and undrained behaviors as well as the
concept of total stress and effective stress analyses. The correlation equations (or transforma-
tion equations) that can estimate design soil parameters are provided. The working stress design
(WSD), ultimate strength design (USD), and limit state design (LSD) methods are explained in
this chapter. A summary and general comments regarding this chapter are as follows:
1. In foundation design, the target soils should be classified as either drained or undrained mate-
rial, and effective stress analysis or total stress analysis should be carried out accordingly.
2. The effective stress analysis method treats the soil as a two-phase material, that is, soil parti-
cles and pore water, during the analysis process. The groundwater level and pore water pres-
sure in the soil should be taken into account. All soil parameters are expressed in terms of
effective stress. The total stress analysis method regards fine-grained soil and pore water as a
one-phase material or a soil‒water mixture, as if pore water does not exist in the material. The
groundwater level is not considered. All soil parameters are expressed in terms of total stress.
3. As the soil is stressed, under undrained or short-term conditions, a positive or negative excess
pore water pressure is generated, which is dissipated under long-term conditions. The criti-
cal shear strength of the soil should be smaller between short-term and long-term conditions.
According to the characteristics of the stress‒strain behavior of clays, the critical shear strength
for normally consolidated or slightly overconsolidated clay is the undrained shear strength,
where the undrained soil parameters su and ϕ = 0 should be adopted for analysis or design. The
critical shear strength for highly overconsolidated clay is the drained shear strength, where the
effective strength parameters c' and ϕ' are adopted for analysis and design.
4. Direct determination of design soil parameters (such as shear strengths, moduli, at-rest lateral
earth pressure coefficient, etc.) usually requires undisturbed soil specimens and sophisticated
tests. In practice, transformation equations are often adopted to estimate these design soil
parameters based on the results of convenient tests, such as SPT N values, CPT cone tip
resistance, and Atterberg limits. These transformation equations are not exact, and there is
usually a significant amount of transformation uncertainty.
5. The WSD and USD methods are force-based. The design criteria are used to evaluate whether
the design load exceeds the load capacity of a foundation system (or component). In contrast,
the limit state design method is both “force-based” and “displacement-based.” In addition to
checking for the load capacity, the LSD method also evaluates whether the displacement of a
foundation system (or component) under the design load exceeds its displacement capacity.
6. The design load for the WSD method is a combination of unfactored loads. Uncertainties in
loads and resistances are lumped in FS. For the USD method, the design load is a combina-
tion of factored loads. Uncertainties in loads and resistances are reflected in load and resist-
ance factors, respectively. Because both methods are force-based, additional checking for
foundation settlement and displacement is needed.
7. The LSD method introduces a more complete view to design a system considering differ-
ent limit states. A limit state is a set of critical conditions to be avoided. Each state has
corresponding performance requirements. Two basic limit states are the serviceability and
ultimate limit states. Recently, the performance-based design (PBD) method was proposed
based on the concept of the LSD method by considering the types of actions and importance
of structures. To perform PBD, a complete capacity curve of a foundation system for the
relationship between the foundation load capacity and displacement is needed.
8. In Eurocode 7, the characteristic value is defined as a cautious estimate of the value affect-
ing the occurrence of the limit state. The characteristic value explicitly addresses various
uncertainties related to the estimation of design soil parameters (such as spatial variability,
transformation uncertainty, and statistical uncertainty). In Eurocode 7, the required safety
level is separately addressed for partial factors.
References
AASHTO. (2012), LRFD Bridge Design Specifications, 6th edition, Washington, DC: American Associa-
tion of State Highway and Transportation Officials (AASHTO).
Bjerrum, L. (1972), Embankments on Soft Ground. ASCE Conf. on Performance of Earth and Earth-
Supported Structures, ASCE, New York, NY, 2, 1–54.
CEN. (2004), Eurocode 7: Geotechnical Design. Part 1: General Rules, EN 1997–1, Brussels: European
Committee for Standardisation.
Chen, J.R. (2004), Axial Behavior of Drilled Shafts in Gravelly Soils, Ph.D. dissertation, Cornell Univer-
sity, Ithaca, New York.
Ching, J. (2020), Personal Database.
Ching, J., Lin, G.H., Chen, J.R. and Phoon, K.K. (2017), Transformation models for effective friction
angle and relative density calibrated based on a multivariate database of coarse-grained soils. Canadian
Geotechnical Journal, Vol. 54, No. 4, pp. 481–501.
Ching, J. and Phoon, K.K. (2014), Transformations and correlations among some clay parameters—the
global database. Canadian Geotechnical Journal, Vol. 51, No. 6, pp. 663–685.
Ching, J., Phoon, K.K. and Wu, C.T. (2022), Data-centric quasi-site-specific prediction for compressibility
of clays. Canadian Geotechnical Journal. Vol. 59, No. 12, pp. 2033–2049.
CSA (Canadian Standards Association). (2014), Canadian Highway Bridge Design Code. CAN/CSA-
S6-14, Mississauga, Ontario, Canada: CSA.
Duncan, J.M. and Buchignani, A.L. (1976), An Engineering Manual for Settlement Studies. Department of
Civil Engineering, Berkeley: University of California at Berkeley.
Hatanaka, M. and Uchida, A. (1996), Empirical correlation between penetration resistance and internal
friction angle of sandy soils. Soils and Foundations, Vol. 36, No. 4, pp. 1–9.
Holtz, R.D., Kovacs, W.D. and Sheahan, T.C. (2011), An Introduction to Geotechnical Engineering, New
York: Pearson Education, Inc.
Honjo, Y. (2003), Comprehensive Design Codes Development in Japan: Geo-code 21 ver. 3 and Code
PLATFORM ver. 1, LSD2003: International Workshop on Limit State Design in Geotechnical Engineer-
ing Practice.
Idriss, I.M. and Boulanger, R.W. (2008), Soil Liquefaction during Earthquakes. Monograph MNO-12,
Oakland, CA: Earthquake Engineering Research Institute.
Jamiolkowski, M., Ladd, C.C., Germain, J.T. and Lancellotta, R. (1985), New developments in field and
laboratory testing of soils. In Proceedings of the 11th International Conference on Soil Mechanics and
Foundation Engineering, AA Balkema, Rotterdam, Netherlands, Vol. 1, pp. 57–153.
Kulhawy, F.H. and Mayne, P.W. (1990), Manual on Estimating Soil Properties for Foundation Design,
Report EL-6800, Palo Alto: Electric Power Research Institute.
Ladd, C.C. and Edgers, L. (1972), Consolidated-Undrained Direct-simple Shear Tests on Saturated Clays.
MIT Research Report R72-82, Cambridge, MA: Massachusetts Institute of Technology.
Ladd, C.C., Foott, R., Ishihara, K., Schlosser, F. and Poulos, H.G. (1977), Stress-deformation and strength
characteristics. In Proceedings of 9th International Conference on Soil Mechanics and Foundation
Engineering, Japanese Society of Soil Mechanics and Foundation Engineering, Tokyo, pp. 421–494.
Liao, S.S. and Whitman, R.V. (1986), Overburden correction factors for SPT in sand. Journal of Geotech-
nical Engineering, ASCE, Vol. 112, No. 3, pp. 373–377.
Locat, J. and Demers, D. (1988), Viscosity, yield stress, remolded strength, and liquidity index relation-
ships for sensitive clays. Canadian Geotechnical Journal, Vol. 25, No. 4, pp. 799–806.
Marchetti, S. (1980), In-situ tests for flat dilatometer. Journal of the Geotechnical Engineering Division,
ASCE, Vol. 106, No. GT3, pp. 299–321.
Mayne, P.W. (2007), NCHRP Synthesis 368: Cone Penetration Test. Transportation Research Board,
Washington, DC: National Academies Press.
Mayne, P.W., Christopher, B.R. and DeJong, J. (2001), Manual on Subsurface Investigations. National
Highway Institute Publication No. FHWA NHI-01-031, Federal Highway Administration, Washing-
ton, DC.
Mayne, P.W. and Kulhawy, F.H. (1982), K0-OCR relationships in soil. Journal of the Geotechnical Engi-
neering Division, ASCE, Vol. 108, No. GT6, pp. 851–872.
Mesri, G. and Huvaj, N. (2007), Shear strength mobilized in undrained failure of soft clay and silt depos-
its. In Advances in Measurement and Modeling of Soil Behaviour (GSP 173), Ed., D.J. DeGroot et al.,
Reston, VA: ASCE, pp. 1–22.
Meyerhof, G.G. (1957), The Ultimate Bearing Capacity of Foundations on Slopes. In Proc., 4th Interna-
tional Conference on Soil Mechanics and Foundation Engineering, Butterworths, London.
Munfakh, G., Arman, A., Collin, J.G., Hung, J.C.-J. and Brouillette, R.P. (2001), Shallow Foundations
Reference Manual, FHWA-NHI-01-023. Federal Highway Administration, U.S. Department of Trans-
portation, Washington, DC.
NAVFAC (1982), Soil Mechanics (DM 7.1), Naval Facilities Engineering Command, Alexandra.
Ohya, S., Imai, T. and Matsubara, M. (1982), Relationships between N value by SPT and LLT pressurem-
eter results. In Proceedings, 2nd European Symposium on Penetration Testing, CRC Press, Amsterdam,
Vol. 1, pp. 125–130.
Phoon, K.K. and Kulhawy, F.H. (1999), Evaluation of geotechnical property variability. Canadian Geo-
technical Journal, Vol. 36, No. 4, pp. 625–639.
Poulos, H.G. (1975), Settlement of isolated foundations. In Soil Mechanics—Recent Developments, Eds.,
Valliappan et al., Australia, NSW: Unisearch, 181–212.
Railway Technical Research Institute (RTRI). (2012), Design Standards for Railway Structures and Com-
mentary (Foundation Structures). (in Japanese). Tokyo: Railway Technical Research Institute.
Skempton, A.W. (1986), Standard penetration test procedures and the effect in sands of overburden
pressure, relative density, particle size, aging, and overconsolidation. Geotechnique, Vol. 36, No. 3,
pp. 425–447.
Terzaghi, K. and Peck, R.B. (1967), Soil Mechanics in Engineering Practice, 2nd edition, New York: John
Wiley and Sons.
Chapter 3
Shallow foundations
3.1 Introduction
All structures, such as buildings and bridges, need foundations to transfer structure loads to soil.
According to the embedment depth of foundations, structure foundations are classified into two
main categories: shallow and deep. Shallow foundations transfer structure loads to the near-
surface soils, as shown in Figure 3.1. They are the simplest and oldest foundation type because
of their easy construction and low construction cost.
In general, shallow foundations are situated on sturdy ground. However, for sites with soft
soil, because of the low stiffness and strength of the soil, shallow foundations under loading may
undergo unfavorable settlement and even cause ground failure. Under these conditions, some
approaches, such as enlargement of the foundation base, inclusion of a basement, and ground
improvement, can be used to improve the performance of shallow foundations. Alternatively,
deep foundations may be more appropriate to transfer loading to deep stronger soil.
Figure 3.1 Types of shallow foundations: (a) isolated footing, (b) combined footing, (c) continu-
ous footing, and (d) mat foundation.
DOI: 10.1201/9781003350019-3
Shallow foundations 91
To ensure that a structure achieves its anticipated performance, foundations should be well
designed and constructed. In this chapter, we will introduce the design principles and basic
design methods of shallow foundations.
3.2 Types of shallow foundations

Shallow foundations include spread footings and mat foundations, as shown in Figure 3.1.
A spread footing is an enlargement at the bottom of a column or bearing wall that is used to
spread structure loads over a large soil area. Spread footings are often used in small- to medium-
sized structures on sites with moderate to good soil conditions. A footing that supports a single
column is called an isolated footing (Figure 3.1a). In some cases, when columns are close, a
combined footing (Figure 3.1b) may be used to support two or more columns. When a footing
supports a load-bearing wall, it is referred to as a continuous footing (Figure 3.1c).
For very weak and highly compressible soils, it is more appropriate and economical to use
mat foundations (also called raft foundations) (Figure 3.1d). A mat foundation is a very large
spread footing that supports an entire structure. The large size of the footing base can help
reduce the contact pressure between the bottom of the footing and the soil below, and the rigid-
ity of the footing slab can limit excessive differential settlement among columns.
3.3 Components of shallow foundation design

Designing a shallow foundation involves the following four main items:
1. Evaluation of vertical bearing capacity

2. Evaluation of eccentricity
3. Evaluation of settlement
4. Structural design of foundation components
The first two items concern the capacity issue to avoid bearing failure. Generally, a founda-
tion is designed for centric loading. However, in many cases, the foundation may also be sub-
jected to eccentric loading, such as slabs of retaining walls and footings under moment loading.
The eccentric loading will cause nonuniform contact pressure and reduce the bearing capacity.
The third item concerns the foundation settlement, which is essential for the functionality of
structures. Finally, the fourth item is to perform a detailed structural design of foundation com-
ponents, including foundation slabs and slab–column connections, to ensure sufficient strength
for transmitting structure loads to the ground.
In the following sections, we will introduce the details of items 1–3. Item 4 is not addressed
since it concerns reinforced concrete design.
3.4 Vertical bearing capacity
3.4.1 Ultimate bearing capacity

The bearing pressure qb is the contact pressure between the foundation and soil, which is
defined as:
Q
qb = (3.1)
A
92 Shallow foundations
Figure 3.2 Failure mechanism of a footing under vertical loading
where Q is the load carried by the footing and A is the bottom area of the footing.
As the load increases, the bearing pressure of the footing may reach its ultimate capacity,
called the ultimate bearing capacity qu, leading to bearing failure. Shear failure occurs within
the surrounding soil, as shown in Figure 3.2, generally accompanied by excessive foundation
settlement. Therefore, to prevent shear failure, a footing should be well designed to prevent the
bearing pressure from reaching its ultimate bearing capacity.
3.4.2 Bearing failure modes

A complete shear failure for a footing normally has three failure zones, as shown in Figure 3.2.
As the bearing pressure increases, zones I to III gradually develop. Zone I is initiated first, and
the soil in this zone is displaced downward. Then, zone II is formed, and the soil in this zone is
pushed laterally. Eventually, zone III is generated and pushed upward. Depending on the devel-
opment of the three failure zones, there are three main bearing failure modes: general shear fail-
ure, local shear failure, and punching shear failure, as shown in Figure 3.3. The bearing failure
mode is dependent on the soil type and density and the embedment depth.
General shear failure, as shown in Figure 3.3a, occurs in strong soil, such as dense sand or
stiff clay. The aforementioned three failure zones are fully mobilized. The failure surface is well-
defined and global. Failure occurs suddenly (brittle failure) when the bearing pressure reaches a
peak, called the ultimate bearing capacity in the bearing pressure–settlement plot, as displayed
in Figure 3.4. The settlement for the ultimate bearing capacity is approximately 4–10% of the
width of the footing. For this type of failure, a clear bulge appears on the ground surface.
Local shear failure, as shown in Figure 3.3b, occurs in medium strong soil, such as medium
dense sand. The failure surface in the soil will gradually extend outward from the foundation.
Unlike general shear failure, only zones I and II (solid lines) are generated. The failure surface
is well-defined only below the footing but is not obvious near the ground surface. The bearing
pressure–settlement plot is gentler than that under general bearing failure, as displayed in Fig-
ure 3.4. A sudden failure does not occur (ductile failure) without producing a peak in the bearing
pressure–settlement plot. When zones I and II are mobilized, the bearing pressure reaches qy,
Figure 3.3 Types of bearing failure modes: (a) general shear failure, (b) local shear failure,
(c) punching shear failure.
Source: Adapted from Vesic (1963).
generally referred to as the yield bearing capacity. Beyond this point, settlement significantly
increases as the bearing pressure increases. Considerable movement is required for the failure
surface in the soil to extend to the ground surface (dashed lines, zone III). The bearing pres-
sure at which this development happens is referred to as the ultimate bearing capacity qu. The
load–settlement plot after this point is steep and linear. The settlement for the ultimate bearing
capacity is approximately 12–22% of the width of the footing.
The punching shear failure, as shown in Figure 3.3c, occurs in loose sand or soft clay. The
shear failure develops downward gradually with increasing settlement, with little or no bulging
Figure 3.4 Bearing pressure–settlement plots for the three bearing failure modes.
at the ground surface. The failure surface in the soil does not extend to the ground surface (only
zone I and/or zone II can be mobilized). The bearing pressure–settlement plot is similar to that
under local shear failure (no peak is observed); however, it is softer, as displayed in Figure 3.4.
The bearing pressure reaches qy as zone I or a part of zone II is mobilized. Beyond the yield
point, settlement significantly increases as the bearing pressure increases. Generally, when the
settlement reaches 15–25%, the width of the footing under bearing pressure reaches its ultimate
value. Beyond the ultimate bearing capacity qu, the pressure–settlement plot is steep and linear.
The difference between qy and qu is small since zone III (even zone II) is not mobilized.
For footings on sand, Vesic (1963) found from model tests that the failure modes are depend-
ent on the density of sand and the embedment depth of the footing, as shown in Figure 3.5.
In this figure, B indicates the footing width, and Df indicates the embedment depth. For foot-
ings with shallow embedment, general shear failure occurs on dense sand, local shear failure
occurs on medium dense sand, and punching shear failure occurs on loose sand. With increasing
embedment depth, global shear failure is unlikely to occur. For embedment depths greater than
five times the width of the footing, punching shear failure always occurs.
3.4.3 Terzaghi’s bearing capacity theory

Terzaghi (1943) developed a theory to estimate the ultimate bearing capacity of footings. Ter-
zaghi defined foundations with embedment depths less than or equal to the foundation width
as shallow foundations. The theory considers a rough continuous or strip foundation (width-to-
length ratio approaches 0) and the general failure mode. The following failure zones developed
below and around the footing are assumed, as shown in Figure 3.6.
Figure 3.5 Types of shear failure for various embedment depths and relative densities.
Source: Vesic (1973).
Figure 3.6 Shear failure mechanism of Terzaghi’s bearing capacity theory.
Zone I: The triangular (elastic) zone A'OA immediately under the foundation. The angles
A'AO and AA'O are assumed to be equal to the soil friction angle ϕ. Zone II: The radial shear
zones AOC and A'OC', with the curves OC and OC', respectively, being arcs of logarithmic
spirals. Zone III: Two triangular Rankine passive zones ACE and A'C'E'.
The shear resistance offered by the soil and footing sides within depth Df is considered not
significant and is ignored. The overburden effect of the soil above the bottom of the foundation
is replaced by an equivalent surcharge q = γ Df (γ = unit weight of soil).
The ultimate bearing capacity of the foundation can be deduced by considering the equilib-
rium of the triangular wedge A'OA. Consider that AO and A'O are two walls that push the soil
wedges AOCE and A'OC'E', respectively, to cause passive failure. Passive force Pp is inclined at
an angle ϕ to the perpendicular drawn to the wedge faces (AO and A'O), assuming the angle of
wall friction equals the angle of friction of the soil. If the bearing pressure is applied to the foun-
dation to cause general shear failure, the passive force will act on each face of the soil wedge.
According to the free body diagram of zone I shown in Figure 3.7, considering the equilibrium
of vertical forces:
qu . 2b .1 = -W + 2C sin 0 + 2 Pp (3.2)
where qu is the ultimate bearing capacity, b = B/2, W is the weight of the wedge A'OA carried by
the footing, and C is the cohesion on the faces of the wedge A'OA.
The passive force Pp is the sum of three components: Ppγ produced by the weight of the shear
zone AOCE, Ppc produced by the soil cohesion, and Ppq produced by the surcharge, as expressed
in eq. 3.3.
Pp =Ppq + Ppc + Ppy (3.3)
Figure 3.7 Free body diagram of zone I.

Combining eqs. 3.2 and 3.3 gives:
Ppq{ Ppc } { Ppy y b }

qu = +| + c tan 0 | + | - tan 0 | (3.4)
b { b } { b 2 }
The three terms of eq. 3.4 can be further expressed in the form:
1
qu = qN q + cN c + y BNy (3.5)
2
where Nq, Nc, and Nγ are the bearing capacity factors.

Comparing eqs. 3.4 and 3.5, Nq, Nc, and Nγ are the bearing capacity factors representing the
contributions of surcharge, cohesion, and soil weight to the ultimate bearing capacity, respec-
tively. Once Ppq, Ppc, and Ppγ are given, the factors Nq, Nc, and Nγ can be determined accordingly.
Considering the force equilibrium of zones 2 and 3, Ppq, Ppc, and Ppγ are derived individually
by ignoring the contributions of the other two components. Therefore, Nq, Nc, and Nγ are deter-
mined as functions of ϕ, as shown in Table. 3.1 and Figure 3.8, in which Nq and Nc have analyti-
cal forms as follows:
e 2 (3- / 4 -0 / 2 ) tan 0
Nq = (3.6)
2 cos 2 (45 + 0 / 2)
{ e 2 (3- / 4 -0 / 2 ) tan 0 }
N c = cot 0 | -1| = cot 0 ( N q - 1) (3.7)
{ 2 cos 2
( 4 5 + 0 / 2) }
As the angle of friction influences the size of the failure zones, a larger value of the angle of
friction induces a larger failure surface and therefore results in larger bearing capacity factors.
Figure 3.8 Terzaghi’s bearing capacity factors.

Table 3.1 Values of Terzaghi’s bearing capacity factors
ϕ Nc Nq Nγ ϕ Nc Nq Nγ
0 5.70 1.00 0.00 26 27.09 14.21 9.84

1 6.00 1.10 0.01 27 29.24 15.90 11.60
2 6.30 1.22 0.04 28 31.61 17.81 13.70
3 6.62 1.35 0.06 29 34.24 19.98 16.18
4 6.97 1.49 0.10 30 37.16 22.46 19.13
5 7.34 1.64 0.14 31 40.41 25.28 22.65
6 7.73 1.81 0.20 32 44.04 28.52 26.87
7 8.15 2.00 0.27 33 48.09 32.23 31.94
8 8.60 2.21 0.35 34 52.64 36.50 38.04
9 9.09 2.44 0.44 35 57.75 41.44 45.41
10 9.60 2.69 0.56 36 63.53 47.16 54.36
11 10.16 2.98 0.69 37 70.07 53.80 65.27
12 10.76 3.29 0.85 38 77.50 61.55 78.61
13 11.41 3.63 1.04 39 85.97 70.61 95.03
14 12.11 4.02 1.26 40 95.66 81.27 115.31
15 12.86 4.45 1.52 41 106.81 93.85 140.51
16 13.68 4.92 1.82 42 119.67 108.75 171.99
17 14.56 5.45 2.18 43 134.58 126.50 211.56
18 15.52 6.04 2.59 44 151.95 147.74 261.60
19 16.56 6.70 3.07 45 172.29 173.29 325.34
20 17.69 7.44 3.64 46 196.22 204.19 407.11
21 18.92 8.26 4.31 47 224.55 241.80 512.84
22 20.27 9.19 5.09 48 258.29 287.85 650.67
23 21.75 10.23 6.00 49 298.72 344.64 831.99
24 23.36 11.40 7.08 50 347.51 415.15 1072.80
25 25.13 12.72 8.34
*From Kumbhojkar (1993).
In addition, when the angle of friction becomes larger than 30º, there is a rapid increase in the
values of the factors.
The preceding bearing capacity factors are derived in general form for a c-ϕ soil. For the
drained, long-term condition, effective stress analysis is applied using c = c', ϕ = ϕ', γ = γmoist
(above groundwater table) or γ' (below groundwater table) in eq. 3.5. However, for saturated
clay under the undrained condition (short-term condition), total stress analysis based on the
“ϕ = 0 concept” is applied, and the values c = cu, ϕ = ϕu = 0, and γ = γsat are used in eq. 3.5. For
this condition, Nq = 1, Nc = 5.7, and Nγ = 0.
Eq. 3.5 is used for strip foundations. For square and circular foundations, empirical modifica-
tions are made from model tests to consider the effect of footing shape as follows:
For square foundations,
qu = 1.3cN c + qN q + 0.4y BNy (3.8)
For circular foundations with diameter D:
qu = 1.3cN c + qN q + 0.3y DNy (3.9)

Example 3.1
A square footing is 2 m × 2 m in plan view, situated on a dry sandy stratum with an embedment
depth of 1 m. The angle of friction of sand is 32°. The unit weight of sand is 18 kN/m3. Apply
Terzaghi’s bearing capacity equation to estimate the ultimate bearing capacity of the footing.
Solution
Eq. 3.8 is used.
For this sandy layer:
c = c' = 0
ϕ = ϕ' = 32°, Nc = 44.04, Nq = 28.52, Nγ = 26.87 (see Table 3.1)
qu = (1.3)(0)(44.04) + (1)(18)(28.52) + (0.4)(18)(2)(26.87) = 900.29 kN/m2
Example 3.2
A square footing is 2 m × 2 m in plan view, situated on a saturated clay stratum with an embed-
ment depth of 1 m. The groundwater table is located at the ground surface. The saturated unit
weight of the soil is 18 kN/m3. The effective angle of friction and cohesion of the soil are 25°
and 20 kN/m2, respectively. The undrained shear strength of the clay is 100 kN/m2. Apply Ter-
zaghi’s bearing capacity equation to estimate the ultimate bearing capacity of the footing for
long-term and short-term conditions.
Solution
Eq. 3.8 is used.
For the long-term condition, effective analysis is conducted.
c = c' = 20 kN/m2, ϕ = ϕ' = 25°, γ = γ' = 18-9.81 = 8.19 kN/m3

Nc = 25.13, Nq = 12.72, Nγ = 8.34 (see Table 3.1)
qu (long-term)
= (1.3)(20)(25.13) + (1)(8.19)(12.72) + (0.4)(8.19)(2)(8.34) = 812.2 kN/m2
For the short-term condition, total stress analysis is conducted under the ϕ = 0 condition.
c = cu = 100 kN/m2, ϕ = 0°, γ = γsat = 18 kN/m3

Nc = 5.7, Nq = 1, Nγ = 0
qu (short-term)
= (1.3)(100)(5.7) + (1)(18)(1) + (0.4)(18)(2)(0) = 759 kN/m2
Because qu (short-term) < qu (long-term), the short-term bearing capacity governs the design.
3.4.4 General bearing capacity equation

There are three major limitations when using Terzaghi’s bearing capacity equation mentioned
in Section 3.4.3. Terzaghi’s equation does not address cases of rectangular footings of B/L ≠ 1.
It does not consider the shear resistance of the soil above the foundation base. The load on the
foundation is vertical without any inclination. To improve the aforementioned limitations, Mey-
erhof (1963) proposed a general bearing capacity equation by introducing shape, depth, and load
inclination factors.
1
qu = cN c Fcs Fcd Fci +qN q Fqs Fqd Fqi + y BNy Fy s Fy d Fy i (3.10)
2
where:
c = cohesion.
q = overburden pressure at the bottom of the foundation.
y = unit weight of soil.
B = width of foundation (= diameter for a circular footing).
Fcs, Fqs, Fys = shape factors.
Fcd, Fqd, Fyd = depth factors.
Fci, Fqi, Fyi = load inclination factors.
Nq, Nc, Ny = bearing capacity factors.
The shape factors Fqs, Fcs, and Fys are for rectangular footings of B/L ≠ 1 and square and
circular footings (B/L = 1). The depth factors Fqd, Fcd, and Fyd reflect the shear resistance above
the base of the embedded footings. The load inclination factors Fqi, Fci, and Fyi are for inclined
loading on the footing. These factors are listed in Table 3.2.
Table 3.2 Meyerhof ’s shape, depth, and inclination factors
Shape factors
For q = 0
Fcs = 1+ 0.2(B / L )
Fqs = Fy s = 1
For q ≥ 10°
Fcs = 1+ 0.2(B / L ) tan2 ( 45 + q / 2)
Fqs = Fy s = 1+ 0.1(B / L ) tan2 ( 45 + q / 2)
Depth factors
For q = 0
Fcd = 1+ 0.2(Df / B)
Fqd = Fy d = 1
For q ≥ 10°
Fcd = 1+ 0.2(Df / B) tan( 45 + q / 2)
Fqd = Fy d = 1+ 0.1(Df / B) tan( 45 + q / 2) D f: embedment depth of footing
Inclination factors
Fci = Fqi = 1− (B / 90. )2
Fy i = 1− (B / q)2 B: inclination of the load on the footing with respect to the vertical
Source: Meyerhof (1963).
Different from Terzaghi’s assumption, the angle of triangular zones A'AO and AA'O (zone I)
of the failure surface with respect to the ground surface is assumed to be 45 + φ/2, as shown in
Figure 3.9, considering that the footing base is smooth. This zone is usually referred to as the
Rankine active zone. Upon this modified failure surface, another set of bearing capacity factors
is derived as follows, with the values presented in Table 3.3.
N q = tan 2 (45 + 0 / 2)e- tan 0 (3.11)
N c = cot 0 ( N q - 1) (3.12)
Ny = tan 1.40 ( N q - 1) (3.13)
Similar to Terzaghi’s equation, Meyerhof’s equation is presented in general form for a c-ϕ
soil. For the drained long-term condition, effective stress analysis is applied using the values
Figure 3.9 Shear failure mechanism for Meyerhof ’s bearing capacity equation.
Table 3.3 Values of Meyerhof ’s bearing capacity factors
0 5.14 1.00 0.00 31 32.67 20.63 18.56

1 5.38 1.09 0.00 32 35.49 23.18 22.02
2 5.63 1.20 0.01 33 38.64 26.09 26.17
3 5.90 1.31 0.02 34 42.16 29.44 31.15
4 6.19 1.43 0.04 35 46.12 33.30 37.15
5 6.49 1.57 0.07 36 50.59 37.75 44.43
6 6.81 1.72 0.11 37 55.63 42.92 53.27
7 7.16 1.88 0.15 38 61.35 48.93 64.07
8 7.53 2.06 0.21 39 67.87 55.96 77.33
9 7.92 2.25 0.28 40 75.31 64.20 93.69
10 8.34 2.47 0.37 41 83.86 73.90 113.99
11 8.80 2.71 0.47 42 93.71 85.37 139.32
12 9.28 2.97 0.60 43 105.11 99.01 171.14
13 9.81 3.26 0.74 44 118.37 115.31 211.41
14 10.37 3.59 0.92 45 133.87 134.87 262.74
15 10.98 3.94 1.13 46 152.10 158.50 328.73
16 11.63 4.34 1.37 47 173.64 187.21 414.33
17 12.34 4.77 1.66 48 199.26 222.30 526.45
18 13.10 5.26 2.00 49 229.92 265.50 674.92
19 13.93 5.80 2.40 50 266.88 319.06 873.86
20 14.83 6.40 2.87 51 311.75 385.98 1,143.95
21 15.81 7.07 3.42 52 366.66 470.30 1,516.08
22 16.88 7.82 4.07 53 434.42 577.50 2,037.29
23 18.05 8.66 4.82 54 518.80 715.07 2,781.13
24 19.32 9.60 5.72 55 624.92 893.48 3,865.77
25 20.72 10.66 6.77 56 759.79 1,127.44 5,487.58
26 22.25 11.85 8.00 57 933.17 1,437.96 7,986.27
27 23.94 13.20 9.46 58 1,158.83 1,855.52 11,979.47
28 25.80 14.72 11.19 59 1,456.54 2,425.08 18,664.42
29 27.86 16.44 13.24 60 1,855.10 3,214.14 30,570.95
30 30.14 18.40 15.67
Source: Meyerhof (1963).
c = c', ϕ = ϕ', and γ = γmoist (above groundwater table) or γ' (below groundwater table) in eq. 3.10.
However, for saturated clay under the undrained condition (short-term condition), total stress
analysis based on the ϕ = 0 condition is applied, and the values c = cu, ϕ = ϕu = 0, and γ = γsat are
used in eq. 3.10. For this condition, Nq = 1, Nc = 5.14, and Nγ = 0.
Example 3.3
A rectangular footing is 2 m × 4 m in plan view, situated on a dry sandy stratum with an embed-
ment depth of 1 m. The angle of friction of sand is 32°. The unit weight of sand is 18 kN/m3.
Apply Meyerhof’s bearing capacity equation to estimate the ultimate bearing capacity of the
footing.
Solution
Eq. 3.10 is used.
c = c' = 0
ϕ = ϕ' = 32°, Nc = 35.49, Nq = 23.18, Nγ = 22.02
B = 2 m, L = 4 m
Fcs = 1 + (0.2)(2/4)tan2(45 + 32/2) = 1.325
Fqs = Fγs = 1 + (0.1)(2/4)tan2(45 + 32/2) = 1.163
Fcd = 1 + (0.2)(1/2)tan(45 + 32/2) = 1.180
Fqd = Fγd = 1 + (0.1)(1/2)tan(45 + 32/2) = 1.090
Fci = Fqi = 1
Fγi = 1
qu = (0)(35.49)(1.325)(1.180)(1) + (1)(18)(23.18)(1.163)(1.09)(1) + (1/2)(18)(2)(22.02)
(1.163)(1.09)(1) = 1031.38 kN/m2
3.4.5 Factor of safety and allowable bearing capacity

Calculating the allowable load-bearing capacity of shallow foundations requires applying a fac-
tor of safety (FS) to the ultimate bearing capacity. The magnitude of the factor of safety depends
on the design load conditions. An FS ≥ 3 is usually adopted for the long-term load condition. The
following two approaches for the design inequality and associated allowable bearing capacity
are commonly adopted in design codes:
1. In terms of the gross ultimate bearing capacity qu
qb ≤ qall (qall = qu/FS) (3.14)
where qall = gross allowable bearing capacity
2. In terms of the net ultimate bearing capacity qu(net)
In general, the embedment depth of a footing is a given parameter in design, and therefore,
the surcharge q is regarded as a certain value without uncertainty. This approach does not apply
an FS to the surcharge. To address the net contribution of ultimate bearing capacity and bear-
ing pressure exceeding the surcharge above the foundation level before the foundation is con-
structed, the net ultimate bearing capacity qu(net) and net bearing pressure qb(net) are defined as
follows.
qu(net) = qu–q (3.15)

qb(net) = qb–q (3.16)
Therefore, the design inequality is:
qb(net) ≤ qall(net) (qall(net) = qu(net)/FS) (3.17)
where qall(net) = net allowable bearing capacity.
Example 3.4
ment depth of 1 m. The angle of friction of sand is 32°. The unit weight of sand is 18 kN/m3.
1. Apply Meyerhof’s bearing capacity equation to estimate the net ultimate bearing capacity of
the footing.
2. Compute a factor of safety in terms of the net ultimate bearing capacity, given that the footing
is subjected to a total centric vertical load of 3,200 kN on its base.
Solution
1. According to example 3.3, qu = 1031.38 kN/m2.
The surcharge q = γDf = 18 × 1 = 18
qu(net) = 1031.38 – 18 = 1013.38 kN/m2
2. qb(net) = 3200/(2×4) – 18 = 382 kN/m2

FS = 1013.38/382 = 2.65
3.4.6 Modification of bearing capacity equations for

the groundwater table
This section describes the considerations for the groundwater table when using the bearing
capacity equation in effective stress and total stress analyses.
1. Effective stress analysis
The position of the groundwater table may alter the effective stress in the soil and further the
ultimate bearing capacity of footings. Different groundwater levels may yield different values
of the parameter q in the second term and parameter γ in the third term of the bearing capacity
equations.
Figure 3.10 illustrates the three groundwater table conditions. Case 1 is for the groundwater
table located above the bottom of the footing (the depth of groundwater table Dw < the embed-
ment depth Df). Cases 2 and 3 are for the groundwater table below the bottom of the footing.
Assuming that the main influence depth of bearing failure of the footing is the same as the width
of footing B, case 2 indicates that the groundwater table is within the limit of the influence zone
(Dw < Df + B), whereas case 3 indicates that the groundwater table is on or outside the limit of
the influence zone (Dw ≥ Df + B).
Assume γ2 is the unit weight of the soil above the bottom of the footing, which is used to com-
Df
pute q in the second term of the bearing capacity equation, that is, q = | y 2 dz, and γ1 is the unit
0
weight of the soil below the bottom of the footing, which is used in the third term of the bearing
capacity equation. γ1 and γ2 are computed as follows for these three cases.
Figure 3.10 Three cases for groundwater table depth.

For case 1, γ1 uses the submerged unit weight of soil, and γ2 is dependent on the location of
the groundwater table:
y 1 = y 1' = y 1,sat - y w
y 2 = y 2,moist (above groundwater table)
y 2 = y 2' = y 2,sat - y w (below groundwater table)
where γ 1', γ 1,sat are the effective and saturated unit weights of the soil below the bottom of the
footing, respectively, and γ 2', γ 2, moist ,γ 2, sat are the effective, moist, and saturated unit weights of
the soil above the bottom of the footing, respectively.
For case 2, γ2 is not influenced by the location of the groundwater table. However, since the
groundwater table is located within the influence zone of bearing failure of the footing, a depth-
weighted average γ1 as what follows is used for partially considering the effect of water.
{ Dw - D f } { B - ( Dw - D f ) } { Dw - D f }
y1 = | | y 1, moist +(y 1, sat - y w ) | | =y 1'+ | | (y 1,moist -y 1')
{ B } { B } { B }
y 2 = y 2,moist
For case 3, neither q nor γ is influenced by the location of the groundwater table.
y 1 = y 1,moist
y 2 = y 2,moist
where γ 1, moist , γ 2, moist are the moist unit weight of the soil below and above the bottom of the
footing, respectively.
2. Total stress analysis
This analysis is applied for the saturated clay condition and is applicable only to case 1 in Figure
3.10. γ1 is the saturated unit weight, but its value has no effect on the bearing capacity as Nγ = 0;
γ2 needs to consider the position of the groundwater table. That is:
y 1 = y 1,sat
y 2 (above water table) = y 2,moist
y 2 (below water tabble) = y 2,sat
Example 3.5
A rectangular footing is 2 m × 4 m in plan view, situated on a sandy stratum with an embed-
ment depth of 1 m. The angle of friction of sand is 32°. The groundwater table is located at a
depth of 0.5 m. The unit weights of sand above and below the groundwater table are 18 and
20 kN/m3, respectively. Apply Meyerhof’s bearing capacity equation to estimate the ultimate
bearing capacity of the footing.
Solution
Eq. 3.10 is used.
c = c' = 0
ϕ = ϕ' = 32°, Nq = 23.18, Nc = 35.49, Nγ = 22.02
B = 2 m, L = 4 m
Fcs = 1 + (0.2)(2/4)tan2(45 + 32/2) = 1.325
Fqs = Fγs = 1 + (0.1)(2/4)tan2(45 + 32/2) = 1.163
Fcd = 1 + (0.2)(1/2)tan(45 + 32/2) = 1.180
Fqd = Fγd = 1 + (0.1)(1/2)tan(45 + 32/2) = 1.09
Fci = Fqi = 1
Fγi = 1
q = 18 × 0.5 + (20 − 9.81) × 0.5 = 14.095
γ1 = γ1' = (20 − 9.81) = 10.19
qu = (0)(35.49)(1.325)(1.180)(1) + (14.095)(23.18)(1.163)(1.09)(1) + (1/2)(10.19)(2)
(22.02)(1.163)(1.09)(1) = 698.62 kN/m2
3.5 Eccentrically loaded foundations

In several instances, as with the base of a retaining wall or structures under lateral loading,
foundations are subjected to moments in addition to vertical loading. In this situation, the dis-
tribution of bearing pressure beneath the foundation is no longer uniform. As shown in Figure
3.11, the distribution bearing pressure along the footing base varies with different eccentricities
e. When e < B/6, the footing is fully in contact with the soil, and the shape of the bearing pres-
sure distribution is trapezoidal. The bearing pressure at one edge is larger than that at the other
edge. However, when e = B/6, the shape of the bearing pressure distribution is triangular, with 0
bearing pressure at one edge. At this time, the footing begins to lift off. When e > B/6, the shape
of the bearing pressure distribution is also triangular; however, because the soil cannot sustain
tension, the footing lifts off without full contact with the soil below. To avoid a large amount
of eccentricity from affecting the serviceability and stability of the footing, a common design
requires that the maximum eccentricity should not be larger than 1/6 the width of the foundation
slab under long-term load conditions and should not be larger than 1/3 the width of the founda-
tion slab under short-term load conditions.
3.5.1 Bearing pressure distributions of eccentrically loaded

foundations
According to the equilibria of the vertical force and moment, the bearing pressure distribution
can be determined for different eccentricities.
1. When the eccentricity e < B/6 (Figure 3.11), the maximum and minimum of the trapezoidal
bearing pressure distribution are:
Q { 6e }
qb,max = |1 + | (3.18.1)
BL { B}
Figure 3.11 Bearing pressure of eccentrically loaded foundations.

Q { 6e }
qb,min = |1 - | (3.18.2)
BL { B}
The preceding equations are derived based on the equilibria of the vertical force and moment
at the footing base considering a trapezoidal distribution of bearing pressure.
2. When the eccentricity e = B/6 (Figure 3.11), the footing begins to lift off, and the maximum
and minimum of the bearing pressure distribution become:
2Q
qb,max = (3.19.1)
BL
qb,min = 0 (3.19.2)
The preceding equations are derived by substituting e = B/6 into eqs. 3.18.1 and 3.18.2. At
this state, the trapezoidal distribution of the bearing pressure becomes triangular.
3. When e > B/6 (Figure 3.11), because the soil cannot sustain any tension, a separation between
the foundation and underlying soil will occur.
4Q
qb,max = (3.20.1)
3L( B - 2e)
qb,min = 0 (3.20.2)
{B }
The contact area of the footing Bc is reduced to Bc = 3 | - e |.
{ 2 }
The preceding equations are derived based on the equilibria of the vertical force and moment
at the footing base, considering a triangular distribution of bearing pressure within the contact
area.
3.5.2 Ultimate bearing capacity of eccentrically loaded foundations

Compared with a centric footing, an eccentric footing has a nonuniform bearing pressure distri-
bution and even a reduced contact area, which may influence the ultimate bearing capacity of
the footing. To account for this influence, the effective area concept is applied to compute the
ultimate bearing capacity of a footing under eccentric loading. In this approach, the ultimate
bearing capacity equation is modified using an effective foundation size as follows.
1
qu = cN c Fcs Fcd Fci +qN q Fqs Fqd Fqi + y B 'Ny Fy s Fy d Fy i (3.21)
2
where B' is the effective width of the footing (short side), and the shape factors Fqs, Fcs, and Fγs
are determined based on the effective width B' and length L' of the footing. Note that the deter-
mination of Fqd, Fcd, and Fγd uses the original footing width for B, without using B'.
Figure 3.12 demonstrates the determination of the effective size of a footing for one-way and
two-way eccentric loading. For one-way eccentric loading, assuming the eccentricity is in the
direction of the footing width:
B ' = B - 2e
L' = L
Figure 3.12 Effect size of eccentrically loaded foundations.

If the eccentricity is in the longitudinal direction, B' always refers to the short side, and
therefore:
B ' = min( B, L - 2e)

L' = max( B, L - 2e)
For two-way eccentric loading, eccentricities eB and eL occur in the B and L directions, respec-
tively. The effective width and length of the footing are:
B ' = min( B - 2eB , L - 2eL )

L' = max( B - 2eB , L - 2eL )
Once the effective size of the footing is determined, qu is computed based on eq. 3.21. In the
same way, the factor of safety of qu/qmax is checked if its value satisfies the design requirement.
Example 3.6
ment depth of 1 m. The footing is subjected to a total vertical load of 3,200 kN and a longitudi-
nal moment 640 kN-m (toward the transverse direction) on its base at the centerline. The angle
of friction of the sand is 32°. The unit weight of sand is 18 kN/m3. Compute qb,max/qb,min and the
ultimate bearing capacity.
Solution
B = 2 m, L = 4 m, Q = 3,200 kN, ML = 640 kN-m, eL = 640/3,200 = 0.2 m (< L/6 (= 4/6))
qb,max/min = [(3200)/(2 × 4)](1 ± (6)(0.2)/4) = 520 kN/m2/280 kN/m2
Using eq. 3.21:
B' = min(2, 4 – 2 × 0.2) = 2 m, L' = max(2, 4 − 2 × 0.2) = 3.6 m
c = c' = 0
ϕ = ϕ' = 32°, Nq = 23.18, Nc = 35.49, Nγ = 22.02
Fcs = 1 + (0.2)(2/3.6)tan2(45 + 32/2) = 1.362
Fqs = Fγs = 1 + (0.1)(2/3.6)tan2(45 + 32/2) = 1.181
Fcd = 1 + (0.2)(1/2)tan(45 + 32/2) = 1.180
Fqd = Fγd = 1 + (0.1)(1/2)tan(45 + 32/2) = 1.09
Fci = Fqi = Fγi = 1
qu = (0)(35.49)(1.362)(1.18)(1) + (1)(18)(23.18)(1.181)(1.09)(1) + (1/2)(18)(2)(22.02)
(1.181)(1.09)(1) = 1047.34 kN/m2
Example 3.7
ment depth of 1 m. The footing is subjected to a total vertical load of 3,200 kN and a transverse
moment 640 kN-m (toward longitudinal direction) on its base at the centerline. The angle of
friction of the sand is 32°. The unit weight of sand is 18 kN/m3. Compute qb,max/qb,min and the
ultimate bearing capacity.
Solution
B = 2 m, L = 4 m, Q = 3,200 kN, MT = 640 kN-m, eB = 640/3,200 = 0.2 m (< B/6 (= 2/6))
qb,max/min = [(3,200)/(2 × 4)](1 ± (6)(0.2)/2) = 640 kN/m2/160 kN/m2
Using eq. 3.21:
B' = min(2 – 2 × 0.2, 4) = 1.6 m, L' = max(2 − 2 × 0.2, 4) = 4 m
c = c' = 0
ϕ = ϕ' = 32°, Nq = 23.18, Nc = 35.49, Nγ = 22.02
Fcs = 1 + (0.2)(1.6/4)tan2(45 + 32/2) = 1.26
Fqs = Fγs = 1 + (0.1)(1.6/4)tan2(45 + 32/2) = 1.13
Fcd = 1 + (0.2)(1/2)tan(45 + 32/2) = 1.18
Fqd = Fγd = 1 + (0.1)(1/2)tan(45 + 32/2) = 1.09
qu = (0)(35.49)(1.362)(1.18)(1) + (1)(18)(23.18)(1.13)(1.09)(1) + (1/2)(18)(1.6)(22.02)
(1.13)(1.09)(1) = 904.47 kN/m2
3.6 Foundation settlement

Checking for the bearing capacity of a footing cannot ensure the actual performance of the
footing. It is also necessary to check the settlement of the footing upon loading. When the set-
tlements of footings at different locations are different, differential settlement between footings
may occur and cause additional stresses in the structure members.
There are two major types of foundation settlement: immediate and delayed settlement.
Immediate settlement occurs during or immediately after the construction of the structure.
Delayed settlement occurs over time. The total settlement of a foundation is the sum of the
immediate settlement and delayed settlement. For footings on sand upon loading, the settlement
is an immediate settlement. For footings on saturated clay upon loading, the settlement com-
prises both an immediate settlement and a delayed settlement. Generally, the delayed settlement
is contributed by a consolidation of clay from the dissipation of excess pore water pressure.
Consolidation settlement occurs in two phases: primary and secondary. Primary consolidation
settlement is more significant than secondary settlement in inorganic clays and silty soils. Sec-
ondary consolidation settlement occurs after the primary consolidation caused by the slippage
and reorientation of soil particles under a sustained load (σ' is unchanged), and it is significant
in organic soils.
The following sections discuss how to compute immediate and delayed settlements.
3.6.1 Immediate settlement

The immediate settlement of a shallow foundation can be estimated using the mechanics
of elasticity. For example, as shown in Figure 3.13, the settlement at a position immedi-
ately beneath the foundation can be computed by integrating the vertical strains induced by
stress increments from the net bearing pressure of the foundation along the depth below the
position.
- 1 -
Si = | e z dz = | (-- z - u s -- x - u s -- y ) (3.22)
0 Es 0
Figure 3.13 Stress increments in a stratum due to net bearing pressure q b (net).
where:
Si = immediate settlement.
Es = secant modulus of elasticity of the soil (refer to Section 2.3.2).
μs = Poisson’s ratio of soil (refer to Section 2.3.2).
Δσx, Δσy, and Δσz = stress increments in the x, y, and z directions, respectively, due to the net
bearing pressure.
Based on this theory, Steinbrenner (1934) proposed the following equation to calculate the
settlement of a perfectly flexible foundation under the net bearing pressure qb(net). Referring to
Figure 3.14:
{ 1- u s2 }
Si = qb ( net ) (a B ') | | Is I f (3.23)
{ Es }
Where Es is the average secant modulus of elasticity of the soil under the foundation, generally
measured from z = 0 to z = 5B, considering the major influence range of stress.
Figure 3.14 A flexible foundation under net bearing pressure q b(net).
Table 3.4 The depth factor I f
μs D f /B B/L
0.2 0.5 1
0.3 0.2 0.95 0.93 0.9

0.4 0.9 0.86 0.81
0.6 0.85 0.8 0.74
1.0 0.78 0.71 0.65
0.4 0.2 0.97 0.96 0.93
0.4 0.93 0.89 0.85
0.6 0.89 0.84 0.78
1.0 0.82 0.75 0.69
0.5 0.2 0.99 0.98 0.96
0.4 0.95 0.93 0.89
0.6 0.92 0.87 0.82
1.0 0.85 0.79 0.72
B' = B and a = 1 at the corner of the foundation.

B' = B/2 and a = 4 at the center of the foundation.
Is is the shape factor and is a function of width B and length L.
If is the depth factor and is a function of the depth of the foundation base Df (refer to Table 3.4).
For the shape factor Is, define m = L/B and n = H/B', where H is the thickness of the soil below
the foundation:
1 - 2us
I s =F1 + F2 (3.24)
1- u s
1
F1 = ( A0 + A1 ) (3.25)
-
n
F2 = tan -1 A2 (3.26)
2-
(1 + m 2 + 1) m 2 + n 2
A0 =m ln (3.27)
m(1 + m 2 + n 2 + 1)
(m + m 2 + 1) 1+ n 2
A1 = ln (3.28)
m + m 2 + n 2 +1
m
A2 = (3.29)
n m + n2 + 1
2
Figure 3.15 Values of I s and I f.

Source: After Christian and Carrier (1978).
The immediate settlement of a rigid footing can be approximated as:
Si (rigid) ≈ 0.93Si (flexible, center) (3.30)
where Si (rigid) and Si (flexible, center) are the immediate settlement of a rigid footing and the immediate
settlement of a flexible footing at the center, respectively.
For saturated clay under loading in the short-term condition, the soil is regarded as in the
undrained condition without volume change. For this state, Poisson’s ratio is set to 0.5 for the no
volume strain condition. Similar to Steinbrenner’s solution, Janbu et al. (1956) derived an equa-
tion for the average settlement of flexible foundations on saturated clay soils under net bearing
pressure qb(net), as follows. Referring to Figure 3.14:
{ qb ( net ) B }
Si = I s I f | | (3.31)
{ Es }
where Is the shape factor and is a function of width B and length L (refer to Figure 3.15), and If
is the depth factor and is a function of the depth of the foundation base Df (refer to Figure 3.15).
Note that the modulus of elasticity for saturated clays is for the undrained condition. Section
2.3.2 describes the secant undrained Young’s modulus as follows.
Es = Eus (3.32)
Eus/su is a function of the plasticity index and overconsolidation ratio (OCR), as shown in Figure 2.24.
Example 3.8
A rigid footing of 2 m × 4 m in plan view is situated on a dry sandy stratum with a thickness of
6 m. The embedment depth of the footing is 1 m. The average modulus of elasticity is 10,000
kN/m2, and Poisson’s ratio is 0.3. Calculate the immediate settlement of the footing under a net
bearing pressure of 200 kN/m2.
Solution
Eq. 3.23 is used.
For the center of the footing, a = 4:
m = L/B = 4/2 = 2, n = H/(B/2) = 5/(2/2) = 5 => F1 = 0.526, F2 = 0.058 (A0 = 0.593, A1 =

1.061, A2 = 0.073)
Is = 0.526 + [(1 − 2(0.3))/(1 − 0.3)](0.058) = 0.559
B/L = 2/4 = 0.5, Df/B = 1/2 = 0.5 => If = (0.86 + 0.80)/2 = 0.83
Si (flexible) = (200)(4)(2/2)[(1 − 0.32)/10000](0.559)(0.83) = 0.0338 m = 33.8 mm
Si (rigid) = (0.93) (33.8) = 31.4 mm
Example 3.9
A footing of 2 m × 4 m in plan view is situated on a saturated clay stratum with a thickness of
6 m. The embedment depth of the footing is 1 m. The undrained shear strength of the clay is
150 kN/m2, OCR = 2, PI = 30. Calculate the average immediate settlement of the footing under
a net bearing pressure of 200 kN/m2.
Solution
Eq. 3.31 is used.
For OCR = 2, PI = 30 => Eus/su = 550 (according to Figure 2.24).
Es = Eus = (550)(150) = 82,500 kN/m2

L/B = 2, H/B = 5/2 = 2.5 => Is = 0.7
Df/B = 1/2 = 0.5 => If = 0.95
Si (av) = (0.7)(0.95)[(200)(2)/82,500] = 0.003224 m = 3.22 mm
3.6.2 Consolidation settlement

Consolidation settlement occurs over time in saturated clay subjected to an increased load caused
by the construction of the foundation and the supported structure. According to the one-dimen-
sional consolidation theory, the following approach can be applied to compute the consolidation
settlement, considering the initial soil stress condition, stress increment, and overconsolidation
ratio (OCR) of clay. For example, a clay with a thickness of Hc is subjected to the net bearing
pressure qb(net), as shown in Figure 3.16. First, the simplified 2 (vertical): 1 (horizontal) rule can
be used to approximately estimate the average stress increment due to qb(net). The average stress
increment -- av' at the midpoint of the clay layer is computed as follows:
qb ( net ) B 2 qb ( net ) BL
' =
-- av or (3.33)
( B + z0 ) 2
( B + z0 )( L + z0 )
where z0 is the depth from the foundation base to the midpoint of the clay layer.
Figure 3.16 Stress increment in saturated clay due to the net bearing pressure q b(net).
The settlement of the clay layer Sc upon the stress increment can be computed for different
soil conditions.
For normally consolidated clay:
Cc H c - ' + -- av'
Sc = log 0 (3.34)
1+ e0 - 0'
where e0 and Cc are the initial void ratio and compression index of the clay, respectively, and - 0'
is the initial overburden pressure.
For overconsolidated clay, two scenarios are considered:
1. When the stress in the clay layer upon the stress increment is less than the preconsolidation
stress pc′ (i.e., -- 0' + -- av' < pc'), then:
Cs H c - ' + -- av'
Sc = log 0 (3.35)
1+ e0 - 0'
where Cs is the swell index of the clay.
2. When the stress in the clay layer upon the stress increment is larger than the preconsolidation
stress pc′ (i.e., -- 0' + -- av' > pc'), then:
Cs H c p' C H - ' + -- a'v

Sc = log c + c c log 0 (3.36)
1 + e0 - 0' 1 + e0 pc'
Example 3.10
A footing of 2 m × 4 m in plan view is situated on a saturated clay stratum with a thickness of
6 m. The embedment depth of the footing is 1 m. Assume that γ = 16 kN/m3, e0 = 0.8, Cc = 0.3,
Cs = 0.1, and pc′ = 100 kN/m2. Calculate the average consolidated settlement of the footing
under a net bearing pressure of 200 kN/m2.
Solution
Hc = 5 m
σ0' (at the middle of the clay layer below the footing) = (16 − 9.81)(1 + 5/2) = 21.67 kN/m2
< pc′ = 100 kN/m2 => OC clay
qb (net) = 200 kN/m2
Δσ'av = (200)(4)(2)/[(4 + 5/2)(2 + 5/2)] = 54.7 kN/m2
σ0' + Δσ'av = 21.67 + 54.7 = 76.37 < pc′
By using eq. 3.35:
Sc (av) = [5/(1 + 0.8)](0.1)log(76.37/21.67) = 0.1520 m = 152 mm
Much larger than the immediate settlement!

3.6.3 Allowable bearing capacity considering allowable settlement

(example of saturated clay)
The allowable bearing capacity determined from the factor of safety does not provide informa-
tion about the settlement of footings. To consider the serviceability of a footing, the allowable
bearing capacity also needs to account for the foundation settlement. This section illustrates an
example to determine the allowable bearing capacity of a surface footing on saturated clay con-
sidering the bearing capacity and foundation settlement. As displayed in Figure 3.17, two lines
that describe the relationships between the allowable net bearing pressure qall (net) and the footing
width B are built based on the criteria of ultimate bearing capacity and settlement individually.
Line 1 is the relationship of the allowable bearing capacity based on the ultimate bearing
capacity qu (net) divided by a factor of safety FS. That is:
qu ( net ) = cN c (3.37)
qu ( net ) cN c
qall=
( net ),1 = (3.38)
FS FS
From eq. 3.37, the curve is a horizontal line, implying that the allowable bearing capacity is
independent of the footing width B.
Line 2 (blue curve) is the relationship of the allowable bearing capacity based on the allow-
able foundation settlement Sall. Consider the immediate settlement, according to eq. 3.31:
S all Es
qall ( net ),2 = (3.39)
Is I f B
Figure 3.17 Schematic diagram of the interaction curve for allowable bearing capacity in satu-
rated clay.
From eq. 3.39, for a given allowable settlement, the allowable bearing capacity decreases
with an increase in the footing width. More precisely, it is inversely proportional to the footing
width.
The final relationship of the allowable bearing capacity versus the footing size is the
lesser of qall (net),1 and qall (net),2, as shown by the dashed line in Figure 3.17. The intersec-
tion of the two lines indicates the critical size of the footing. Below this critical size, the
allowable bearing capacity is governed by the bearing capacity, but for a footing larger
than the critical size, the allowable bearing capacity is governed by the allowable footing
settlement.
Example 3.11
A rigid footing of 2 m × 4 m in plan view is situated on a dry sandy stratum with a thick-
ness of 6 m. The embedment depth of the footing is 1 m. The average modulus of elasticity is
10,000 kN/m2, Poisson’s ratio is 0.3, the angle of friction of sand is 32°, and the unit weight of
sand is 18 kN/m3. Calculate the allowable net bearing pressure of the footing, considering an
allowable settlement of 40 mm and FS = 3.
Solution
1. With the settlement controlled:
Use eq. 3.23.

1- u s2
Si = qb ( net ) (a B ') Is I f
Es
For the center of the footing, α = 4:
m = L/B = 4/2 = 2, n = H/(B/2) = 5/(2/2) = 5 => F1 = 0.526, F2 = 0.058

Is = 0.526 + [(1 − 2(0.3))/(1 − 0.3)](0.058) = 0.559
B/L = 2/4 = 0.5, Df/B = 1/2 = 0.5 => If = (0.86 + 0.80)/2 = 0.83
Si (rigid) = (0.93)qall(net)(4)(2/2)[(1 − 0.32)/10,000](0.559)(0.83) = 0.04 m = 40 mm
qall(net) = 254.6 kN/m2
2. With the bearing capacity controlled:
Referring to example 3.3, qu = 1031.38 kN/m2

qu (net) = 1031.38 − 18(1) = 1013.38 kN/m2
qall(net) = 1013.38/3 = 337.79 kN/m2
Considering conditions 1 and 2, choose the smaller one:
qall(net) = min(254.6, 337.79) = 254.6 kN/m2 (Condition 1 governs!)
3.6.4 Tolerable settlement of buildings

In most conditions, the subsoil is not homogeneous, and the load carried by foundations of a
structure may vary at different locations. Varying degrees of settlement may occur in different
parts of a building. The differential settlement of the parts of a building can lead to damage to
the structure. Figure 3.18 presents schematic settlement profiles at the base of a building with
length L. Figure 3.18a shows the building settlement without tilt, whereas Figure 3.18b shows
the building settlement with tilt. In the figure, points A, B, C, D, and E indicate points on the
building base before settlement, and points A', B', C', D', and E' indicate their positions after
settlement. According to this figure, the following parameters are defined to quantify the settle-
ment condition of the building:
Figure 3.18 Definitions of settlement parameters.

Source: Redrawn following Wahls (1981).
ρi = total settlement of point i.

δij = differential settlement between points i and j (= ρi – ρj).
ω = rigid body rotation.
ηij = angular distortion between points i and j (= δij/lij for the building settlement without tilt,
= δij/lij – ω for the building settlement with tilt, where lij is the distance between points i and j).
In the preceding parameters, the angular distortion is directly related to the degree of damage
to a building. As defined, when the foundation settlement exhibits tilt (refer to Figure 3.18b),
the tilt ω is removed to deduce an actual angular distortion as the tilt is a rigid-body rotation
that does not induce additional stress in the structure. Table 3.5 relates the potential structural
damage to the degree of angular distortion. As angular distortion increases, more damage is
anticipated to occur in structures. According to this table, if the first cracks in structures are not
allowed, the allowable angular distortion is limited to less than 1/300.
Alternatively, differential settlement and total settlement are usually used to determine the
allowable settlement. Similar to angular distortion, differential settlement can also be related
to structural damage. However, it is not dimensionless. An angular distortion may correspond
Table 3.5 Limiting angular distortion to potential damage of structures
Category of potential damage η
Danger to machinery sensitive to settlement 1/750

Danger to frame with diagonals 1/600
Safe limit for no cracking of buildings 1/500
First cracking of panel walls 1/300
Difficulties with overhead cranes 1/300
Tilting of high rigid buildings becoming visible 1/250
Considerable cracking of panel and brick walls 1/150
Danger of structural damage to general buildings 1/150
Safe limit for flexible brick walls, L/H > 4 b 1/150
Source: Bjerrum a (1963).
a Following Wahls (1981).
b Safe limits include a factor of safety, H = height of building.
Table 3.6 Allowable total and differential settlements
Foundation type Soil type Total settlement (cm) Differential settlement (cm) Note
Isolated footing Sand 2.5 2.0 T

5.0 3.0 S
3.0 – J
Isolated footing Clay 7.5 – S
10.0 – J
Mat foundation Sand 5.0-7.5 3.0 S
6.0-8.0 – J
– 3.0 G
Mat foundation Clay 7.5-12.5 4.5 S
20.0-30.0 – J
5.6 G
Note: T: Terzaghi and Peck (1967); S: Skempton and MacDonald (1957) for 1/300 angular distortion; J:
Architectural Institute of Japan (1988); G: Grant et al. (1974) for 1/300 angular distortion.
to different values of differential settlement for different column spans. For a general building,
the distance of two neighboring columns is approximately 6 m. According to this distance and
Table 3.5 (η = 1/300 for the first cracking of panel walls), the differential settlement between two
columns can be limited to be less than 2 cm to avoid cracking in a building. Differential settle-
ment is generally not easy to measure. A large differential settlement is usually accompanied by
a large total settlement of the footing. Based on past experience, the differential settlement for a
building on sand is approximately 3/4 of the total settlement; thus, the allowable total settlement
of an isolated footing on sand is set to 2.5 cm. However, footings on clay or buildings with mat
foundations generally settle more uniformly, and the allowable total settlement can be larger.
Table 3.6 lists some allowable total and settlement values suggested in the literature. Note that
these values are applicable for the column spacing of approximately 6 m; however, they may not
be appropriate for span distances larger or less than 6 m.
Considering uncertainties in settlement evaluation and measurement, in practical applica-
tions, two parameters, angular distortion and total settlement, or differential settlement and total
settlement, are used to simultaneously evaluate the settlement condition of a building.
3.7 Applications of plate loading tests

The estimation of bearing capacity and foundation settlement using the aforementioned theo-
ries may lead to uncertainties in the theoretical models and the determination and selection of
material parameters. To reduce uncertainties and confirm results in design, plate load testing
performed in the field is beneficial to estimate the bearing capacity and settlement of footings.
Figure 1.25 demonstrates a general setup and a field picture of vertical plate load testing. First,
a test pit is excavated to the foundation bottom. The width of the pit is generally larger than
four times the width of the plate to ignore the effect of the overburden above the foundation
base. Reaction beams are set to react to the jack when applying vertical loading to the plate.
With this testing, the pressure–settlement relationship of the plate can be recorded, as pre-
sented in Figure 1.26. Based on this relationship, the settlement and ultimate pressure of the
plate can be used to estimate the settlement and ultimate bearing capacity of the footing for
design reference.
3.7.1 Estimation of q u
For tests in clays:
qu ( F ) = qu ( P ) (3.40)
where qu(F) is the ultimate bearing capacity of the proposed foundation, and qu(P) is the ultimate
bearing capacity of the test plate.
This equation implies that the ultimate bearing capacity in clay is practically independent of
the plate size.
For tests in sandy soils:
BF
qu ( F ) = qu ( P ) (3.41)
BP
where BF is the width of the foundation, and BP is the width of the test plate.
This equation implies that the ultimate bearing capacity in sand is linearly proportional to the
plate size.
3.7.2 Estimation of settlement

A footing generally has a larger size than the plate. Because of their different influence depths
of stress Di and nDi, the footing and plate may have different settlements under the same bear-
ing pressure. As shown in Figure 3.19, considering the difference in size between the footing
and plate:
For tests in clayey soils (overconsolidated clay), assume that the elastic modulus of the soil
is constant with depth:
BF
Si ( F ) = Si ( P ) (3.42)
BP
This equation is derived based on eq. 3.31, which implies that the settlement of a footing in clay
linearly increases with the plate size.
Figure 3.19 Influence zone of stress for the plate and foundation.
For tests in sandy soils:

2 2
{ B } { B + 0.3 }
Si ( F ) = Si ( P ) | F | | P | (3.43)
{ BP } { BF + 0.3 }
where BF and BP are in meters.

Because the elastic modulus of sandy soil is dependent on the confining stress, this equation
was empirically formulated by Terzaghi and Peck (1967) based on several plate load tests with
different plate sizes.
3.8 Mat foundations

A mat foundation (raft foundation) is a combined footing that covers the entire area under
a structure, supporting several columns, and walls. It is sometimes applied to soil that has a
low bearing capacity but supports large column or wall loads. Mat foundations might be more
economical when spread footings need to cover a large area of the building. Figure 3.20 dis-
plays three common types of mat foundations: (a) the foundation is a flat plate with uniform
thickness, (b) the foundation is composed of a slab with girder beams, in which the columns
are located at the intersections of the beams, and (c) the foundation is the basement with
slabs and walls. The foundation of type (c) is often referred to as a compensated foundation
because the weight of soil in the basement area is excavated to counterbalance a part of or all
the structure loads.
3.8.1 Bearing capacity of mat foundations

The net ultimate bearing capacities of a mat foundation can be determined by eq. 3.15. A fac-
tor of safety is used to calculate the net allowable bearing capacity. A factor of safety of 3 is
normally used.
3.8.2 Compensated foundations

The net bearing pressure under a mat foundation can be reduced by increasing the excavation
depth of the foundation. As shown in Figure 3.21, for a mat foundation subjected to the dead
Figure 3.20 Common types of mat foundations: (a) flat plate, (b) beams and slab, and (c) slab
with basement walls.
Figure 3.21 Compensated foundation design concept.

weight of the structure and the live load Q, when its base is embedded at depth Df and the soil
within this depth (highlighted zone) is removed, the net bearing pressure is reduced as follows:
Q
qb ( net ) = - y Df (3.44)
A
where A is the area of the foundation.

According to the earlier equation, increasing the embedment depth of the foundation implies
that the more that is removed, the smaller the net bearing pressure becomes. Because of the
decrease in the net bearing pressure, the factor of safety against bearing capacity failure
increases, and the foundation settlement decreases. This approach is known as compensated
foundation design and is useful when structures are built on soft clays. For this design, a base-
ment is made below the structure, as shown in Figure 3.20(c).
Furthermore, for no increase in the pressure in the soil below a foundation, qb(net) should be
zero. The foundation for this condition is called a fully compensated foundation, and its exca-
vation depth is:
Q
Df = (3.45)
yA
Example 3.12
A mat foundation has dimensions of 20 m × 30 m in plan view. The total load on the mat is
100 MN. The mat is situated on a saturated clay with a unit weight of 18 kN/m2. Determine the
excavation depth for this mat to be a fully compensated foundation.
Solution
Use eq. 3.45, Df = 100,000/[(18)(20)(30)] = 9.26 m.
3.8.3 Stress analysis of mat foundations

Compared with spread footings, mat foundations are generally more flexible because of a
large foundation base. Figure 3.22 displays the distribution of bearing pressure for a mat
foundation for different soil consistencies. When the mat is on soft soil, it may behave like
a rigid footing, and the bearing pressure is linearly distributed because of a linear footing
displacement profile, as presented in Figure 3.11. Therefore, for a rigid mat, a linear bearing
pressure profile is assumed to calculate the stress in the mat. However, when the same mat
is on much stiffer soil, the mat may behave flexibly, the bearing pressure profile is nonlinear,
and a larger pressure occurs around the location of the load. For this condition, the analysis
method for a rigid footing cannot be applied. To simulate a more accurate distribution of
bearing pressure and stress in the foundation for design, the slab on the Winkler founda-
tion model is commonly used in practice, as presented in Figure 3.23. In this model, the
mat is no longer rigid, and the flexibility of the mat is simulated in the model; therefore,
the method is called the flexible method. In addition, this model simulates the soil as a bed
of springs (Winkler springs), each with a stiffness kv. kv is called the coefficient of vertical
Figure 3.22 Distributions of bearing pressure for a footing on soft and stiff soil.
subgrade reaction, which defines the relationship between bearing pressure and settlement.
kv is expressed as follows:
p
kv = (3.46)
p
where p is the bearing pressure and ρ is the settlement.

Since discrete springs are adopted in numerical modeling, the stiffness of the springs is:
K v = kv A (3.47)
where A is the tributary area of the spring, as shown in Figure 3.23.

In the slab on the Winkler foundation model, the determination of kv is an important and
difficult task. Theoretical analysis and empirical judgment are usually needed to determine an
appropriate value. In essence, the coefficient of subgrade reaction is not a fundamental soil prop-
erty but a fictitious property that relates the bearing pressure to the settlement. It varies with the
properties of the subgrade and the beam or slab and even the size of the footing. In practice, kv
can be estimated by field load testing, the results of boring and laboratory tests, or past experi-
ence. Terzaghi (1955) proposed an approach to determining kv by using field plate load testing
with a plate size of 0.3 m (1 ft). Considering the settlement characteristics of a footing on clay
and sandy soils and the difference in size between the foundation and the plate, Terzaghi (1955)
Figure 3.23 Slab on Winkler foundation model.
further related the coefficient of subgrade reaction of a rigid footing with that of a rigid plate as
follows, based on eqs. 3.48 and 3.49 for clayey and sandy soils, respectively. These two equa-
tions clearly show that kv decreases as the footing size increases.
For clayey soils:
B1
kvB = kv1 (3.48)
B
For sandy soils:
{ B + B1 }
2
kvB = | | kv1 (3.49)

{ 2B }
where kv1 and kvB are the coefficients of the subgrade reaction of the plate of size 0.3 m (1 ft) and
the rigid foundation, respectively, and B1 and B are the sizes of the rigid plate and rigid founda-
tion, respectively. In addition, Terzaghi (1955) suggested empirical values of kv1 for different
types of soils, as listed in Table 3.7.
Vesic (1961) compared the solutions of an infinite beam on a continuum model and a beam
on an elastic foundation and proposed:
0.65Es 12 Es B 4
kv = (3.50)
1- u s2 EI
where B is the width of the footing, E is the modulus of elasticity of the foundation, and I is the
moment of inertia of the foundation’s cross section.
Table 3.7 Suggested values for k v1
Clay Stiff Very stiff Hard
c u (kN/m 2) 47.5–95 95–190.5 >190.5

k v1 (MN/m 3) 15.6–31.3 31.3–62.7 >62.7
Sand Loose Medium Dense
Unsaturated k v1 (MN/m 3) 6.3–18.8 18.8–94 94–313.4

Saturated k v1 (MN/m 3) 7.8 25.1 94
Source: Terzaghi (1955).
To evaluate the rigidity of a foundation system, a parameter λ that compares the rigidities of
the foundation and soil can be used. It is defined as follows:
kv B
-= 4 (3.51)
4EI
With this parameter, when λL < 1.0 (where L is the foundation length), the foundation can be
regarded as rigid.
The actual distribution of kv is not uniform because of the nonuniform distributions of bear-
ing pressure and foundation displacement. To consider a detailed variation of kv, a more delicate
approach considering the actual interaction between the bearing pressure and foundation dis-
placement can be performed. This approach requires iterations. First, assuming an initial distri-
bution of kv, the structure model is built, and the bearing pressure is calculated after the structure
is loaded. Second, according to the obtained bearing pressure, the ground surface settlement
is analyzed with numerical analysis, in which the adopted soil model can consider the nonlin-
ear behavior of soil and complex soil and boundary conditions. Fourth, based on the bearing
pressure and ground surface settlement, a variation in kv can be formulated and used to update
the original kv distribution. With the updated kv distribution, the aforementioned procedure is
repeated until the distribution of kv converges.
3.8.4 Piled mat foundations

A mat foundation may have a sufficient bearing capacity to resist structure loads because of a
large foundation size, but it may still be subjected to excessive settlement when it supports a
high-rise building on soft soil. To overcome this excessive settlement issue, a piled mat founda-
tion can be an option. A piled mat is a hybrid foundation in which piles are used to support the
mat to reduce the total and differential settlements of the mat; in this foundation type, the piles
serve as settlement reducers. This foundation type differs from a conventional capped group pile
foundation (which is introduced in Chapter 7). For a piled mat, most of the bearing capacity of
the foundation is provided by the mat, but for a capped group pile foundation, its bearing capac-
ity is supposed to be provided entirely by the piles, considering that the pile cap is rigid and the
cap is not in contact with the soil.
Designing a piled mat foundation requires more comprehensive analyses than those when
mat or pile foundations are used alone. The interaction among the piles, soil, and mat needs to
(a)
Kv
Pile
Kvp
(b)
Figure 3.24 Numerical models for piled mat foundations: (a) finite element model and (b) slab
on spring foundation model.
be simulated in the analyses. Issues such as the proportions of loads carried by the piles and
mat, how the total and differential settlements are reduced by the inclusion of the piles, and how
the internal forces (shear forces and moments) in the mat are reduced by the piles need to be
addressed. The answers to these issues determine the final design parameters, for example, the
mat thickness and reinforcement, the pile arrangement, and the pile diameter and length. The
piles in the mat are not necessarily uniformly distributed; an efficient design can be achieved by
placing the piles at the locations where the settlement is considerable to reduce the differential
settlement and therefore the internal forces in and the thickness of the mat. No simple design
charts for addressing the preceding concerns are available. The response of piled rafts is often
analyzed numerically either by using finite element simulation (Figure 3.24a) or by the slab on
the spring foundation model (Figure 3.24b). The finite element analyses that use solid elements
to model the mat and soil are the most common because they can directly simulate the effect of
pile‒soil–mat interactions; however, the computational cost is large. Similar to mat foundations,
the slab on the spring foundation model, in which the soil reactions and piles are modeled by
springs Kv and Kvp, respectively, can be applied for piled mats and is more efficient. However,
the determination of spring properties of the pile and soil reactions should consider the interac-
tion of the piles and the surrounding soil, which is not an easy task.

This chapter introduces shallow foundation design methods. The summary of and general com-
ments on this chapter are as follows:
1. Foundations with embedment depths less than or equal to the foundation width are called
shallow foundations. Designing a shallow foundation involves four main items: evaluation of
vertical bearing capacity, evaluation of eccentricity, evaluation of settlement, and structural
design of foundation components.
2. The bearing failure modes of a footing include general, local, and punching modes. The
bearing failure mode is dependent on the soil type and density and the embedment depth. For
the general bearing failure mode, Terzaghi’s and Meyerhof’s bearing capacity formulas are
introduced, which show that the ultimate bearing capacity of a footing is influenced by three
parts: cohesion, surcharge, and soil weight. The bearing capacity is sensitive to the angle of
friction of the soil. For ϕ > 30º, the bearing capacity increases rapidly. For footing on satu-
rated clay, effective stress and total stress analyses with appropriate soil parameters can be
applied to evaluate the long-term and short-term bearing capacities, respectively.
3. The ultimate bearing capacity is also influenced by the foundation eccentricity and location
of the groundwater table. The parameters in the bearing capacity formulas need to be modi-
fied to consider their influences. In addition to the stability of the footing, the eccentricity not
only increases the bearing pressure but also reduces the bearing capacity because of reduced
effective contact area between the footing and soil. The location of the groundwater table
may alter the effective stress in the soil and thus the ultimate bearing capacity of footings.
4. The total settlement of a foundation is the sum of the immediate settlement and delayed
settlement. Immediate settlement occurs during or immediately after the construction of a
structure. Delayed settlement occurs over time. For footings on sand upon loading, the settle-
ment is an immediate settlement. For footings on saturated clay upon loading, the settlement
comprises both an immediate settlement and a delayed settlement. The immediate settlement
can be computed based on the theory of elasticity. The delayed settlement of saturated clay
is due to the consolidation of clay, which can be computed based on the theory of consolida-
tion. The angular distortion (differential settlement divided by the distance of two points) is
directly related to the degree of damage to a building. Both total settlement and differential
settlement (or angular distortion) are used in the design to evaluate the settlement condition
of the building.
5. The allowable bearing capacity determined from the factor of safety does not provide infor-
mation about the settlement of a footing. To determine the serviceability of a footing, the
allowable bearing capacity also needs to account for the foundation settlement. Interactive
curves can be built to address the relationship between the allowable bearing capacity and
footing size, considering the settlement and bearing failure. Generally, the allowable bearing
capacity for a small footing is governed by the bearing failure, but that for a large footing is
governed by the foundation settlement. Field plate load testing can be used to estimate the
bearing capacity and settlement of a footing for modifying and confirming foundation design
parameters and results.
6. A mat foundation is commonly applied to soil that has a low bearing capacity but supports
large structure loading. To increase the factor of safety against vertical bearing capacity and
reduce settlement, a type of mat foundation called a compensated foundation can be applied.
A compensated foundation uses basements as the foundation body and excavates the soil
within the basement to reduce the net bearing pressure. Compared with spread footings, mat
foundations are more flexible because of their large foundation base, especially when the
mat is on stiff soil. To simulate the actual distribution of bearing pressure and stress in the mat
foundation for design, the beam-on-elastic foundation model is commonly used in practice.
7. A piled mat is a hybrid foundation in which piles are used to support a mat to reduce the
total and differential settlement of the mat. This design can reduce excessive settlement in
the mat foundation when supporting a high-rise building. The main role of the piles in this
foundation type is a settlement reducer. The piles in the mat can be efficiently arranged by
allocating them at the locations where the settlement is significant. The response of piled
rafts is often analyzed numerically either by using finite-element simulation or by the slab-
on-a-spring foundation model. The finite element analyses that use solid elements to model
the mat and soil are the most common because they can directly simulate the pile‒soil‒mat
interaction.
Problems
3.1 A footing is situated on a dry sandy stratum with an embedment depth of 1 m. The angle
of friction of sand is 34°. The unit weight of sand is 18.5 kN/m3. Apply Terzaghi’s bearing
capacity equation to estimate the ultimate bearing capacity of the footing in the following
cases:
a. 3 m wide strip footing
b. 3 m × 3 m square footing
c. circular footing 3 m in diameter
3.2 A square footing is 3 m × 3 m in plan view, situated on a saturated clay stratum with an
embedment depth of 1 m. The groundwater table is located at the ground surface. The
saturated unit weight of the soil is 18 kN/m3. The effective angle of friction and cohesion
of the soil are 24° and 10 kN/m2, respectively. The undrained shear strength of the clay
is 80 kN/m2. Apply Terzaghi’s bearing capacity equation to estimate the ultimate bearing
capacity of the footing for long-term and short-term conditions.
3.3 Same as problem 3.1. Use Meyerhof’s bearing capacity to estimate the ultimate bearing
capacity of the footing.
3.4 Same as problem 3.2. Use Meyerhof’s bearing capacity to estimate the ultimate bearing
capacity of the footing for long-term and short-term conditions.
3.5 A rectangular footing is 3 m × 4 m in plan view, situated on a dry sandy stratum with an
embedment depth of 1 m. The angle of friction of sand is 34°. The unit weight of sand is
18.5 kN/m3.
a. Apply Meyerhof’s bearing capacity equation to estimate the gross and net ultimate
bearing capacity of the footing.
b. Compute a factor of safety in terms of the net ultimate bearing capacity, given that the
footing is subjected to a total centric vertical load of 6,000 kN on its base.
3.6 Redo problem 3.5 when the groundwater table is located at a depth of 0.5 m. The unit
weights of sand above and below the groundwater table are 16.5 and 19.5 kN/m3,
respectively.
3.7 A rectangular footing is 3 m × 4 m in plan view, situated on a dry sandy stratum with an
embedment depth of 1 m. The footing is subjected to a total vertical load of 6,000 kN
and a moment of 1,200 kN-m in the transverse direction of the footing on its base at the
centerline. The angle of friction of the sand is 34°. The unit weight of sand is 18.5 kN/m3.
Compute qb,max/qb,min and the ultimate bearing capacity.
3.8 Redo problem 3.7 when the moment is applied in the longitudinal direction of the footing.
3.9 A rigid footing of 3 m × 4 m in plan view is situated on a dry sandy stratum with a thick-
ness of 5 m. The embedment depth of the footing is 1 m. The average modulus of elastic-
ity is 12,000 kN/m2, and Poisson’s ratio is 0.3. Calculate the immediate settlement of the
footing under a net bearing pressure of 250 kN/m2.
3.10 A footing of 3 m × 4 m in plan view is situated on a saturated NC clay stratum with a thick-
ness of 5 m. The embedment depth of the footing is 1 m. The undrained shear strength of
the clay is 20 kN/m2, PI = 30. Calculate the average immediate settlement of the footing
under a net bearing pressure of 200 kN/m2.
3.11 A footing of 3 m × 4 m in plan view is situated on a saturated clay stratum with a thickness
of 5 m. The embedment depth of the footing is 1 m. Assume that γ = 18 kN/m3, e0 = 0.8,
Cc = 0.25, Cs = 0.05, and pc' = 80 kN/m2. Calculate the average consolidated settlement of
the footing under a net bearing pressure of 200 kN/m2.
3.12 A rigid footing of 3 m × 4 m in plan view is situated on a dry sandy stratum with a thick-
ness of 5 m. The embedment depth of the footing is 1 m. The average modulus of elasticity
is 12,000 kN/m2, Poisson’s ratio is 0.3, the angle of friction of sand is 34°, and the unit
weight of sand is 18.5 kN/m3. Calculate the allowable net bearing pressure of the footing,
considering an allowable settlement of 25 mm and FS = 3.
3.13 Redo problem 3.12 for an allowable settlement of 40 mm and FS = 2.
3.14 A mat foundation has dimensions of 25 m × 30 m in plan view. The total load on the mat
is 120 MN. The mat is situated on a saturated clay having a unit weight of 17.5 kN/m2.
Determine the excavation depth for this mat to be a fully compensated foundation.
References
Architectural Institute of Japan (AIJ). (1988), Recommendation for the Design of Building Foundations (in
Japanese), Tokyo: Architectural Institute of Japan.
Bjerrum, L. (1963), Allowable settlement of structures. In Proceedings, 3rd European Conference on Soil
Mechanics and Foundation Engineering, Wiesbaden, Germany, Vol. 3, pp. 135–137.
Christian, J.T. and Carrier, W.D. (1978), Janbu, Bjerrum and Kjaernsli’s chart reinterpreted. Canadian
Geotechnical Journal, Vol. 15, pp. 123–128.
Grant, R., Christian, J.T. and Vanmarcke, E.H. (1974), Differential settlement of buildings of buildings.
Journal of the Geotechnical Engineering Division, ASCE, Vol. 100, No. 9, pp. 973–991.
Janbu, N., Bjerrum, L. and Kjaernsli, B. (1956), Veuledning ved losning av fundamenteringsoppgaver.
Norwegian Geotechnical Institute Publication, Vol. 16, pp. 30–32 (in Norwegian).
Kumbhojkar, A.S. (1993), Numerical evaluation of Terzaghi’s Nγ. Journal of the Geotechnical Engineer-
ing Division, ASCE, Vol. 119, No. 3, pp. 598–607.
Meyerhof, G.G. (1963), Some recent research on the bearing capacity of foundations. Canadian Geotech-
nical Journal, Vol. 1, No. 1, pp. 16–26.
Skempton, A.W. and McDonald, D.H. (1957), Allowable settlement of buildings. Proceedings, Institute of
Civil Engineers, Part III, Vol. 5, pp. 727–768.
Steinbrenner, W. (1934), Tafeln zur setzungsberechnung. Die Strasse, Vol. 1, pp. 121–124.
Terzaghi, K. (1943), Theoretical Soil Mechanics, New York: John Wiley & Sons, Inc.
Terzaghi, K. (1955), Evaluation of coefficient of subgrade reaction. Geotechnique, Vol. 5, No. 4,
pp. 297–326.
Terzaghi, K. and Peck, R.B. (1967), Soil Mechanics in Engineering Practice, New York: John Wiley and
Sons.
Vesic, A.S. (1961), Bending of beam resting on isotropic elastic solid. Journal of the Engineering Mechan-
ics Division, ASCE, Vol. 87, No. EM2, pp. 35–53.
Vesic, A.S. (1963), Bearing capacity of deep foundations in sand. Highway Research Record, No. 39,
pp. 112–153.
Vesic, A.S. (1973), Analysis of ultimate loads of shallow foundations. Journal of Soil Mechanics and
Foundations Division, ASCE, Vol. 99, No. SM1, pp. 45–73.
Wahls, H.E. (1981), Tolerable settlement of buildings. Journal of the Geotechnical Engineering Division,
ASCE, Vol. 107, No. 11, pp. 1489–1504.
Chapter 4
Lateral earth pressure
4.1 Introduction
Lateral earth pressure is an important design element in several foundation engineering prob-
lems, such as retaining walls, excavations, slopes, and even pile foundations. Thus, it is neces-
sary to estimate the lateral earth pressure acting on structural members, which is a function of
several factors, including the unit weight of soil, shear strength parameters, and the types and
amount of structural member displacement.
Figure 4.1 shows three types of movement of a retaining wall. When the wall is restrained
from movement, the lateral earth pressure is called the at-rest earth pressure, σ h0 (Figure 4.1a).
Under the action of lateral earth pressure on the wall, the wall rotates with respect to the wall
toe. When the rotation is sufficiently large, a failure surface forms behind the wall, and the
lateral earth pressure acting on the wall is called the active earth pressure, σ a (Figure 4.1b).
The soil behind the wall settles or moves downward. On the other hand, if the wall is pushed
backward, it induces the formation of a failure surface (Figure 4.1c). Under such conditions, the
lateral earth pressure is referred to as passive earth pressure, σ p. The soil behind the wall heaves
or moves upward.
+tH -tH
Figure 4.1 Types of lateral earth pressure: (a) at-rest state, (b) active state, and (c) passive
state.
DOI: 10.1201/9781003350019-4
Lateral earth pressure 135
Figure 4.2 Direction of active and passive forces on a vertical wall.
As shown in Figure 4.2, in the problem of retaining walls, the soil behind the wall is initially
in the at-rest state. As the wall moves, the earth pressure evolves into active earth pressure.
Assuming the friction angle between the wall and soil is δ , the active earth pressure, Pa, inclines
at an angle δ to the normal to the wall surface as the soil moves forward and downward with
the forward movement of the wall. Conversely, the forward movement of the wall compresses
the soil in front of the wall, and the earth pressure gradually transitions into a passive state. The
passive earth pressure Pp inclines at an angle δ to the normal to the wall surface as the soil is
compressed and heaves with the forward movement of the wall.
This chapter addresses the estimation of those lateral earth pressures based on various earth
pressure theories.
4.2 Lateral earth pressure at rest

Figure 4.3a shows a vertical retaining wall with height H. Assume that friction does not exist
between the retaining wall and that the soil and the wall are infinitely long. When the wall is
restrained from movement, the stresses at depth z below the ground surface are under elastic
equilibrium with no shear stress. Supposing - v'0 represents the effective vertical overburden
pressure, the effective lateral pressure - h' 0 at rest is:
- h' 0 = K 0- v'0 (4.1)
where K 0 is the coefficient of lateral earth pressure at rest.

136 Lateral earth pressure
Figure 4.3 Lateral earth pressure at rest: (a) stress of soil at depth z; (b) distribution of the
at-rest earth pressure.
The total lateral earth pressure at rest is:
- h 0 = - h' 0 + uw (4.2)
where uw is the pore water pressure, which is the summation of the hydrostatic water pressure
and excess pore water pressure.
For cohesionless soil, K 0 can be estimated by Jaky’s (1944) equation:
K 0 = 1 - sin 0 ' (4.3)
where 0 ' is the effective internal angle of friction, also called the drained angle of friction.
When cohesionless soil is in the preconsolidated state, that is, overconsolidated, K 0 can be
estimated by the following equation (Schmidt, 1967, Alpan, 1967):
K 0, OC = K 0, NC (OCR )a (4.4)
where:
K 0,OC = coefficient of lateral earth pressure at rest for overconsolidated soil with the overcon-
solidation ratio, OCR
K 0,NC = coefficient of lateral earth pressure at rest for normally consolidated soil.
a = empirical coefficient, a = sin 0 '.
Ladd et al. (1977) suggested that K 0 can also be estimated by eq. 4.3 for normally consoli-
dated cohesive soil and by eq. 4.4 for overconsolidated cohesive soil.
Eqs 4.1 and 4.2 can be used to determine the variation in the at-rest earth pressure with depth,
as shown in Figure 4.3b. The resultant force, P0, per unit length of the wall can be obtained from
the area of the pressure diagram in Figure 4.3b as:
1 1
P0 = K 0y H x H x = K 0y H 2 (4.5)
2 2
The location of the line of action of the resultant force can be obtained by computing the
moment about the bottom of the wall, which is z = H / 3 for this case.
In addition to estimating the lateral earth pressure on the wall that is restrained from move-
ment, the at-rest earth pressure or K0 value is treated as the initial stress in the numerical analysis.
Example 4.1
A retaining wall 6 m high in sand is shown in Figure EX 4.1. The groundwater level is located
2 m below the surface. A uniformly distributed load q = 30 kN/m2 is applied on the surface. The
unit weight of the soil above and below the groundwater level is 17 and 20 kN/m3, respectively.
The effective strength parameters (c', 0 ') are shown in the figure. Compute the at-rest earth pres-
sure acting on the wall, including the pore water pressure.
Solution
According to eq. 4.3, K 0 = 1 - sin 0 ' = 1 - sin 34o = 0.44 .
Figure EX 4.1 (a) A 6-m high retaining wall in sand, (b) variation in the effective lateral earth
pressure with depth, (c) variation in the pore water pressure with depth, and
(d) variation in the total lateral earth pressure with depth.
As stated in Section 4.7, as a uniformly distributed load, q, is applied on the entire surface,
the increase in the vertical stress at any depth below the surface is equal to q, and the increase
in the lateral stress is K0 q.
At z = 0 m:
- v'0 = - v 0 = q = 30 kN/m2
- h' 0 = - h 0 = K 0 x - v'0 = 30 x 0.44 = 13.2 kN/m2
At z = 2 m:
- v'0 = - v 0 = q + y z = 30 + 17 x 2 = 64 kN/m2
- h' 0 = K 0 x - v'0 = 64 x 0.44 = 28.16 kN/m2
uw = 0
- h 0 = - h' 0 + uw = 28.16 kN/m2
At z = 6 m:
- v0 = 64 + 20 x 4 = 144 kN/m2
uw = 9.81x 4 = 39.24 kN/m2
- h' 0 = K 0 (- v 0 - uw ) = 46.09 kN/m2
- h 0 = - h' 0 + uw = 85.33 kN/m2
Figures EX 4.1b, c, and d show the variation in effective lateral earth pressure, pore water
pressure, and total lateral earth pressure at a given depth, respectively. The total resultant force
per unit length of the wall (Figure EX 4.1d) can then be calculated as:
P1 = 13.2 x 2 = 26.4 kN/m, P2 = (28.16 - 13.2) x 2 / 2 = 14.96 kN/m,

P3 = 28.16 x 4 = 112.64 kN/m, P4 = (85.33 - 28.16) x 4 / 2 = 114.34 kN/m
P0 = P1 + P2 + P3 + P4 = 268.34 kN/m
Location of the line of action of the total resultant force measured from the bottom of the wall:
P1 (2 / 2 + 4) + P2 (2 x 1 / 3 + 4) + P3 (4 x 1 / 2) + P4 (4 x 1 / 3)
z= = 2.16 m
P1 + P2 + P3 + P4
Example 4.2
Figure EX. 4.2a shows a retaining wall 6 m high in normally consolidated clay. The ground-
water level is located 2 m below the surface. The unit weight of the soil above and below
the groundwater level is 15 and 18 kN/m3, respectively. Both the effective strength parameters
(c', 0 ') and total strength undrained parameters (cu , φu) are also shown in the figure. Compute the
at-rest earth pressure acting on the wall, including the pore water pressure.
Solution
According to eq. 4.3, K 0 = 1 - sin 0 ' = 1 - sin 30o = 0.5 .
Since the soil is in the at-rest state, the wall does not move, and no shear stress is induced
in the soil. Therefore, no excess pore water pressure is generated in the soil, and the effective
Figure EX 4.2 (a) A 6-m high retaining wall in clay, (b) variation in the effective lateral earth
pressure with depth, (c) variation in the pore water pressure with depth, (d)
variation in the total lateral earth pressure with depth.
stress analysis should be conducted. The effective strength parameters will be used in the
analysis.
At z = 2 m:
- v0 = 15 x 2 = 30 kN/m2
- h' 0 = K 0 x - v'0 = 30 x 0.5 = 15 kN/m2, uw = 0
- h 0 = - h' 0 + uw = 15 kN/m2
At z = 6 m:
- v 0 = 15 x 2 + 18 x 4 = 102 kN/m2, - h' 0 = K 0 (- v 0 - uw ) = 31.38 kN/m2,
uw = 9.81x 4 = 39.24 kN/m2
- h 0 = - h' 0 + uw = 70.62 kN/m2
The variations in the effective lateral earth pressure, pore water pressure, and total lateral earth
pressure with depth are shown in Figures EX 4.2b, c, and d, respectively. The total resultant
force per unit length of the wall (Figure EX 4.2d) can then be calculated as:
P1 = 15 x 2 / 2 = 15 kN/m, P2 = 15 x 4 = 60 kN/m, P3 = (70.62 - 15) x 4 / 2 = 111.24 kN/m

P0 = P1 + P2 + P3 = 15 + 60 + 111.24 = 186.24 kN/m
Location of the line of action of the total resultant force measured from the bottom of the wall:
P1 (2 x 1 / 3 + 4) + P2 (4 x 1 / 2) + P3 (4 x 1 / 3)
z= = 1.82 m
P1 + P2 + P3
4.3 Rankine’s earth pressure theory
4.3.1 Active earth pressure

Rankine (1857) developed a theory of lateral earth pressure in front of and behind a retaining
wall on the basis of the concept of plastic equilibrium.
Assume that the retaining wall is infinitely long. As shown in Figure 4.4a, the parameters
for the Mohr‒Coulomb failure line in the back of the retaining wall are c and φ . Assuming that
there is no friction between the retaining wall and the soil in the back of the retaining wall, the
stress state is in the K0 state before the retaining wall moves. The stress state of the soil at depth
z below the ground surface can be represented by Mohr’s circle a in Figure 4.4b. Due to the
earth pressure acting on the back of the retaining wall, the wall rotates with respect to the wall
toe and moves to AB' from AB. The horizontal stress thus decreases with the movement (ΔH)
at the top of the wall, while the vertical stress remains unchanged, as shown by Mohr’s circle
b. As the movement (ΔH) or Mohr’s circle grows sufficiently large, causing Mohr’s circle to be
tangential to the Mohr‒Coulomb failure line, the soil is in failure. This type of failure is called
active failure, and the lateral earth pressure on the retaining wall is called the active earth pres-
sure, which is represented by σ a on circle c. Therefore:
AB AB (- v - - a ) / 2
sin 0 = = =
' '
O A O O+OA c cot 0 + (- v + - a ) / 2
Which can be simplified as follows:
1- sin 0 cos 0
-a = -v - 2c (4.6)
1+ sin 0 1+ sin 0
0 0
= - v tan 2 (45o - ) - 2c tan(45o - )
2 2
= - v K a - 2c K a (4.6a)
where K a is the coefficient of Rankine’s active earth pressure, K a = tan 2 (45o - 0 2).
According to Mohr’s failure theory, a failure zone forms behind the retaining wall that is
called the active failure zone. The soil in the failure zone is all in failure. The failure surfaces all
form an angle of (45o + 0 2) with the horizontal (see Figure 4.4c).
For effective stress analysis or drained analysis, the parameters of the Mohr‒Coulomb fail-
ure line should be expressed in terms of the effective cohesion (c’) and the effective internal
angle of friction (0 '). Eqs. 4.6 and 4.6a should be rewritten in terms of effective stress as
follows:
0' 0'
- a' = - v' tan 2 (45o - ) - 2c' tan(45o - ) (4.7)
2 2
= - v' K a - 2c' K a (4.7a)
where K a = tan 2 (45o - 0 ' 2).

According to Mohr‒Coulomb theory, the angle between the active failure surface and the
horizontal plane is 45o + 0 ' 2.
tH
Figure 4.4 Rankine’s active earth pressure: (a) stress state at depth z, (b) evolution of Mohr’s
circle, (c) variation in the active earth pressure with depth.
For total stress undrained analysis, the pressure (or stress) and strength parameters should be
expressed in terms of total stress parameters. For example, σ a, σ v, σ h, c, and φ . The equation has
exactly the same form as that shown in eq. 4.6.
For saturated clay under the short-term or undrained condition, the “ϕ = 0” concept is applied.
Assuming the undrained shear strength is su, the total active earth pressure (σ a) on the wall is:
- a = - v - 2su (4.8)
where K a = tan 2 (45o - 0 2) = 1.

According to eq. 4.6, we can obtain the theoretical distribution of earth pressure as shown in
Figure 4.4c. As shown in this figure, there is a tension zone in the cohesive soil behind the wall.
Soil is seldom subjected to tension stress, and therefore, tension cracks often occur. Assuming
the soil is unable to bear tension stress, let the lateral earth pressure be 0 in the tension crack
zone. According to eq. 4.6a, we have:
- a = y zK a - 2c K a = 0 (4.9)
The depth of a tension crack is:

2su
zc = (4.10)
y Ka
where zc is the depth of tension cracks.

Once tension cracks occur, rain and other environmental factors will further fill the cracks
with water, which will increase the total pressure on the wall.
The long-term behavior of cohesive soil should be analyzed on the basis of the complete
dissipation of excess pore water pressure. The distribution of earth pressure is estimated to be
similar to that of cohesionless soil, as discussed in the following section.
4.3.2 Passive earth pressure

Similarly, Mohr’s circle a in Figure 4.5a shows the initial stress state of the soil at depth z below
the ground surface before the movement of the wall. When the wall rotates backward and moves
to AB' from AB, the horizontal stress increases with the movement (ΔH), while the vertical
stress remains unchanged. The corresponding Mohr circle decreases, as represented by Mohr’s
circle b. When the horizontal stress is greater than the vertical stress, Mohr’s circle begins to
enlarge with the increase in ΔH until Mohr’s circle is tangent to the Mohr‒Coulomb failure line,
which indicates the soil in failure. This type of failure is called passive failure, and the lateral
earth pressure acting on the retaining wall is called passive earth pressure, as represented by σ p
on Mohr’s circle c.
Again, as shown in Figure 4.5b, we have:
QD QD (- p - - v ) / 2
sin 0 = = =
O'D O'O+OD c cot 0 + (- v + - p ) / 2
The preceding equation can be simplified as:

0 0
- p = - v tan 2 (45o + ) + 2c tan(45o + ) (4.11)
2 2
Figure 4.5 Rankine’s passive earth pressure: (a) stress state at depth z, (b) evolution of Mohr’s
circle, and (c) variation in the passive earth pressure with depth.
= - v K p + 2c K p (4.11a)
where K p is the coefficient of Rankine’s passive earth pressure, K p = tan (45 + 0 2).
2 o
Similarly, according to Mohr’s failure theory, the soil within the failure zone where pas-
sive failures occur is called the passive failure zone. The soil in the area is all in failure,
whose failure surfaces form angles of 45o - 0 2 with the horizontal plane, as shown in
Figure 4.5c.
Similar to the active earth pressure, for effective stress analysis or drained analysis, eqs. 4.11
and 4.11a should be rewritten in terms of effective stress as:
0' 0'
- 'p = - v' tan 2 (45o + ) + 2c' tan(45o + ) (4.12)
2 2
= - v' K p + 2c' K p (4.12a)
where K p = tan (45 + 0 ' 2).

2 o
For saturated clay under the short-term or undrained condition, the “ϕ = 0” concept is applied.
Assuming the undrained shear strength is su , the total passive earth pressure (σ p) acting on the
wall is:
- p = - v + 2su (4.13)
4.3.3 Earth pressure with sloping ground

According to Rankine’s earth pressure theory, the coefficients of the earth pressure for sloping
ground (Bowles, 1988) can be computed as follows:
cos B - cos 2 B - cos 2 0

Ka = (4.14)
cos B + cos 2 B - cos 2 0
cos B + cos 2 B - cos 2 0

Kp = (4.15)
cos B - cos 2 B - cos 2 0
where β = slope of the ground.

Thus, the active and passive earth pressures and their resultants acting on the retaining wall,
as shown in Figure 4.6, are:
- a = y zK a (4.16)
1
Pa = y H 2 K a (4.17)
2
- p = y zK p (4.18)
1
Pp = yh H 2 K p (4.19)
2
Figure 4.6 Rankine’s earth pressures with sloping ground.
where:
γ = unit weight of soil.
H = height of the retaining wall.
Pa = resultant of active earth pressure.
Pp = resultant of passive earth pressure.
The horizontal and vertical components of the resultants are:
- a , h = y zK a cos B , - a ,v = y zK a sin B (4.20)
1 1
Pa , h = y H 2 K a cos B , Pa ,v = y H 2 K a sin B (4.21)
2 2
- p , h = y zK p cos B , - p ,v = y zK p sin B (4.22)
1 1
Pp , h = Y H 2 K p cos B , Pp ,v = y H 2 K p cos B (4.23)
2 2
Example 4.3
Figure EX. 4.3a shows a 6 m high retaining wall in sand. The groundwater level is located 3
m below the surface. The unit weight of the soil above and below the groundwater level are 16
and 21 kN/m3, respectively. The effective stress parameters ( c', 0 ' ) are also shown in the figure.
Compute the active earth pressure on the wall using Rankine’s earth pressure theory.
Figure EX 4.3 (a) A 6 m high retaining wall in sand, (b) variation in the effective active earth
pressure with depth, (c) variation in the pore water pressure with depth, and
(d) variation in the total active earth pressure with depth.
Solution
0' 33o
For 0 ' = 33o , K a = tan 2 (45 - ) = tan 2 (45 - ) = 0.294
2 2
At z = 3 m:
- a' = 16 x 3 x 0.294 = 14.11kN/m2, uw = 0
- a = - a' + uw = 14.11 kN/m2
At z = 6 m:
- a' = [16 x 3 + (21 - 9.81) x 3] x 0.294 = 23.98 kN/m2, uw = 9.81x 3 = 29.43 kN/m2
- a = - a' + uw = 53.41 kN/m2
Figures EX 4.3b, c, and d show the variations in the effective active earth pressure, pore water
pressure, and total active earth pressure, respectively.
The total resultant force (Figure EX 4.3d) per unit length of the wall can be calculated as
follows:
Pa1 = 14.11x 3 / 2 = 21.17 kN/m, Pa 2 = (53.41 - 14.11) x 3 / 2 = 58.95 kN/m

Pa3 = 14.11x 3 = 42.33 kN/m
Pa = Pa1 + Pa 2 + Pa 3 = 21.17 + 58.95 + 42.33 = 122.45 kN/m
Location of the line of action of the resultant force measured from the bottom of the wall:
Pa1 x (3 + 3x 1 / 3) + Pa 2 (3 x 1 / 3) + Pa 3 (3 x 1 / 2) 207.13
z= = = 1.69 m
Pa1 + Pa 2 + Pa3 122.45
Example 4.4
Figure EX 4.4 shows a 6 m high retaining wall in clay where the soil condition is the same as
in example 4.2. Compute the active earth pressure on the wall using Rankine’s earth pressure
theory.
Solution
0 0o
The “0 = 0 concept” is applied for clay, K a = tan 2 (45 - ) = tan 2 (45 - ) = 1.
2 2
1. Tension cracks are not formed, that is, the soil is able to bear tension stress.
At z = 0 m:
- a = -2c K a = -40 kN/m2
At z = 2 m-:
- a = 15 x 2 - 2 x 20 = 30 - 40 = -10 kN/m2
At z = 2 m+:
- a = 15 x 2 - 2 x 30 = 30 - 60 = -30 kN/m2
At z = 6 m:
- a = 15 x 2 + 18 x 4 - 2 x 30 = 42 kN/m2
The variation in earth pressure with depth is shown in Figure EX 4.4a. From this figure, we
have:
Pa1 = 30 x 2 x 0.5 = 30 kN/m, Pa 2 = 30 x 4 = 120 kN/m

Pa3 = 72 x 4 x 0.5 = 144 kN/m
Pa 4 = 2 x 40 = 80 kN/m
Pa5 = 4 x 60 = 240 kN/m
The resultant force per unit length of the wall is:
Pa1 + Pa 2 + Pa 3 - Pa 4 - Pa5 = 30 + 120 + 144 - 80 - 240 = -26 kN/m
2. Tension cracks occur.
The depth of the tension crack can be calculated as:
0 = [2 x 15 + ( zc - 2) x 18] x K a - 2 x 30 , zc = 3.67 m
Figure EX 4.4 (a) Variation in the active earth pressure without the occurrence of tension
cracks in a 6 m high retaining wall in clay.
The active earth pressure is:
At z = 3.67 m:
-a = 0
At z = 6 m:
- a = 2 x 15 + 4 x18 - 2 x 30 = 102 - 60 = 42 kN/m 2
If the tension cracks are full of water due to the original groundwater, then the pore water pres-
sure in the tension crack is:
At z = 2 m:
uw = 0
At z = 3.67 m:
uw = (3.67 - 2) x 9.81 = 16.38 kN/m2
The variation in active earth pressure and pore water pressure with depth is shown in Figure
EX 4.4b.
Pa = 42 x (6 - 3.67) x 0.5 = 48.93 kN/m, Pw = 16.38 x (3.67 - 2) / 2 = 13.68 kN/m

Pa + Pw = 62.61 kN/m
Figure EX 4.4 (b) Variation in the active earth pressure with tension cracks in a 6 m high
retaining wall in clay.
Location of the resultant force from the bottom of the wall:

Pa x (2.33 x 1 3) + Pw x (2.33 + 1.67 x 1 3) 77.44
z= = = 1.24 m
Pa + Pw 62.61
Example 4.5
Figure EX. 4.5a shows a 9 m high retaining wall. The sand and clay deposits are behind the
wall. The groundwater level is located on the surface. The saturated unit weight of the soils and
strength parameters are indicated in the figure. Compute the active earth pressure on the wall
using Rankine’s earth pressure theory.
Solution
0' 31o
Sand(1) : K a = tan 2 (45 - ) = tan 2 (45 - ) = 0.32
2 2
0' 33o
Sand(2) : K a = tan 2 (45 - ) = tan 2 (45 - ) = 0.3
2 2
0 0o
Clay : K a = tan 2 (45 - ) = tan 2 (45 - ) = 1
2 2
The earth pressure (- a' or σ a) and pore water pressure ( uw ) at each depth are calculated as
follows.
The effective stress analysis with pore water pressure consideration is used for sand, while the total
stress undrained analysis without pore water pressure consideration is used for clay. Therefore:
At z = 3 m − :
- a' = (21 - 9.81) x 3 x 0.32 = 10.74 kN/m2, uw = 9.81x 3 = 29.43 kN/m2
- a = - a' + uw = 40.17 kN/m2
Figure EX 4.5 (a) A 9 m high retaining wall in a sand and clay deposit.
At z = 3 m + :
- a = 21x 3 x 1 - 2 x 25 = 13 kN/m2
At z = 6 m − :
- a = 21x 3 + 18 x 3 - 2 x 25 = 67 kN/m2
At z = 6 m + :
- a' = (117-6 x 9.81) x 0.3 = 17.44 kN/m2, uw = 9.81x 6 = 58.86 kN/m2
- a = - a' + uw = 76.3 kN/m2
At z = 9 m :
= a' = [117+3 x 21-(9 x 9.81)] x 0.3 = 27.51kN/m2, uw = 9.81x 9 = 88.29 kN/m2
- a = - a' + uw = 115.8 kN/m2
The variations in earth pressure and pore water pressure are shown in Figure EX 4.5b. The
resultant force per unit length of the wall is then calculated as:
Pa = Pa1 + Pa 2 + Pa 3 + Pa 4 + Pa 5
= 40.17 × 3 / 2 +13 × 3 + (67 -13) × 3 / 2 + 76.3 × 3 + (1115.8 - 76.3) × 3 / 2
= 60.26 + 39 + 81+ 228.9 + 59.25 = 468.41 kN / m
Figure EX 4.5 (b) Variation in the active earth pressure and pore water pressure with depth.
Location of the resultant force measured from the bottom of the wall:
60.26x (3 / 3 + 6) + 39 x (3 / 2 + 3) + 81x (3 / 3 + 3) + 228.9 x 3 / 2 + 59.25x 3 / 3

z=
4668.41
1323.92
= = 2.83 m
468.41
Example 4.6
The wall and soil conditions are the same as those shown in Figure EX 4.3a. Compute the pas-
sive earth pressure on the wall using Rankine’s earth pressure theory.
Solution
0' 33o
For 0 ' = 33o , K p = tan 2 (45 + ) = tan 2 (45 + ) = 3.39
2 2
The effective earth pressure ( - a' ) and pore water pressure ( uw ) at each depth are calculated as
follows:
At z = 3 m:
- p = - 'p = 16 x 3 x 3.39 = 162.72 kN/m2, uw = 0
At z = 6 m:
- 'p = [16 x 3 + (21 - 9.81) x 3]x 3.39 = 276.52 kN/m2, uw = 9.81x 3 = 29.43 kN/m2
- p = - 'p + uw = 305.95 kN/m2
The variations in passive earth pressure and pore water pressure are shown in Figure EX 4.6.
The resultant force per unit length of the wall can be calculated as:
Pp1 = 162.72 x 3 / 2 = 244.08 kN/m, Pp 2 = (305.95 - 162.72) x 3 / 2 = 214.85 kN/m
Figure EX 4.6 Variation in passive earth pressure and pore water pressure with depth.
Pp3 = 162.72 x 3 = 488.16 kN/m

PP = PP1 + PP 2 + PP 3 = 947.09 kN/m
Location of the resultant force from the bottom of the wall:

244.08 x (3 + 3x 1 / 3) + 214.85x (3 x 1 / 3) + 488.16 x (3 x 1 / 2)
z= = 2.03 m
947.09
4.4 Coulomb earth pressure theory
4.4.1 Active earth pressure

Coulomb’s earth pressure theory (1776) assumes that the soil in the back of the retaining wall is
homogeneous and cohesionless, the failure surface is a plane, the wedge between the wall and
the failure surface is rigid material, and the weight of the wedge, the reaction of the soil, and the
reaction of the wall are in equilibrium.
Figure 4.7a shows a retaining wall of height H, retaining a soil with an angle of friction φ
and an angle of friction between the wall and soil δ . The BC1 line is an assumed failure surface
that intersects the horizontal direction at an angle of a1. Figure 4.7b illustrates the force polygon
formed by the reaction (P) of the wall against the wedge, the reaction (R) of the soil against the
wedge, and the weight (W) of the soil wedge. The directions and magnitudes of P and R can then
be determined from the force polygon.
By assuming different failure surfaces, for example, BC2 and BC3, and with a similar proce-
dure, we can yield the corresponding reaction forces (P) of the wall. The largest reaction force,
Pmax, is then the active force on the wall, as shown in Figure 4.7a.
If the failure surface is treated as a variable, expressed by a1, according to the force polygon
in Figure 4.7b, we can derive the following equation:
W P
= (4.24)
sin(90 + 0 + - - a + p ) sin(a - 0 )
o
W sin(a - 0 )
P= (4.25)
sin(90o + 0 + - - a + 0 )
1 [ cos(0 - a ) cos(0 - B ) sin(a - 0 ) ]

= yH2 | 2 | (4.25a)
2 [ cos 0 s in(a - B ) sin(90 o
+ 0 + - - a + 0 ) ]
where γ is the unit weight of the soil; the parameters γ , φ , δ , θ , β , and H are constants; a is a
variable because BC is an assumed failure surface; P varies with a ; and the maximum P repre-
sents the active force Pa, that is:
dP
=0 (4.26)
da
Substituting the critical a value derived from eq. 4.26 into eq. 4.25, we obtain the active force
(Pa ) as follows:
1
Pa = y H 2 K a (4.27)
2
Figure 4.7 Coulomb’s active earth pressure: (a) assumed failure surface and the corresponding
reaction of the wall against the failure wedge, (b) force polygon, and (c) distribution
and resultant of active earth pressure.
cos 2 (0 - 0 )
Ka = 2
[ sin(d + 0 ) sin(0 - B ) ] (4.27a)
cos 0 cos(d + 0 ) |1 +
2
|
[ cos(d + 0 ) cos(0 - B ) ]
where K a is the coefficient of Coulomb’s active earth pressure or active force.

The direction of the active earth pressure and its distribution are shown in Figure 4.7c. The
horizontal components of the active force and active earth pressure are Pah = Pa cos(d + 0 ) and
- ah = - a cos(d + 0 ), respectively. When 0 = 0, B = 0, and d + 0, eq. 4.27a can be rewritten as
K a = tan 2 (45o - 0 2), which is identical to Rankine’s.
4.4.2 Passive earth pressure

Figure 4.8 illustrates the passive soil failure in the back of the retaining wall, which is pushed
backward by an external force. The BC line is an assumed failure surface. The directions of the
soil reaction force (R) and the reaction (P) of the wall are determined on the basis that the wall
is pushed backward and the wedge (ABC) moves upward. Following a similar method used to
obtain the active force, the passive force (Pp) can be derived as shown in Figure 4.8b, which can
also be expressed by the following equation:
1
Pp = y H 2 K p (4.28)
2
cos 2 (0 + 0 )
Kp = 2
[ sin(0 + d ) sin(0 + B ) ] (4.28a)
cos 0 cos(d - 0 ) |1 -
2
|
[ cos(d - 0 ) cos( B - 0 ) ]
where K p is the coefficient of Coulomb’s passive earth pressure or force.
The direction of the passive force and its distribution are shown in Figure 4.8c. The horizon-
tal components of the passive force and the passive earth pressure are Pph = Pp cos(d - 0 ) and
- ph = - p cos(d - 0 ), respectively. When 0 = 0, B = 0, and d = 0, eq. 4.28a can be rewritten as
K p = tan 2 (45o + 0 2), which is identical to Rankine’s.
Example 4.7
Similar to example 4.3. Compute the active earth pressure and its line of action of the resultant
force using Coulomb’s earth pressure theory with d = 20 ' / 3.
Solution
According to eq. 4.27a, K a = 0.26.
The horizontal component of the coefficient of active earth pressure is:
2
K ah = K a x cos d = 0.26 x cos( x 33) = 0.24
3
At z = 3 m:
- a ,h = - a' ,h = 16 x 3 x 0.24 = 11.52 kN/m2
At z = 6 m:
- a' ,h = [16 x 3 + (21 - 9.81) x 3] x 0.24 = 19.58 kN/m2, uw = 9.81x 3 = 29.43 kN/m2
- a ,h = - a' ,h + uw = 49.01 kN/m2
Figure 4.8 Coulomb’s passive earth pressure: (a) assumed failure surface and the correspond-
ing reaction of the wall against the failure wedge, (b) force polygon, and (c) distribu-
tion and resultant of passive earth pressure.
Figure EX 4.7 Variation in the active earth pressure and pore water pressure with depth.
The variations in acitve earth pressure and pore water pressure with depth are shown in Figure EX 4.7.
The resultant force per unit length of the wall can be calculated as:
Pa1 = 11.52 x 3 / 2 = 17.28 kN/m

Pa 2 = (49.01 - 11.52) x 3 / 2 = 56.24 kN/m
Pa3 = 11.52 x 3 = 34.56 kN/m
Pa = Pa1 + Pa 2 + Pa 3 = 108.08 kN/m
17.28x (3 + 3 x 1 / 3) + 56.24 x (3 x 1/ 3) + 34.56 x (3 x 1 / 2)
z= = 1.64 m
108.08
Example 4.8
Similar to example 4.3. Compute the passive earth pressure and its line of action of resultant
force using Coulomb’s earth pressure theory with d = 20 ' / 3 .
Solution
According to eq. 4.28a, K p = 8.08 .
The horizontal component of the coefficient of active earth pressure is:
2
K ph = K p x cos d = 8.08 x cos( x 33) = 7.49
3
At z = 3 m:
- p , h = - 'p , h = 16 x 3 x 7.49 = 359.52 kN/m2
At z = 6 m:
- 'p , h = [16 x 3 + (21 - 9.81) x 3] x 7.49 = 610.96 kN/m2, uw = 9.81x 3 = 29.43 kN/m2
- p , h = - 'p , h + uw = 640.39 kN/m2
Figure EX 4.8 Variation in the passive earth pressure and pore water pressure with depth.
The variations in passive earth pressure and pore water pressure with depth are shown in Figure
EX 4.8.
The resultant of passive earth pressure per unit length can be computed as:
Pp1 = 359.52 x 3 / 2 = 539.28 kN/m

Pp 2 = (640.39 - 359.52) x 3 / 2 = 421.31 kN/m
Pp3 = 359.52 x 3 = 1078.56 kN/m
Pp = Pp1 + Pp 2 + Pp 3 = 2039.15 kN/m
539.28x (3 + 3x 1 / 3) + 421.31x (3 x 1/ 3) + 1078.56 x (3 x 1 / 2)
z= = 2..06 m
2039.15
4.5 Displacement and earth pressure

According to the loading condition of the soil, when the strain state of the soil changes from the
K 0 state to the active state, the direction of its principal stresses remains unchanged. However, if
the direction is from the K 0 state to the passive state, the principal direction rotates by 90o. That
is, σ 1 is originally vertical (- 1 = - v) and then becomes horizontal (- 1 = - h). Both the major and
minor principal stresses rotate by 90o.
Thus, the strain of the soil required to reach passive failure should be greater than that
required to reach active failure. Regarding the problem of retaining walls, many experiments
have explored the relationships between wall displacement and earth pressure. Figure 4.9 shows
the results from experiments on rigid gravity walls provided by NAVFAC DM7.2 (1982). As
shown in Figure 4.9, the necessary wall displacement inducing the passive condition for loose
and dense sands is much larger than that inducing active conditions. Similar behavior can be
found for cohesive soil.
Figure 4.9 Effect of wall movement on earth pressure (ΔH is the lateral movement at the top
of the wall, and H is the wall height).
Source: Redrawn from NAVFAC DM7.2 (1982).
4.6 Caquot and Kerisel’s solution

Rankine’s earth pressure theory assumes no friction between the retaining wall and soil. Cou-
lomb’s earth pressure theory takes into account the friction between the wall and soil but assumes
that its failure surface is a plane. In fact, the wall surface may be rough, and friction between the
wall and soil does exist, causing the failure surface to no longer be a plane. Whether the earth
pressure theories are sufficiently accurate to represent the actual earth pressure depends on how
close the assumed failure surfaces are to the actual condition, that is, the nonplane condition
(Peck and Ireland, 1961; Rowe and Peaker, 1965; Mackey and Kirk, 1967; James and Bransby,
1970; Rehnman and Broms, 1972). As Figure 4.10a shows, Coulomb’s active failure surface is
close to the actual failure surface. Thus, Coulomb’s active earth pressure is close to the actual
earth pressure. Coulomb’s earth pressure theory can then be used to estimate the active earth
pressure.
However, as shown in Figure 4.10b, Coulomb’s passive failure surface is rather different from
the actual surface. The larger the friction between the retaining wall and the surrounding soil,
the farther the failure surface to a plane and the greater the difference between Coulomb’s pas-
sive earth pressure from the actual one. In addition to Rankine’s and Coulomb’s theories, several
earth pressure theories are available in the literature. Figure 4.11 shows a commonly used model
for calculating the passive earth pressure, which considers the friction between the wall and the
soil. BC can be assumed to be an arc of a circle, an ellipse, or a logarithm spiral, while CD can
be a plane. Then, following a method similar to Coulomb’s earth pressure theory, the forces
Figure 4.10 Real failure surface and assumed failure surface in Coulomb’s earth pressure the-
ory: (a) active condition and (b) passive condition.
Figure 4.11 Passive earth pressure with curved failure surface.

in the wedge, including the reaction force of the wall, the reaction force of the soil, and the
weight of the wedge, should be in equilibrium and form a close force polygon. The passive earth
pressure can then be solved. Caquot and Kerisel (1948) assumed that BC is a function of the
logarithm spiral and derived the passive earth pressure for various conditions. NAVFAC DM7.2
(1982) simplifies Caquot‒Kerisel’s solutions by figures and reduction factors. Figures 4.12 and
4.13 show Caquot‒Kerisel’s Kp for d = 0 '. For d = 0 ', a reduction factor is provided, as shown in
Table 4.1. Figure 4.14 shows the values of Kp from Rankine’s, Coulomb’s, and Caquot‒Kerisel’s
earth pressure theories for various 0 ' for d = 0 '. It can be found that the Rankine’s Kp are the
smallest, Coulomb’s the largest, while Caquot‒Kerisel’s Kp are in between. Such a difference
decreases with the reduction of the δ value.
The equation for Caquot–Kerisel’s passive earth pressure has the same form as Coulomb’s
(i.e., eq. 4.28), but with different Kp values, as shown here:
1
Pp = y K p H 2
2
Figure 4.12 Caquot and Kerisel’s passive earth pressure with vertical wall and sloping backfill
for δ/φ' = 1.0.
Figure 4.13 Caquot and Kerisel’s passive earth pressure with a sloping wall and horizontal
backfill for δ/φ’ = 1.0.
Table 4.1 Reduction factor (R) for Caquot and Kerisel’s passive pressure calculation
φ′ δ / φ′
0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0
10 0.978 0.962 0.946 0.929 0.912 0.898 0.881 0.864

15 0.961 0.934 0.907 0.881 0.854 0.830 0.803 0.775
20 0.939 0.901 0.862 0.824 0.787 0.752 0.716 0.678
25 0.912 0.860 0.808 0.759 0.711 0.666 0.620 0.574
30 0.878 0.811 0.746 0.686 0.627 0.574 0.520 0.467
35 0.836 0.752 0.674 0.603 0.536 0.475 0.417 0.362
40 0.783 0.682 0.592 0.512 0.439 0.375 0.316 0.262
45 0.718 0.600 0.500 0.414 0.339 0.276 0.221 0.174
Caquot–Kerisel
Figure 4.14 Comparison of earth pressures from Rankine’s, Coulomb’s, and Caquot–Kerisel’s
earth pressure theories.
Source: Ou (2022).
If d = 0 ', Kp can be obtained directly from Figures 4.13 or 4.14, while if d = 0 ', Kp can be esti-
mated as:
Kp (δ ≠ ϕ′) = R × Kp (δ = ϕ′ ) (4.29)
Example 4.9
Similar to example 4.3. Compute the passive earth pressure and its line of action of resultant
force using Caquot–Kerisel’s earth pressure theory with d = 20 ' / 3.
Solution
According to Figure 4.12 or 4.13, K p = 8.5 for d / 0 ' = 1.0 and 0 ' = 33o .
Table 4.1 shows that the reduction factor R = 0.83 for d / 0 ' = 2 / 3 . Therefore,
K p = 8.5 x 0.83 = 7.05 .
K ph = K p x cos d = 7.05 x cos( 2 / 3 x 33) = 6.54

At z = 3 m:
- p , h = - 'p , h = 16 x 3 x 6.54 = 313.92 kN/m2
At z = 6 m:
- 'p , h = [16 x 3 + (21 - 9.81) x 3] x 6.54 = 533.47 kN/m2, uw = 9.81x 3 = 29.43 kN/m2
- p , h = - 'p , h + uw = 562.9 kN/m2
The variations in passive earth pressure and pore water pressure with depth are shown in Figure
EX 4.9.
Pp1 = 313.92 x 3 / 2 = 470.88 kN/m

Pp 2 = (562.9 - 313.92) x 3 / 2 = 373.47 kN/m
Pp3 = 313.92 x 3 = 941.76 kN/m
Pp = Pp1 + Pp 2 + Pp 3 = 1786.11 kN/m
Resultant force location from the bottom of the wall:

470.88 x (3 + 3 x 1 / 3) + 373.47 x (3 x 1/ 3) + 941.76 x (3 x 1/ 2)
z= = 2.006 m
1786.11
Figure EX 4.9 Variation in the passive earth pressure and pore water pressure with depth.
4.7 Earth pressure due to surcharge

Figure 4.15 shows a uniformly distributed load, q, acting on a very large area of the ground sur-
face. If the wall is restrained from movement, the stress state should be in the at-rest condition.
The lateral earth pressure is:
- h = qK 0 (4.29)
where K 0 is the coefficient of earth pressure at rest.

If the wall moves forward or backward toward the soil, the lateral earth pressure induced
by the uniformly distributed loads can be estimated as either active or passive earth pressure,
respectively:
- a = qK a (4.30)
- p = qK p (4.31)
If a line load or strip load is applied to the surface, the wall will certainly move, but the
movement may not be sufficiently large to make the soil achieve an active or passive state.
Moreover, the solutions to lateral pressures induced by a line load or strip load in the active or
passive state have not yet been derived. Under such conditions, the equations obtained from the
theory of elasticity, with some modification, are adopted to estimate the lateral earth pressure
induced by surface loads. Figure 4.16 illustrates the lateral earth pressure distribution caused
by a line load of intensity Q per unit length parallel to the retaining wall. The lateral earth
Figure 4.15 Lateral earth pressure produced by a uniformly distributed load.

pressure at any depth z and its resultant can be expressed as follows (NAVFAC DM7.2, 1982;
USS Steel, 1975):
m ≤ 0.4
0.203Q n
-h = (4.32)
H (0.16 + n 2 ) 2
m > 0.4
1.28Q m2 n
-h = (4.33)
H (m 2 + n 2 ) 2
And the resultant is:

0.64Q
Ph = (4.34)
(m 2 +1)
Figure 4.16 illustrates the dimensionless diagram of the earth pressure distribution derived
from the preceding equation, where m = 0.1, 0.3, 0.5, and 0.7. The corresponding point of action
R is also marked in the figure.
Figure 4.17 illustrates a strip load with an intensity Qs parallel to the retaining wall. The earth
pressure acting on the wall caused by the strip load can be determined by the following equation
(USS Steel, 1975):
2Qs
-h = ( B - sin B cos 2a ) (4.35)
H
where α and β are angles in radians.
Figure 4.16 Lateral earth pressure produced by a line load.

Figure 4.17 Lateral earth pressure produced by a strip load.
4.8 Earth pressure due to earthquakes

Earthquakes will generate lateral and vertical inertia forces on a soil mass. Let a represent the
earthquake acceleration. Then, the inertia force (F) is:
W
F = ma = (kh or kv x g ) = Wkh or Wkv (4.36)
g
where m is the soil mass; W is the soil weight; g is the acceleration of gravity; a is the accelera-
tion, which is a fraction of the acceleration of gravity, such as a = kv g or kh g; kh is the coefficient
of horizontal acceleration; and kv is the coefficient of vertical acceleration.
In the ultimate condition, earthquakes will increase the active earth pressure behind the
retaining wall and decrease the passive earth pressure in front of the wall. The Mononobe–
Okabe equation is generally adopted (Mononobe, 1929; Okabe, 1926) to compute the active
and passive earth pressures under the influence of earthquakes. As shown in Figure 4.18, the
total (static + dynamic) active force (Pae) during an earthquake can be computed by the follow-
ing equations:
1
Pae = y H 2 (1 - kv ) K ae (4.37)
2
cos 2 (0 - 0 - a )
K ae = 2
{ 1
} (4.37a)
| [ sin(0 + d ) sin(0 - B - a ) ] 2 |
cos a cos 0 cos(d + 0 + a ) {1+ |
2
| }
| [ cos(d + 0 + a ) cos( B - 0 ) ] |
{ }
Figure 4.18 The active force due to earthquakes.
where a = tan -1 [ kh (1 - kv ) ] .
In addition, based on many tests, Bowles (1988) suggested that δ be assumed to be 0 under
dynamic conditions.
Seed and Whitman (1970) proposed that the total (static + dynamic) active force (Pae) could
be separated into two components: the static active force (Pa) and the dynamic increment (ΔPae)
due to an earthquake, as illustrated in Figure 4.19, which can be expressed as:
Pae = Pa + -Pae (4.38)
The static active force as computed from Coulomb’s theory is applied at H/3 from the base
of the wall in a homogeneous cohesionless soil, resulting in a triangular distribution of earth
pressure. Seed and Whitman (1970) also proposed that the dynamic increment was an inverted
triangular pressure distribution with a force resultant acting at 0.6H or 2H/3 from the base of the
wall. The location of the total active force can be determined as follows:
1. Compute the dynamic increment of the active force ΔPae:
-Pae = Pae - Pa (4.39)
2. Locate Pa at H/3 from the base of the wall.

3. Locate ΔPae at 2H/3 (or 0.6H) from the base of the wall.
4. Compute the location (z ) of the total active force (Pae):
Pa ( H 3) + -Pae (2 H / 3)
z= (4.40)
Pae
Figure 4.19 Total seismic earth pressure consisting of a static active earth force and a dynamic
increment.
Moreover, the total (static + dynamic) passive force (Ppe) under the influence of earthquakes
can be computed by the following equations (Figure 4.20):
1
Ppe = y H 2 (1 - kv ) K pe (4.41)
2
cos 2 (0 + 0 - a )
K pe = 2
{ 1
} (4.41a)
| [ sin(0 + d ) sin(0 + B - a ) ] 2 |
cos a cos 0 cos(d - 0 + a ) {1- |
2
| }
| [ cos(d - 0 + a ) cos( B - 0 ) ] |
{ }
The distribution and characteristics of the static passive force, dynamic increment, and total
passive force are similar to those displayed in Figure 4.19. The locations of the static passive
force and the dynamic increment are also at H/3 and 2H/3 (or 0.6 H) from the base of the wall,
respectively. Therefore, the location of the total passive force can be determined in a way similar
to the total active force.
In design, the influence of earthquakes on the safety of a retaining wall as a permanent struc-
ture should be considered. In deep excavations, the outer walls of a basement are permanent
structures. The design should thereby take into consideration the influence of earthquakes. If the
diaphragm walls also serve as the outer walls of a basement, the influence of earthquakes should
be considered accordingly. Otherwise, the influence of earthquakes can be ignored. For tempo-
rary retaining structures, such as soldier piles, steel sheet piles, and column piles, the effects of
earthquake are ignorable.
Figure 4.20 The passive force due to earthquakes.

This chapter can be summarized as follows:
1. For retaining walls where displacement does not occur, such as the outer wall of a base-
ment, the lateral earth pressure at rest is adopted for design. The coefficients of the at-rest
lateral earth pressure for normally consolidated cohesive soils and cohesionless soils can be
estimated by Jaky’s equation. For overconsolidated soils, this chapter provides an empirical
equation estimation.
2. Rankine’s earth pressure theory cannot consider the friction between the retaining wall and
soil. Coulomb’s earth pressure theory, on the other hand, can. Although the two theories are
based on different assumptions, the obtained earth pressures are identical when a vertical
and smooth wall surface is used. Both theories assume the failure surfaces as planes, not
conforming to reality, and thus, their results cannot represent actual earth pressures.
3. As the friction between the retaining wall and soil is considered, the real failure surface
is a curved surface. Thus, Rankine’s active earth pressure would overestimate the real
earth pressure very slightly, whereas its passive earth pressure would underestimate the
real value. Coulomb’s active earth pressure is usually treated as the real earth pressure
because the assumed failure surface is close to the actual value. However, Coulomb’s
passive earth pressure significantly overestimates the real passive earth pressure for a
large friction angle between the wall and soil. Caquot‒Kerisel’s earth pressures, both
active and passive, are the closest to the real values and are thus regarded as the real
earth pressures.
4. In retaining wall problems, the active earth pressure is the main force engendering failure.
Caquot‒Kerisel’s active earth pressure, regarded as the real earth pressure, is adopted for
design. Coulomb’s active earth pressure is close to Caquot‒Kerisel’s; therefore, Coulomb’s

active earth pressure can be used for design.
5. The passive earth pressure is often the force used to resist failure. Caqout‒Kerisel’s earth
pressure is the first choice for analysis and design since it is regarded as the real earth pres-
sure. Rankine’s earth pressure is too small and differs significantly from the real value. Thus,
Rankine’s is not adopted. When d < 0.50 ', Coulomb’s coefficient of passive earth pressure
is similar to that of Caqout‒Kerisel’s and can also be used for analysis and design. When
d > 0.50 ', Coulomb’s coefficient of passive earth pressure is obviously larger than Caqout‒
Kerisel’s and will lead to an unsafe design.
Problems
4.1 A 9.5 m deep (measured from the ground surface to the bottom of the foundation) two-
story basement is located in a normally consolidated clay. The depth of the groundwater
level is H = 2.5 m, and the unit weights above and below the groundwater level are y = 16
kN/m3 and y sat = 20 kN/m3 (saturated unit weight), respectively. The effective strength
parameters are c′ = 0 and 0 ' = 32o. The undrained shear strength is su = 55 kN/m 2. Compute
the total lateral force per unit length of the basement wall and the location of the resultant.
4.2 Similar to problem 4.1. The depth of the foundation is 9.5 m, and the groundwater level
H = 3 m. The soil is silty sand. Assume the following parameters: y = 18 kN/m3, y sat = 22
kN/m3 (saturated unit weight), c′ = 0, 0 ' = 34o. Compute the total lateral force per unit
length of the outer wall of the basement and the location of the resultant.
4.3 Assume the ground shown in Figure P4.3 is clay. The groundwater level location H1 = 4 m
and H 2 = 5 m. Above the groundwater level, su1 = 20 kN/m 2 and y 1 = 17 kN/m3, and below
the groundwater level, y sat 2 = 18 kN/m3 and su 2 = 30 kN/m 2. Determine the following:
a. The depth of the tension cracks
b. The variation in Rankine’s active pressure with depth
c. The magnitude and location of the total active thrust per unit length of the wall, assum-
ing tension cracks have occurred
Figure P4.3
Figure P4.5
Figure P4.7
4.4 Redo problem 4.3 with the following parameters: H1 = 3.5 m, H 2 = 4.5 m; su1 = 15kN/m 2,
y 1 = 16 kN/m3; su 2 = 25 kN/m 2, and y sat 2 = 17 kN/m3.
4.5 Figure P4.5 shows a wall in sand where the groundwater level H1 = 2.0 m and H2 = 4.0 m.
The effective strength parameters are c′ = 0 and 0 ' = 34o . The unit weight is y = 16 kN/m3 ,
y sat = 20 kN/m3. Use Rankine’s earth pressure theory to compute the total active force per
unit length of the wall and the location of the resultant.
4.6 Redo problem 4.5 with H1 = 3.0 m and H2 = 4.0 m. The effective strength parameters are
c′ = 0 and 0 ' = 35o. The unit weight is y = 15 kN/m3, y sat = 22 kN/m3.
4.7 Figure P4.7 shows a wall in two layers of clay with H1 = 4.0 m and H2 = 4.0 m. The ground-
water level is on the surface. The undrained shear strengths and saturated unit weight are
su1 = 20 kN/m 2 and y sat1 = 16 kN/m3, respectively, for the upper clay and su 2 = 25 kN/m 2
and y sat 2 = 18 kN/m3, respectively, for the lower clay. Determine the following:
a. The depth of the tension cracks
c. The magnitude and location of the total active thrust per unit length of the wall, assum-
ing tension cracks do not occur
4.8 Redo problem 4.7 with the following parameters: H1 = 3.0 m, H2 = 4.0 m. su1 = 10 kN/m 2,
y sat1 = 16 kN/m3, su 2 = 20 kN/m 2, and y sat 2 = 17 kN/m3.
4.9 Figure P4.9 shows a wall in two layers of sand with H1 = 2.0 m and H2 = 3.0 m. The
groundwater level is on the surface. The effective strength parameters and saturated unit
weight are c1′ = 0, 01' = 30o, and y sat1 =19 kN/m3, respectively, for the upper sand and c2′ = 0,
02' = 32o, and y sat 2 = 21, respectively, for the lower sand. Determine the following:
a. The variation in Rankine’s active pressure and pore water pressure with depth
b. The magnitude and location of the total active thrust per unit length of the wall
4.10 Redo problem P4.9 with the following parameters: H1 = 3.0 m, H2 = 3.0 m, c1′ = 0, 01' = 32o,
y sat1 = 18 kN/m3, c2′ = 0, 02' = 34o, and y sat 2 = 22 kN/m3.
4.11 Figure P4.11 shows a wall in sand and clay with H1 = 3.0 m and H2 = 4.0 m. The
groundwater level is on the surface. The effective strength parameters and saturated
unit weight for the sand are c′ = 0, 0 ' = 32o, and y sat1 = 20 kN/m3, respectively. The total
stress undrained shear strength and saturated unit weight for the clay are su = 20 kN/m 2
and y sat 2 =16 kN/m3, respectively. Use Rankine’s earth pressure theory to compute the
following:
a. The variation in pore water pressure with depth
c. The magnitude and location of the total active thrust per unit length of the wall
Figure P4.9
Figure P4.11
4.12 Redo problem 4.11 with the following parameters: H1 = 4.0 m, H2 = 3.0 m. c′ = 0, 0 ' = 30o,
y sat1 = 18 kN/m3; su = 15 kN/m 2, and y sat 2 = 17 kN/m3.
4.13 Same as problem 4.5. Assuming the friction angle between the wall and sand δ , use Cou-
lomb’s earth pressure theory to compute the total active force (horizontal component) per
unit length of the wall and the location of the resultant for d = 0 ' / 2 and d = 20 ' / 3.
4.14 Calculate Coulomb’s Ka for d = 0, d = 0 ' / 3, d = 0 ' / 2, d = 20 ' / 3, and d = 0 ', assuming
0 ' = 33o. For convenience, you can use software tools such as a spreadsheet to perform the
calculation. Compare the results with Rankine’s Ka and comment on their difference.
4.15 Same as problem 4.3. Assuming the retained clay is in the passive state, use Rankine’s
earth pressure theory to compute the magnitude and location of the total passive force per
unit length of the wall against the wall.
4.16 Same as problem 4.4. Assuming the retained clay is in the passive state, use Rankine’s
earth pressure theory to compute the magnitude and location of the total passive force per
unit length of the wall against the wall.
4.17 Same as problem 4.5. Assuming the retained sand is in the passive state, use Rankine’s
earth pressure theory to compute the magnitude and location of the total passive force
(horizontal component) per unit length of the wall against the wall.
4.18 Same as problem 4.6. Assuming the retained sand is in the passive state, use Rankine’s
earth pressure theory to compute the magnitude and location of the total passive force
(horizontal component) per unit length of the wall against the wall.
4.19 Redo problem 4.17. Use Coulomb’s earth pressure theory for d = 0 ' / 2 and d = 20 ' / 3.
4.20 Calculate Coulomb’s and Caquot‒Kerisel’s Kp for d = 0, d = 0 ' / 3 d = 0 ' / 2, d = 20 ' / 3,
and d = 0 ', assuming 0 ' = 30o and 0 ' = 35o. For convenience, you can use software tools
such as a spreadsheet to perform the calculation. Compare the results with Rankine’s Kp
and comment on their difference.
4.21 In problem 4.1, if there is a surcharge of 30 kN/m 2 on the ground surface level, what would
be the total lateral force per unit of the wall?
4.22 In problem 4.3, if there is a surcharge of 20 kN/m 2 on the ground surface level, what would
be the total lateral force per unit of the wall?
4.23 In problem 4.15, if there is a surcharge of 40 kN/m 2 on the ground surface level, what
would be the total horizontal force per unit of the wall?
References
Alpan, I. (1967), The empirical evaluation of the coefficient K0 and K0R. Soils and Foundations, Vol. VII,
No. 1, pp. 31–40.
Bowles, J.E. (1988), Foundation Analysis and Design, 4th edition, New York: McGraw-Hill Book
Company.
Caquot, A. and Kerisel, J. (1948), Tables for the Calculation of Passive Pressure, Active Pressure, and
Bearing Capacity of Foundations, Paris: Gauthier-Villars.
Coulomb, C.A. (1776), Essai sur une application des regles de maximis et minimis a quelques problemes
de statique, relatifs a l’architecture. Memoires Royale des Sciences, Paris, Vol. 3, p. 38.
Jaky, J. (1944), The coefficient of earth pressure at rest. Journal of the Society of Hungarian Architects and
Engineers (in Hungarian), Vol. 8, No. 22, pp. 355–358.
James, R.G. and Bransby, P.L. (1970), Experimental and theoretical investigation of a passive earth pres-
sure problem. Geotechnique, Vol. 20, No. 1, pp. 17–37.
Ladd, C.C., Foote, R., Ishihara, K., Schlosser, F. and Poulous, H.G. (1977), Stress-deformation and
strength characteristics, state-of-the-ART report. In Proceedings of the Ninth International Conference
on Soil Mechanics and Foundation Engineering, Tokyo, Vol. 2, pp. 421–494.
Mackey, R.D. and Kirk, D.P. (1967), At rest, active and passive earth pressures. In Proceedings of South-
east Asia Regional Conference on Soil Engineering, Bangkok, pp. 187–199.
Mononobe, N. (1929), Earthquake-proof construction of masonary dams. In Proceedings, World Engineer-
ing Conference, Vol. 9, pp. 274–280.
NAVFAC DM7.2 (1982), Foundations and Earth Structures, Design Manual 7.2, USA: Department of
the Navy.
Okabe, S. (1926), General theory of earth pressure. Journal of the Japanese Society of Civil Engineers,
Vol. 12, No. 1.
Ou, C.Y. (2022), Fundamentals of Deep Excavations, London: CRC Press, Taylor and Francis Group.
Peck, R.B. and Ireland, H.O. (1961), Full-scale lateral load test of a retaining wall foundation. Pro-
ceedings of 5th International Conference on Soil Mechanics and Foundation Engineering, Vol. 2,
pp. 453–458.
Rankine, W.M.J. (1857), On Stability on Loose Earth, London: Philosophic Transactions of Royal Society,
Part I, pp. 9–27.
Rehnman, S.E. and Broms, B.B. (1972), Lateral pressures on basement wall: Results from full-scale tests.
Proceedings of 5th European Conference on Soil Mechanics and Foundation Engineering, Vol. 1,
pp. 189–197.
Rowe, P.W. and Peaker, K. (1965), Passive earth pressure measurements. Geotechnique, Vol. 15, No. 1,
pp. 57–78.
Schmidt, B. (1967), Lateral Stresses in Uniaxial Strains, Bulletin, No. 23, Danish Geotechnical Institute,
pp. 5–12.
Seed, H.B. and Whitman, R.V. (1970), Design of earth retaining structures for dynamic loads. In ASCE
Specialty Conference, Lateral Stresses in the Ground and Design of Earth Retaining Structures, Cornell
University, Ithaca, New York, pp. 103–147.
USS Steel (1975), Sheet-Pile Design Manual. July, ADUSS.
Chapter 5
Earth retaining structures
5.1 Introduction
Earth retaining structures are constructed to stabilize an unstable soil mass by providing exter-
nal support or internal reinforcement to resist lateral earth pressure from the soil. Figure 5.1
illustrates the conditions in which the use of earth retaining structures should be consid-
ered for cohesionless slopes. When a slope is gentle, the slope inclination angle is less than
the angle of repose of the soil (i.e., the angle of friction): β < ϕ'. The mobilized soil shear
strength is adequate to maintain slope stability, and the soil mass is stable (Figure 5.1a). When
a slope is steep, β ≥ ϕ', the soil mass is unstable because the mobilized soil shear strength is
inadequate to maintain slope stability. Under this condition, the construction of earth retain-
ing structures is necessary to maintain slope stability (Figure 5.1b). In general, constructing
a retaining wall is more cost-intensive than forming a slope. Therefore, the need for a retain-
ing wall should be assessed carefully during preliminary design, and efforts should focus on
limiting the height of retaining structures to minimize construction costs and potential system
instability.
Earth retaining structures are typically used for grade separation, bridge abutments, slope
stabilization, and excavation support (Sabatini et al., 1997; Tanyu et al., 2008). Some applica-
tions include highway embankments, foundation walls, basement excavations, and cut slopes.
Earth retaining structures are also widely used for slope stabilization or restoration to prevent
and mitigate landslide hazards.
Figure 5.1 Conditions in which earth retaining structures should be considered for cohesion-
less slopes: (a) stable and (b) unstable conditions.
DOI: 10.1201/9781003350019-5
176 Earth retaining structures
This chapter is organized as follows. Various earth retaining structures commonly used in engi-
neering practice are introduced. These earth retaining structures are further classified based on
their construction methods and stabilization mechanisms. The advantages and disadvantages of
earth retaining structures are compared, and the wall selection flowchart and factors that govern the
selection and use of various earth retaining structures are discussed. Subsequently, the design and
analysis of three types of earth retaining structures, namely, gravity and semi-gravity walls, rein-
forced walls, and nongravity walls, are discussed. The typical dimensions, drainage systems, failure
modes, stability analyses, and improvement measures for each wall type are discussed in detail.
5.2 Types of earth retaining structures

Numerous earth retaining structures have been developed and used in engineering practice. Despite
the many options of earth retaining structures, often, only a few types of structures are the most
efficient or economical for a certain project. Therefore, selecting the most suitable earth retaining
structure for a specific case is important. Because retaining wall systems are diverse and have many
benefits and limitations, this section provides information for readers to help them (1) identify the
many retaining wall types on the basis of their construction methods and stabilization mechanisms;
(2) understand the applications, limitations, advantages, and disadvantages of different earth retain-
ing systems; and (3) select the most technically appropriate and cost-effective earth retaining struc-
ture for an application based on comprehensive knowledge of available wall systems.
5.2.1 Classification
A classification system for earth retaining structures is presented in Figure 5.2. According
to Sabatini et al. (1997), earth retaining structure systems are classified by their construction
Figure 5.2 Classification of earth retaining structures.

Earth retaining structures 177
methods (i.e., fill or cut construction) and stabilization mechanisms (i.e., external or internal
stabilization). Accordingly, each earth retaining structure in Figure 5.2 is classified into four
phases. For instance, a gravity wall is classified as an externally stabilized fill wall, whereas a
soil nail wall is classified as an internally stabilized cut wall.
Figure 5.3 illustrates the construction of a fill wall, which involves a sequence of backfill-
ing and compacting soil from the base to the top of the wall. Figure 5.4 presents the con-
struction of a cut wall, which often involves excavating and stabilizing soil from the top to
the base of the wall. Fill and cut walls typically cannot be used interchangeably because of
differences in their construction methods and sequence. For example, a gravity wall cannot
be used for a deep excavation project, and constructing a fill wall is infeasible for a cut-type
project. However, the classification may have exceptions; for example, the sheet-pile wall
for the waterfront structure can sometimes be constructed as a fill wall by backfilling the soil
behind the wall.
Figure 5.5 presents the load transfer mechanism of an externally stabilized wall. The sta-
bility of externally stabilized walls is provided by the external resistance through (1) base
shear friction induced by the weight of gravity walls or (2) passive earth pressure acting on
the embedded wall length and the additional lateral load from supports for nongravity walls.
Figure 5.6 presents the load transfer mechanism of an internally stabilized wall. The stability of
internally stabilized walls is provided by internal resistance through the reinforcement tensile
force mobilized by the soil–reinforcement interaction. Reinforcements should have sufficient
embedment length, often extending beyond the potential failure surface, to prevent walls from
Figure 5.3 Fill wall construction: (a) retaining wall construction, (b) soil backfilling and com-
pacting, and (c) repeated backfilling and compacting until the desired height is reached.
(a) (b ) (c)
Figure 5.4 Cut wall construction: (a) original slope profile, (b) soil excavation and stabilization,
and (c) repeated excavation and stabilization until the desired depth is reached.
Figure 5.5 Externally stabilized mechanism: (a) gravity walls; (2) nongravity walls.
Figure 5.6 Internally stabilized mechanism.
losing anchorage efficiency and shifting with the soil failure mass (i.e., pullout failure). Because
externally and internally stabilized walls have different load transfer mechanisms, their design
methods are fundamentally different.
Figures 5.7–5.11 present illustrations and photographs of various types of fill walls. The
typical dimensions and primary components of each fill wall are indicated in the figures.
Gravity-type walls include gravity walls and semi-gravity cantilever walls. Prefabricated
modular gravity walls include crib or bin walls and gabion walls. Gravity walls and pre-
fabricated modular gravity walls rely on their own weight to resist overturning and slid-
ing due to lateral pressure from retained soil behind the wall. Semi-gravity walls consist of
steel-reinforced concrete wall stems and base slabs to form an inverted “T” shape. Triangular
buttresses are sometimes installed at regular intervals along the length of a wall. The main
purpose of both measures is to enhance wall stiffness and, hence, reduce wall deformation.
Figure 5.7 Gravity wall.
Figure 5.8 Semi-gravity cantilever wall.
Figure 5.9 Crib/bin wall.

Figure 5.10 Gabion wall.
Figure 5.11 Reinforced wall.
Semi-gravity walls rely on their own weight plus the soil weight above the base slab to resist
overturning and sliding caused by lateral earth pressure. Reinforced walls, generally known
as mechanically stabilized earth (MSE) walls, employ either metallic reinforcements or geo-
synthetics (polymer reinforcements) in the backfill soil. Walls reinforced with geosynthetics
are called geosynthetic-reinforced soil (GRS) walls. The lateral earth pressure within the rein-
forced wall is resisted internally by the reinforcement tensile force mobilized by the soil–rein-
forcement interaction. Reinforced walls are relatively flexible and sustain large deformations
without considerable structural distress.
Fill walls typically require the use of granular, free-draining soil as backfill to prevent the
accumulation of water pressure. Most fill walls are used for permanent applications because
they are cost-intensive, and most of the wall components are nonreusable. Few MSE and GRS
walls without permanent facings can be used for temporary applications. Compared with per-
manent walls, walls used for temporary applications typically have less-restrictive requirements
on material durability, factors of safety (Fs), performance, and overall appearance. The service
life of walls for temporary applications is based on how long they need to support the soil
before permanent walls are installed. Temporary systems are commonly used for 18–36 months,
depending on actual project conditions. Table 5.1 summarizes the cost-effective height ranges,
right-of-way (ROW) requirements, advantages, and disadvantages of fill wall systems. The
ROW is defined as the easement (or space) required to construct the retaining wall.
Figures 5.12–5.18 present illustrations and photographs of various types of cut walls. The
primary components for each cut wall are indicated in these figures. Nongravity walls include
sheet-pile walls, soldier pile and lagging walls, diaphragm walls, and tangent and secant pile
walls. The resisting force of nongravity walls is derived mainly from passive earth pressure act-
ing on the embedment length of the wall at the excavation side and the additional lateral resist-
ance from several layers of support. Anchored walls derive resistance from ground anchors. One
end of the ground, anchor is fixed and prestressed at the anchor head at the wall face, and the
other end of the ground anchor is extended to a grouted zone beyond the potential failure sur-
face. The prestressed load transfers from the ground anchor to the wall face to provide substan-
tial lateral support. In situ reinforced walls include soil nail and micropile walls. For these walls,
reinforcements, typically steel bars or beams, are drilled or driven to a depth beyond the failure
Table 5.1 Summary of fill wall types
Types Temporary/ Cost-effective Required Advantages Disadvantages

permanent height range ROW
Gravity Permanent 1–4 m 0.5–0.7H • Easy design • Differential

wall • Design method settlement
is well-developed • Bearing capacity
Semi-gravity Permanent 2–9 m 0.5–1.0H • Easy design • Differential
cantilever • Design method settlement
wall is well-developed • Bearing capacity
Crib/bin Permanent 2–11 m 0.5–0.7H • Easy design and • Difficult for
wall fast construction compaction
• Design method • Difficult to adjust
is well-developed the geometry of
• High drainage the wall
capacity • Bearing capacity
Gabion wall Permanent 2–8 m 0.5–0.7H • Erosion control • Labor
• Ecology-friendly • Gabion wire
system backer corrosion
• Flexible system • Source of rock
• High drainage and stone
capacity
Reinforced Temporary/ 2–20 m 0.7–1.0H • Rapid • Granular soil for
wall permanent construction backfill
• Different types • Metallic
of facing reinforcement
• Flexible system corrosion
• High drainage • Reinforcement
capacity strength
reduction due to
creep, durability,
and installation
damage
Figure 5.12 Sheet-pile wall.
soldier pile
Figure 5.13 Soldier pile and lagging wall.
Figure 5.14 Diaphragm (slurry) wall.

Figure 5.15 Tangent/secant pile wall.
Figure 5.16 Soil anchor wall.
Figure 5.17 Soil nail wall.

Figure 5.18 Micropile wall.
surface and grouted to prevent long-term corrosion of the metal. Steel reinforcements with high
tensile strength prevent the development of a critical failure surface within the reinforced zone,
thereby increasing the overall stability of the retaining wall system.
Nongravity walls typically must be watertight to prevent groundwater leakage through the
wall. For some less-impermeable wall systems, such as sheet-pile walls, soldier pile and lag-
ging walls, and tangent pile walls, a waterproofing layer must be installed behind the wall.
In contrast to nongravity walls, soil anchor and soil nail walls must install drainage pipes to
guide the groundwater out. If the metallic reinforcing elements of cut walls are submerged
in groundwater, long-term corrosion protection measures are crucial to prevent the loss of
tensile strength or even breakage. Some cut walls, such as sheet-pile walls, soldier pile and
lagging walls, and soil nail walls, can be used for temporary support during construction.
After excavation is complete, a permanent structure or facing is placed in front of these tem-
porary walls. Soil anchor walls, soil nail walls, and micropile walls often require consider-
able underground space (or ROW) to enable the reinforcing elements to extend beyond the
potential failure surface to achieve adequate anchorage. The feasibility of using these walls
is influenced by the presence of buried utility lines and current foundations nearby and the
cost of permanent underground easement for the placement of reinforcing elements. Table 5.2
summarizes the cost-effective height ranges, required ROWs, advantages, and disadvantages
of cut wall systems.
5.2.2 Wall selection

The selection of a retaining wall for a certain construction project is complex because it
involves matching several objectives of wall selection with the numerous retaining walls
available to determine the most appropriate wall type. The objectives of wall selection
include cost-effectiveness, practicality of construction, stability, watertightness, aesthetics
and consistency with the surroundings, and environmental friendliness. First, the design
engineer should be familiar with the applications, limitations, advantages, and disadvantages
of various earth retaining systems, as discussed in the previous section. A rigorous and sys-
tematic evaluation and selection process should be conducted to determine the most appro-
priate wall type. Figure 5.19 presents a flowchart of the wall selection process suggested by
Sabatini et al. (1997).
Table 5.2 Summary of cut wall types
Types Temporary/ Cost-effective Required Advantages Disadvantages

permanent height range ROW
Sheet-pile Temporary/ <5 m None • Can be • Noisy and

wall permanent used below vibrates
groundwater • Difficult to
table drive into hard,
• Design dense soil and
method is gravels
well-developed • Vibration-
• Rapid induced
construction settlement in
loose ground
Soldier pile Temporary/ <5 m None • Rapid • Difficult to
and lagging permanent construction maintain vertical
wall • Less piles are tolerances in
driven than a hard ground
sheet-pile wall • Waterproof
Diaphragm Permanent 6–25 m None • Rigid structure • Disposal
wall • Impermeable of slurry
• Suitable for all may follow
soils environmental
restrictions
Tangent/ Permanent 3–9 m/6 m None • Rigid and • Requires
secant pile above for impermeable specialized
wall wall with • Suitable for all equipment
anchors soils • Significant spoil
or struts • Can control disposal
wall stiffness
Soil anchor Permanent 5–20 m 0.6H + • Can resist • Requires
wall anchor large skilled labor
bond horizontal and specialized
length pressures equipment
• Adaptable to • Requires
varying site permanent
conditions easement
• Anchors
may require
corrosion
protection
Soil nail wall Temporary/ 3–20 m 0.6–1.0H • Rapid • Nails may
permanent construction require
permanent
easements
• Difficult to
construct design
below water
table
Micropile Permanent N/A Varies • Rapid • Design
wall construction method is not
• Suitable for well-developed
slope stabilized
and building
protection
Figure 5.19 Wall selection flowchart.

Source: Redrawn from Sabatini et al. (1997).
The first step in wall selection is to identify the need for an earth retaining wall system. If the
construction site has restricted or congested space (i.e., limited ROW), then an earth retaining
system that uses abrupt changes in slope grades is warranted to stabilize the soil for projects. If
space restrictions are not a problem at the construction site, simply grading a slope can achieve
the same effect.
The second step is to identify project requirements and site constraints by performing a review
of the specific site. Items that must be identified during the site review include (1) subsurface
ground and groundwater conditions, (2) site accessibility and space restrictions, (3) locations
of buried utilities and current foundations, (4) aesthetic requirements, and (5) environmental
concerns. These items should be weighted by their priority during wall selection. On the basis
of the site review results, several wall systems may be eliminated from consideration, and others
may be recommended for further consideration.
In the third step, the remaining candidate wall systems are evaluated individually in detail
against the priority items identified in the previous step. Specific factors for wall system evalu-
ation include (1) wall geometry, (2) performance criteria, (3) construction consideration, (4),
aesthetics, (5) environmental concerns, and (6) cost. Each factor is evaluated on the basis of its
relevance and importance for each candidate wall system. The wall geometry factor is used to
evaluate whether the cost-effective wall height and width of each wall candidate fulfill the pro-
ject requirements and site ROW restrictions. The performance criteria factor ensures that the per-
formance of the wall candidate (e.g., allowable differential settlement for a fill wall or allowable
lateral movement for a cut wall) meets project requirements and legal regulations. The construc-
tion consideration factor is used to evaluate constructability, site accessibility, the availability
of construction material and equipment, and on-site wall material storage, temporary dewater-
ing requirements, construction speed, local contracting practices, and labor considerations. The
aesthetic factor is used to ensure that the permanent retaining wall is aesthetically pleasing and
consistent with its surroundings, particularly the natural environment. The environmental fac-
tor is used to assess the potential environmental impact of the candidate wall during and after
construction, including the excavation and disposal of soil, use of construction techniques and
materials that are environmentally detrimental or cause high CO2 emissions, the discharge of
large quantities of water or slurry fluids, and effects of construction noise, vibration, and dust.
Regarding the cost factor, in addition to material and construction costs, long-term maintenance
expenses should be considered in determining the total cost of a retaining wall. If a retaining
wall is used for slope stabilization to mitigate landslide hazards, the effectiveness of the wall
type on protecting different scales of landslides, as suggested by Liao (2008), should also be
considered during wall system evaluation (Figure 5.20).
The final step is to select acceptable wall systems. This selection is typically achieved by ana-
lyzing the evaluation results in the third step in a wall selection matrix, and the wall (or walls)
with the highest score is selected for the project (Tanyu et al., 2008). However, an introduction
to the wall selection matrix is beyond the scope of this chapter. Based on the comparison results,
the wall with the highest score is selected as the most appropriate wall system, and other high-
scoring walls can be recommended as acceptable alternatives.
Mid- to deep depth
Figure 5.20 Effectiveness of different retaining walls for protecting against different scales of
landslides.
Source: Modified from Liao (2008).
Example 5.1
Identify the type and classification of the retaining wall in the figure.
1 Which type of retaining wall is pictured?

(a) soil anchor wall, (b) gravity wall, (c) soil nail wall
2 How would the retaining wall be classified?
(a) internally stabilized cut wall, (b) externally stabilized cut wall, (c) internally stabi-
lized fill wall
Figure EX 5.1 Photo of a retaining wall.
Solution
ANS: 1: (a); 2: (b)
The figure depicts a soil anchor wall because it features anchor heads and an RC grid. Com-
pared with a soil nail wall, the anchor heads of a soil anchor wall are much larger. In addition,
soil nail walls typically have shotcrete facings.
The soil anchor wall is classified as an externally stabilized cut wall. A slope is stabilized by
the tensile force of the soil anchor. The tensile force is mobilized at the fixed end of the soil
anchor and transferred to the anchor head through the anchor tendons. This force acting on the
slope face suppresses slope displacement, therefore improving slope stability.
Example 5.2
A 13 m excavation is supported by a retaining wall at a construction site in an urban area. A site
investigation reveals that the subsurface soil consists of mainly NC clay with an SPT-N value
of 2–7. The water table is 2 m below the ground surface. According to the ground conditions
in the question statement, which type of retaining wall is suitable for this construction project?
(a) sheet-pile wall, (b) diaphragm wall, (c) tangent pile wall, (d) gravity wall
Solution
ANS: (b) and (c)
In this case, a cut wall should be selected because an excavation is needed. Thus, a gravity
wall, which is classified as a fill wall, is unsuitable.
Because the construction of a sheet-pile wall could cause noise and vibration, it is also unsuit-
able, especially for construction in an urban area. Furthermore, in deep excavation, a sheet-pile
wall, depending on its stiffness, may experience large displacement in the soft clay layer. Large
ground deformation would affect adjacent buildings and might cause damage.
A diaphragm wall and tangent pile wall are most suitable for this construction project. Com-
pared with the sheet-pile wall, these two retaining walls cause less noise and vibration during
construction. In addition, these walls have high stiffness, producing relatively small lateral wall
displacement when the subsurface soil is excavated.
Example 5.3
A typhoon and heavy rainfall caused many roads to collapse in a mountainous area. The roads were
closed because of slope failures. A project was conducted to reconstruct retaining walls by using
the nearby colluvium soil. The colluvium soil consists of mainly sand from weathered sandstone.
According to the construction specifications, the relative compaction of backfill soil is regulated
to be at least 90% to ensure the stability of the retaining wall. On the basis of these conditions,
which type of retaining wall is suitable for this reconstruction project in the mountainous area? (a)
tangent pile wall, (b) reinforced wall, (c) crib wall, (d) gravity wall
Solution
ANS: (b) and (d)
In this case, a fill wall is needed. Therefore, a tangent pile wall, which is classified as a cut
wall, is unsuitable. A crib wall is constructed using precast components that must be transported
by trucks from a factory. In this scenario, the road collapsed in many places; thus, transporta-
tion might be difficult for large trucks. This situation makes a crib wall unsuitable. Therefore,
reinforced and gravity walls are the most suitable for the situation. In addition, the original slope
was damaged by a typhoon and heavy rain. An improved drainage system should be planned and
implemented in the reconstruction project to prevent the accumulation of water pressure behind
the retaining walls, which could cause further retaining failure.
5.3 Gravity and semi-gravity walls

This section introduces the design of gravity and semi-gravity cantilever walls. The design
methods and performance criteria of these walls are well-developed. This section also presents
the procedures of stability analyses of wall stability against overturning, sliding, and bearing
capacity failures.
5.3.1 Typical dimensions and drainage design

Gravity and semi-gravity cantilever walls are the most common earth retaining structures for
various applications, including roadway embankments, bridge abutments, riverbank protection,
and slope stabilization. The most desirable advantage of gravity and semi-gravity cantilever
walls is their durability in many environments because they are made of mainly concrete.
Figures 5.21 and 5.22 illustrate the typical dimensions and drainage design of gravity and
semi-gravity walls, respectively. Engineers initially use these typical dimensions as trial wall
sections for stability analyses. If the stability analyses yield undesirable results, the wall sec-
tions can be adjusted until the calculated Fs satisfy the design requirements. The wall aspect
D H
L H
Figure 5.21 Typical dimensions and drainage layout of a gravity wall.
D
H
H
L H
Figure 5.22 Typical dimensions and drainage layout of a semi-gravity cantilever wall.
ratios (L/H) are typically 0.5–0.7 for gravity walls and 0.5–1.0 for semi-gravity walls. The wall
embedment depth D should be at least 0.6 m. The thickness at the top of the retaining wall
should be >0.3 m to provide adequate space for proper concrete placement.
A drainage system is considered a primary component of gravity and semi-gravity walls.
Because of rainfall infiltration and a high groundwater level, the backfill of a retaining wall
might become saturated with water, and the water pressure can accumulate and act on the retain-
ing wall as a driving force in addition to the lateral earth pressure. A properly constructed and
maintained drainage system minimizes the possibility of poor wall performance resulting from
water pressure accumulation. Typically, a layer of drainage aggregate with high permeability is
placed behind the wall. If any water enters the drainage aggregate layer, it can easily drain out
via the drainage pipe at the base. The outlet of the drainage pipe is connected to a stormwater
drainage system outside the retaining wall. Weep holes are installed at the stem of the retaining
wall to accelerate water drainage when the groundwater level is high. Weep holes should have
a minimum diameter of approximately 0.1 m and be adequately spaced. Filters (i.e., geotextile)
should be installed inside or behind the weep holes to prevent internal soil erosion by seepage,
which could cause the loss of backfill soil or clog the weep holes in the long term.
5.3.2 Failure modes

A gravity or semi-gravity cantilever wall can fail if (a) it overturns about its toe (Figure 5.23a), (b) it
slides along its base (Figure 5.23b), (c) the foundation soil has insufficient bearing capacity (Figure
5.23c), (d) global failure occurs (Figure 5.23d), or (e) excessive settlement or differential settlement
Figure 5.23 Failure modes: (a) overturning, (b) sliding, (c) bearing capacity, and (d) global failure.
occurs. Stability analyses to evaluate wall stability against overturning, sliding, and bearing capac-
ity failure are described, and improvement measures are suggested in the subsequent sections.
Global failure, also known as deep shear failure, can occur when a soil layer with low shear
strength is behind or under a wall (Figure 5.23d). A failure surface might develop along this
weak soil layer, eventually causing global instability. Global failure is often analyzed with slope
stability software by using limit equilibrium methods, which will be introduced in Chapter 8.
Global failure can be avoided by conducting a detailed site investigation to identify weak soil
layers and consider them in slope stability analysis. The principles of evaluating foundation set-
tlement are discussed in Section 3.5. In addition, foundation settlement can be evaluated using
advanced numerical approaches, such as the finite element method. If the calculated foundation
settlement is excessively large, an improvement measure, such as ground improvement or the
use of lightweight fill material (i.e., geofoam), should be implemented.
In seismic zones, walls can fail when subjected to seismic loadings during earthquakes. Con-
sequently, the influence of earthquakes on the stability of permanent retaining walls should be
considered. The dynamic earth pressure, as discussed in Section 4.8, should be used in stability
analyses. When seismic loadings are included, the Fs under seismic conditions are typically
required to be 0.75 times those under static conditions.
5.3.3 Force components for stability analyses

Figure 5.24 presents the force components of the stability analyses. The force components con-
sidered in the analyses include the weight of the soil above the heel (W1 and W2), the weight
of the retaining wall (W3 to W6), the active lateral earth force behind the wall (Pa), the passive
lateral earth force in front of the wall (Pp), and the interface shear resistance (S).
Figure 5.24 Force components of stability analyses.

To calculate the weight of the retaining wall, the polygon of the wall is divided into several
triangles and rectangles to calculate their areas and centroids easily. Pa is calculated using the
active lateral earth pressure, as expressed in eq. 5.1, and Pav and Pah are the vertical and horizon-
tal force components of Pa.
1
Pa = y 1 H '2 K a (5.1)
2
where:
γ1 = unit weight of backfill.

H’ = height of retaining wall plus height of sloped backfill (as indicated in Figure 5.24).
Ka = active earth pressure coefficient.
The base shear, S, is estimated using the Mohr–Coulomb theory to consider the interface shear
strength between the base of the wall and the foundation soil. S can be derived as follows:
On the basis of the Mohr–Coulomb failure criteria:
t f = ca' + - ' tan d (5.2)
where:
τf = interface shear strength.

σ' = normal stress.
δ and ca = interface friction angle and cohesion, respectively, between the wall and the founda-
tion soil.
The values of δ and ca can be obtained experimentally from the direct shear test on the
concrete–foundation soil interface. The values of δ and ca typically range from 1/2 to 2/3 of ϕ′
and c', respectively. Multiplying eq. 5.2 by the unit area of the wall base (B × 1) converts stress
to force as follows:
t f ( B x1) = Bca + B- ' tan d (5.3)
which yields:
S = Bca + V tan d (5.4)
where:
τf (B × 1) = S = interface shear resistance.

Bσ' = V = ΣW + Pav = sum of vertical force acting at the base of the wall.
Tables 5.3 and 5.4 list the equations for horizontal and vertical force components and their
perpendicular distances to the toe of the wall. The perpendicular distance to the toe of each force
component listed in the tables represents the moment arm measured from the toe of the wall.
Table 5.3 Summary of horizontal force components
Horizontal force Perpendicular distance to toe
1 LP
Pp = K py 2 D2 + 2c '2 K p D
2
1 H' h1 + h2 + t
Pah = Pa cos a = y1H'2 K a cos a =
2 3 3
S = Bc a + V tan d 0
Table 5.4 Summary of vertical force components
Vertical force Perpendicular distance to toe

1 b52 - (b6 - b5 )2 3
W1= y1A1= y1 ( b5 + b6 ) h2 x1 = b1 + b2 + b3 + b4 +
2 b6 +b5
1 b6
W2 = y1A2 = y1 b6 h1 x2 = B -
2 3
1 2b2
W3 = y c A3 = y c b2 h2 x3 = b1 +
2 3
b3
W4 = y c A4 = y c b3h2 x 4 = b1 + b2 +
2
1 b4
W5 = y c A5 = y c b4 h2 x 5 = b1 + b2 + b3 +
2 3
B
W6 = y c Ac = y c Bt x6 =
2
1
Pav = Pa sin a = y1H'2 K a sin a B
2
Note: γ 1 = unit weight of backfill; γ 2 = unit weight of foundation soil; γ c = unit weight of concrete.
These distance values are later used to calculate the resisting or driving moments in stability
analyses.
5.3.4 Evaluation for overturning

The factor of safety against overturning (Fo) is defined as:
Fo =
EM R
=
EW x + P
i i av B
(5.5)
EM D Pah x H ' / 3
where:
ΣMR = sum of the resisting moments with respect to the toe.

ΣMD = sum of the driving moments with respect to the toe.
The moment caused by Pp is disregarded in the calculation. The wall embedment depth is typ-
ically shallow; therefore, the developed passive earth pressure resistance may be nonsignificant
because little overburden pressure acts on the soil adjacent to the toe of the wall. In addition, the
effect of Pp is typically ignored due to the potential for the soil to be removed through natural or
manmade processes during the service life of the structure.
In general, Fo ≥ 2 is required for design. If the required Fo value is not achieved, extending the
wall base (i.e., increasing B) can increase it. Other improvement measures, as illustrated in Fig-
ure 5.25, can be implemented to increase system stability against overturning failure. A tier con-
figuration (Figure 5.25a) can effectively reduce the lateral earth pressure acting on the walls as
the offset distance between tier walls increases. When anchor or tieback systems (Figure 5.25b)
are used, the prestressed load of the reinforcing element can increase the resisting moments of
the wall system, thereby increasing the FS against overturning failure.
5.3.5 Evaluation for sliding

The factor of safety against sliding failure along the base of the wall (Fs) is defined as:
Fs =
EF H ,R
=
S
(5.6)
EF H ,D Pah
where:
ΣFH,R = sum of the horizontal resisting forces.

ΣFH,D = sum of the horizontal driving forces.
As presented in Figure 5.24, S and Pah are the only horizontal resisting and driving forces,
respectively. In the calculation, the resisting force from Pp is disregarded for the same reasons
described in Section 5.3.4.
Figure 5.25 Improvement measures against overturning failure: (a) construction in a tier con-
figuration and (b) use of anchor or tieback systems.
In general, Fs ≥ 1.5 is required for design. If the required Fs value is not achieved, extend-
ing the wall base (i.e., increasing B) can increase S at the wall base, thus increasing the Fs
value. Another effective improvement measure is implementing a shear key under the wall
(Figure 5.26).
If a shear key is designed, the resistance from the passive earth pressure in front of the wall
can be included in the calculation. The passive lateral force (Pp) is calculated as:
1 (5.7)
Pp = K p y 2 D '2 + 2c2' D ' K p
2
where:
Κp = passive earth pressure coefficient.

γ2 = unit weight of foundation soil.
c2′ = cohesion of foundation soil.
D’ = depth of the shear key.
The Fs for a wall with a shear key can be expressed as:
S + Pp
Fs = (5.8)
Pah
Figure 5.26 Shear key under the wall.

5.3.6 Evaluation for bearing capacity

The evaluation for bearing capacity failure uses the same calculation procedures for a shallow
foundation subject to eccentric loading discussed in Section 3.5. In a retaining wall, eccentric
loading at the wall base is caused by the lateral earth pressure, and the resultant force R acts on
the wall base with a distance e from the centerline (Figure 5.27). This e, called eccentricity, can
be evaluated using the following procedures:
The net moment Mnet with respect to the toe is:
M net = E M R - E M D (5.9)
where ΣMR and ΣMo are the sums of the resisting and driving moments about the toe, respec-
tively. The calculation of ΣMR and ΣMo is discussed in Section 5.3.4. The eccentricity e can be
expressed as:
B M net
e= - (5.10)
2 V
where Mnet/V is the distance between the toe and the acting point of R at the wall base.
Because of the effect of eccentric loading, the vertical pressure distribution along the base of
the wall is nonuniformly distributed (Figure 5.27). As the moment acts in a counterclockwise
direction about the wall base, the maximum vertical pressure qmax occurs at the toe of the wall,
centerline
Figure 5.27 Vertical pressure distribution at the base.

and the minimum vertical pressure qmin occurs at the heel of the wall. The vertical stress distri-
bution under the wall base can be determined using the following flexural formula for bending
stress from the mechanics of materials:
V M y
q= + net (5.11)
B x1 I
where:
y = perpendicular distance from any point to the neural plane of the wall base.
Ι = moment of inertia per unit area of the wall base.
For a rectangular wall base, I can be expressed as:
B3 x1
I= (5.12)
12
To calculate qmax at the toe and qmin at the heel, eqs. 5.10, 5.11, and 5.12 are combined, and y is
substituted with B/2.
V ( 6e )
qmax = |1 + | (5.13a)
B( B)
V ( 6e )
qmin = |1 - | (5.13b)
B( B)
When e > B/6, qmin changes from a positive to a negative value, meaning, that the vertical pres-
sure changes from compression to tension. Thus, tensile stress occurs at the heel of the wall,
meaning, that contact between the wall base and the underlying foundation soil is not tight. This
situation is undesirable because the effective width of the wall base is reduced and vertical stress
might become more concentrated. Accordingly, the following criterion for eccentricity should
be fulfilled in the design of a retaining wall:
B
e< (5.14)
6
If the required e value is not achieved, the wall should be redesigned by extending its base.
On the basis of eq. 3.22, which expresses the ultimate bearing capacity of an eccentrically
loaded foundation, the ultimate bearing capacity of a wall is:
1
qu = c2' N c Fcs Fcd Fci + q 'N q Fqs Fqd Fqi + y 2' B 'Ny Fy s Fy d Fy i (5.15)
2
where:
c2′ = cohesion of foundation soil.

q’ = effective overburden pressure at the level of the wall base.
B’ = effective width (= B – 2e).

Nc, Nq, Nγ = bearing capacity factors as defined in Chapter 3.
Fcs, Fqs, Fγs = shape factors.
Fcd, Fqd, Fγd = depth factors.
Fci, Fqi, Fγi = load inclination factors.
The equations for the shape, depth, and load inclination factors are listed in Table 3.4. Notably,
Fcs = Fqs = Fγs = 1 because the retaining wall is considered a continuous foundation. To calcu-
late Fcd, Fqd, and Fγd, the original wall width, B is used instead of the effective width B′. B is
not replaced with B′ in the depth factor equations because B′ is less than B and the use of B′ in
the denominator of the depth factor equations would cause large and unconservative values.
Because the wall base is subject to an inclined loading R (the resultant force of V and Pah in
Figure 5.27), when Fci, Fqi, and Fγi are calculated, the load inclination angle can be determined
as βi = tan−1(Pah/V) in degrees.
The factor of safety against bearing capacity failure (Fb) can then be evaluated as:
qu
Fb = (5.16)
qmax
or
qu - q '
Fb = (5.17)
qmax - q '
Eq. 5.16 is defined on the basis of the gross bearing capacity, and eq. 5.17 is defined on
the basis of the net bearing capacity. In general, Fb ≥ 3 is required for design. Because
gravity and semi-gravity walls have low tolerance to total and differential settlement, a
high Fb value is required to ensure that wall settlement presumably meets performance
criteria. However, there are exceptions to this assumption. Wide walls (i.e., those with
a large B) might fail because of substantial foundation settlement before bearing capac-
ity failure occurs. Foundation settlement should always be evaluated using the methods
discussed in Section 3.5. If the required Fb value is not achieved, improvement measures,
such as increasing the wall width (B), improving the shear strength of the foundation soil
(i.e., ground improvement), using lightweight fill material, or changing the foundation type
(e.g., using a pile foundation), should be implemented to increase the bearing capacity of
the foundation soil.
Example 5.4
A cross section of a semi-gravity cantilever retaining wall is presented in the figure. Evaluate the
stability of the retaining wall by using Coulomb’s lateral earth pressure method.
a Determine all the horizontal and vertical force components within the retaining wall and the
moment arm of all the force components from the wall toe (assume γc = 24 kN/m3 and δ =
2ϕ'/3).
b Calculate the factors of safety against overturning, sliding, and bearing capacity failure
(ignore the passive lateral earth pressure in front of the wall).
Figure EX 5.4a Cross section of a semi-gravity cantilever wall.
Figure EX 5.4b Active lateral earth force behind the wall (Coulomb’s method).
Solution
Figure EX 5.4b indicates the active lateral earth force behind the wall. In Coulomb’s method,
the active lateral earth force Pa is typically assumed to directly act on the back face of the wall
(gravity wall case) or on a vertical plane passing through the heel of the wall base (semi-gravity
wall case). In this exercise, because Pa acts on the soil along the vertical plane above the wall
heel, the soil–soil interface friction angle is considered (δ = ϕ'). The weight of concrete (W1, W2,
and W3) and the weight of the soil above the heel (W4 and W5) are also considered.
ANSWER TO PART A
Calculate the vertical force components:
W1 = y c A1 = 24x 3 = 72 kN/m
[1 ]
W2 = y c A2 = 24 x | x (0.7 - 0.5) x 6 | = 14.4 kN/m
[ 2 ]
W3 = y c A3 = 24x [ (0.7 + 0.7 + 2.6) x 0.7 ] = 67.2 kN/m
W4 = y 1 A4 = 18x (2.6 x 6) = 280.8 kN/m
1
W5 = y 1 A5 = 18x ( x 2.6x 0.458) = 10.72 kN/m
2
Calculate Coulomb’s active earth pressure:
B = 10o, 0 = 0o, 0 ' = 40o, a = 0 ' = 40o

cos 2 (0 ' - 0 )
Ka = 2
= 0.221
[ sin(d + 0 ') sin(0 ' - B ) ]
cos 0 cos(d + 0 ) |1 +
2
|
[ cos(d + 0 ) cos(0 - B ) ]
1
Pav = Pa sina = y 1 H '2 K a sina
2
1
= x18x (0.458+6+0.7)2 x 0.2214x sin40 n o = 65.63 kN/m
2
Calculate the horizontal force components:
1
Pah = Pa cosa = y1 H 2 K a cosa
2
1
= x18x (0.458+6+0.7)2 x 0.2214 x coss40o = 78.2 kN/m
2
S = ca B + Vtand = Kc2' B + VtanK 02'
2 2
= x 40x (0.7+0.7+2.6)+510.75 x tann( x 20o) = 227.72 kN/m
3 3
Table Ex 5.4a Vertical force components
Vertical force (kN/m) Perpendicular distance to toe (m) Moment (kN ∙ m/m)
1
W1 = 72 0.7 + 0.7 - 0.5 x =1.15 82.8
2
2
W2 =14.4 0.7 + ( 0.7 - 0.5 ) x = 0.833 12
3
1
W3 = 67.2 ( 0.7 + 0.7 + 2.6) x = 2 134.4
2
1
W4 = 280.8 0.7 + 0.7 + 2.6 x + 2.7 758.16
2
2
W5 =10.72 0.7 + 0.7 + 2.6 x = 3.13 33.55
3
Pav = 65.63 0.7 + 0.7 + 2.6 = 4 262.52
V = 510.75 EM R =1283.43
Table Ex 5.4b Horizontal force components
Horizontal force (kN/m) Perpendicular distance to toe (m) Moment (kN ∙ m/m)
Pah = 78.2 ( 0.458 + 6 + 0.7 ) 186.59
= 2.38
3
S = 227.72 0 0
ANSWER TO PART B
Calculate the factor of safety against overturning:
Fo =
EM R
=
1283.43
= 6.87 > 2 ok
EM D 186.59
Calculate the factor of safety against sliding:
Fs =
EF H ,R
=
227.72
= 2.91 > 1.5 ok
EF H ,D 78.2
Calculate the factor of safety against bearing capacity failure:
M net = E M R - E M D = 1283.43 - 186.59 = 1096.84 kN . m/m

M net B 1096.84 4 B
e= - = - = 0.14 m < = 0.67 ok
V 2 510.75 2 6
V [ 6e }
qmax = |1+ | = 154.5 kN/m
2
B x1 { B}
0 ' = 20o, N c = 14.83, N q = 6.4, N r = 2.87
F=
cs F=
qs Frs = 1
=Fcd 1=
.107, Fqd 1.054
Pah 78.2
B1 = tan -1 ( ) = tan -1 ( ) = 8.7o
V 510.75
2
9B ) ( bi )
2 2 2
9 8.7 ) ( 8.7 )
Fci = Fqi = 1- | i | = 1 - | | = 0.991, Fri = 1 - | ' | = 1 - | | = 0.81
( 90 ) ( 90 ) (0 ) ( 20 )
qu = 40 x 14.83x 1x 1.107 x 0.991 + 20x 1.5 x 6.4 x 1x 1.054 x 0.99
1
+ x 20 x (4 - 2 x 0.14) x 2.87 x 1x 1.054 x 0.81 = 942.45 kN/m 2
2
qu 942.45
Fb = = = 6.1 > 3 ok
qmax 154.5
5.4 Reinforced walls

This section introduces the design of reinforced walls, generally known as mechanically
stabilized earth (MSE) walls. Walls reinforced by geosynthetics are specifically called geo-
synthetic-reinforced soil (GRS) walls. Because reinforced walls are classified as internally
stabilized fill walls, the internal failure modes and soil–reinforcement interaction mechanisms
are discussed in this section. The steps of the stability analyses used to evaluate a wall’s
internal stability against reinforcement breakage and pullout and facing connection failure
are presented.
5.4.1 Typical dimensions and drainage design

Reinforced walls are widely used in various projects, including highway embankment and
bridge abutments, to improve load-bearing capacity and system stability. The use of rein-
forced walls is driven by numerous factors, including aesthetics, compatibility with the natu-
ral environment, cost-effectiveness, seismic performance, easy and rapid construction, and
the ability to tolerate large deformations without structural distress. In addition to general
applications of vertical loadings, reinforced walls have been used as barriers against lateral
impact forces from natural disasters, such as floods, tsunamis, rockfalls, debris flows, and
avalanches.
A reinforced wall consists of four elements: the reinforcement, the facing, the backfill, and
the drainage system (Figure 5.28). A reinforced wall has typical dimensions of L/H = 0.7–1.0,
where H is the wall height and L is the length of the reinforcement. The wall embedment depth
D should be at least 0.6 m. Reinforcements have two types: metallic reinforcement (e.g., steel
L H
Wrap-around
Figure 5.28 Typical dimensions and drainage layout of a reinforced wall: (a) concrete panel
facing and (b) wraparound facing.
strips, steel grids, and welded wire mesh) and geosynthetic (e.g., geogrids and geotextiles)
(Figure 5.29). Different reinforcement types have different extensibilities and cause different
earth pressure distributions within the wall, which will be discussed later. Facing elements have
various types, including segmental precast concrete panels, modular blocks, welded wire grids,
gabion facings, and geosynthetic wraparound facings. The type of wall facing affects wall
aesthetics and wall settlement tolerances and provides different levels of protection against
backfill sloughing and erosion. Figure 5.30 shows the facing connection. The metallic rein-
forcement is connected to the facing element by means of mechanical or frictional connections
(Figure 5.30a). The wraparound facing is formed by wrapping reinforcement around the soil
bags at the wall face (Figure 5.30b). The wall facing can be vegetated for natural appearance.
A reinforced wall typically requires well-graded granular soil as backfill for constructability,
durability, high drainage capacity, and favorable soil–reinforcement interaction. Table 5.5 lists
suggested backfill properties in the design guidelines (Berg et al., 2009; Elias et al., 2001). In
areas where granular backfill is not readily available, locally available soils containing certain
(a) (b)
Figure 5.29 Reinforcement types: (a) metallic reinforcement and (b) geosynthetics.
(a) (b)
Figure 5.30 Facing connection: (a) metallic reinforcement connected to concrete facing panel,
(b) geosynthetic wraparound facing.
Table 5.5 Backfill properties for reinforced walls suggested for design
Item Symbol Value
Soil classification USCS GW, GP, GM

SW, SP, SM
Maximum particle size D max (mm) ≤125 mm
Percent passing (%) 125 mm 100
4.75 mm (no. 4) 20–100
0.425 mm (no. 40) 0–60
0.075 mm (no. 200) 0–15
Plasticity index PI (%) ≤6
Relative compaction R c (%) 90–95
fines (typically referred to as marginal soils) have been used as alternative backfills to minimize
the construction costs and environmental impact of transporting granular soil to a construction
site. Fill materials outside the gradation and plasticity index requirements listed in Table 5.5
have been used successfully; however, problems, including significant distortion and structural
failure, have been observed because the low permeability of marginal soils can cause pore water
pressure to accumulate upon rainfall infiltration or seepage. Considerable care in drainage design
is required when marginal soils are adopted as backfills to prevent undesirable accumulation of
pore water pressure within the wall.
The drainage system is a crucial component of reinforced walls to dissipate the accumula-
tion of pore water pressure. To ensure proper long-term functionality, drainage layers beneath
and behind the reinforced and retained zone are strongly recommended. As illustrated in Fig-
ure 5.29, the back and base drains (also called chimney drains) are used to collect and guide
rainfall or groundwater out into a longitudinal drain at the toe of the wall. The top of the back
drain should be higher than the maximum groundwater level. An impermeable layer should
cover the top of the wall to prevent rainfall infiltration, and a drainage ditch near the wall crest
should be designed to collect surficial runoff.
5.4.2 Failure modes

A reinforced wall can fail because of (a) internal failure modes (Figure 5.31a–c), (b) exter-
nal failure modes (Figure 5.31d–f), (c) global failure modes (Figure 5.31g), or (d) excessive
wall deformation. A wall can fail internally when the reinforcement breaks, is pulled out, or
is disconnected from the facing elements. Stability analyses for wall internal stability against
reinforcement breakage, reinforcement pullout, and facing connection failure are described, and
corresponding improvement measures are suggested in the subsequent sections.
External failure modes include overturning, sliding, and bearing capacity failures (Figure
5.31d–f). Stability analyses for the external stability of a wall consider the reinforced soil mass
to be a rigid body, and active earth pressure from retained soils acts laterally on the reinforced
wall. The calculation procedures are the same as those for gravity and semi-gravity walls in Sec-
tion 5.3.4. Notably, the flexibility of reinforced walls makes overturning failure highly unlikely.
However, overturning criteria aid in lateral deformation control by limiting tilting and should
still be satisfied. Moreover, because reinforced walls can tolerate large deformations without
structural distress, the FS against bearing capacity failure is 2.5 for reinforced walls rather than
the 3.0 used for gravity and semi-gravity walls.
Figure 5.31 Failure modes: (a) reinforcement breakage, (b) reinforcement pullout, (c) facing
disconnection, (d) overturning, (e) sliding, (f ) bearing capacity failure, and (g)
global failure.
Consistent with gravity and semi-gravity walls, global failure (Figure 5.31g) is often analyzed
on the basis of limit equilibrium methods (introduced in Chapter 8) by using slope stability soft-
ware. For excessive wall deformation, an evaluation of the anticipated wall deformation with
respect to horizontal and vertical displacement should be performed. Vertical wall displacement
is obtained through conventional settlement computations with particular emphasis on differ-
ential settlements longitudinally along the wall face and transversely from the face to the end
of the reinforced soil volume. Horizontal wall displacement, especially wall-facing deflection,
is obtained using the empirical correlation between the reinforcement type and the wall aspect
ratio L/H. Advanced numerical approaches, such as the finite element method, can also be used.
Finally, for walls located in seismic zones, the dynamic earth pressure, as discussed in Section
4.8, should be included in stability analyses.
5.4.3 Force components for stability analyses

Figure 5.32 illustrates the force components within a reinforced wall. The earth pressure σh
pushes the soil mass within the potential failure surface forward and causes wall deformation,
whereas the shear stress developed at the soil–reinforcement interface τsr beyond the potential
failure surface pulls the soil mass backward to prevent failure. These two forces acting in oppo-
site directions mobilize the reinforcement tensile force. The maximum reinforcement tensile
force Tmax is the peak of the reinforcement tensile force, where the reinforcement segment inter-
cepts the potential failure surface.
Figure 5.32 Force components within reinforced wall.
Tmax, which is related to the lateral earth pressure, is a crucial factor in internal stability anal-
yses of reinforced walls. For stability analyses of reinforcement breakage, engineers should
select reinforcement with a tensile strength greater than Tmax; otherwise, reinforcement rupture
can occur. For stability analyses of reinforcement pullout, the reinforcement pullout resistance
(i.e., anchorage strength) generated from τsr should be greater than Tmax to prevent the reinforce-
ment from being pulled out. For stability analyses of facing connection failure, the connection
strength between the facing and reinforcement should be larger than the mobilized reinforce-
ment force at the wall face. However, no simplified method has been developed to evaluate the
mobilized reinforcement force at the wall face. In conservative designs, the mobilized reinforce-
ment force at the wall face is often assumed to be Tmax.
Tmax at each reinforcement layer can be calculated on the basis of the horizontal force equilib-
rium at each reinforcement tributary area (Figure 5.33).
For planar reinforcement (geogrid, geotextile, and welded wire mesh):
Tmax = - h x Sv (kN/m) (5.18a)
For strip reinforcement (steel strip and steel grid):
Tmax = - h x Sv x S h (kN) (5.18b)
where:
σh = lateral earth pressure.

Sv = vertical spacing of reinforcement.
Sh = horizontal spacing of reinforcement.
Figure 5.33 Horizontal force equilibrium at a reinforcement layer.
The value and distribution of σh depend on the extensibility of the reinforcement. For a wall
reinforced with extensile reinforcements (i.e., geosynthetics), sufficient wall deformation can
develop for the soil to reach active failure. Therefore, the active earth pressure is considered
in the design. For a wall reinforced with inextensible reinforcements (i.e., metallic reinforce-
ments), wall deformation is restrained by the reinforcements. Therefore, active soil failure can-
not develop. In this case, the earth pressure between the at-rest and active conditions is assumed.
Figure 5.34 presents the normalized earth pressure coefficient for a wall with different types
of reinforcements (Berg et al., 2009; Elias et al., 2001). The figure was prepared through back
analysis of the lateral stress ratio by using available field data. Stresses in the reinforcements
were measured and normalized as a function of an active earth pressure coefficient Ka. On the
basis of Figure 5.34, σh is calculated as:
(K )
- h = | r | K a (y z + q ) (5.19)
( Ka )
where:
Kr/Ka = normalized earth pressure coefficient.

Kr = earth pressure coefficient of the reinforced wall.
Ka = active earth pressure coefficient.
γ = backfill unit weight.
z = depth from the wall top.
q = surcharge.
K K
Figure 5.34 Normalized earth pressure coefficient.
5.4.4 Evaluation for reinforcement breakage

Because reinforcements are buried in the soil, the tensile properties of reinforcements, especially
geosynthetics, can degrade because of factors such as creep, long-term aging, and mechanical
damage. Given the aforementioned factors, the long-term reinforcement tensile strength is cal-
culated as:
Tult
Tal = (5.20)
RFcr x RFD x RFID
where:
Tal = long-term reinforcement tensile strength.

Tult = ultimate reinforcement tensile strength (obtained experimentally from tensile test).
RFCR = reduction factor for creep (obtain from creep test; typically ranges from 1.5 to 5, depend-
ing on polymer type).
RFD = reduction factor for long-term durability (obtain from durability test; typically ranges
from 1.1 to 2, depending on microorganism, chemical, thermal, oxidation, hydrolysis, and
stress cracking).
RFID = reduction factor for installation damage (obtain from field installation damage test; typi-
cally ranges from 1.5 to 5, depending on the backfill gradation and product unit weight).
After the long-term reinforcement tensile strength is obtained, the factor of safety against
reinforcement breakage (Fbr) can be calculated as:
Tal
Fbr = (5.21)
Tmax
In general, an Fbr ≥ 1.5 is required for design. If the required Fbr value is not achieved, effec-
tive improvement measures include using a reinforcement with high tensile strength (increasing
Tult), reducing reinforcement spacing (reducing Sv or Sh), or using a backfill with high soil shear
strength (increasing c' and ϕ').
5.4.5 Evaluation for reinforcement pullout

The reinforcement pullout resistance Pr is generated from the shear stress that develops at the
soil–reinforcement interface τsr. As presented in Figure 5.35, Pr can be derived by integrating τsr
with respect to the reinforcement embedment length Le (the part of the reinforcement beyond the
potential failure surface). In practice, Pr is calculated simply by using the laws of friction (i.e.,
normal stress × coefficient of friction × surface area), expressed as follows:
Figure 5.35 Horizontal force equilibrium for calculating reinforcement pullout.

For planar reinforcement (geogrid, geotextile, and welded wire mesh):
Pr = C x F * x a x - v' x Le (kN/m) (5.22a)
For strip reinforcement (steel strip and steel grid):
Pr = C x F * x a x - v' x Le x b (kN) (5.22b)
where:
C = reinforcement effective unit perimeter (C = 2 for strips, grids, and sheets).

F* = pullout resistance (or friction-bearing-interaction) factor (obtained experimentally from
pullout test; tan2ϕ'/3 if no pullout test is performed).
α = correction factor to account for reinforcement extensibility (1 for metallic reinforcement,
0.8 for geogrid, 0.6 for geotextile).
- v' = effective vertical stress on reinforcement (kPa) (live loads, such as traffic load, are excluded
as resistance).
Le = reinforcement embedment length (m) (calculated from failure surface, Le ≥ 1.0 m).
b = width of strip reinforcement (m).
After the reinforcement pullout resistance is obtained, the factor of safety against reinforce-
ment pullout (Fpo) can be calculated as:
Pr
Fpo = (5.23)
Tmax
In general, Fpo ≥ 1.5 is required for design. If the required Fpo value is not achieved, effective
improvement measures include increasing the reinforcement length (increasing L), reducing the
reinforcement spacing (reducing Sv or Sh,), or using reinforcement with high pullout resistance
(increasing F*).
5.4.6 Evaluation for facing connection

Reinforcements can be connected to the facing element by mechanical and frictional means. For
example, geogrid reinforcements may be structurally connected to segmental precast panels by
using a bodkin joint or frictionally connected to modular blocks by reinforcement and facing
interface friction. As presented in Figure 5.36, facing connection failure occurs when the con-
nection strength between the facing and reinforcement is insufficient for a reinforced wall with a
concrete panel or modular block facing. Evaluations for facing connection failure are used only
for reinforced walls with concrete panels or modular block facings. Facing connection failure
is not a concern for reinforced walls with wraparound facing if the wraparound length is long
enough (typically ≥ 1.25 m, as indicated in Figure 5.28b).
The long-term facing connection strength, considering creep and long-term aging, is calcu-
lated as:
Tc
Talc = (5.24)
RFcr x RFD
Figure 5.36 Horizontal force equilibrium for calculating facing connection: (a) concrete panel
facing and (b) modular block facing.
where:
Talc = long-term facing connection strength.

Tc = ultimate facing connection strength (obtained experimentally from connection test).
RFCR = reduction factor for creep.
RFD = reduction factor for durability.
Notably, the environment at the connection may not be the same as the environment in the
backfill in the reinforced zone. Therefore, the reduction factor for the durability RFD may be
different than that used to calculate the long-term reinforcement tensile strength Tal. Moreo-
ver, installation damage is not considered (i.e., RFD = 1) in the calculation of Talc because
heavy trucks and compaction rollers are not typically driven on the facing element during
construction.
After the long-term facing connection strength is obtained, the factor of safety against facing
connection failure (Fc) can be calculated as:
Talc
Fc = (5.25)
Tmax
In general, an Fc ≥ 1.5 is required for design. If the required Fc value is not achieved, effec-
tive improvement measures include increasing the facing connection strength (increasing Talc),
reducing the reinforcement spacing (reducing Sv or Sh), or using backfill with high soil shear
strength (increasing c' and ϕ').
Example 5.5
A reinforced wall is planned to provide an access road in a mountainous area. A traffic surcharge
q of 15 kPa is considered in the design. A cross section of the reinforced wall is illustrated in the
figure. The wall dimensions, reinforcement layout, and material properties are also provided in
the figure. Use the reduction factors of RFcr = 2, RFD = 1.5, and RFID = 1.5 to calculate the long-
term reinforcement tensile strength.
a Calculate the factor of safety against reinforcement breakage and identify the most critical
layer.
b Calculate the factor of safety against reinforcement pullout and identify the most critical layer.
Figure EX 5.5 Reinforced wall cross section.
Solution
ANSWER TO PART A
Calculate the maximum reinforcement load at the top and bottom layers.
Top layer:
Kr
= 1 for geosyntheitcs (Figure 5.34)
Ka
(K ) 40
- h = | r | x K a x (y z + q ) = 1x tan 2 (45 - )(18x 0.25 + 15) = 4.24 kPa
K
( a) 2
Tmax = - h x Sv = 4.24 x 0.5 = 2.12 kN/m
Bottom layer:
(K ) 40
- h = | r | x K a x (y z + q ) = 1x tan 2 (45 - )(18 x 4.75 + 15) = 21.85 kP
Pa
( Ka ) 2
Tmax = - h x Sv = 21.8 x 0.5 = 10.93 kN/m
Calculate the long-term reinforcement tensile strength.
Tult 90
Tal = = = 20 kN/m
RFCR x RFD x RFID 2 x 1.5 x 1.5
Calculate the factor of safety against breakage.

Top layer:
Tal 20
Fbr = = = 9.43 > 1.5 ok
Tmax 2.12
Bottom layer:
Tal 20
Fbr = = = 1.83 > 1.5 ok
Tmax 10.93
Table EX 5.5a Results of F br for all reinforcement layers
No. of layer z (m) - v' (kPa) σ h (kPa) Tmax (kN/m) F br
10 (top layer) 0.25 4.5 4.24 2.12 9.43

9 0.75 13.5 6.20 3.10 6.45
8 1.25 22.5 8.15 4.08 4.90
7 1.75 31.5 10.11 5.06 3.95
6 2.25 40.5 12.07 6.03 3.32
5 2.75 49.5 14.03 7.01 2.85
4 3.25 58.5 15.98 7.99 2.50
3 3.75 67.5 17.94 8.97 2.23
2 4.25 76.5 19.90 9.95 2.01
1 (Bottom layer) 4.75 85.5 21.85 10.93 1.83
The bottom layer is the most critical for reinforcement breakage because the lateral earth
pressure increases linearly with depth.
ANSWER TO PART B
Calculate the pullout resistance at the top and bottom layers.

Top layer:
Pr = C x F * x a x - v' x Le = 2 x 0.56x 0.8 x 4.5 x 1.29 = 5.2 kN/m

where:
C=2
2 2
F* = tan 0 ' = tan 40 = 0.56
3 3
a = 0.8
- v' = y z = 18x 0.25 = 4.5 kPa
(H - z) (5 - 0.25)
Le = L - = 3.5 - = 1.29 m
0' 40
tan (45 + ) tan(45 + )
2 2
Bottom layer:
Pr = C x F * x a x - v' x Le = 2 x 0.56x 0.8 x 85.5 x 3.38 = 258.9 kN/m
where
- v' = y z = 18 x 4.75 = 85.5 kPa
(H = z) (5 = 4.75)
Le = L = = 3.5 = = 3.38 m
0 ' 40
tan(45 + ) tan(455 + )
2 2
Calculate the factor of safety against pullout.

Top layer:
Pr 5.2
Fpo = = = 2.45 > 1.5 ok
Tmax 2.12
Bottom layer:
Pr 258.9
Fpo = = = 23.69 > 1.5 ok
Tmax 10.93
Table EX 5.5b Results of F po for all reinforcement layers
No. of layer z (m) Le (m) - v' (kPa) P r (kN/m) Tmax (kN/m) F po
10 (top layer) 0.25 1.29 4.5 5.2 2.12 2.45

9 0.75 1.52 13.5 18.3 3.10 5.90
8 1.25 1.75 22.5 35.3 4.08 8.65
7 1.75 1.98 31.5 56.0 5.06 11.07
6 2.25 2.22 40.5 80.4 6.03 13.33
5 2.75 2.45 49.5 108.6 7.01 15.49
4 3.25 2.68 58.5 140.5 7.99 17.58
3 3.75 2.92 67.5 176.2 8.97 19.64
2 4.25 3.15 76.5 215.7 9.95 21.68
1 (bottom layer) 4.75 3.38 85.5 258.9 10.93 23.69
The top layer is the most critical for reinforcement pullout because the reinforcement embed-
ment length is the shortest and the overburden pressure is the lowest in this layer.
5.5 Nongravity walls

This section introduces the design of nongravity walls, which include sheet-pile walls, soldier
pile and lagging walls, diaphragm walls, and tangent and secant pile walls. These walls are con-
structed to stabilize the soil for excavation support. Nongravity walls typically have two types:
cantilevered and supported walls. Accordingly, these two types of nongravity walls have dif-
ferent earth pressure distributions and analytical methods. The stability analysis procedures for
evaluating wall stability against rotational failure and determining whether a wall has sufficient
embedment depth are presented in this section.
5.5.1 Types and applications

Nongravity walls are used for (a) cofferdams on land and in water for foundation construction;
(b) waterfront structures for building wharfs, quays, and piers; (c) temporary excavation sup-
port; (d) excavation support for cut-and-cover roadway tunnels, culverts, and utility trenches;
and (e) excavation support for building basements and foundations.
Figure 5.37 illustrates the types of nongravity walls. Depending on whether additional sup-
ports (e.g., strut, anchor, deadman anchor) are used, nongravity walls are typically categorized
into cantilevered and supported walls. Cantilevered walls have no support during excavation
(i.e., free unsupported conditions). The resistance is derived from the passive earth pressure on
the wall embedment length on the excavation side. Wall deformation depends primarily on the
stiffness of the structural elements. Because cantilevered walls have no additional support, they
typically require longer embedment depths below the excavation surface. Supported walls have
additional support from several layers of struts, anchors, or deadman anchors to minimize the
Figure 5.37 Nongravity wall types.

size (i.e., depth and thickness) of the wall and maintain wall deformation within certain required
limits. The ratio of the embedment length Hp to the excavated depth (or unembedded length) He
varies depending on subsurface soil conditions. In general, Hp/He should be equal to or greater
than 0.8 for sandy soils, 1.0 for stiff clays, and 1.2–1.4 for soft clays.
5.5.2 Failure modes

Figure 5.38 illustrates the failure modes of a nongravity wall. The first failure mode is rota-
tional failure, which occurs when a cantilevered wall has insufficient embedment length (Fig-
ure 5.38a). The wall can rotate forward considerably, causing the entire retaining wall system
to fail. The second and third failure modes involve rotational failure in the supported walls
(Figures 5.38b and c). The second failure mode occurs when the support resistance is insuf-
ficient, for example, insufficient anchor lengths or insufficient strut axial resistance. Supports
can lose their anchorage efficiency and move together when soil mass failure occurs. Conse-
quently, the wall rotates forward about the toe, and the wall system ultimately fails. Because
(a) (b)
(c) (d)
Figure 5.38 Failure modes: rotational failure for (a) a cantilevered wall, (b) a supported wall
with insufficient support length, (c) a supported wall with insufficient wall embed-
ment length, and (d) a supported wall with excessive bending deformation
the support is ineffective, wall deformation in the second failure mode is similar to that in the
first failure mode. In design, support length should be evaluated to ensure that the support is
sufficiently long, which often necessitates extending the support beyond the potential failure
surface. The third failure mode occurs when the wall has insufficient embedment length. This
failure occurs in the lower part of the wall. The wall rotates backward about the supported
point, and the wall toe is kicked out.
In general, the first three failure modes are caused by inequilibrium in the earth pressure on
both sides of the wall. The wall is then moved a large distance toward the excavation zone,
causing the soil to move so much that the entire excavation site collapses. This rotational failure
is also called base shear failure because it involves the generation of shear stress that reaches
the soil shear strength in most of the soil below the excavation bottom, causing considerable
soil displacement or heave and leading to the failure of the entire excavation system. A detailed
discussion of base shear failure is presented in Chapter 6.
The fourth failure mode is excessive wall deformation by bending (Figure 5.38d). Excessive
wall deformation occurs when a wall has low stiffness or when the space between support layers
is large. Excessive wall deformation must be analyzed with the finite element method or beam–
spring model (Ou, 2022). If the excavation site encounters groundwater, the water pressure may
cause sand boiling failure in sandy soils and upheaval failure in clayey soils. Stability analyses
for evaluating wall stability against rotational failure are described in the subsequent sections.
Other stability analyses for sand boiling failure, upheaval failure, and excessive wall deforma-
tion are discussed in detail in Chapter 6.
For walls located in seismic zones, the influence of earthquakes on the stability of perma-
nent retaining walls should be considered. The influence of earthquakes on temporary retaining
structures can be ignored. For example, if diaphragm walls serve as the permanent outer walls
of a basement, the influence of earthquakes should be considered accordingly. For sheet-pile
walls and soldier pile and lagging walls that serve as temporary retaining structures, the effects
of earthquakes can be disregarded.
5.5.3 Earth pressure distribution

Cantilevered and supported walls have different earth pressure distributions because the wall
embedment length and additional supports influence the magnitude and shape of wall deforma-
tion. Accordingly, two analytical methods have been developed, fixed earth and free earth sup-
port methods (Tanyu et al., 2008; Ou, 2022; Padfield and Mair, 1984), as illustrated in Figures
5.39 and 5.40 using excavations in sandy soils as examples.
The design of cantilevered walls is based on the fixed earth support method. Figure 5.39 illus-
trates the deformation and earth pressure distribution of a cantilevered wall using the fixed earth
support method, which assumes that the wall embedment length is sufficiently long and that a
fixed point (no wall displacement) below the excavation surface can be identified. The embed-
ded part may rotate about the fixed point (Figure 5.39a). Thus, at the soil limit state, the lateral
earth pressure near the fixed point on both sides of the retaining wall varies gradually between
active and passive conditions (Figure 5.39b). The simplified earth pressure distribution is used
for stability analyses (Figure 5.39c).
The design of supported walls is based on the free earth support method. Figure 5.40 illus-
trates the deformation and earth pressure distribution of a supported wall, where the anchor
behind the wall offers the anchorage force. The anchor force comes from the prestressed load
transfers from the ground anchor to the wall face to provide substantial lateral support. The
Figure 5.39 Cantilevered wall: (a) wall deformation, (b) real earth pressure distribution, and
(c) simplified earth pressure distribution.
analysis of the supported wall is based on the free earth support method, which assumes that the
wall embedment length is relatively short, resulting in insufficient embedment to prevent rota-
tion of the toe of the wall. The embedded part of the wall can move a certain distance before the
soil reaches the limit state (Figure 5.40a). Therefore, passive earth pressure on the excavation
side and active earth pressure on the other side can be assumed to act on the wall (Figure 5.40b).
Figure 5.40 Supported wall: (a) wall deformation and (b) earth pressure distribution.
The applications of these two analytical methods are not interchangeable. If the free earth
support method is used to design a cantilevered wall, external forces on the wall, which are only
passive and active forces without the loads from the supports, may not reach equilibrium. There-
fore, the free earth support method is not applicable to cantilevered walls. However, if the fixed
earth support method is used to design a supported wall, the design might be too conservative,
leading to a wall embedment depth that is too long to be economical.
In addition to earth pressure, nongravity walls generally must be watertight to prevent ground-
water leakage through the wall into the construction site. Therefore, the water pressure might
act on the wall if groundwater is encountered during excavation. This consideration should be
included in stability analyses of excavations in sandy soil with groundwater on both sides of the
wall. In clayey soils, even when groundwater is present, water pressure can be ignored because
it is considered implicitly in the undrained analyses.
5.5.4 Stability analyses for cantilevered walls

This section introduces a simplified calculation procedure for a cantilevered wall commonly
adopted in engineering designs to determine the Hp value to ensure that the wall has a sufficient
embedment depth. The design of cantilevered walls is based on the fixed earth support method.
The earth pressure distribution of the cantilevered wall is shown in Figure 5.39 and discussed
in the previous section. Figure 5.41 presents the force components and free-body diagram for
the stability analyses. The force components are assumed on the basis of the fixed earth support
method.
In Figure 5.41, Pau and Ppu are the resultant forces of the active and passive earth pressures,
respectively, above the fixed point O.
1
Pau = y ( H e + H o ) K ah (5.26)
2
2
Figure 5.41 Stability analyses for a cantilevered wall: (a) force components and (b) free-body
diagram.
1
Ppu = y H o 2 K ph (5.27)
2
where:

He = excavated depth (or unembedded length).
Ho = distance from the excavation surface to the fixed point O.
Kah = horizontal component of the active earth pressure coefficient.
Kph = horizontal component of the passive earth pressure coefficient.
La and Lp represent the distance from Pau and Ppu to point O, respectively.
1
La = ( H e + H o ) (5.28)
3
1
Lp = H o (5.29)
3
The moment with respect to point O is used to solve for Ho as follows:
1
Ppu x H o
M r Ppu x L p 3
FS = = = (5.30)
M d Pau x La P x 1 H + H
au ( e o)
3
In general, an FS ≥1.5 is required for design. Because the simplification of this analysis does
not rigorously satisfy all equilibrium conditions, the required embedment depth for design Hp
should be slightly (typically 20%) larger than Ho. Therefore:
H p = 1.2H o (5.31)
The assumption in eq. 5.31 should be evaluated by comparing the forces above and below
point O to ensure that the wall embedment length is sufficient. R in Figure 5.41 is the imbalance
force between Ppu and Pau above point O.
R = Ppu - Pau (5.32)
To maintain the equilibrium of the wall, the resistant force from Ppl and Pal below point O should
be equal to or larger than R.
Ppl - Pal > R (5.33)
where Ppl and Pal are the resultant forces of active and passive earth pressures, respectively,
below point O.
[ ( Ho + He ) + ( H p + He ) ]
Ppl = K phy | | ( H p - Ho ) (5.34)
|[ 2
]|
Ho + H p
Pal = K ahy
2
(H p - Ho ) (5.35)
When Ppl - Pal > R, the assumption in eq. 5.31 is valid, and the wall embedment length is suf-
ficient to maintain the equilibrium of the wall.
5.5.5 Stability analyses for supported walls

This section introduces a calculation procedure for a supported wall. The design of supported
walls is based on the free earth support method. The earth pressure distribution of the sup-
ported wall is shown in Figure 5.40 and discussed in the previous section. Figure 5.42 illus-
trates the force components for stability analyses based on the free earth support method. The
method introduced here is applied only to walls with one level of support. The design of mul-
tilevel supported walls (more than two levels) involves different considerations and is detailed
in Chapter 6.
In Figure 5.42, Pa and Pp are the resultant forces of the active and passive earth pressures on
the wall, respectively.
Pa = y ( H e + H p ) K ah
1 2
(5.36)
2
1 (5.37)
Pp = y H p2 K ph
2
Figure 5.42 Force components of a supported wall for stability analyses.
where Hp is wall embedment (penetration) length below the excavation surface, and the other
parameters were defined previously. La and Lp are the distances from Pa and Pp to point S,
respectively.
2
La = (He + H p ) - H s (5.38)
3
2
Lp = H p + (He - H s ) (5.39)
3
where Hs is the distance from the top of the wall to the supported point S.
To determine the wall embedment length, take the moment with respect to point S and then
solve for Hp. The moment induced by T, the load of the support per unit length, is not included
in eq. 5.40 because T acts directly on point S. Similar to cantilevered wall designs, an FS ≥ 1.5
is typically required in the design guidelines for supported walls.
[2 ]
Pp x | H p = ( H e - H s ) |
M Pp x L p [3 ]
FS = r = = (5.40)
M d Pa x La [2 ]
Pa x | ( H e = H p ) - H s |
[3 ]
To determine the load of the support, take the moment with respect to point O at the end of
the wall, and the same FS value is applied. The moment induced by the horizontal component of
T (i.e., Tcosα) is included in eq. 5.41.
1
Pp (LT - L p ) Pp x H p
Mr 3
FS = = = (5.41)
M d Pa ( LT - La ) - TLT P x 1 ( H + H ) - T cos a x ( H + H - H )
a e p e p s
3
where LT is the distance from point S to point O.

Then, solve for T:
Pp 1 Pp 1
Pa ( LT - La ) - ( LT - L p ) Pa x ( H e + H p ) - x H p
FS 3 FS 3
T= = (kN/m) (5.42)
LT cos a x ( H e + H p - H s )
Moreover, eq. 5.41 can be rearranged as:
M r Pp ( LT - L p )
M d = Pa ( LT - La ) - TLT = = (5.43)
FS FS
In the preceding equation, the FS is only applied to Pp (or Kp) to account for the uncertainty in
the soil parameter and the associated passive earth pressure.
To rigorously satisfy all the equilibrium conditions, T can also be evaluated based on the
horizontal force equilibrium. The FS is also applied on Pp for the same reasons as discussed
previously.
Pp
T cos a = Pa - (5.44)
FS
And then:
Pp
Pa -
FS
T= (kN/m) (5.45)
cos a
The load of the support can be obtained as follows:
FT = T x S h (kN) (5.46)
where Sh is the horizontal spacing of supports.

The calculated FT is the minimum required load of the support, that is, anchor force, for
maintaining the equilibrium of the retaining wall system. For design, the allowable load of the
support is determined by applying an FS to the calculated FT to account for uncertainties due to
the influence of the construction method, soil properties, groundwater, and design life on the
material of the support. This step is also taken to ensure that the designed supports perform
under working stress conditions.
FT
Fd = (5.47)
Fanchor
The FS value for design depends on the materials of the support. For a ground anchor, an
Fanchor ≥ 3 for a permanent anchor and an Fanchor ≥ 2 for a temporary anchor are generally
required for design. Importantly, the Fanchor applied to the support in eq. 5.47 and the FS
applied to maintain the stability of the wall system in eqs. 5.40 and 5.41 have different values
and meanings.
To determine the length of the support, the embedment length of the support should extend
beyond the potential failure surface to achieve adequate anchorage, as shown in Figure 5.43.
0'
L cos a > ( H e + H p - H s - L sin a ) tan(45 - ) (5.48)
2
Figure 5.43 Length of the support related to the potential failure surface.
Because the inclination of the support is usually gentle (10° ≤ α ≤ 20° in general), for sim-
plicity, Lsinα can be ignored in the calculation. The embedment length of the support can be
calculated as:
0'
(H e + H p - H s ) tan(45 -
2
)
(5.49)
L>
cos a
Finally, the total length of the support can then be determined as L plus the length of the fixed
part of the support.
Example 5.6
A cross section of a cantilevered wall is presented in the figure. Assume that the groundwater
level is very deep. Determine the embedment length of this wall by using the fixed earth support
method with a required FS of 1.5.
Figure EX 5.6 Cross section of a cantilevered wall.

Solution
Calculate Rankine’s active and passive earth pressure coefficients:
0' 30
K ah = tan 2 (45 - ) = tan 2 (45 - ) = 0.333
2 2
0 ' 30
K ph = tan 2 (45 + ) = tan 2 (45 + ) = 3
2 2
Calculate the lateral earth forces:
1 1
Pau = y ( H e + H o ) K ah = x 20 ( 4 + H o ) x 0.333
2 2
2 2
= 3.33H o 2 + 26.64H o + 53.28
1 1
Ppu = y H o 2 K ph = x 20 x H o 2 x 3 = 30H o 2
2 2
Determine the wall embedment length to satisfy the required FS:

Take the moment on point O.
1
Ppu x H o
Mr 3
FS = =
Md P x 1 H + H
au ( e o)
3
1
30H o 2 x H o
= 3 > 1.5
( 3.33H o
2
+ 26.64H o + 53.28 ) x
1
3
(4 + Ho )
H o ≥ 4.9 m
H p 1=
= .2 H o 5.88 m
Compare the forces above and below point O to ensure that the embedment length of the wall
is sufficient:
1 1
R = Ppu - Pau = K phy H o2 - K ahy ( H e + H o ) 2
2 2
1 1
= x 3 x 20 x 4.92 - x 0.333x 20 x (4 + 4.9) 2
2 2
= 456.5 kN/m
[ ( Ho + He ) + ( H p + He ) ] H + Hp
Ppl - Pal = K ph y | | ( H p - H o ) - K ah y o ( H p - Ho )
|[ 2 |] 2
[ ( 4.9 + 4 ) + (1.2 x 4.9 + 4 ) ] 4.9 + 1.2 x 4.9

= 3 x 20 | | (1.2 x 4.9 - 4.9 ) - 0.333 x 20 x x (1.2 x 4.9 - 4.9 )
[ 2 ] 2
= 516.9 kN/m
Ppl – Pal ≥ R ok
Example 5.7
A supported wall is built to support the soil in an excavation project. A cross section of the sup-
ported wall is presented in the figure. The total excavated depth is 5 m below the ground surface.
The subsurface in this construction site consists of two soil layers: the top fill layer that is 0–3 m
deep, and the underlying original soil layer below 3 m deep. Determine the embedment length
of this wall by using the free earth support method with a required FS of 1.5. Additionally, find
the anchor load for the anchor with a horizontal spacing of 3 m.
Figure EX 5.7a Cross section of a supported wall.
Figure EX 5.7b Earth pressure distribution on both sides of the wall.

Solution
Figure EX 5.7b shows the lateral earth pressure distribution acting on both sides of the wall.
Calculate Rankine’s active and passive earth pressure coefficients:
0' 40
K ah1 = tan 2 (45 - ) = tan 2 (45 - ) = 0.217
2 2
0' 20
K ah2 = tan (45 - ) = tan (45 - ) = 0.490
2 2
2 2
0' 20
K ph2 = tan (45 + ) = tan (45 + ) = 2.040
2 2
2 2
Calculate the lateral earth forces:
1 1
Pa1 = y 1h12 K ah1 = x 18 x 32 x 0.217 = 17.57 kN/m
2 2
( )
Pa 2 = y 1h1 K ah 2 - 2c2 K ah 2 x ( H p +2 )
= (18 x 3 x 0.49 - 2 x 10 x 0.49 ) x ( H +2 ) p
= 12.46H p + 24.92
Pa 3 =
1
2
( (
(y 1h1 + y 2 ( H p + 2)) K ah 2 - 2c2 K ah 2 - y 1h1 K ah2 - 2c2 K ah2 x ( H p +22 ) ))
( (
= (18x 3 + 20( H p + 2)) x 0.49 - 2 x10x 0.49 - 18 x 3 x 0.49 - 2 x 10 x 0.49 x ( H p +2 )
1
2
))
= 4.9 H p2 + 19.6H p +19.6
Pp1 = 2c2 K ph 2 H p
= 2 x 10x 2.04H p
= 28.57H p
Pp 2 =
1
2
(y 2 H p K ph 2 ) H p
= ( 20 x H p x 2.04 ) H p
1
2
= 20.4H p2
Determine the wall embedment length to satisfy the required FS:

Take the moment on point S and solve the equation to obtain Hp:
Mr
FS =
Md
Pp1 [[ 12 H p + ( H e - H s ) ]] + Pp 2 [[ 32 H p + ( H e - H s ) ]]
=
Pa1 ( 32 h1 - H s ) + Pa 2 [[ 12 ( H p + 2 ) + ( h1 - H s ) ]] + Pa 3 [[ 32 ( H p + 2 ) + ( h1 - H s ) ]]
Pp1 ( 12 H p + 3.2 ) + Pp 2 ( 32 H p + 3.2 )

=
Pa1 x 0.2 + Pa 2 ( 12 H p + 2.2 ) + Pa 3 ( 32 H p + 2.53)
Mr
FS =
Md
Pp1 [[ 12 H p + ( H e - H s ) ]] + Pp 2 [[ 32 H p + ( H e - H s ) ]]
= Earth retaining structures 231
Pa1 ( 32 h1 - H s ) + Pa 2 [[ 12 ( H p + 2 ) + ( h1 - H s ) ]] + Pa 3 [[ 32 ( H p + 2 ) + ( h1 - H s ) ]]
Pp1 ( 12 H p + 3.2 ) + Pp 2 ( 32 H p + 3.2 )

=
Pa1 x 0.2 + Pa 2 ( 12 H p + 2.2 ) + Pa 3 ( 32 H p + 2.53)
H p > 2.43 m = 2.5 m
Summary of the resultant forces of the active and passive earth pressures:
Pa1 = 17.57 kN/m

Pa 2 = 12.46H p + 24.92 = 56.07 kN/m
Pa 3 = 4.9 H p2 + 19.6H p +19.6 = 99.22 kN/m
Pa = Pa1 + Pa 2 + Pa 3 = 172.86 kN/m
=Pp1 28.
= 57 H p 71.42 kN/m
Pp 2 20.
= = 4 H p2 127.5 kN/m
Pp = Pp1 + Pp 2 = 198.92 kN/m
Determine the anchor load.

Take the moment on point O, and solve the equation to obtain T.
Mr
FS =
Md
1 1
Pp1 x H p + Pp 2 x H p
= 2 3
{1 } 1 1
Pa1 x | h1 + H p + 2 | + Pa 2 x ( H p + 2) + Pa 3 x ( H p + 2)-T cos a x ( H e + H p - H s )
{3 } 2 3
[ {1 } 1 1 ] { 1 1 }
| Pa1 x | 3 h1 + H p + 2 | + Pa 2 x 2 ( H p + 2) + Pa 3 x 3 ( H p + 2) | - | Pp1 x 2 H p + Pp 2 x 3 H p | /FS
[ { } ] { }
T=
cos a x ( H e + H p - H s )
=
[17.57 x 5.5 + 56.07 x 2.255 + 99.225 x1.5] - ( 71.425 x1.25 + 127.5 x 0.83) /FS
cos 20 x (5 + 2.5 - 1.8)
= 45.09 kN/m
Solve T again based on the horizontal force equilibrium:
Pp
T cos a = Pa -
FS
and then
Pp 198.92
Pa - 172.86 -
FS 1.5 = 42.83 kN/m
T= =
cos a cos 20
Notably, the T values calculated using both methods (moment and horizontal force equilibrium)
are close.
For the anchor load with an anchor horizontal spacing of 3 m:
FT = 45.09 x 3 = 135.18 kN

This chapter introduced various types of earth retaining structures commonly used in engineer-
ing. The designs of gravity, semi-gravity, reinforced, and nongravity walls were described. For
each wall type, the typical dimensions, failure modes, and stability analyses were discussed in
detail. A summary and general comments regarding this chapter are as follows:
1 Earth retaining systems are classified by construction method (i.e., fill or cut construction)
and by stabilization mechanism (i.e., external or internal stabilization). In general, fill and cut
walls cannot be used interchangeably because they involve different construction methods.
Because externally and internally stabilized walls have different load transfer mechanisms,
their design methods are fundamentally different. The applications, limitations (e.g., cost-
effective height and required ROW), advantages, and disadvantages of various earth retain-
ing systems are compared and summarized in Tables 5.1 and 5.2.
2 Selecting a retaining wall for a construction project can be complex because it involves
matching several objectives of wall selection with the numerous retaining walls available in
practice. The objectives of wall selection are cost-effectiveness, practicality of construction,
stability, watertightness, aesthetic appropriateness and consistency with the surroundings,
and environmental friendliness. To determine the most appropriate wall type, a rigorous and
systematic wall system evaluation and selection process should be conducted.
3 A drainage system, typically comprising a layer of drainage aggregate with high permeability
placed behind the wall and drainage pipes at the bottom, is considered a primary component of
gravity and semi-gravity walls. A properly constructed and maintained drainage system minimizes
the possibility of poor wall performance resulting from the accumulation of pore water pressure.
4 For the design of gravity and semi-gravity cantilever walls, stability analyses should be per-
formed to evaluate wall stability against overturning, sliding, bearing capacity, global failure,
and excessive settlement. In the calculation of the FS against sliding failure, the resisting force
from passive earth pressure Pp acting in front of the wall is disregarded if the shear key is not
included. In the calculation of the FS against bearing capacity failure, because of the influence
of lateral earth pressure, the eccentric loading at the wall base should be considered, and eccen-
tricity should be evaluated to ensure that it is less than one-sixth of the wall width (e < B/6).
5 For the design of reinforced walls, stability analyses should be performed to evaluate the
internal, external, and global stabilities as well as wall deformation. Internal stability is used
to assess reinforcement breakage, reinforcement pullout, and facing connection failure. The
earth pressure distributions are dependent on the type and extensibility of the reinforcement.
In the calculation of wall internal stability, active earth pressure is assumed for reinforced
walls with extensile geosynthetics, whereas earth pressure between the at-rest and active
conditions is assumed for reinforced walls with inextensible metallic reinforcements.
6 Nongravity walls are categorized into cantilevered and supported types, depending on
whether they have additional supports. Accordingly, the earth pressure distributions and
analytical methods for these two types of walls are different. The analysis of cantilevered
walls should use the fixed earth support method, which assumes that a fixed point on the
wall below the excavation surface can be identified. At the soil limit state, the lateral earth
pressure near the fixed point on both sides of the wall varies between active and passive con-
ditions. Analysis of supported walls should use the free earth support method, which assumes
that the wall rotates about its toe. Passive earth pressure on the excavation side and active
earth pressure on the other side can be assumed to act on the wall.
Problems
5.1 An earth embankment is planned for highway transportation. The geometry of the earth
embankment is shown in Figure P5.1. The height of the embankment is 5 m, and the width
at the top of the embankment is 8 m. The ROW for the embankment construction is 12 m.
The embankment is backfilled with sand, with ϕ' = 40° and c' = 0 kPa. Please determine
whether a retaining wall is needed to support the soil of the embankment.
Figure P 5.1
5.2 Identify and classify the type of the retaining wall in the figure. Which type of retaining
wall is pictured? How would the retaining wall be classified?
a Which type of retaining wall is pictured? (1) reinforced wall, (2) diaphragm wall, (3)
gabion wall
b How would the retaining wall be classified? (1) internally stabilized fill wall, (2) exter-
nally stabilized cut wall, (3) externally stabilized fill wall
Figure P 5.2
5.3 Identify the type and classification of the retaining wall in the figure. Which type of retain-
ing wall is pictured? How would the retaining wall be classified?
a Which type of retaining wall is pictured? (1) reinforced wall, (2) gravity wall, (3) crib
wall
b How would the retaining wall be classified? (1) internally stabilized fill wall, (2) exter-
nally stabilized cut wall, (3) externally stabilized fill wall
Figure P 5.3
5.4 A cross section of a gravity wall is shown in Figure P5.4. The unit weight of concrete is
γ = 24 kN/m3. Assume δ is 2ϕ'/3. Use Rankine’s earth pressure and consider the passive
resistance.
a Find the resultant of active earth pressure and passive earth pressure.
b Determine the FS against sliding and overturning.
c Check if the bearing capacity of the retaining wall is sufficient.
5.5 A cross section of a cantilever wall is shown in Figure P5.5. The sand is used as a backfill.
The unit weight of concrete is γ = 24 kN/m3. Assume δ is 2ϕ'/3. Answer the following
questions by using Rankine’s lateral earth pressure method and ignore the passive lateral
earth pressure in front of the wall.
a Determine all the horizontal and vertical force components within the retaining wall
and the moment arm of all the force components from the wall toe.
b Calculate the factor of safety against sliding and check whether this design meets the
specification requirements. If not, please propose an improvement measure.
c Calculate the factor of safety against overturning and check whether this design meets
the specification requirements. If not, please propose an improvement measure.
Figure P 5.4
Figure P 5.5
5.6 Redo problem 5.5 using Coulomb’s lateral earth pressure method.
5.7 Redo problem 5.5, considering a backslope behind the retaining wall with an inclination
angle of β = 10°, as shown in Figure P5.7. Use Coulomb’s method to calculate the lateral
earth pressure.
5.8 Consider a reinforced wall, as shown in Figure P5.8, given H = 8 m, γ = 16.8 kN/m3, ϕ' =
34°, L = 4.5 m, Sv = 0.5 m, and Tult = 150 kN/m. The sand is used as backfill. The geogrid is
Figure P 5.7
Figure P 5.8
used as reinforcement with reduction factors RFCR = 2, RFD = 1.5, and RFID = 1.5 obtained
from laboratory tests. The top reinforcement layer is installed 0.25 m below the top of the
wall.
a Calculate the factor of safety against reinforcement breakage and identify the most
critical layer.
b Calculate the factor of safety against reinforcement pullout and identify the most criti-
cal layer.
5.9 Redo problem 5.8, considering a reinforced wall backfilled with a soil with soil strength
properties of ϕ' = 31° and c' = 5 kPa.
a Calculate the factor of safety against reinforcement breakage, and identify the most
critical layer.
b Calculate the factor of safety against reinforcement pullout and identify the most criti-
cal layer.
5.10 Consider the design of a reinforced wall, given H = 6 m, γ = 15.7 kN/m3, ϕ' = 36°, and
Tult = 90 kN/m. The geotextile is used as reinforcement with reduction factors RFCR = 2.5,
RFD = 1.5, and RFID = 1.2 obtained from laboratory tests. The required Fs against rein-
forcement breakage and pullout are Fbr = 1.5 and Fpo = 1.5. Determine the optimal Sv and
L that satisfy the required Fs.
5.11 Consider the design of a reinforced wall, given H = 6 m, γ = 16.5 kN/m3, ϕ' = 31°, c' =
5 kPa, and Tult = 120 kN/m. The geogrid is used as reinforcement with reduction factors
RFCR = 2.5, RFD = 1.5, and RFID = 1.2 obtained from laboratory tests. The required Fs
against reinforcement breakage and pullout are Fbr = 1.5 and Fpo = 1.5. Determine the
optimal Sv and L that satisfy the required Fs.
5.12 Figure P5.12 shows a 4 m deep excavation with a cantilevered wall. The groundwater level
is very deep. The unit weight of the soil is γ = 20 kN/m3. The effective soil shear strength
properties are ϕ' = 25° and c' = 0 kPa. Determine the embedment length of this wall by
using the fixed earth support method with a required FS of 1.5. Adopt Rankine’s earth pres-
sure method.
5.13 A cross section of a cantilevered wall is presented in the figure. Assume that the ground-
water level is very deep. Determine the embedment length of this wall by using the fixed
earth support method with a required FS of 1.5.
5.14 A cross section of a cantilevered wall is presented in the figure. Determine the embed-
ment length of this wall by using the fixed earth support method with a required FS
of 1.5.
5.15 The groundwater table lies at a depth of 10 m below the ground surface and consists
entirely of sands. A 7 m deep excavation is to be constructed with anchored sheet piles to
support the excavation. The tie rod of the anchor inclined at an angle of 15° is placed 1 m
below the surface, with horizontal spacing of the tie rods being 3 m. The unit weights of
sand above and below the water table are γ1 = 17.0 kN/m3 and γ2 = 20.0 kN/m3. The sand
has ϕ' = 35°.
a Determine the required depth of the sheet pile.
b Calculate the force in the tie rods if they are placed 3 m apart (inclined at an angle
of 15°).
c Determine the length of the tie rods.
Figure P 5.12
Figure P 5.13
Figure P 5.14
Figure P 5.15
5.16 A cross section of the sheet pile is presented in the figure. The subsurface soil consists of
two soil layers: the depth of the top fill layer is 0–6 m, and the depth of the underlying
original clay layer is below 6 m.
a Determine the embedment length of this wall by using the free earth support method
with a required FS of 1.5.
b Find the anchor load for the anchor with a horizontal spacing of 3 m.
Figure P 5.16
References
Berg, R., Christopher, B.R. and Samtani, N. (2009), Design and Construction of Mechanically Stabilized
Earth Walls and Reinforced Soil Slopes, Vol. 1. FHWA-NHI-10-024, Federal Highway Administration,
Washington, DC, p. 306.
Elias, V., Christopher, B.R. and Berg, R. (2001), Mechanically Stabilized Earth Walls and Reinforced Soil Slopes
Design and Construction Guidelines, FHWA-NHI-00-043, Federal Highway Administration, Washington,
DC, p. 394.
Liao, R.T. (2008), Countermeasures and Case Studies of Landslide Disaster Prevention and Mitigation,
Taipei: Techbook Ltd, p. 234.
Ou, C.Y. (2022), Fundamentals of Deep Excavation, London: CRC Press, p. 478.
Padfield, C.J. and Mair, R.J. (1984), Design od Retaining Walls Embedded in Stiff Clay. CIRIA Report 104,
London: Construction Industry Research and Information Association, p. 146.
Sabatini, P.J., Elias, V.E., Schmertmann, G.R. and Bonaparte, R. (1997), Geotechnical Engineering Circular No.
2, Earth Retaining Systems, FHWA-SA-96-038, Federal Highway Administration, Washington, DC, p. 161.
Tanyu, B.F., Sabatini, P.J. and Berg, R. (2008), Earth Retaining Structures, FHWA-NHI-07-071, Federal
Highway Administration, Washington, DC, p. 764.
Chapter 6
Excavation
6.1 Introduction
The construction of building basements or foundations or subways involves excavating soil
down to several meters or even 40 m below the ground surface. To prevent the unexcavated soil
from collapsing toward the excavation zone, a supported retaining system is needed. Figure 6.1
shows various structural components of a typical supported retaining system, which includes
retaining walls, struts, wales, and center posts. As noted in Section 5.2, soldier piles with and
without laggings, sheet piles, column piles, and diaphragm walls can be used as retaining walls.
Figure 6.1 Components of a strutted retaining system.
DOI: 10.1201/9781003350019-6
242 Excavation
When the soil is excavated, an unbalanced force acting on the excavation bottom and retain-
ing wall is generated. The unbalanced force may cause an excavation collapse and sometimes
a large amount of lateral deflection of retaining walls and ground settlement. An analysis of
the excavation is necessary before the design. Excavation analysis is a typical soil–structure
interaction problem. Soil is also a nonlinear, inelastic, and anisotropic material. Theoreti-
cally, the analysis of deep excavation involves simulations of the elastoplastic behavior of
the soil, interface behavior between the soil and retaining walls, and the excavation process.
However, for practical purposes, excavation analysis must be simplified. In this chapter, exca-
vation analysis is simplified and divided into several individual subjects. The first important
subject is stability analysis to prevent the excavation from collapse. Other subjects, such as
deformation and stress analyses of retaining walls, wales, struts, and monitoring systems, are
also introduced. In addition, the center post, dewatering system, excavation procedure, and
deformation control should be analyzed or designed. However, considering the page limita-
tion of this book, these subjects are not introduced here, and interested readers are advised to
refer to Ou (2022).
6.2 Excavation methods

Excavation methods include the full open cut method, the braced excavation method, the island
excavation method, the anchored excavation method, and the top-down construction method.
The selection of an appropriate excavation method must consider many factors, such as the
construction budget, construction period, existence of adjacent excavations, availability of
construction equipment, area of construction site, conditions of adjacent buildings, foundation
types of adjacent buildings, and so on. Experienced engineers are able to make good selections
based on these factors. This section will introduce the two most commonly used methods: the
braced excavation and top-down construction methods. For other methods, interested readers
can refer to related references, such as Ou (2022).
6.2.1 Braced excavation

Installing struts in front of retaining walls to resist the earth pressure acting on the back of
walls is called the braced excavation method. Figure 6.2 shows a typical arrangement of struts.
Figure 6.3 shows a photo of a braced excavation. As shown in Figures 6.1 and 6.2, the braced
excavation method includes struts, wales, and center posts. The function of wales is to transfer
the earth pressure on the back of retaining walls to horizontal struts. Center posts support the
strut, so the strut does not fall off due to its own weight.
In deep excavations, construction is often carried out in stages. The following is the construc-
tion procedure for the braced excavation method:
1 Place center posts in the construction area.

2 Proceed to the first stage of excavation.
3 Install wales above the excavation surface, then install struts and preload them.
4 Repeat procedures 2 to 3 until the designed depth is reached.
5 Build the foundation of the building.
6 Demolish the struts above the foundation.
7 Construct floor slabs.
8 Repeat procedures 6 to 7 until the construction of the ground floor is completed.
Excavation 243
Figure 6.2 Braced excavation method: (a) profile and (b) plan.
244 Excavation
Figure 6.3 Photo of a braced excavation.
As excavation is carried out to the designed depth, after which raft foundations or founda-
tion slabs are built, struts can then be removed level by level, and floor slabs are built accord-
ingly. Thus, the underground construction is finished. As such, the underground structure is
constructed from the bottom to the top, and the braced excavation method is thus also called the
bottom-up construction method.
6.2.2 Top-down construction

The top-down construction method is to erect molds and construct floor slabs after each excava-
tion. The floor slabs are permanent structures, which replace temporary steel struts in the braced
excavation method to counteract the earth pressure acting on the back of the retaining wall. In
this way, the underground structure construction is finished upon completion of the excavation
process. The construction of the underground structure proceeds from top to bottom, the oppo-
site of conventional foundation construction methods. The method is thus called the top-down
construction method.
The floor slabs used in the top-down construction method are heavier than the steel struts
used in conventional excavation methods. In addition, the superstructure, which is constructed
simultaneously during excavation, places more weight on the column. Thus, the bearing capac-
ity of the column must be examined. Consequently, pile foundations are often chosen as build-
ing foundations in the top-down construction method. The typical construction procedure of the
top-down construction method is as follows (see Figure 6.4):
1 Construct the retaining wall.

2 Construct piles. Place the steel columns on top of the piles.
3 Proceed to the first stage of excavation.
4 Cast the floor slab of the first basement level (B1 slab).
5 Begin to construct the superstructure.
Excavation 245
Figure 6.4 Top-down construction method.
6 Proceed to the second stage of excavation. Cast the floor slab of the second basement level
(B2 slab).
7 Repeat the same procedures until the designed depth is reached.
8 Construct foundation slabs and ground beams. Complete the basement.
9 Keep constructing the superstructure until finished.
6.3 Stability analysis

An excavation may fail because the stress on the support system exceeds the strength of its
materials, for example, when the strut load exceeds the buckling load of struts or the bending
moment of the retaining wall exceeds the limiting bending moment. Failure can also arise from
soil instability. The method for analyzing soil instability is called stability analysis. Stability
analysis considers base shear failure, upheaval failure, and sand boiling.
6.3.1 Base shear failure

When the shear stress generated in most of the soil below the excavation bottom reaches the
shear strength, the soil will be subject to a large amount of displacement or heave, leading to
the failure of the entire strutted retaining system. This phenomenon is called base shear failure.
Base shear failure may occur in either clayey or sandy soil. When the shear stress is equal to
the shear strength, this phenomenon is called “failure” in conventional soil mechanics and the
“plastic state” in the finite element method. Base shear failure in clayey soil will cause a large
amount of soil to heave at the bottom of the excavation, and therefore, it is also called “plastic
246 Excavation
basal heave,” or simply, “basal heave” or “plastic heave.” Figure 1.1 shows an excavation case
with base shear failure or basal heave.
As shown in Figure 6.5a, in the ultimate state, the earth pressures acting on the front and back
of the retaining wall in excavations will reach the passive and the active earth pressures, respec-
tively. Taking the retaining wall below the lowest level of the strut as a free body and conducting
a force equilibrium analysis (Figure 6.5b), we can then calculate the factor of safety against base
shear failure as follows:
Mr Pp L p
Fb = = (6.1)
M d Pa La - M s
where:
Fb = factor of safety against base shear failure.

M r = resistant moment.
M d = driving moment.
qs = surcharge.
Pa = resultant of the active earth pressure on the back of the wall below the lowest level of the
strut.
La = length from the lowest level of strut to the point of action, Pa.
M s = allowable bending moment of the retaining wall.
Pp = resultant of the passive earth pressure on the front of the retaining wall below the excava-
tion surface.
L p = length from the lowest level of strut to the point of action, Pp .
Figure 6.5 Earth pressure equilibrium method for the analysis of base shear failure: (a) earth
pressure distribution and (b) forces on the free body.
Excavation 247
Eq. 6.1 is computed based on the gross pressure distribution. The required Fb should be
equal to or greater than 1.5. Nevertheless, when assuming M s = 0, Fb ≥ 1.2. Eq. 6.1 can be
used to obtain either the factor of safety for a certain depth of wall or the required penetration
depth of a retaining wall with a certain value of the safety factor. The method is often called
the load factor earth pressure equilibrium method or the earth pressure equilibrium method
(load factor). The earth pressure equilibrium method (load factor) can be applied to sandy soil,
clayey soil, or layers of sandy and clayey soils. Since all the uncertainties are lumped into a
single factor of safety, which is not necessarily reasonable, the results for some scenarios from
the earth pressure equilibrium method (load factor), such as stiff clay with su = constant, are
unreasonable (Ou, 2022).
If the factor of safety is located at the source where the largest uncertainty arises, that is, the
shear strength, as shown in eq. 6.2 or 6.3, then 0m' and cm′ are used to calculate the distributions
of earth pressure on both sides of the wall.
For the effective stress analysis:
tan 0 ' c'

tan 0m' = cm' = (6.2)
Fb Fb
For the undrained analysis:
su
su , m = (6.3)
Fb
Then, a factor of safety (Fb) is selected to ensure that the resistance moment is greater than or
equal to the driving moment:
Pp L p > Pa La - M s (6.4)
where the definitions of Pp , Pa , L p , La , and M s are the same as those in eq. 6.1.
Similar to eq. 6.1, Ms can be assumed to equal 0. The method is called the strength factor
earth pressure equilibrium method, or the earth pressure equilibrium method (strength factor).
The earth pressure equilibrium method (strength factor) can be applied to all kinds of soils, and
normally, it can yield reasonable results for all scenarios.
Terzaghi (1943) assumes that the failure surface of base shear failure in clayey soil, that is,
plastic heave, initiates with a circular arc failure surface below the excavation surface, then
develops outward to the excavation surface level ab, extending upward to the ground surface.
The soil weight within the width of B / 2 acting on plane ab is treated as a driving force that
causes the excavation to fail.
According to Terzaghi’s bearing capacity theory (Section 3.4.3), the bearing capacity of the
clayey soil below plane ab can be denoted as Pmax = 5.7 su . When the soil weight above plane ab
is greater than the soil bearing capacity, the excavation will fail. In addition, the failure surface
will be constrained by stiff soil. Let D represent the distance between the excavation surface and
the stiff soil. Terzaghi’s method can then be discussed in two parts.
When D ≥ B 2:
248 Excavation
As shown in Figure 6.6a, the formation of the failure surface is not constrained by the stiff
soil. Suppose the unit weight of the soil is γ . The soil weight (containing the surcharge qs ) range
B / 2 on plane ab is:
B
W = (y H e + qs ) (6.5)
2
The bearing capacity, Qu , of the saturated clay below plane ab is:
B
Qu = (5.7 su 2 ) (6.6)
2
When a basal heave failure occurs, vertical failure plane bc can offer shear resistance (su1 H e ),
and the factor of safety against basal heave (Fb ) is:
Qu 5.7 su 2 B 2
Fb = = (6.7)
W - su1 H e (y H e + qs ) B 2 - su1 H e
Figure 6.6 Terzaghi’s method for the analysis of base shear failure in clayey soil: (a) D ≥ B 2,
(b) D < B 2 .
Excavation 249
where su1 and su 2 represent the undrained shear strengths of the soils above and below the exca-
vation surface, respectively, and qs denotes the surcharge on the ground surface.
When D < B 2, under such conditions, the failure surface is constrained by the stiff soil, as
shown in Figure 6.6b, and its factor of safety (Fb ) is:
Qu 5.7 su 2 D
Fb = = (6.8)
W - su1 H e (y H e + qs ) D - su1 H e
Terzaghi’s method can be applied only to excavations in clay. Moreover, Terzaghi’s method
does not take the influence of the penetration depth and stiffness of the retaining wall into account.
That is, it assumes that the retaining wall does not exist. Terzaghi’s method is applicable to wide
excavations rather than narrow excavations because it overestimates the shear resistance along
the vertical failure plane for narrow excavations. For most excavation cases, Terzaghi’s factor of
safety (Fb ) should be greater than or equal to 1.5 (Mana and Clough, 1981; JSA, 1988).
Bjerrum and Eide (1956) assumed that the failure surface induced by excavation is similar
to that of a pile foundation subjected to uplift loading, considering that the unloading behavior
of the soil below the excavation surface caused by excavation can be analogous to that of a pile
foundation subjected to uplift loading. Then, using the unloading bearing capacity equation for
the pile foundation, we can obtain the unloading bearing capacity. The factor of safety is the
ratio of the unloading bearing capacity to the unloading pressure. As shown in Figure 6.7, the
failure surface initiates with a circular arc with a radius B 2 . The factor of safety against base
shear failure can be calculated as follows:
N c . su
Fb = (6.9)
y . H e + qs
Figure 6.7 Bjerrum and Eide’s method for the analysis of base shear failure in clayey soil:
(a) D ≥ B / 2 , (b) D < B / 2 .
250 Excavation
where:
qs = the surcharge on the ground surface.

γ = unit weight of the soil.
H e = excavation depth.
su = undrained shear strength of the clayey soil.
N c = Skempton’s bearing capacity factor (1951), which can be determined from Figure 6.8 or
calculated by the following equation:
B
N c ( rectangular ) = N c (square ) (0.84 + 0.16 ) (6.10)
L
where:
B = the excavation width.

L = the excavation length.
Since N c takes into account the effects of the excavation depth and excavation shape, eq. 6.9
is equally valid for shallow and deep excavations, as well as rectangular excavations.
Based on Reddy and Srinivasan’s study (1967), NAVFAC DM 7.2 (1982) extended Bjerrum
and Eide’s method to excavations with stiff soils below the excavation surface (Figure 6.7b).
The extension of Bjerrum and Eide’s method can be expressed as follows:
su N c f d f s
Fb = (6.11)
y He
where:
N c = bearing capacity factor that considers the stiff soil, which can be determined from
Figure 6.9a.
Figure 6.8 Skempton’s bearing capacity factor.

Source: Skempton (1951).
Excavation 251
f d = depth modification factor, which can be found in Figure 6.9b.

f s = shape modification factor, which can be estimated by the following equation:
B
f s = 1 + 0.2 (6.12)
L
where B refers to the excavation width and L is the excavation length.
50
40
30
Nc 20
10
5
0.7 0.6 0.5 0.4 0.3 0.2 0.1 0
D/B
(a)
Figure 6.9 Extended Bjerrum and Eide’s method for the analysis of base shear failure in clayey
soil: (a) the bearing capacity factor considering stiff soil and (b) the depth modifica-
tion factor.
252 Excavation
Similar to Terzaghi’s method, Bjerrum and Eide’s method does not consider the influence of
the penetration depth and stiffness of the retaining wall. That is, it assumes that the retaining
wall does not exist. For most excavations, the factor of safety obtained with Bjerrum and Eide’s
method ( Fb ) should be larger than or equal to 1.2 (JSA, 1988).
The slip circle method assumes the main part of the trial failure surface below the excavation
bottom is a circular arc, with its center at the lowest level of the strut. The failure surface then
grows outward to the excavation surface level and extends vertically to the ground surface, as
shown in Figure 6.10. The shear strength on the vertical failure plane (line bc in Figure 6.10a)
is ignored. Take the retaining wall and soil below the lowest level of struts as well as above the
circular arc as a free body (Figure 6.10b). The soil weight above the excavation surface in the
back of the retaining wall can be treated as the driving force, and the shear strength along the
failure surface generates the resistant force. The ratio of the resistance moment to the driving
moment with respect to the lowest level of the strut is:
-
+a
2
X | su ( Xd0 ) s
Mr
Fb = = 0
(6.13)
Md X
W . - Ms
2
where:
M r = resistance moment.
M d = driving moment.
M s = allowable bending moment of the retaining wall.
Figure 6.10 The slip circle method for the analysis of base shear failure in clayey soil: (a) devel-
opment of the failure surface and (b) forces on the free body.
Excavation 253
su = undrained shear strength of clay.

s = distance between the lowest level of the strut and the excavation surface.
X = radius of the failure circle, X = Hp + s.
W = total weight of the soil in front of the vertical failure plane and above the excavation sur-
face, including the surcharge (qs) on the ground surface.
The original source of the slip circle method is untraceable. Nevertheless, TGS (2022) and
JSA (1988) adopt the method in their building codes. Both assume that M s = 0 and recommend
that the factor of safety (Fb ) against base shear failure or basal heave should be greater than or
equal to 1.2. According to design experience in some countries, the value is quite reasonable.
In fact, the allowable bending moment value of the retaining wall M s is far less than the resist-
ance moment provided by shear strength. Thus, to simplify the computation, it is reasonable to
assume M s = 0.
When the safety factor computed as the soil shear strength increase with depth is insuffi-
cient, the penetration depth of the wall should be increased to ensure the computed safety fac-
tor meets the requirements. However, as the soil shear strength increases little with depth, an
increase in the wall penetration depth contributes slightly to the safety factor. Under such condi-
tions, soil improvement is an effective way to increase the safety factor.
Example 6.1
Figure EX. 6.1a shows an excavation in sandy soil with two levels of the strut. The groundwater
levels behind the wall and in front of the wall are on the ground surface and on the excavation
surface, respectively. The properties of sandy soil are as follows: the effective cohesion c' = 0,
the effective angle of friction 0 ' = 33o , and the saturated unit weight y sat = 20 kN / m3. The
excavation width B = 35 m. The excavation depth H e = 10 m, and the wall penetration depth
H p = 8 m. The strut locations are h1 = 2 m, h2 = 4 m, and h3 = 4 m. The distance to the imperme-
able soil D = 5 m. Because of the difference between the levels of groundwater, seepage will
occur. Assume that the friction angles (δ ) between the retaining wall and soil on both the active
and passive sides are equal to 0. Compute the factor of safety (Fb) against base shear failure
using the earth pressure equilibrium method (load factor).
Solution
Since the groundwater level in front of the wall is different from the water pressure behind the
wall, seepage will occur. The pore water pressure due to seepage can be analyzed with a flow
net or numerical simulation. According to Figure EX 6.1a, the pore water pressure on the wall
is evaluated, as shown in Figure EX 6.1b. The active and passive earth pressures can then be
computed as follows.
According to Rankine’s earth pressure theory, K a = 0.295, K p = 3.392.
Active earth pressure:
At GL-6 m:
uw = 36.2 kPa
- v' = - v -uw =20 x 6 - 36.2 = 83.8 kPa, - a' = - v' K a = 24.7 kPa
254 Excavation
Figure EX 6.1 A 10 m excavation in sandy soil: (a) excavation profile, (b) pore water pressure
distribution, and (c) lateral earth pressure distribution (including pore water
pressure).
At GL-18 m:
uw = 108.7 kPa
- v' = - v - uw =20 x 18 - 108.7 = 251.3 kPa, - a' = - v' K a = 74.1 kPa
Passive earth pressure:
At GL-10 m:
uw = 0, - v' = - v - uw = 0 , - 'p = - v' K p = 0
- 'p + uw = 0
Excavation 255
At GL-18 m:
uw = 108.7 kPa
- v' = - v - uw = 20 x 8 - 108.7 = 51.3 kPa, - 'p = - v' K p = 174 kPa
Figure EX 6.1c shows the distribution of effective earth pressure. A similar concept to
eq. 5.43 in Chapter 5, the factor of safety, Fb, is only applied to Pp (or Kp) to account for the
uncertainty in soil parameters related to passive earth pressure. Therefore, the net water pressure
is treated as the driving force. Figure EX6.1d shows the net water pressure distribution. The
resistance moment and driving moment are calculated as follows:
Pa1 = 24.7 x 12 = 296.4 kN/m, La1 = 6 m

Pa 2 = (74.1-24.7) x 12/2 = 296.4 kN/m, La 2 = 8 m
Pw1 = 60.4 x 8 / 2 = 241.6 kN/m, Lw1 = 6.67 m
Pw2 = 36.2 x 4 = 144.8 kN/m, Lw2 = 2 m
Pw3 = (60.4 - 36.2) x 4 / 2 = 48.4 kN/m, Lw3 = 2.67 m
Pp = 174 x 8 / 2 = 696 kN/m, L p = 9.3 m
M r = Pp L p = 6473 kN-m/m
M d = Pa1 La1 + Pa 2 La 2 + Pw1 Lw1 + Pw2 Lw 2 + Pw3 Lw3 = 6180 kN-m/m
Mr
Fb
= = 1.05
Md
Example 6.2
Figure EX 6.2a shows an excavation in clayey soil. The excavation depth H e = 15 m, and the
wall penetration depth H p = 15 m. The excavation width and length are 30 m and 500 m, respec-
tively. The groundwater levels behind the wall and in front of the wall are on the ground surface
and on the excavation surface, respectively. The properties of clayey soils are:
Clay 1: y sat = 15kN/m3 , su = 20kN / m 2

Compute the factor of safety against base shear failure using the earth pressure equilibrium
method (load factor).
Solution
Use the Rankine’s earth pressure theory to calculate the earth pressure on the wall.
Active earth pressure, σ a = σ v K a - 2c K a

256 Excavation
Figure EX 6.2 A 15 m excavation in clayey soil: (a) excavation profile and (b) lateral earth
pressure distribution.
Excavation 257
At GL-12 m (in clay 3):

- v = 20 x 9 + 15 x 3 = 225 kPa
- a = 225 - 2 x 60 = 105 kPa
At GL-30 m:
- v = 225 + 18x 18 = 549 kPa
- a = 549 - 2 x 60 = 429 kPa
Passive earth pressure: σ p = σ v K p +2c K p
At GL-15 m:
-v = 0
- p = 0+2 x 60 = 120 kPa
At GL-30 m:
- v = 15 x 18 = 270 kPa
- p = 270 + 2 x 60 = 390 kPa
Figure EX 6.2b shows the distribution of earth pressures on the front and back of the retain-
ing wall.
3 + 15 3 + 15 2 x (3 + 15)
M d = 105 x (3 + 15) x + (429 - 105) x x = 52, 002 kN-m/m
2 2 3
15 1 2
M r = (120 x 15) x ( + 3) + (390 - 120) x 15 x x (15 x + 3) = 45, 225 kN-m/m
2 2 3
Mr
Fb =
= 0.87
Md
Example 6.3
Same as example 6.2. Compute the factor of safety against base shear failure (basal heave) using
the earth pressure equilibrium method (strength factor).
Solution
Assuming F is a strength reduction factor and following eq. 6.3, we have the reduced undrained
shear strength su,m = su/F, for clay 3.
By trying various F values, the corresponding distribution of active and earth pressures,
similar to Figure EX 6.2b, can be obtained with a similar computation procedure to that
in example 6.2. The relationship between PpLp/PaLa and the F value can then be obtained
as shown in Figure EX 6.3. As PpLp/PaLa = 1.0, then the F value is the factor of safety,
Fb = 0.85.
258 Excavation
Figure EX 6.3 Variation in the ratio of the resistance moment to the driving moment with the
strength reduction factor.
Example 6.4
Same as example 6.2. Compute the factor of safety against base shear failure (basal heave) using
the slip circle method, Terzaghi’s method, and Bjerrum and Eide’s method.
Solution
1 SLIP CIRCLE METHOD
Assuming that the failure surface passes the wall toe, the radius of the failure surface is equal
to 18 m.
a = cos -1 (3 / 18) = 80.4o = - / 2.24. The total vertical pressure at a depth of GL-15 m outside
the excavation is:
- v = 15 x 3 + 20 x 9 + 18 x 3 = 279 kPa
The weight of the soil column is:
W = - v,GL-15m x 18 = 279 x 18 = 5,022 kN/m

- / 2 + - / 2.24
Mr = |
0
18 x 60x (18 x d0 ) = 57,800 kN-m/m
Excavation 259
or
p p
M r = 18 x ( + ) x 60 x 18 = 57, 800 kN-m/m
2 2.24
M d = 5022 x 18 / 2 = 45,198 kN-m/m
M r 57, 800
Fb
= = = 1.28
M d 45,198
2 TERZAGHI’S METHOD
D = 22 m. According to Terzaghi’s method, the radius of the failure surface is:
B / 2 = 21.2 m < D
Qu 5.7 su 2 B / 2 5.7 x 60 x 30 / 2
Fb = = = =1
W - su1 H e (y H e + qs ) B / 2 - su1 H e (15 x 3 + 20 x 9 + 18 x 3) x 30 / 2 - (20 x 3 + 30 x 9 + 60 x 3)
.7 su 2 B / 2 5.7 x 60 x 30 / 2
= = 1.34
qs ) / 2 1 H (15 x 3 + 20 x 9 + 18 x 3) x 30 / 2 - (20 x 3 + 30 x 9 + 60 x 3)
3 BJERRUM AND EIDE’S METHOD
According to Figure 6.8, we can find N c = 5.8 . The radius of the failure surface = B/ 2 = 21.2.
The failure surface will initiate from the intersection point of the wall and excavation bot-
tom down to a depth of GL-(15 + 21.2) m or GL-36.2 m and then develop upward to the
surface. The failure surface will pass through soil layers 1, 2, and 3. The average undrained
surface along the failure surface is [60 × (36.2 – 12) + 30 × 9 + 20 × 3]/36.2 = 1782/36.2
= 49.2 kN/m2.
Qu ( N c su ) B N c su 5.8 x 49.2
Fb = = = = = 1.02
Q (y H e + qs ) B y H e + qs 15 x 3 + 20 x 9 + 18 x 3
As shown in examples 6.2, 6.3, and 6.4, the Fb values obtained from the earth pressure equi-
librium method (load factor), the earth pressure equilibrium method (strength factor), the slip
circle method, Terzaghi’s method, and Bjerrum and Eide’s method are equal to 0.87, 0.85, 1.28,
1.34, and 1.02, respectively. Among those methods, the Fb value obtained from the earth pres-
sure equilibrium methods is smaller than that obtained from Terzaghi’s method and the slip
circle method. This difference is because the shear strength or adhesion between the wall and
clay in the earth pressure equilibrium methods is assumed to be 0 in examples 6.2 and 6.3.
However, if it is included in the computation using the method as stated in Ou (2022), the Fb
value obtained from the earth pressure equilibrium methods is close to that from the slip circle
method and Terzaghi’s method.
260 Excavation
6.3.2 Upheaval failure

If there is a permeable layer (such as sandy or gravelly soils) underlying an impermeable layer
below the excavation surface, the impermeable layer tends to be lifted up by the pore water pres-
sure in the permeable layer. The safety against upheaval failure should be examined. As shown
in Figure 6.11, the factor of safety against upheaval is:
Ey ti . hi
Fup = i (6.14)
Hw .y w
where:
Fup = factor of safety against upheaval failure.

γ ti = unit weight of soil in each layer above the bottom of the impermeable layer.
hi = thickness of each soil layer above the bottom of the impermeable layer.
H w = head of the pore water pressure in the permeable layer, which can be measured in
piezometers.
γ w = unit weight of the groundwater.
The factor of safety against upheaval failure, Fup , should be larger than or equal to 1.2.
When the computed safety factor does not satisfy the requirement, it is necessary to reduce
the pore water pressure below the impermeable layer by dewatering. Improving the soil below
the impermeable layer to increase the weight of the soils, that is, increase the numerator in
eq. 6.14, is another alternative.
Example 6.5
Same as example 6.2. According to site investigation, the pore water pressure in the deposit is
in the hydrostatic state. Compute the factor of safety against upheaval failure.
Figure 6.11 Stability analysis against upheaval failure.

Excavation 261
Solution
Since the pore water pressure is in the hydrostatic state, the pore water pressure at a depth of
GL-37.0 m is:
uw = 9.81x 37 = 363 kPa
The total vertical pressure above the permeable layer is:
- v = 18 x 22 = 396 kPa
The factor of safety against upheaval failure is:
396
Fup
= = 1.09
363
6.3.3 Sand boiling

If a retaining wall in sandy soil that has good permeability does not penetrate into the impermeable
soil, dewatering in an excavation will cause a pressure difference behind the wall and in front of the
wall, thereby inducing seepage. When the upward seepage water pressure is greater than or equal
to the effective stress of the soil inside the excavation, the soil strength drops to 0 and cannot resist
the earth pressure on the back of the wall, resulting in excavation failure. This phenomenon is called
sand boiling. The safety of the excavation in sandy soil must be examined against sand boiling.
Commonly used methods for the analysis of sand boiling include Harza’s method and Ter-
zaghi’s method. Figure 6.12 shows a watertight retaining wall, in which the maximum hydraulic
Figure 6.12 Stability analysis against sand boiling.

262 Excavation
gradient at the exit of seepage normally occurs at the location near the wall, that is, point A.
When the hydraulic gradient at point A is equal to or close to the critical value, sand boiling
occurs. Harza (1935) calculates the safety factor against sand boiling as:
icr
Fs = (6.15)
imax(exit )
where icr is the critical hydraulic gradient and imax(exit ) is the maximum “exit” hydraulic gradient,
which can be obtained with the flow net method or numerical analysis.
According to the phase relationship, the critical hydraulic gradient can be calculated as:
y ' Gs -1
icr = = (6.16)
y w 1+ e
Since the Gs value of sand is approximately 2.65 and its e value is between 0.57 and 0.95,
the critical hydraulic gradient for most sands is close to 1 according to the preceding equation.
Terzaghi (1922) found from the results of many model tests with watertight sheet piles that
sand boiling often occurs within a distance of approximately H p / 2 from the sheet piles (H p
refers to the penetration depth of the sheet piles). Thus, to analyze the stability of sheet piles
against sand boiling, we can take the soil prism H p × H p / 2 in front of the sheet pile as the ana-
lytic object, as shown in Figure 6.12. The uplift force throughout the soil prism is:
1 2
U = (the volume of the soil prism) x(iavg y w ) = H p iavg y w (6.17)
2
where iavg is the average hydraulic gradient throughout the soil prism. The downward force of
the soil prism (i.e., the submerged weight) is:
1 2 1
W' = H p (y sat - y w ) = H p2y ' (6.18)
2 2
Therefore, the factor of safety is:
W' y'
Fs = = (6.19)
U iavg y w
According to Pratama et al. (2019), the required Fs for medium sand and very dense sand
are 2.0 and 1.6, respectively, for Harza’s method and 1.75 and 1.40, respectively, for Terzaghi’s
method. As the computed factor of safety is insufficient, sand boiling is likely to occur. Possible
remedial measures are increasing the wall penetration depth, penetrating the wall into the imper-
meable layer, and lowering the water level behind the wall.
Example 6.6
Same as example 6.1. Compute the factor of safety against sand boiling using Harza’s and Ter-
zaghi’s methods.
Excavation 263
Solution
Figure EX 6.6a shows the flow net with 6 flow channels and 20 equipotential drops.
The critical hydraulic gradient is calculated as:
y ' 20 - 9.81
icr = = = 1.039
yw 9.81
The maximum hydraulic gradient at the exits is the one adjacent to the wall. The head loss at
the exit is estimated as:
10 -h 0.5
-h = = 0.5 imax(exit ) = = = 0.325
20 , L 1.54
With Harza’s method:
icr 1.039
=Fs = = 3.20
imax(exit ) 0.325
With Terzaghi’s method, the soil prism with the potential sand boiling zone has a cross sec-
tion of 8 m × 4 m. Figure EX 6.6b shows the soil prism on an enlarged scale. By using the flow
net, we can calculate the head loss across the prism as follows:
7
Along section ba, the head loss = ×10m=3.5 m.
20
3.8
Along section cd, the head loss = ×10m=1.9 m.
20
Figure EX 6.6a Analysis of sand boiling for Figure EX 6.1: (a) flow net and (b) soil prism.
264 Excavation
Figure EX 6.6b (Continued)
For other intermediate sections along bc, the approximate head loss can be similarly calcu-
lated. The average value of the head loss throughout the entire soil prism is 0.27 ´ 10 = 2.7 m,
and the average hydraulic gradient is:
2.7
i=
avg = 0.3375
Hp
y' 20 - 9.81
Fs = = = 3.08
iavg y w 0.3375x 9.81
6.4 Stress analysis

A typical supported retaining system is a highly indeterminate structure. The stress analysis can
resort to finite element analysis or the beam–spring model. Basically, the finite element method
with advanced soil models can obtain the necessary data for design, including the strut load,
bending moment of the wall, wall deflection, ground settlement, and even the factor of safety
against base shear failure. The beam–spring model assumes the retaining wall to be a beam sup-
ported by elastic soil. As shown in Figure 6.13, the interaction between the retaining wall and
adjacent soils is simulated by a series of elastic soil springs. The lateral strut is also simulated
by springs. Numerical analysis can then be conducted by simulating each construction activity.
With the beam–spring model, one can obtain the strut load, bending moment, and shear force of
the wall, but not the ground settlement. The detailed finite element method and the beam–spring
Excavation 265
Figure 6.13 Beam–spring model: (a) profile of an excavation and (b) a beam–spring model.
model can be found in Ou (2022). Considering the page limit, this section presents stress analy-
sis using simple structural mechanics only.
6.4.1 Struts
Figure 6.14 shows diagrams of the apparent earth pressure as proposed by Peck (1969). As
shown in the figure, when the soil in the back of the wall consists of mainly sandy soils, the
apparent earth pressure, pa , is:
pa = 0.65y H e K a (6.20)
where:
γ = unit weight of sandy soils.

0 ' = effective friction angle.
K a = coefficient of Rankine’s earth pressure = tan 2 (45o - 0 ' 2).
The effective stress analysis method should be adopted for sand in eq. 6.20.
If the soil in the back of the wall is soft to medium clayey soil (i.e., y H e / su > 4), the apparent
earth pressure, pa , is the larger of
pa = K a y H e or pa = 0.3y H e (6.21)
4su
Ka = 1 - m (6.21a)
y He
266 Excavation
Figure 6.14 Apparent earth pressure diagram: (a) sand, (b) soft to medium clay ( y H e su > 4 ) ,
and (c) stiff clay ( y H e su < 4 ) .
where:
4 su
K a = coefficient of earth pressure, K a = (1 - m ).
y He
m = parameter considering a possible large deformation of the wall below the excavation bottom.
su = average undrained shear strength of clayey soil over the depth of excavation.
When N b = y H e / sub < 5.14 , m = 1.0 (Terzaghi et al., 1996), where sub is the undrained shear
strength at the base of the excavation, that is, the undrained shear strength of the soil between
the excavation bottom and the influence depth of the excavation. However, when N b > 5.14, a
large displacement at the bottom of the retaining wall is expected, m = 0.4 (Peck et al., 1977).
Eq. 6.21 should be used based on the total stress analysis, that is, assume 0 = 0 without consider-
ing the pore water pressure.
If the soil in the back of the wall is stiff clayey soil (y H e / su < 4), the apparent earth pressure,
pa , is:
pa = 0.2y H e ~ 0.4y H e (the average is 0.3 γ H e ) (6.22)
Similarly, eq. 6.22 should be used based on total stress analysis. It does not consider the
pore water pressure. In addition to Peck’s diagrams of the apparent earth pressure, Peck et al.
(1977), Terzaghi et al. (1996), and many other investigators have recommended similar types
of diagrams.
For alternating layers of sandy and clayey soils, one can apply the concept of the equivalent
cohesion and unit weight, respectively, to calculate the apparent earth pressure (Peck, 1943).
Excavation 267
As shown in Figure 6.15a, where a sandy soil is above a clayey soil, the equivalent cohesion of
alternating layers of sandy and clayey soils can be calculated as follows:
1
su , eq = [( K s y s H s tan d ) H s / 2 + H c su ]
He
(6.23)
1
= [y s K s H s 2 tan d + 2 H c su ]]
2H e [
where:
γ s = unit weight of sandy soil.
H s = height of the sandy soil.
H c = height of the clayey soil.
K s = coefficient of lateral earth pressure.
φs = friction angle of sandy soil.
δ = friction angle between the wall and the soil.
su = undrained shear strength of clayey soil.
The equivalent unit weight of the alternating layers is:
1
y eq = [y s H s + H c y c ] (6.24)
He
where γ c is the unit weight of clay.
Figure 6.15 Multiple soil layers in excavations: (a) sand and clay and (b) clay.
268 Excavation
Similarly, for layered clayey soils, the concept of equivalent values can also be used to calcu-
late the strut load, as shown in Figure 6.15b. The equivalent cohesion of clayey soil is:
1
su , eq = ( su1 H1 + su 2 H 2 ) (6.25)
He
where:
su1 and su 2 = undrained shear strength of the first and second clayey layers, respectively.
H1, H 2 = height of the first and second clayey layers, respectively.
The equivalent unit weight of clayey soil would be:
1
y eq = (y 1 H1 + y 2 H 2 ) (6.26)
He
where γ 1, γ 2 is unit weight of the first and second clayey layers, respectively.
Providing the equivalent cohesion (or the undrained strength) and the equivalent unit
weight are derived as earlier, Figure 6.14 can be used to choose a proper distribution of
earth pressures. The strut load for each strut level can then be calculated using an appropri-
ate simplified model. For example, the load on each level of struts can be the result of the
earth pressure within the range covering half the vertical span between the upper and cur-
rent struts and half the vertical span between the current and lower struts. This method can
be referred to as the half method. The calculation procedure for the half method is shown
in Figure 6.16.
Figure 6.16 Computation of strut load.

Excavation 269
6.4.2 Wales
The function of wales is to transfer the earth pressure acting on the retaining wall to the struts.
For the purpose of analysis, the earth pressure can therefore be assumed to act on the wale
directly. The earth pressure distribution can be obtained from the apparent earth pressure distri-
bution. To compute the maximum bending moment and shear force of wales, the wales can be
considered as simply supported beams with struts as supporting hinges or viewed as continuous
beams, as shown in Figure 6.17. Details of the analysis can be found in Example 6.7.
If a wale is viewed as a simply supported beam, then:
1 2
M max = pL (6.27a)
8
1
Qmax = pL (6.27b)
2
If viewed as a continuous beam, then:
1
M max = pL2 (6.28a)
12
1
Qmax = pL (6.28b)
2
where:
M max = maximum bending moment of the wale.

Qmax = maximum shear of the wale.
L = distance between struts.
p = earth pressure acting on the wall, which can be derived from the apparent earth pressure
diagram (Figure 6.14) or can be transformed from the strut load (please see example 6.7).
Figure 6.17 Computation of the bending moment of wales.

270 Excavation
If a simply supported beam is assumed, it may not work out as economically. Nevertheless, if
a continuous beam model is assumed, it tends to not be conservative. The real condition should
be somewhere between a continuous beam and a simply supported beam. Further discussion can
be found in Ou (2022).
6.4.3 Retaining walls

The assumed support method is a commonly used stress analysis method for supported
retaining walls. Theoretically, the earth pressure on the back of the wall should be the
active earth pressure rather than the apparent earth pressure. Therefore, in the computation
of the wall bending moment and shear force, the active earth pressure is used. To simplify
analysis, struts are assumed to be fulcrums, and the reaction force of the soil in front of the
wall can be viewed as a concentrated force or roller, which is called the assumed support.
When the moment formed by the active earth pressure and the moment formed by the pas-
sive earth pressure are balanced, the action point of the resultant passive earth pressure is
the assumed support, as shown in Figure 6.18. The location of the assumed support (  ) can
be determined as:
Pa  a
= -s (6.29)
Pp
After determining the distribution of earth pressures and location of the assumed support, the
retaining wall can be seen as a simply supported beam. Figure 6.19 shows the computed wall
bending moment and shear force using the assumed support method. The wall bending moment
and shear force can also be computed using another beam model (Figure 6.20), but the active
earth pressure should be used rather than the apparent earth pressure. Details of the analysis can
be found in example 6.7.
Figure 6.18 Estimation of the location of the assumed support.

Excavation 271
Figure 6.19 Computation of the wall bending moment and shear force.
Active earth pressure assumption Apparent earth pressure assumption

(incorrect)
Figure 6.20 Alternative computation of the wall bending moment.

272 Excavation
Example 6.7
Same as example 6.2. Assume that the sheet pile is used as the retaining wall. The average hori-
zontal spacing between struts is 5 m. The allowable stress of the steel is - all = 170 x 103 kN/m2.
1 Draw the earth pressure envelope.

2 Compute the strut loads at the first, second, and third levels.
3 Determine the section modulus required for the wale at the second level.
4 Determine the section modulus required for the sheet pile.
Solution
1 Draw the earth pressure envelopes.
According eqs. 6.21 and 6.22, the average values (0–15 m) are:
su(avg) = 34 kPa, γsat (avg) = 18.6 kN/m3

y H e 15x 3 + 20 x 9 +18x 3
Nb = = = 4.65
sub 60
Since Nb > 5.14, m = 1.0.
4su 4 x 34
Ka = 1 - m = 1- = 0.51
y He 18.6 x 15
pa = K a y H e = 0.51x 18.6 x 15 = 142 kN/m2
Alternatively, pa = 0.3y H e = 84 kN/m2.
Consider pa = 142 kN/m2.
The apparent earth pressure is shown in Figure EX 6.7a.
2 The strut load.
With the half method:
Q1 = (3 = 1.5 = 0.75) x 0.5 x 142 x 5 = 1864 kN

Q2 = (1.5 + 1.5) x 142 x 5 = 2130 kN
Q3 = (1.5 + 1.5) x 142 x 5 = 2130 kN
Q4 = (1.5 + 1.5) x 142 x 5 = 2130 kN
3 Section modulus for the wale.
The strut load at the second level of the strut Q2 = 2130 kN. The strut spacing L = 5 m. The
uniform pressure acting on the wale p = 2,130/L = 426 kN/m.
Excavation 273
Figure E X 6.7 Design of various components of the retaining support system for Figure EX 6.2:
(a) apparent earth pressure diagram, (b) analysis of bending moment and shear
force using the assumed support method, and (c) calculation of the location
of zero shear force.
274 Excavation
Figure E X 6.7 (Continued)
If the simply supported model is adopted:
1 2 1
M max = pL = x 426 x 52 = 1331 kN-m
8 8
1 1
Qmax = pL = x 426 x 5 = 1065 kN
2 2
M max 1331
The section modulus S = = = 7.83 x 10-3 m3.
- all 170x1000
If the continuous end beam model is employed:
1 1
M max = pL2 = x 426 x 52 = 888 kN-m
12 12
M max 888
The section modulus S + + + 5.22 x 10-3 m3.
- all 170x1000
4 Section modulus for the sheet pile.
Use the Rankine’s earth pressure to calculate the bending moment and shear force of the
wall K=
a K=
p 1.
2cu 2 x 20
Depth of the tension crack, zc = = = 2.67 m.
y sat 15
Excavation 275
Active side:
At GL-2.67 m, - a = 0.
At GL-3.0 m-, - a = - v K a - 2cu K a = ( 3 x 15 ) - 2 ( 20 ) = 5 kPa.
At GL-3.0 m+, o a = o v K a - 2cu K a = ( 3 x 20 ) - 2 ( 30 ) = 0 .
At GL-12 m-, o a = o v K a - 2cu K a = (12 x 20 ) - 2 ( 30 ) = 180 kPa.
At GL-12 m+, o a = o v K a - 2cu K a = (12 x 18 ) - 2 x 60 = 96 kPa.
At GL-15 m, o a = o v K a - 2cu K a = (15 x 18 ) - 2 x 60 = 150 kPa.
At GL-30 m o a = o v K a - 2cu K a = ( 30 x 18 ) - 2 x 60 = 420 kPa.
Passive side:
At GL-15 m, o p = o v K p + 2cu K p = ( 0 x 18 ) + 2 x 60 = 120 kPa.

At GL-30 m, o p = o v K p + 2cu K p = (15 x 18 ) + 2 x 60 = 390 kPa.
According to the method introduced in Figure 6.18, the location of the assumed support can be
estimated as:
Pa  a
= - s = 5.64 m
Pp
Figure EX 6.7b shows the earth pressure distribution that will be used for the analysis of the
wall bending moment with the assumed support method. The maximum bending moment will
occur at the location of the shear force equal to zero. Figure EX 6.7b shows that zero shear force
should occur within span D–E. As shown in Figure EX 6.7c, the reaction force at point D is
RD = 300 kN/m.
The shear force at distance x from point D within span D–F is calculated as:
1 | (150 - 96) x |
Vx = 300 - 96 x - | |x
2| 3 |
When Vx = 0, x = 2.53 m.
The maximum bending moment at x = 2.53 m is calculated to be 404 kN-m.
M max 404
The section modulus S = = = 2.38 x 10-3 m3.
o all 170x1000
The size and type of the sheet pile can then be selected according to the product manual.
6.5 Deformation analysis

The objective of deformation analysis is to analyze the deflection of the retaining wall and the
ground settlement caused by excavation to ensure the safety of the retaining wall and adjacent
276 Excavation
building around the excavation. The items of deformation analysis include the lateral deflection
of the retaining wall and ground settlement.
6.5.1 Wall deflection

Two types of deflection of the retaining wall usually occur in excavations: cantilever deflec-
tion and deep inward deflection, as shown in Figure 6.21. The cantilever deflection usually
occurs in stiff soil or excavations with weak strut stiffness, and the maximum deflection
occurs at the top of the retaining wall. The deep inward deflection normally occurs in soft
clay or excavation with high strut stiffness, and the maximum deflection occurs at a certain
depth of the retaining wall, usually near the excavation surface. To accurately predict the
deflection pattern of the retaining wall, it must be analyzed by the finite element method
or a beam–spring model (Ou, 2022). In this section, only the maximum wall deflection is
discussed.
For soft to medium stiff clay, the maximum deflection of the retaining wall can be obtained
by Clough and O’Rourke’s chart (1990), as shown in Figure 6.22. The parameters used to
derive the maximum wall deflection are the safety factor against base shear failure or basal
heave and the system stiffness of the retaining wall. The safety factor must be calculated by
Terzaghi’s method (eq. 6.4 or 6.5). The system stiffness is a dimensionless parameter that is
defined as:
EI
Sw = (6.30)
y w ha4vg
here γ w is the unit weight of water, EI is the stiffness of the retaining wall, and havg is the average
vertical spacing of the struts.
Figure 6.21 Types of wall deflection and ground surface settlement.

Excavation 277
1 m thick
Figure 6.22 Estimation of the maximum wall deflection.

Source: Clough and O’Rourke (1990).
For sandy soil, the maximum deflection of the retaining wall can be estimated by an empirical
equation s hm = (0.2 - 0.3)H e , where H e is the excavation depth.
Notably, the wall deflection is also affected by the excavation width. The greater the excava-
tion width, the greater the unloading force, which in turn causes greater wall deflection. How-
ever, Clough and O’Rourke’s chart does not take the excavation width into account. Based
on the authors’ experience, Clough and O’Rourke’s chart is suitable for excavation widths of
approximately 40 m. If the excavation width is different from B = 40 m, the maximum wall
deflection obtained from Figure 6.22 can be modified based on the fact that the maximum wall
deflection is inversely proportional to the excavation width because the soil inside the excava-
tion has unloading behavior, that is, elastic behavior.
6.5.2 Ground settlement

The shapes or types of ground surface settlement engendered by excavation can be categorized
into the spandrel type and the concave type, as shown in Figure 6.21 (Hsieh and Ou, 1998). The
main factors responsible for these two types of ground surface settlements are the magnitude
and shape of the deflection of a retaining wall.
The spandrel type of settlement is more likely to relate to the cantilever type of wall deflection,
while the concave type is more related to the deep inward type of wall deflection. Under normal
construction conditions, excavation in soft clay will produce more wall deflection, which tends
to bring about deep inward movement, leading to the concave type of settlement. Excavation in
278 Excavation
sandy soil or stiff clay, on the other hand, will produce less deflection of the retaining wall, and
the spandrel type of settlement is more likely to occur (Ou and Hsieh, 2011).
Since the factors affecting the deflection of a retaining wall also affect the ground surface set-
tlement, there should be a certain relationship between the maximum wall deflection (δ hm) and
the maximum ground surface settlement (δ vm). Ou et al. (1993) found that s vm = (0.5 ~ 0.75)s hm
for most cases, with the lower limit for sandy soils, the upper limit for clays, and a limit some-
where between the two for alternating layers of sandy and clayey soils.
The ground surface settlement at the various distances behind the wall can be estimated using
Peck’s method (1969), which was derived from excavations with soldier piles and sheet piles.
Peck’s method classifies soil into three types according to the soil characteristics (Figure 6.23):
Type I: Sandy and soft to stiff clayey soil, with average workmanship
Type II (a): Very soft to soft clayey soil
1 Limited depth of clayey soil below the excavation bottom
2 Significant depth of clayey soil below the excavation bottom but with an adequate factor
of safety against base shear failure, or N b < N cb
(b): settlements affected by construction difficulties
Type III: Very soft to soft clayey soil to a significant depth below the excavation bottom and
with a low factor of safety against base shear failure, or N b ≥ N cb.
where N b , the stability number of the soil, is defined as γ H e / su, where γ is the unit weight of the
soil, H e is the excavation depth, su is the undrained shear strength of the soil below the base of the
excavation; N cb is the critical stability number against base shear failure, corresponding to the factor
of safety against base shear failure equal to 1. Under such conditions, N cb = 5.7 (Terzaghi et al., 1996).
Clough and O’Rourke (1990) proposed various envelopes of settlements for different
soils. Basically, excavations in sandy or stiff clayey soil tend to produce triangular (spandrel)
Figure 6.23 Peck’s method for the estimation of ground surface settlement.
Source: Peck (1969).
Excavation 279
Figure 6.24 Clough and O’Rourke’s method for the estimation of ground surface settlement:
(a) sand, (b) stiff to very stiff clay, and (c) soft clay.
Source: Clough and O’Rourke (1990).
ground surface settlement. The maximum settlement will be found near the retaining wall.
The envelopes of the ground surface settlement are shown in Figures 6.25a and 6.25b. Exca-
vation in soft to medium clayey soil will generate a trapezoidal envelope of ground surface
settlement, as shown in Figure 6.24c. However, as studied by Ou and Hsieh (2011), excava-
tions in soft clayey soil normally produce the concave type of ground surface settlement, and
the settlement predicted from Figure 6.24c is not very consistent with field observation, as
shown in Figure 6.21. Regarding the prediction of spandrel and concave types of settlement
profiles, please refer to Ou and Hsieh (2011).
6.6 Control of excavation-induced movement

Excessive wall deflections may damage the retaining wall itself, which in turn causes excessive
ground settlements, leading to damage to adjacent buildings or public facilities. Wall deflections
and/or ground settlement in excavations must be reduced to a tolerable value by improvement
measures. Strengthening the bracing system and/or increasing the passive resistance are two
commonly adopted improvement measures in engineering practice, and many successful case
histories have been reported.
Soil improvement has been proven to be an effective way to increase the passive resistance
of soil in front of the wall and therefore reduce the wall deflection and ground settlement.
Figure 6.25 shows that many soil improvement piles are implemented in front of the wall to
increase the passive resistance of the soil in front of the wall. Figure 6.26 shows a photo of an
excavation with ground improvement piles. During excavation, both the improved soil and in
280 Excavation
Figure 6.25 Nonoverlapping soil improvement piles in an excavation.
Figure 6.26 Photo of an excavation with ground improvement piles.
situ soil inside the excavation are excavated. Soil improvement can also be performed with
overlapping improvement piles, which is certainly more effective to reduce movements but
more expensive than nonoverlapping piles.
In addition, cross walls and buttress walls have been used extensively to reduce the wall
deflection and ground settlement in excavations. Figure 6.27 shows a schematic arrangement
Excavation 281
Figure 6.27 Schematic arrangement of the cross walls: (a) plan and (b) profile.
of the cross walls in excavations. As shown in this figure, prior to excavation, a concrete
wall is constructed to connect two opposite retaining walls. The cross wall can be viewed as
a lateral support or strut, which exists before excavation and bears a great deal of compres-
sive strength. Therefore, the cross wall can reduce the lateral wall deformation and ground
settlement to a very small value as long as suitable spacing and depth of the cross walls are
designed. The case histories, performance, design, and analysis for cross walls have been
reported in the literature (e.g., Ou et al., 2011; Hsieh et al., 2012; Hsieh et al., 2013; Ou et al.,
2013). Figure 6.28 shows a possible wall deformation mode for an excavation with and with-
out cross walls.
Moreover, the concrete wall is constructed with limited length, that is, not to the opposite
side of the retaining wall. This concrete wall is called the buttress wall. Figure 6.29 shows a
schematic arrangement of buttress walls in excavations. Buttress walls can be demolished step-
by-step as excavation progresses; however, for safety reasons, they may be maintained until
excavation is completed and then demolished. They may be maintained even after excavation
and can be used as permanent structures or partition walls. Figure 6.30 shows a photo of an
excavation with buttress walls where the buttress walls are maintained during excavation. Dif-
ferent treatments of buttress walls result in different mechanisms in reducing the wall deflection.
In general, buttress walls are less expensive and less effective in reducing movements than cross
walls. Figure 6.31 shows a possible deformation mode for an excavation with and without but-
tress walls. It is beyond the scope of this chapter to introduce those measures in detail. Interested
readers are advised to refer to related references (e.g., Ou et al., 2011; Hsieh and Ou, 2018; Lim
et al., 2019; Ou, 2022).
282 Excavation
Figure 6.28 Possible wall deformation modes for an excavation with and without cross walls
(in plan).
Figure 6.29 Schematic arrangement of buttress walls: (a) plan and (b) profile.
Excavation 283
Figure 6.30 Photo of an excavation with buttress walls.
Figure 6.31 A possible wall deformation mode for an excavation with and without buttress walls.
284 Excavation
6.7 Monitoring system

Although the technologies of analysis, design, and construction for excavations have advanced
considerably, they are not capable of contending with all the changes or uncertainty in an
excavation. A well-arranged monitoring system can be helpful to monitor the safety of exca-
vations. Before a failure or damage occurs, there are signs, such as an extraordinary increase
in the wall or soil displacement or stress. A monitoring system can issue an immediate warn-
ing to help engineers adopt effective measures to forestall a failure or damage when these
signals appear.
6.7.1 Monitoring items

Monitoring field performance is basically the measurement of physical quantities of some
objects, such as displacement, stress, and strain. The common monitoring items in excavations
are (1) displacement of a structure or soil, (2) stress or strain of a structure or soil, and (3) pore
water pressure and groundwater level.
The displacement of a retaining wall relates closely to the ground settlement or the building
settlement. The inclinometer is a commonly used device for measuring the lateral displacement
of retaining walls. An inclinometer is a device for measuring the displacement across a refer-
ence line in a ground. As shown in Figure 6.32, an inclinometer contains a tilt sensor inside a
rigid tube. The tilt sensor measures the angle that the tube makes with the reference line in the
vertical direction. As measurement is conducted, the inclinometer is placed inside a guide tube
(known as an inclinometer casing) that is normally installed inside a borehole. An inclinometer
Figure 6.32 An inclinometer.

Excavation 285
Figure 6.33 Displacement measured by inclinometers: (a) casing in a diaphragm wall and (b)
principle of measurement of wall displacement by inclinometers.
casing has two pairs of internal tracks. Figure 6.33a shows a photo of an inclinometer casing
installed in a diaphragm wall. To ensure accurate measurement, the space between the incli-
nometer casing and soil/concrete should be filled with a cement/bentonite grout. Measurement
can be performed by running an inclinometer along a pair of tracks to measure the deviation at
intervals equal to the length of the inclinometer. Figure 6.33b shows how the relative displace-
ment between the top and bottom of an inclinometer casing is measured. Figure 6.34a shows
the typical displacements for the diaphragm wall at various stages of excavation that were
measured using an inclinometer in an excavation, namely, the TNEC excavation project (Ou
et al., 1998).
For the ground settlement, the simplest method is the installation of steel nails or settlement
nails and measuring the settlement with a level. Buildings will tilt as a result of ground settle-
ment and thereby be damaged. A typical ground surface settlement that was measured with level
and settlement nails is shown in Figure 6.34b.
The tilt of a building can be estimated by the relative settlement between two reference points
with plane surveying, such as a level or theodolite. It can also be monitored using a tiltmeter,
which provides the tilt angle. Figure 6.35 shows the photo of a datum plate that is fixed on a
building or structure. Similar to an inclinometer, a tiltmeter contains a sensor that can measure
the tilt of the datum plate or the building/structure.
The load on the struts must be monitored constantly during excavation so as not to exceed
the allowable value and endanger the safety of the excavation. Strain gauges are widely used
286 Excavation
Figure 6.34 A typical measurement of movements in an excavation project, TNEC project: (a)
lateral wall displacement and (b) ground surface settlement.
Figure 6.35 A datum plate of tiltmeters fixed on the wall of a building.

Excavation 287
Figure 6.36 Photo of a strain gauge on a strut.
devices to measure strut loads. Figure 6.36 shows a photo of a strain gauge installed on the web
of a strut.
Piezometers are usually installed in sand or gravel in excavations to measure the pore water
pressure that may cause upheaval failure below the impermeable layer. The piezometer can also
be used to monitor changes in water seepage during dewatering in an excavation. The instru-
ment for measuring groundwater level is the water observation well, where a perforated stand-
pipe is often used. A typical configuration for a piezometer and water observation well can be
found in Figures 1.22 and 1.23, respectively, in Section 1.8.1.
In addition to the monitoring instruments mentioned previously, other types of monitor-
ing instruments can also be employed in excavations. For example, crackmeters can be used
to monitor the cracks on the walls or columns of buildings. The concrete stressmeter can
be used to monitor the stress of concrete, earth pressure cells for the earth pressure on the
retaining wall, and so on. Interested readers can refer to related literature for details (e.g.,
Ou, 2022).
6.7.2 Arrangement of the monitoring system

Monitoring items or devices should be set according to excavation characteristics. The fre-
quency of measurements or readings should be determined based on the size of the excavation,
soil conditions, uncertainty, and risk. It is beyond the scope to introduce these devices and meas-
urements in detail. Interested readers are advised to refer to related references (e.g., Ou, 2022).
An example of a monitoring system in an excavation project, namely, the TNEC excavation
288 Excavation
Figure 6.37 An example of the arrangement of monitoring devices in the TNEC excavation
project (a) plan and (b) profile.
Excavation 289
project, is illustrated in Figure 6.37 (Ou et al., 1998). The excavation depth was 19.7 m, and
a diaphragm wall was used as an earth retaining wall. The top-down construction method was
adopted. As shown in this figure, there were six buildings, A, B, C, D, E, and F, adjacent to the
site. Buildings A, B, D, and E were installed with tiltmeters. Five inclinometers were installed
in the diaphragm wall. Piezometers were installed in sandy and gravel soils. Several settlement
nails were set around the excavation site.

This chapter introduces two commonly used excavation methods, various stability analysis
methods, deformation analysis, stress analysis, control of excavation-induced movements, and
monitoring systems. The summary and general comments of this chapter are as follows.
1 Stability analysis of excavations against base shear failure can be conducted with the earth
pressure equilibrium method. In addition, Terzaghi’s method, Bjerrum and Eide’s method,
and the slip circle method are commonly used for clayey soil.
2 For excavations in sandy soil, stability analysis of base shear failure and sand boiling should
be carried out. In clayey soil, stability analysis includes base shear failure or basal heave. In
alternating sandy and clayey soil, analysis of base shear failure, upheaval failure, and some-
times, sand boiling should be conducted.
3 The strut load can be calculated by the apparent earth pressure method. The apparent earth
pressure is the earth pressure obtained by back-calculating the strut load obtained from the
field measurement. Generally, the apparent earth pressure method should be applied only to
a small scale or shallow depth of excavation.
4 A typical supported retaining system is a highly indeterminate structure. The stress
analysis can resort to a sophisticated finite element analysis or beam–spring model.
With the beam–spring model, one can obtain the strut load, bending moment diagram,
and shear force diagram. However, for simplicity, each structural component can be
analyzed with simple structural mechanics and based on experience. It is generally sug-
gested that these simplified methods be applied only to small-scale or shallow depths of
excavations.
5 There are two types of deflection for a retaining wall: cantilever type and deep inward type.
The cantilever deflection usually occurs in stiff soil or excavations with weak strut stiffness,
and the maximum deflection occurs at the top of the retaining wall. Deep inward deflection
normally occurs in soft clayey soil or excavations with high strut stiffness, and the maximum
deflection normally occurs a certain depth below the ground surface. The maximum deflec-
tion can be estimated using Clough and O’Rourke’s chart (1990).
6 There are two types of ground surface settlement caused by excavation: spandrel type and
concave type. The former is related to cantilever wall deflection, while the latter is associ-
ated mostly with deep inward wall deflection. The maximum ground surface settlement can
be estimated using the methods recommended by Peck (1969) and Clough and O’Rourke
(1990).
7 Soil improvement in an excavation is a common measure to increase the factor of safety
against base shear failure, upheaval failure, or sand boiling. In addition, soil improve-
ment on the passive side can reduce the lateral wall deflection and ground settlement. To
reduce excavation-induced movement, cross walls and buttress walls have recently been
adopted.
290 Excavation
Problems
6.1 Figure P6.1 shows an excavation in layers of sandy and clayey soils. The groundwater
level behind the wall is the same as that in front of the wall. Assume the following param-
eters: He = 9.0 m, Hp = 8.0 m; h1 = 2.0 m, h2 = 3.5 m, h3 = 3.5 m; y = 16 kN/m3 , y sat = 20
kN/m3 ; c' = 0, o ' = 30o ; and D = 5 m. Compute the factor of safety (Fb) against base shear
failure using both the load factor and strength factor earth pressure equilibrium methods
with Coulomb’s earth pressure theory, assuming s / o ' = 0.67.
6.2 Redo problem 6.1 with the following parameters: He = 10.0 m, Hp = 11.0 m; h1 = 2.0 m,
h2 = 4.0 m, h3 = 4.0 m; y = 16 kN/m3, y sat = 18 kN/m3 ; c' = 0, o ' = 33o ; and D = 5 m.
6.3 Same as problem 6.1, but the penetration depth Hp is unknown. Assuming the factor of
safety (Fb) against base shear failure is equal to 1.2, compute the required Hp using both
the load factor and strength factor earth pressure equilibrium methods.
6.4 Same as problem 6.2, but the penetration depth Hp is unknown. Assuming the factor of
safety (Fb) against base shear failure is equal to 1.2, compute the required Hp using both
the load factor and strength factor earth pressure equilibrium methods.
6.5 Figure P6.5 shows an excavation in three layers of clay. The groundwater levels behind
the wall and in front of the wall are on the ground surface and on the excavation surface,
respectively. The excavation width is 20 m. Assume the following parameters: He = 9.0 m,
Hp = 11.0 m; h1 = 2.0 m, h2 = 3.0 m, h3 = 4.0 m; su1 = 10 kN/m 2 , y sat1 = 16 kN/m3 ; su 2 =
20 kN/m 2 , y sat 2 = 17 kN/m3 ; su3 = 30 kN/m 2 , y sat 3 = 18 kN/m3 ; and D = 5 m. Compute the
factor of safety (Fb) against base shear failure using both the load factor and strength factor
earth pressure equilibrium methods.
Figure P6.1
Excavation 291
Figure P6.5
6.6 Redo problem 6.5 using the slip circle method, Terzaghi’s method, and Bjerrum and Eide’s
method.
6.7 Refer to Figure P6.5. The excavation width is 30 m. Assume the following parameters:
He = 10.0 m, Hp = 11.0 m; h1 = 2.0 m, h2 = 4.0 m, h3 = 4.0 m; su1 = 12 kN/m 2 , y sat1 = 18 kN/m3 ;
su 2 = 25 kN/m 2 , y sat 2 = 19 kN/m3 ; su3 = 30 kN/m 2 , y sat 3 = 18 kN/m3; and D = 8 m. Com-
pute the factor of safety (Fb) against base shear failure using both the load factor and
strength factor earth pressure equilibrium methods.
6.8 Redo problem 6.7 using the slip circle method, Terzaghi’s method, and Bjerrum and Eide’s
method.
He = 9.0 m, Hp = 11.0 m; h1 = 2.0 m, h2 = 3.0 m, h3 = 4.0 m, h4 = 4.0 m; su1 = 10 kN/m 2 ,
y sat1 = 16 kN/m3 ; su 2 = 20 kN/m 2 , y sat 2 = 17 kN/m3 ; su3 = 30 kN/m 2 , y sat 3 = 18 kN/m3 ;
and D = 5 m. Compute the factor of safety (Fb) against base shear failure using both the
load factor and strength earth pressure equilibrium methods, slip circle method, Terzaghi’s
method, and Bjerrum and Eide’s method.
He = 10.0 m, Hp = 11.0 m; h1 = 2.0 m, h2 = 4.0 m, h3 = 4.0 m, h4 = 4.0 m; su1 = 12 kN/m 2 ,
y sat1 = 18 kN/m3 ; su 2 = 25 kN/m 2 , y sat 2 = 19 kN/m3 ; su3 = 30 kN/m 2 , y sat 3 = 18 kN/m3 ;
and D = 8 m. Compute the factor of safety (Fb) against base shear failure using both the
load factor and strength factor earth pressure equilibrium methods, slip circle method and
Terzaghi’s method, and Bjerrum and Eide’s method.
6.11 Refer to problem 6.1. If the groundwater level behind the wall and in front of the wall
are on the ground surface and on the excavation surface, respectively, seepage will occur.
292 Excavation
Figure P6.9
Compute the factor of safety (Fs) against sand boiling using Terzaghi’s method and Har-
za’s method, assuming excavation width equal to 60 m.
6.12 Refer to problem 6.2. If the groundwater level behind the wall and in front of the wall
are on the ground surface and on the excavation surface, respectively, seepage will occur.
Compute the factor of safety against sand boiling (Fs) using Terzaghi’s method and Har-
za’s method, assuming excavation width equal to 52 m.
6.13 Refer to problem 6.5. If the pore water pressure head in sand is also on the ground surface,
compute the factor of safety (Fup) against upheaval failure.
6.14 Refer to problem 6.7. If the pore water pressure head in sand is also on the ground surface,
compute the factor of safety (Fup) against upheaval failure.
6.15 Refer to problem 6.5. If the retaining wall is the sheet pile and the spacing between two
adjacent struts is equal to 5 m, determine the following:
a Draw the apparent earth pressure.
b Compute the strut loads at the struts.
c Determine the section modulus required for the sheet pile.
d Determine the section modulus required for the wale at the second level.
6.16 Refer to problem 6.7. If the retaining wall is the sheet pile and the spacing between two
adjacent struts is equal to 5 m, determine the following:
a Draw the earth pressure envelope.
b Compute the strut loads at the struts.
c Determine the section modulus required for the sheet pile.
d Determine the section modulus required for the wale at the second level.
Excavation 293
6.17 Refer to problem 6.5. Assuming the stiffness of the retaining wall per unit meter is 65,000
kN-m2. Estimate the maximum wall deflection induced by excavation, and plot the pos-
sible settlement profile using Clough and O’Rourke’s method.
6.18 Refer to problem 6.7. Assuming the stiffness of the retaining wall per unit meter is 175,500
kN-m2. Estimate the maximum wall deflection induced by excavation, and plot the pos-
sible settlement profile using Clough and O’Rourke’s method.
References
Bjerrum, L. and Eide, O. (1956), Stability of strutted excavation in clay. Geotechnique, Vol. 6, pp. 32–47.
Clough, G.W. and O’Rourke, T.D. (1990), Construction-induced movements of insitu walls, Design and
Performance of Earth Retaining Structures, ASCE Special Publication, No. 25, pp. 439–470.
Harza, L.F. (1935), Uplift and Seepage Under Dams in Sand, Transaction, ASCE, No. 100, pp. 1352–1406.
Hsieh, P.G. and Ou, C.Y. (1998), Shape of ground surface settlement profiles caused by excavation. Cana-
dian Geotechnical Journal, Vol. 35, No. 6, pp. 1004–1017.
Hsieh, P.G. and Ou, C.Y. (2018), Mechanism of buttress walls in restraining the wall deflection caused by
deep excavation. Tunneling and Underground Space Technology, Vol. 82, pp. 542–553.
Hsieh, P.G., Ou, C.Y. and Lin, Y.L. (2013), Three-dimensional numerical analysis of deep excavations with
cross walls. Acta Geotechnica, Vol. 8, No. 1, pp. 33–48.
Hsieh, P.G., Ou, C.Y. and Shih, C. (2012), A simplified plane strain analysis of the lateral wall deflection
for excavations with cross walls. Canadian Geotechnical Journal, Vo. 49, pp. 1134–1146.
JSA (1988), Guidelines of Design and Construction of Deep Excavations, Japanese Society of Architecture.
Lim, A., Ou, C.Y. and Hsieh, P.G. (2019), An innovative earth retaining supported system for deep excava-
tion. Computers and Geotechnics, Vol. 114, p. 103135.
Mana, A.I. and Clough, G.W. (1981), Prediction of movements for braced cut in clay. Journal of Geotech-
nical Engineering Division, ASCE, Vol. 107, No. 6, pp. 759–777.
NAVFAC DM7.2 (1982), Foundations and Earth Structures, USA: Department of the Navy, Alexandria,
Virginia.
Ou, C.Y. (2022), Fundamentals of Deep Excavations, London: CRC Press, Taylor and Francis Group.
Ou, C.Y. and Hsieh, P.G. (2011, December), A simplified method for predicting ground settlement profiles
induced by excavation in soft clay. Computers and Geotechnics, Vol. 38, pp. 987–997.
Ou, C.Y., Hsieh, P.G. and Chiou, D.C. (1993), Characteristics of ground surface settlement during excava-
tion. Canadian Geotechnical Journal, Vol. 30, pp. 758–767.
Ou, C.Y., Hsieh, P.G. and Lin, Y.L. (2011), Performance of excavations with cross walls. Journal of Geo-
technical and Geoenvironmental Engineering, ASCE, Vol. 137, No. 1, pp. 94–104.
Ou, C.Y., Hsieh, P.G. and Lin, Y.L. (2013, May), A parametric study of wall deflections in deep excava-
tions with the installation of cross walls. Computers and Geotechnics, Vol. 50, pp. 55–65.
Ou, C.Y., Liao, J.T. and Lin, H.D. (1998), Performance of diaphragm wall constructed using top-
down method. Journal of Geotechnical and Geoenvironmental Engineering, ASCE, Vol. 124, No. 9,
pp. 798–808.
Peck, R.B. (1943), Earth pressure measurements in open cuts Chicago (III) subway. Transactions, ASCE,
Vol. 108, p. 223.
Peck, R.B. (1969), Advantages and limitations of the observational method in applied soil mechanics.
Geotechnique, Vol. 19, No. 2, pp. 171–187.
Peck, R.B., Hanson, W.E. and Thornburn, T.H. (1977), Foundation Engineering, New York: John Wiley
and Sons.
Pratama, I.T., Ou, C.Y. and Ching, J. (2019), Calibration of reliability-based safety factors for sand boiling
in excavations. Canadian Geotechnical Journal, Vol. 57, No. 5, pp. 742–753.
Reddy, A.S. and Srinivasan, R.J. (1967), Bearing capacity of footing on layered clay. Journal of the Soil
Mechanics and Foundations Division, ASCE, Vol. 93, No. 2, pp. 83–99.
294 Excavation
Skempton, A.W. (1951), The bearing capacity of clays. Proceeding of Building Research Congress, Vol.
1, pp. 180–189.
Terzaghi, K. (1922), Der Grundbrunch on Stauwerken und Seine Verhutung. Die Wasserkraft, Vol. 17,
pp. 445–449, Reprinted in From Theory to Practice in Soil Mechanics, New York: John Wiley and Sons,
pp. 146–148, 1961.
Terzaghi, K. (1943), Theoretical Soil Mechanics, New York: John Wiley & Sons, Inc.
Terzaghi, K., Peck, R.B. and Mesri, G. (1996), Soil Mechanics in Engineering Practice, New York: John
Wiley and Sons.
TGS (2022), Design Specifications for the Foundation of the Building, Taiwan Geotechnical Society, Taipei.
Chapter 7
Pile foundations
7.1 Introduction
Pile foundations, as deep foundations, are commonly used for high-rise buildings, bridges, and
structures under complex loading conditions (such as retaining walls and offshore wind turbine
foundations), as shown in Figure 7.1. Piles can be driven or bored into the ground. Notably,
bored piles are usually called drilled shafts in some areas, such as North America. The functions
of piles are as follows:
a. Transferring loads from the superstructure through weak compressible strata or through
water onto stiffer or more compact and less compressible soils or onto rocks.
b. Carrying uplift loads when used to support tall structures subjected to overturning forces
(from winds or waves).
c. Carrying combinations of vertical and horizontal loads to support retaining walls, bridge
piers and abutments, and machinery foundations.
The capacity of piles is highly affected by the installation methods, ground conditions, and
loading types. In this chapter, the pile installation, estimation of pile load capacity and settle-
ment, and in situ pile load test will be introduced. Importantly, the estimation methods for pile
load capacity introduced in the following sections are based on theories, field tests, and model
tests, and in situ pile load tests should be carried out if possible. The pile load test result provides
a check of the estimation.
7.2 Classification of piles

Piles can be generally classified in a number of ways, such as by function, material, and instal-
lation method. Each classification will be introduced in detail in the following sections.
7.2.1 Functions of piles

Piles provide resistance to loadings from the superstructures, which include both vertical (com-
pression and tension) and lateral loadings. Thus, piles can be classified based on their functions
as follows:
a. Compression pile. This category represents the majority of pile foundations. Piles resist the
vertical downward loadings from the superstructures, as shown in Figure 7.2. The resistances
DOI: 10.1201/9781003350019-7
296 Pile foundations
Figure 7.1 Functions of piles: (a) foundations for high-rise buildings, (b) foundations for
bridges, and (c) in retaining walls.
Pile foundations 297
Figure 7.1 (Continued)
Figure 7.2 Pile in compression: (a) foundations for high-rise buildings and (b) foundations for
bridges.
of the piles are provided by the mobilized side frictional force and end-bearing capacity. Set-
tlements occur when loading is applied; thus, the allowable settlement is an important design
factor for the compression pile.
b. Tension pile. This pile is also called an uplift pile and provides resistance against vertical
upward forces in transmission line towers and foundations subjected to floating forces, as
shown in Figure 7.3. The resistance is mainly from the tensile strength of the surrounding
soils, which is generally lower than the compressive strength. Thus, the design load capacity
of tension piles is generally lower than that of compression piles.
c. Laterally loaded pile. The laterally loaded pile can be classified into two subcategories, that
is, pile-to-soil and soil-to-pile. As shown in Figure 7.4a, piles can be used as the foundation
of wind turbines and to transmit the lateral loads from the wind turbine to soils. On the other
hand, in Figure 7.4b, the piles are used as the retaining structures, which resist the lateral
forces from the soil. The design considerations of lateral piles are the bending moment and
deflections of piles and the lateral resistance of soils, which determine the ultimate horizontal
force of piles.
7.2.2 Materials
In addition to their function, piles can also be classified by their materials, such as timber, steel,
and concrete. The selection of pile material depends mainly on the required function, durability,
and local supplier capacity. Common materials for piles are described here:
a. Timber piles are light and easy to handle. Timber comprises excellent natural materials in
regions with rich timber resources; thus, they can be quite competitive in terms of cost.
Even though timbers are convenient to collect and manufacture into piles, their dimension
and loading capacity are limited due to their natural texture and defects. A timber pile head
is damaged easily during the pile-driving process. Timber is effective when situated wholly
below the groundwater level.
Figure 7.3 Pile in tension.
Vibrating oscillator
(a)
Figure 7.4 Laterally loaded pile: (a) pile-to-soil and (b) soil-to-pile.
(b)
b. Steel piles are robust, easy to handle, and capable of carrying high compressive loads when
driven onto a hard stratum. Steel piles have many shapes and cross sections, such as plain
tubes, box sections, box piles built up from sheet piles, H-sections, and tapered and fluted
tubes. However, compared to concrete piles, the cost is high.
c. Concrete piles are commonly used in practice due to their cost and performance. The instal-
lation of concrete piles is very flexible. Both precast and cast-in-place are available.
7.2.3 Method of installation

The pile load capacity is highly dependent on its installation. Installation with a driving device
usually causes larger displacement to surrounding soils, namely, displacement piles, and thus
compacts or squeezes the soils. Installation with an opening prior to inserting the pile is usually
called a replacement pile, where the soils are removed first and then the pile is installed. The
installation of piles will have different effects on the surrounding soils, which depend on the soil
Figure 7.5 Method of pile installation.

Source: Redrawn from Poulos and Davis (1980).
types and installation methods. The effects will be introduced in the next section. The classifica-
tion of pile installation is shown in Figure 7.5 and explained in what follows.
a. Displacement pile:
i. Large displacement
All types of driven piles, such as solid section piles or hollow section piles with a closed
end, are driven or jacked into the ground and thus displace the soil.
ii. Small displacement
Also driven or jacked into the ground but have a relatively small cross section, such as a
hollow section with an open end, H- or I-section, pipe, or box steel pile.
b. Replacement pile (non-displacement pile): Remove soils by boring first and then fill the
borehole (either with casing or without casing) with concrete.
7.3 Installation of piles and its impact on soils

The ground condition and installation of piles predominantly determine the load capacity of
piles. The installation of piles affects the status of the surrounding soils and thus results in dif-
ferent load capacities. In this section, installation methods and their effects on both cohesive and
cohesionless soils will be introduced.
7.3.1 Driven pile

Driven installation is commonly used for steel piles and precast concrete piles. Figure 7.6 shows
the precast concrete pile with a sharp tip and corresponding driven equipment. The main process
Figure 7.6 Driven piles and equipment: (a) pile head, (b) pile tip, and (c) pile driving equipment.
Source: Provided by CTCI Resources Engineering Inc.
for a driven pile includes transportation, unloading, placement, lifting, positioning, driving, and
welding if necessary. Handling stress should be considered in driven piles due to the mentioned
handle process. Piles can be driven by a drop hammer or vibrating devices. For prefabricated
driven piles, the advantages are as follows:
a. Good quality on pile material with inspections before driving

b. Not liable to squeezing or necking, especially in soft ground
c. Construction operation not affected by groundwater
Potential problems include the following:
a. Noise and vibration due to driving are high.

b. Need to be jointed for the desired length and may break during driving.
c. May suffer unseen damage.
d. Displacement of soil during driving may lift adjacent piles/structures.
e. Jointed precast piles may not be suitable for tension and lateral loads.
For driven and cast-in-place displacement piles, the advantages and disadvantages are as
follows.
Advantages:
a. Easily adjust length.

b. Driving tube driven with a closed end to exclude groundwater.
c. Enlarging the base is possible.
Disadvantages:
a. Concrete liable to be defective in soft squeezing soils where withdrawable tube types are
used.
b. Noise and vibration due to driving are high.
c. Displacement of soil during driving may lift adjacent piles/structures.
7.3.2 Effect of pile driving in clays

When driving piles into clayey soils, the undrained shear strength of clays, excess pore water
pressures, and soil structures will be affected and thus impact the performance of piles, espe-
cially in the short term. The effects of pile driving in clays can be classified into four major
categories (de Mello, 1969):
a. Remolding or partial structural alteration of the soil surrounding the pile

b. Alteration of the stress state in the soil near the pile
c. Dissipation of the excess pore water pressure around the pile
d. Long-term strength regain in the soil
The phenomena will be introduced in detail next.

Influence on the soil strength and pile capacity

Driving piles into clay will initially cause some loss in undrained shear strength because of
the “remolding” (disturbance due to pile driving) of clays. After the completion of driving, the
strength of the soil and pile capacity will increase because of the following:
a. Regain of undrained shear strength (structural bonds partially restored)

b. Dissipation of excess pore water pressure
As shown in Figure 7.7, the pile load capacity is regained with the elapsed time after pile driv-
ing. For some cases, it takes one hundred hours (one to several weeks) to recover the load capac-
ity to 90% of the maximum value. For this reason, the pile load test is usually performed at one
to two weeks after pile driving.
Pore pressures developed during driving

During the driving of a pile in clays, soils were subjected to shear forces along the pile surface.
When the pile punched into the clay, the surrounding soils were squeezed, and thus, pore water
pressure was generated. According to the excess pore water pressure measurement during pile
driving shown in Figure 7.8, the maximum excess pore water pressure can be 1–2 times the
effective stresses on the pile surface and 3–4 times the effective stresses at the pile base. The
excess pore water pressure gradually decreases with the distance to the center of the pile, r.
The distance r is normalized by the radius of the pile a, as shown in Figure 7.8. Based on field
observations, the excess pore water pressure remains at the maximum value ( ∆um) within a
Figure 7.7 Increase in load capacity over time.

Source: Nishida (1962).
r/a
Figure 7.8 Distribution of excess pore water pressure at different locations.

certain distance, R, which is related to the clay type. Based on the observation shown in Figure 7.8,
a simple model for the distribution of excess pore water pressure ( ∆u ) is proposed in eq. 7.1
and shown in Figure 7.9.
Within the distance R:
Au = Aum (7.1a)
Beyond the distance R (Nishida, 1964; Kishida, 1967):
(r R) (7.1b)
2
Au = Aum
Where:
R ≈ 3a ~ 4a for insensitive clay.

R ≈ 8a for sensitive clay.
a = the radius of the pile.
Figure 7.9 Pore pressures developed while driving beyond the distance R.
Notably, for pile groups, the pore pressure distributions around individual piles may be super-
posed, thus decreasing the pile capacity.
7.3.3 Effect of pile driving in sands
Single piles
Pile driving is a process of vibrating or hammer dropping that causes energy to propagate
through the soil. In loose sands, the load capacity of a pile is increased due to the compaction
effect. The relative density is also increased. The SPT-N and CPT cone resistances of the sands
increase after pile driving, as shown in the field tests. The effects of driving a pile near the pile
tip can be estimated as follows:
o1' + 40o
o2' = (Kishida, 1967) (7.2)
2
o
Where o1' and o2' are shown in Figure 7.10. Eq. 7.2 implies that when o1' > 40 , o2, might
decrease due to pile driving. If the SPT-N of soils after pile driving is known, φ can be esti-
mated by eq. 7.3 (Kishida, 1967) or eq. 2.13 (in Chapter 2).
o ' = 20N + 15o (7.3)

Figure 7.10 Effects of driving a pile on friction angle in sands.

Source: Kishida (1967).
Pile groups
When groups of piles are driven into loose sand, the sand around and between the piles
becomes highly compacted. The densification effect due to single pile driving may over-
lap if the pile spacing is small enough. When the pile spacing is less than 6 times the pile
diameter (6d), the efficiency of the group ( ξ ) will be greater than 1.0, which implies that
the ultimate capacity of the group is greater than the sum of the single pile capacities.
The group effect depends on the position of the observation points and the pile-driving
sequence. These values can be estimated by applying Kishida’s single pile approach (Fig-
ure 7.10 and eq. 7.2).
7.3.4 Bored and cast-in-place pile

The bored pile and cast-in-place pile need to create an opening in the ground and then insert
the pile body to the design depth. Protection of the borehole wall may be needed if the ground
condition is poor or the water level is high. The advantages and disadvantages are as follows.
Advantages
a. Length can be varied to suit variations in the level of the bearing stratum.
b. Material forming piles are not governed by handling or driving stress.
c. No appreciable noise or vibration.
d. No ground heaves.
Disadvantages
a. Squeezing or necking or bulges (as shown in Figure 7.11).
b. Local slumping of open bore due to groundwater seepage.
When the stability of the borehole wall is an issue during boring, the reverse circulation
drilling (RCD) pile and full casing pile are better installation methods for the bored pile. The
construction of the reverse circulation drilling pile is shown in Figure 7.12. During the drilling
process, high-pressure air is introduced to the annulus between the inner tube and the outer rod.
The air flows through the drill steel and powers the drill tool. As the air exhausts, it serves as
a circulating medium by carrying the cuttings from the surface of the bit directly through the
inside of the drill steel. As it exits the top of the drill stack, the air is guided into a cyclone, which
slows the cuttings, separates them from the air, and collects them, while the remaining waste
fluids are captured in an isolated, watertight containment bin. A rebar cage is inserted into the
borehole when the design depth is reached. Concrete is poured into the borehole from the bot-
tom, and then the pile forms.
Necking Crack - Break Bulging
Figure 7.11 Installation defects of bored piles.

Rebar cage
Lead in tube
Concrete
Sediment pond Casing
(a) (b)
Lead in tube
Sediment pond Sediment pond
Rebar cage
Rebar cage
Concrete
(c) (d)
Figure 7.12 Construction of the reverse circulation drilling pile: (a) setting RCD ring and drill-
ing, (b) installation of rebar cage, (c) bottom sludge treatment (air lift), and
(d) withdrawing the casing.
Source: Adapted from Tung Feng Construction Engineering Co. Ltd.
The full casing pile is another bored pile that provides good protection of the borehole wall
and prevents defects in the bored pile (as shown in Figure 7.11). Full casing means the bore-
hole is protected by casing from the ground surface to the bottom of the borehole, as shown
in Figure 7.13. The casing is welded when the total length of the pile is greater than a single
casing length.
Figure 7.13 Construction of the full casing pile: (a) drilling, (b) installation of casing, (c) instal-
lation of rebar cage, and (d) pouring concrete and withdrawing casing.
Source: Adapted from Tung Feng Construction Engineering Co. Ltd.
7.3.5 Effect of installing the bored pile
Effects on the surrounding ground

When installing a bored pile into soils, the effects are different and depend on the soil types. For
clayey soils, the effects are as follows.
a. Softening:
i. Absorption of moisture from the wet concrete.
ii. Migration of the water from the body of the clay toward the less highly stressed zone
around the borehole.
iii. Water is poured into the boring to facilitate the operation of the cutting tool.
b. Disturbance:
Especially the clay just beneath the pile case.
For sandy soils, the borehole wall may collapse if no casing or stabilized fluid is used. Piping is
another issue if there are no protection measures on the wall of the borehole.
Effects on piles
When installing a bored pile into soils, the effects on piles are:
a. Caving of the borehole, resulting in necking or misalignment of the pile.

b. Aggregate separation within the pile.
c. Buckling of the pile reinforcement.
7.4 Axial load capacity of piles
7.4.1 Load transfer mechanism

The load capacity of a pile is determined by the pile‒soil interaction affected by the soil char-
acteristics (soil type and/or existence of hard stratum) and installation method. Loads from the
superstructures are transferred into the soils, and the resistance to loadings comes from both the
frictional force and end bearing, as shown in Figure 7.14. The frictional force, or so-called shaft
resistance, comes from the interface between the pile and surrounding soils, which is dominated
by the shear strength of soils. The end bearing, or so-called base resistance, is considerable
when the pile penetrates into hard strata. The net ultimate load capacity, Qu, of a single pile is
generally expressed as:
Qu = Qsu + Qbu - W (7.4)
Where:
Qu = net ultimate load capacity.

Qsu = ultimate shaft resistance (frictional resistance).
Qbu = ultimate base resistance (point resistance).
W = weight of pile.
Figure 7.14 Load transfer mechanism of a pile: (a) end-bearing pile and (b) friction pile.
When a load Q is applied on the pile top and gradually increased, maximum frictional resist-
ance will be fully mobilized when the relative displacement between the pile and the soil reaches
5–10 mm (irrespective of the pile dimension). The maximum base resistance will not be mobi-
lized until the pile tip has moved approximate 10–25% of the pile width. In other words, the
frictional resistance is mobilized faster at low settlement level than that of base resistance.
The pile is classified as a friction pile when the resistance contribution from frictional force is
much greater than the end bearing, that is, Qbu  Qsu, Qu ≅ Qsu, for instance, bored pile in thick,
soft soils. On the other hand, the pile is treated as an end-bearing pile when the load resistance
contribution from the end bearing is more significant than the frictional force, that is, Qsu  Qbu,
Qu ≅ Qbu, for instance, pile tip in rocks/gravels.
7.4.2 End-bearing resistance

The ultimate base resistance, Qbu , is fully activated when the soil at the bottom of the pile
reaches the failure condition, that is, ultimate bearing failure; the failure surface is as shown in
Figure 7.15. The pile foundations are deep foundations, and the soil fails mostly in a punching
model. A triangular failure zone I is developed at the pile tip first, which is pushed downward
without generating any other slip surface. In dense materials, a radial shear zone II may par-
tially develop. Thus, the ultimate base resistance can be evaluated from bearing capacity theory
similar to that in shallow foundations. All necessary shape and depth factors are included in the
* *
bearing capacity factors, that is N c* N q Nγ .
Qbu = Ab ( cN c* + o vb N a* + 0.5y dNy* ) (7.5)
Where:
Ab = area of pile base.

c = soil cohesion.
σ vb = vertical stress in soil at the pile base.
N c* N q* Nγ* = the bearing capacity factors considering the angle of internal friction 𝜙 of the soil,
the relative compressibility of the soil, and the pile geometry.
d = pile diameter.
Notably, the notations used in eq. 7.5 are general terms; for example, the vertical stress σ vb
can represent either total stress or effective stress, depending on the ground material encoun-
tered. For saturated clay under undrained condition (short-term), all the properties in cal-
culating ultimate load capacity should be values for the undrained condition. For instance,
c ( = su , undrained shear strength), φ (= 0), and all stresses should be total stresses. For drained
or long-term ultimate load capacity, all parameters should correspond to the drained values, and
all stresses should be effective stresses. The difference will be illustrated in what follows when
the ultimate load capacity for clays and sands is calculated.
End-bearing resistance in clay

When a pile is driven into clayey soils, the end-bearing resistance should be calculated by und-
rained parameters. All the frictional angles, including the internal frictional angle φ and the
Figure 7.15 Ultimate bearing capacity failure at the pile tip.

Source: Redrawn from Das (2016).
interface frictional angle between the pile and soil δ , should be 0; the cohesion c should be the
undrained shear strength of the clay su ; and the stresses σ vb are total stresses.
If φ = 0 , then N q* = 1 and Ny = 0 .
*
The end-bearing capacity from eq. 7.5 is as follows:
Qbu = Ab ( cN c* + o vb N a* + 0.5y dNy* ) = Ab ( su N c* + o vb ) (7.6)
In many cases, Abo vb ~ W . Thus, the net end-bearing capacity can be calculated as follows:
Qbu,net = Ab ( su N c* ) (7.7)
*
The bearing capacity factor, N c , in eq. 7.7 has been investigated and proposed by several stud-
ies, as listed in Table 7.1. In general, N c* = 9.0 is a good value for the estimation.
End-bearing resistance in sand

When a pile is driven into sandy soils, the end-bearing resistance should be calculated with the
drained parameters. All the frictional angles, including the internal frictional angle φ and the
interface frictional angle between the pile and soil δ , should be drained values; the cohesion
c should be the drained cohesion ( c′ ); and the stresses o vb' are effective stresses. Based on
field observation (Vesic, 1967a, 1967b), both the average frictional and end-bearing resistances
become constant beyond a certain depth of penetration, as shown in Figure 7.16. The depth
where resistances become constant is called critical depth. The normal stress on the pile surface
is constant beyond the critical depth due to the arching effect. The critical depth varies between
4 to 60 times of pile width and depends on the friction angle of soils and construction method.
*
Table 7.1 Suggestion of bearing capacity factor N c
Proposed N c* value Note Reference
6.14~9.0 For circular area Skempton (1951)

( N c* = 9.0 has been
pile length pile length
( =0~ > 4) confirmed in tests in
pile diameter pile diameter London clay and widely
accepted in practice.)
5.7~8.0 From model test Sowers et al. (1961)
5.7~8.2 For expansive clay Mohan (1963)
7.4~9.3 For insensitive clay Ladanyi (1963)
Figure 7.16 End-bearing resistance with penetration depth.

NAVFAC DM-7.2 (1982) suggested 20 times of pile width as the critical depth. If c' = 0 , the
net end-bearing capacity in sand can be calculated as follows:
Qbu,net = Ab (o vb' N q* + y dNy* ) (7.8)
Where Nγ* and N q* are factors in eq. 7.5 revised by Vesić. When the ratio of pile length to pile
diameter exceeds 5, which is commonly seen in pile foundations, the γ dNγ* term becomes neg-
ligible and may be ignored. The net end-bearing capacity can be calculated as follows:
Qbu,net = Ab (o vb' N q* ) (7.9)
The coefficient N q* is a function of the friction angle and compressibility. The compressibility
effect is defined with a rigidity index, I r , of the soil (Vesic, 1977), as shown in eq. 7.10. N q* is
computed with eq. 7.11.
Shear modulus E
Ir = = (7.10)
Shear strength 2(1+ u )o vb' tan o '
Where:
E = modulus of elasticity of soil within the zone of influence.

µ = Poisson’s ratio of soil within the zone of influence.
o vb' = vertical effective stress at the toe elevation.
(1 + 2 K ) No
N q* =
3
'
( 90-o )T 4 sin o '
3 o'
No = e 180 tan 2 (45 + )I r 3(1+ sin o ') (7.11)
3 - sin o ' 2
Where:
K = coefficient of lateral earth pressure.

o '= effective friction angle, in degrees.
7.4.3 Frictional resistance

The ultimate frictional resistance along the pile surface, Qsu , is fully activated when sliding
displacement between the pile and soil occurs, that is, shear failure at the interface of the pile
and soil. Thus, the ultimate frictional resistance can be evaluated from the shear strength of the
pile and soil interface.
L
Qsu = | Ut a dz (7.12)
0
L
= | U (ca + o n tan s )dz
0
L
= | U (ca + K so v tan s )dz
0
Where:
τ a = pile‒soil shear strength.

ca = adhesion.
σ n = normal stress between pile and soil.
δ = angle of friction between pile and soil.
σ v = vertical stress.
K s = coefficient of lateral pressure.
U = pile perimeter.
L = length of pile shaft.
Notably, the notations used in eq. 7.12 are general terms; for example, the vertical stress σ v
can be either total stress or effective stress, which depends on the ground material encountered.
For saturated clay under undrained condition (short-term), all the properties in calculating
ultimate load capacity should be values for the undrained condition. For instance, ca is und-
rained pile‒soil adhesion, φ (= 0), φ (= 0), and all stresses should be total stresses. For drained
or long-term ultimate load capacity, all parameters should correspond to the drained values, and
all stresses should be effective stresses. The difference will be illustrated in what follows when
the ultimate load capacity for clays and sands is calculated.
Frictional resistance in clay

When a pile is driven into normal-consolidated clayey soils, the short-term frictional resistance
should be calculated by undrained parameters. All the frictional angles, including the internal
frictional angle φ and the interface frictional angle between the pile and soil φa , should be 0; the
adhesion ca should be the undrained pile‒soil adhesion; and the stresses σ v are total stresses.
If o = S = 0 , then tan s = 0 .
The ultimate frictional resistance capacity from eq. 7.12 is:

L L
Qsu = | U (ca + K so v tan s )dz = | U (ca )dz (7.13)
0 0
The pile‒soil adhesion ca is an undetermined parameter in eq. 7.13 when estimating the pile
load capacity. It varies with the soil type, pile type, and installation method. It is generally deter-
mined from pile load tests and some empirical methods. Two empirical methods are introduced
in the following. The undrained pile‒soil adhesion ca is determined based on the undrained
shear strength of soils ( a method) and the combination of mean effective vertical stress and
average undrained shear strength along the pile ( λ method).
a - method
In this method, the soil‒pile adhesion ca is assumed to be related to the undrained shear strength
of soils ca with a coefficient of a . That is:
a = ca / su (7.14)
Table 7.2 Empirical correlations of 𝛼 with s u

su
Pa α
≦0.1 1.00
0.2 0.92
0.3 0.82
0.4 0.74
0.6 0.62
0.8 0.54
1.0 0.48
1.2 0.42
1.4 0.40
1.6 0.38
1.8 0.36
2.0 0.35
2.4 0.34
2.8 0.34
Note: P a = atmospheric pressure ≒100 kN/m 2 or 2,000 lb/ft 2.
Source: Interpolated values from Terzaghi et al. (1996).
Empirical correlations of 𝛼 with su were established, as shown in Table 7.2. In general, softer
clay has a higher 𝛼 value. When su ≤ 20 kPa, 𝛼 is approximately 1.0. Once the soil‒pile adhesion
is determined, then the frictional resistance can be determined via eq. 7.14.
y - method
In this method, the frictional resistance is calculated with the average pile‒soil adhesion, which
takes into account the mean effective vertical stress between the ground surface and pile tip and
the average undrained shear strength along the pile. The average pile‒soil adhesion factor is:
ca o m'
= y( + 2) or ca = y (o m' + 2 su , avg ) (7.15)
su , avg su , avg
Where:
o m' = mean effective vertical stress between the ground surface and pile tip.
su , avg = average undrained shear strength along pile.
λ = coefficient of average pile‒soil adhesion (see Table 7.3).
Once the average soil‒pile adhesion ca is determined, then the frictional resistance can be
determined by substituting eq. 7.15 into eq. 7.13, which yields:
Qsu = y (o m' + 2 su ,avg ) As (7.16)
Where:
AS = pile surface area.

Table 7.3 Determination of the coefficient

of average pile‒soil adhesion
Embedment length, L (m) λ
0 0.5
5 0.336
10 0.245
15 0.200
20 0.173
25 0.150
30 0.136
35 0.132
40 0.127
50 0.118
60 0.113
70 0.110
90 0.110
B - method
In some cases, such as piles in stiff and overconsolidated clay or long-term behavior of NC
clay (refer to Section 2.2.4), the long-term pile capacity is an issue, and the frictional resistance
should be calculated with the drained parameters. An effective stress approach, the β method,
is thus proposed. It assumes that ca' = 0 and applies the effective stress. Then, eq. 7.12 becomes:
L
Qsu = | U ( K S o v' tan s )dz (7.17)
0
A coefficient β is thus defined as:
B = K s tan s (7.18)
To determine the coefficient β , the following suggestions can be used.
1. For NC clay (Burland, 1973):

B = (1 - sin o ') tan s (7.19)
Where o ' is the effective stress friction angle for the clay.
Thus, B = 0.24 ~ 0.29 for o ' = 20 ~ 30 .
2. For OC clay (Burland, 1973):

K S = K 0 = (1- sin o ') OCR (7.20)
s = or' = remolded friction angle.
3. Meyerhof (1976):
K s = 1.5 K 0 for driven pile in stiff clay.
K s (bored pile) = 0.5 K s (driven pile)
Frictional resistance in sand

When a pile is driven into sandy soils, the frictional resistance should be calculated by drained
parameters. All the frictional angles, including the internal frictional angle o ' and the interface
frictional angle between the pile and soil δ , should be the drained values; the cohesion c should
be the drained cohesion (c’); and stresses o v' are effective stresses. As mentioned in the calcu-
lation of the end-bearing capacity of a pile, the resistance force does not increase unlimitedly
along the pile. Due to the arching effect, the normal stress on the pile surface is constant beyond
a certain depth (L’), called the critical depth, as shown in Figure 7.17. The magnitude of the
critical depth is normally in the range of 15 to 20 times the pile diameter. If c' = 0 and ca' = 0 ,
then the frictional resistance in eq. 7.12 becomes:
L
Qsu = | U ( K S o v' tan s )dz (7.21)
0
When the effective stress o v' becomes constant beyond the critical depth, then the frictional
resistance is approximately constant as well.
Some empirical expressions have been proposed for the coefficients K s and tan δ in eq. 7.21.
Typically, the coefficients increase with the friction angle after pile driving. The β method is
also valid for cohesionless soils, where B = K s tan s .
1. tan δ
The coefficient of friction, tan δ , was measured with laboratory tests and correlated with the
effective friction angle of the soil, o ' . Table 7.4 lists some typical values of s o ' .
Figure 7.17 Critical depth of a pile driven in sands.

Table 7.4 Range of s o ' values for the soil‒pile interface
Driven pile type s o'
Concrete 0.8–1.0
Rough steel (i.e., step-taper pile) 0.7–0.9
Smooth steel (i.e., pipe pile or H-pile) 0.5–0.7
Timber 0.8–0.9
Source: From Kulhawy et al. (1983) and Kulhawy (1991).
Table 7.5 Range of K s K 0 values for the driven pile
Method of construction Ks K0
Driven pile—jetted 0.5–0.7

Driven pile—small displacement 0.7–1.2
Driven pile—large displacement 1.0–2.0
Source: From Kulhawy et al. (1983) and Kulhawy (1991).
2. K s
As defined earlier, K s is the coefficient of lateral pressure, that is, the ratio between the hori-
zontal and vertical effective stresses as shown in eq. 7.22.
o h'
KS = (7.22)
o v'
The pile driving process will cause considerable compression to the surrounding soils and
thus will result in a higher coefficient of lateral pressure, K s , than that in the at-rest condition,
K 0 . The ratio of K s K 0 is listed in Table 7.5. The possible maximum value of K s is K p , the
coefficient of passive earth pressure.
3. β
It is sometimes difficult to determine K s and tan δ separately. Thus, engineers often combine
K s and tan δ into β , which is determined from published references or by load test results.
β is approximately 0.3 for loose sands, while for dense sand, it might be as high as 1.0–1.5.
Gravels can have much higher values, perhaps as large as 3. De Nicola and Randolph (1993)
developed a relationship between β and Dr from static load tests on closed-end steel pipe piles
in cohesionless soils.
B = 0.18 + 0.65Dr (7.23)
Where Dr is the relative density of the soil expressed in decimal form.
7.4.4 Design with SPT data

The determination of pile load capacity presented earlier relies greatly on the reliability of soil
properties obtained from lab tests on samples. In this section, the pile load capacity determined
from field tests, including SPT and CPT, will be presented. However, those expressions may
apply only to the regions where the data are collected. One should seek empirical expressions
based on data from the local region where the pile is constructed.
SPT-N (Meyerhof, 1956; Soderberg, 1962)

A. For displacement piles in saturated sand:
N
Qu = Qbu + Qsu = 40 N p Ab + As (unit: force in tons, area in 𝑚2) (7.24)
5
Where:
N
< 10 tf m 2 .
5
N p = standard penetration number, N, at pile base.
N = the average value of N along the pile shaft.
Ab = area of pile base.
B. For small displacement piles (e.g., steel H-pile):
N
Qu = Qbu + Qsu = 40 N p Ab + As (unit: force in tons, area in 𝑚2) (7.25)
10
Where:
N
< 5 tf m 2 .
5
Ab = net sectional area of toe.
As = gross surface area of shaft (area of all surfaces of flanges and web for H-piles).
7.4.5 Uplift capacity
Driven pile
The nominal uplift capacity of driven piles, Qup,n , relies on the side friction capacity only. There
is no end bearing, and the contribution from the weight of the pile is typically small. Thus, the
selection of appropriate values of the side friction capacity is the most critical parameter in the
design of piles subjected to uplift. According to field test results, the side friction capacity in
uplift is 70–85% of the corresponding value for downward loading. The possible reasons are as
follows (Tomlinson, 1969):
• The soil fabric near the pile changes due to the uplift load.
• The soil loosens due to uplift load and the shear surface.
• The Poisson effect during tensile loading decreases the pile diameter, thus slightly reducing
the contact pressure.
Bored pile
Bored piles can be constructed with an enlarged (or belled) base. The purpose of an enlarged
base is usually to increase the base area, and thus, the downward load is transferred through the
end bearing. It also has the advantage of generating a higher upward load capacity by bearing
on top of the enlarged portion. However, the belled bored piles are no longer widely used due to
the safety concerns of workers and the required construction time. The alternative method is to
construct a longer straight bored pile rather than an enlarged base.
Comparison between compression and tension pile load capacity

Two rectangular (barrette) piles with dimensions of 2.7 m × 1.0 m × 49.2 m (width × thickness ×
length) were constructed at the same site in Taipei City, Taiwan. Static pile load tests for both
compression and tension were performed. Figure 7.18 shows the plan of the tested pile and the
soil profile. The pile load test result is shown in Figure 7.19, which clearly shows a lower fric-
tional resistance in the tension (or uplift) pile than that in the compression pile under a similar
settlement.
7.4.6 t-z curve

The t-z method (Kraft et al., 1981) is a widely used static analysis for developing pile load‒set-
tlement response at the pile head and vertical load distribution along the pile length. It considers
three major components (side friction, end bearing, and elastic compression) and the interac-
tions between them. Different side friction mobilization functions along the length of the pile
are employed in the method for stratified soil profiles. The notation for variables t and z is
inconsistent with the notation used in this book and usually causes confusion. The definitions of
notations are listed here:
t: mobilized side friction resistance

z: displacement at a depth of pile
q: mobilized net end-bearing resistance
(a)
Figure 7.18 Pile load test: (a) site plan and (b) soil profile.
(b)
The t-z method divides the pile into finite segments, as shown in Figure 7.20. The elastic prop-
erties for each segment are defined by the cross area, A, modulus of elasticity, E, and segment
length. The side friction resistance between each segment and the soil is modeled using a non-
linear spring whose load‒displacement behavior is defined by a t-z curve. The end bearing is
also modeled using a nonlinear spring, which is defined by a q-z curve.
Functions for t-z and q-z curves can be found in published references, which are usually
derived from static load tests or numerical analyses. If in situ load test data are available, curves
may be custom-fitted and then used in a t-z analysis to extrapolate the load test result. An exam-
ple of the t-z curve from the static pile load test is shown in Figure 7.21, where the site is
that mentioned in Section 7.4.5. The mobilized side friction resistance can be determined by
measuring the axial force taken by the pile along the pile shaft. The difference between the two
sequential measurements at depth is the mobilized side friction resistance in the pile segment.
To perform a t-z analysis, the settlement, δ , at the pile top is set as the independent variable, and
an initial value is chosen. The numerical model is then solved to obtain force equilibrium and a
value of the corresponding applied load, P. Repeated solutions with different δ values produce
a load‒settlement plot and thus calculate the ultimate pile load capacity. Commercial software
Figure 7.19 Pile load test results: (a) Tension test and (b) compression test.
Figure 7.20 Numerical model for the t-z method.

Source: Redrawn from Coyle and Reese (1966).
Figure 7.21 t-z curves at different depths derived from a static pile load test.
capable of performing t-z analysis are available. For more detailed analysis, refer to Coyle and
Reese (1966).
Example 7.1: Pile load capacity in sand

• A pile 20 m in length and 0.6 m in diameter is driven into a sandy layer with the following
properties: y = 18 kN m3, o ' = 33, and Es = 35MPa . The groundwater table is on the ground
surface. Determine the end-bearing resistance of the pile.
Solution
The end-bearing resistance of the pile in sand can be calculated by eq. 7.9.
Qbu,net = Ab (o vb' N q* )
1
The pile base area: Ab = T d 2 = 0.283m 2 .
4
The critical depth is 20 times the pile diameter, that is, 12 m.
The effective stress at the pile base equals to that at critical depth: o vb' + y 'z + (18 - 9.81) x 12 +
98.28 kPa
E 35 MPa
Ir = = = 210.922
2(1 + v )o vb' tan o ' 2 x (1 + 0.3) x 98.28 kPa x tan 33
'
( 90-o )T 4 sin o '
3 o'
No = e 180 tan 2 (45 + ) I r 3(1+ sin o ') = 138.663
3 - sin o ' 2
(1 + 2 K ) No ||1 + 2 (1 - sin o ' ) || No
N q* = = = 88.29
3 3
Qbu,net = Ab (o vb' N q* ) = 0.283 m 2 × 98.28 kPa × 88.29 = 2455 kN
Example 7.2: Pile load capacity in clay

• A pile 20 m in length and 0.6 m in diameter is driven into clayey layers, as shown in the
figure. Determine the end-bearing resistance and frictional resistance of the pile.
Su sat
Su sat
Su sat
Figure EX 7.2a Soil profile.

Figure EX 7.2b Vertical effective stress distribution.
Figure EX 7.2c Vertical effective stress and undrained shear strength distribution.
Solution
1. Qbu
Qbu = 9 su Ab
Ab = T (0.3) 2 = 0.283 m 2
The pile tip is at GL. ‒20 m; thus, su = 60 kPa .
Qbu = 9 su Ab = 9 x 60 x 0.283 = 152.82 kN
2. Qsu
Qsu = E f sU AL
U = T x 0.6 = 1.885 m
f s can be determined by three methods, that is, the a method, β method, and λ method.
a method
Depth (m) ∆L (m) su (kPa) a f s (kPa ) f sU ∆L (kN)
0–3 3 30 0.82 24.60 139.1

3–10 7 45 0.71 31.95 421.6
10–20 10 60 0.62 37.20 701.2
Q su = 1262
β method
Depth (m) ∆L (m) o v' (kPa ) K β f s (kPa ) f sU ∆L (kN)
0–3 3 12.29 0.71 0.41 5.04 28.5

3–10 7 53.24 0.61 0.35 18.63 245.8
10–20 10 122.85 0.5 0.29 35.63 671.6
Q su = 946
λ method
Qsu = y (o m' + 2 su ,avg ) As o m' = 81.9 kPa
;
su , avg = 201 ( su1 x AL1 + su 2 x AL2 + su3 x AL3 ) = 50.25 kPa
y = 0.173 from Table 7.3
Qsu = 0.173 ( 81.9 + 2x 50.25 ) x 1.885 x 20 = 1190 kN
7.5 Lateral loaded piles

Lateral loads from superstructures applied to vertical piles will be carried by the lateral resistance
of surrounding soils. When significant horizontal loads are applied to vertical piles, the pile head
(or the ground surface) should not have severe movements, that is, no failure should occur at the
upper part of the ground. Pile caps could be connected with horizontal concrete pads or beams to
minimize movement and provide sufficient lateral resistance. Responses of single vertical piles
subjected to lateral forces and evaluations of their capacities will be introduced in this section.
7.5.1 Classification of lateral piles

Two categories are used in evaluating piles subjected to lateral loads: short piles and long piles, as
shown in Figure 7.22. A short pile refers to a pile without sufficient embedment length that thus
behaves like a rigid body when subjected to lateral forces. The toe of a short pile tends to rotate
as it is loaded laterally at the pile head. A rotation point is assumed in the pile and results in the
rigid rotation of a short pile, as shown in Figure 7.22. The ultimate lateral load capacity of short
piles is controlled by the soils surrounding the pile. Some soils in front of the piles yielded and
thus contributed to the ultimate lateral resistance before the pile had any flexural deflections. In
contrast, the long pile is the pile embedded deeply into the ground and thus places restraints on
any rotations and/or horizontal movements at the pile toe. Several points in the pile with a deflec-
tion equal to zero will be found in a sufficiently long pile. The ultimate lateral capacity of long
piles is thus controlled by its flexural capacity, which is governed by the pile structural design.
To classify the short pile and long pile, the deflection characteristics governed by the soil‒pile
interaction are employed. The pile is treated as a beam, which is loaded laterally. The responses
of soil and pile interaction are solved by a nonlinear differential equation as well as equilibrium
and compatibility conditions. For detailed information, refer to Chapter 8 in FHWA-HRT-04-043
(2006). The governing equation for pile deflection, y, subjected to later loads is shown in eq. 7.26.
The lateral load and displacement were usually solved numerically. The eigenvalue for eq. 7.26
is β , which controls the deformation pattern of a lateral loaded pile and can be calculated by eq.
7.27. A dimensionless length factor β L (L is the pile length) is employed to classify lateral piles.
When B L < 1, the pile is treated as a short pile or rigid pile; B L > 3 indicates a long pile. Among
long piles, 3 > B L > 1 indicates a semirigid pile, and B L > 3 indicates a nonrigid pile. The lateral
load capacity is of interest to geotechnical engineers and will be introduced in the next section.
d4y
Ep I p + Es y = 0 (7.26)
dz 4
Es
B= 4
4E p I p (7.27)
Figure 7.22 Short and long lateral piles: (a) short pile and (b) long pile.
Source: Redrawn from Zhang et al. (2016).
Where:
E p = modulus of elasticity of the pile.

Es = subgrade modulus of soil reactions.
I p = moment of inertia of the pile in the direction of bending.
y = lateral deflection of the pile.
z == depth below the ground surface.
7.5.2 Lateral load capacity for rigid piles

The evaluation of the ultimate lateral load capacity for a vertical rigid pile follows the pro-
cedures proposed by Broms (1964a, 1964b, 1965), which are based on some assumptions,
for example, the pile behaves like a rigid body after being pushed by a horizontal force. The
prediction from Broms’s method agreed reasonably with field test results, which indicates
(a) free head
(b) fixed head

Figure 7.23 Failure mechanism for short piles.
Source: Redrawn from Broms (1964a, 1964b).
Figure 7.24 Distribution of soil resistance in a laterally loaded pile.

(a) free head
(b) fixed head
Figure 7.25 Suggested distribution of soil resistance in a laterally loaded pile in cohesive soils.
that the assumptions are justified. The evaluation method is divided into piles in cohesive
soils and granular soils with different pile head conditions, that is, free and fixed head.
When a short rigid pile is subjected to horizontal loading, the possible failure mechanisms
for both the free head and fixed head are shown in Figure 7.23. It is clear that the failure
is governed by the soil strength. For the free-head rigid pile, the soils above and below
rotation point A all reach the yielding condition (on different sides of the pile), as shown in
Figure 7.23 (a).
Rigid piles in cohesive soils: Free head and fixed head

Based on Figure 7.24, the ultimate resistance of cohesive soils pushed by a horizontally loaded
pile is in the range of 8 to 12 times the undrained shear strength ( su ). The lateral resistance
reaches the value at a depth of three times the pile diameter. Therefore, Broms (1964a, 1964b,
1965) suggested a distribution of soil resistance, as shown in Figure 7.25. An ultimate soil resist-
ance of 9su appears at a depth of 1.5 times the pile diameter. For a free-head short pile subjected
to a horizontal load, Broms assumed that the maximum bending moment occurs at a depth of
1.5d + f, as shown in Figure 7.25a, where the shear is zero. Thus, f = Pult 9su d , where Pult is
the ultimate applied pressure at the pile head. Assuming that all soils in the pile length yield,
the depth of the rotation point can be determined based on the force equilibrium. By taking
moments about the pile tip and the maximum moment location, the ultimate applied pressure
Pult and maximum moment M max can be calculated by solving eqs. 7.28 to 7.32 (Broms, 1964a,
1964b, 1965).
Pult
f = (7.28)
9su d
M max = Pult ( e + 1.5d + 0.5 f ) (7.29)
M max = 2.25su dg 2
(7.30)
g = L -1.5d - f (7.31)
|
( ) |
- ( 2 ( de ) + ( dL ) +1.5 ) |
0.5
Pult = 9 su d 2 | 4 ( de ) + 2 ( dL ) + 4 ( de ) ( dL ) + 6 ( de ) + 4.5 (7.32)
2 2
| |
For a fixed-head short pile subjected to a horizontal load, the soil resistance is shown in Fig-
ure 7.25b. Based on the force equilibrium and taking the moment at the pile tip, the ultimate
applied pressure Pult and maximum moment M max can be calculated by eq. 7.33 and eq. 34,
respectively. The dimensionless solutions are given in Figure 7.26 based on eqs. 7.32 and 7.33.
For the case of piles with fixed heads, the solutions imply that the moment restraint from the pile
cap is equal to that in the pile just below the cap (Broms, 1964a, 1964b, 1965).
Pult = 9 su d 2 ( L d - 1.5 ) (7.33)
M max =Pult ( L2 + 34d ) = 4.5su d 3 |( L d ) - 2.25| (7.34)

2
| |
e/d
Pult /sud2
L/d
Figure 7.26 Ultimate lateral load for short piles in cohesive soils.
Source: Redrawn from Broms (1965).
Rigid piles in granular soils: Free head and fixed head

Possible failure mechanisms for short piles in granular (cohesionless) soils are shown in Figure
7.27. The following assumptions are made by Broms (1964a, 1964b):
• The active earth pressure acting on the back of the pile is neglected.
• The distribution of passive pressure along the front of the pile is equal to 3 times the Rankine
passive pressure (from certain observations).
• The shape of the pile section has no influence on the distribution of the ultimate soil pressure
or ultimate lateral resistance.
• The full lateral resistance is mobilized at the movement considered.
For a free-head short pile subjected to a horizontal load, Broms (1964a, 1964b) assumed
that the maximum bending moment occurs at a depth of f, as shown in Figure 7.27a, where the
shear will be zero. Thus, f = 2 Pult 3K p y d , where K p is the coefficient of passive pressure
( K p = (1 + sin o ') (1 - sin o ') ). Based on the force equilibrium and taking the moment at the pile
tip, the ultimate applied pressure Pult and maximum bending moment M max can be calculated
by eq. 7.35 and eq. 7.36, respectively (Broms, 1964a, 1964b, 1965).
K p y dL2
Pult = (7.35)
2 (1 + e L )
3K p y dL3
M max = Pult ( e + L ) - = Pult ( e + 32 f ) (7.36)
2x3
For a fixed-head short pile subjected to a horizontal load, the soil resistance is shown in Fig-
ure 7.27b. Based on the force equilibrium and taking the moment at the pile tip, the ultimate
applied pressure Pult and maximum bending moment M max can be calculated by eq. 7.37 and
eq. 7.38, respectively. Based on eqs. 7.35 and 7.37, dimensionless solutions are given in Figure
7.28 (Broms, 1964a, 1964b, 1965).
Pult = 32 K p y dL2 (7.37)
M max = K p γ dL3 (7.38)
Example 7.4: Lateral pile load capacity

• A short pile 3 m in length and 0.6 m in diameter is subjected to a horizontal force at an eleva-
tion of 0.6 m. The soil is soft clay with su = 30kPa . Determine the ultimate lateral capacity
( Pult ) of the pile and the maximum moment ( M max ).
Solution
Based on the given conditions, we have d = 0.6m , e = 0.6m , L = 3.0m , and su = 30kPa .
Pult Pult
=f =
9su d 162
Pult
g = L - 1.5d - f = 2.1-
162
( P )
M max = Pult ( e + 1.5d + 0.5 f ) = Pult |1.5 + ult |
( 324 )
P )
2
(
M max = 2.25su dg 2 = 40.5 | 2.1 - ult |
( 162 )
By solving the preceding equations simultaneously, a quadratic equation can be obtained.
40.5Pult 2 + 66922.2 Pult - 4687309.62 = 0

Pult = 67.3 kN
(a) free head
(b) fixed head
Figure 7.27 Suggested distribution of soil resistance in a laterally loaded pile in cohesionless
soils.
e /L
P/Kpd3y
L/d
Figure 7.28 Ultimate lateral load for short piles in cohesionless soils.
Source: Redrawn from Broms (1965).
or
|
( ) |
- ( 2 ( de ) + ( dL ) + 1.5 ) | = 67.3 kN
0.5
Pult = 9 su d 2 | 4 ( de ) + 2 ( dL ) + 4 ( de ) ( dL ) + 6 ( de ) + 4.5
2 2
| |
( P ) ( 67.3 )
M max = Pult |1.5 + ult | = 67.3 |1.5 + | = 114.9 kN-m
( 324 ) ( 324 )
7.5.3 Lateral load capacity for nonrigid piles

For nonrigid piles, that is, long piles, rigid pile analyses do not provide accurate predictions due
to simplified assumptions used in the rigid-pile method. More theoretical lateral load evaluation
methods that consider the flexural rigidity of the piles, the soil’s responses to the lateral force,
and the soil‒pile interaction have been developed. These methods require more complex analy-
sis and the calculation of more parameters to yield a more precise prediction; thus, there are no
simple closed-form solutions. The finite element method is a good numerical tool to solve the
highly nonlinear soil‒pile interaction problem, but it is beyond the scope of this book. Instead,
the p-y method using the Winkler beam-on-elastic-foundation model is introduced here.
Figure 7.29 p-y method: (a) conceptual model, (b) analytical model, and (c) p-y curve.
Source: Redrawn from Reese et al. (1974).
p-y method
The p-y method considers the soil‒pile interaction by using a series of nonlinear springs, which
is similar to the method introduced for the mat foundation. The only difference is the orientation
of the soil‒pile interaction springs. They are in the horizontal direction, as shown in Figure 7.29a.
The p-y method has been calibrated and verified with existing lateral pile load test data and sup-
ported by numerous commercial software programs. It is a well-accepted and preferred method
for most practical design problems.
The first step of the p-y method is to divide the pile into finite intervals (or elements) with
a node at the end of each interval. The nonlinear springs, that is, the soils, are located at each
node, as shown in Figure 7.29b. The flexural rigidity of each element is defined by E p I p of
the pile, where E p is the modulus of elasticity of the pile and I p is the moment of inertia
of the pile in the direction of bending. The load‒deformation responses of each spring are
defined by a p-y curve, as shown in Figure 7.29c, which implies that the soil‒pile reaction at
any depth is proportional to the lateral displacement. Notably, the springs act independently.
When applying the p-y model with appropriate boundary conditions, the software will apply
lateral loads in increments and compute the shear, bending moment, and lateral deflection at
each element.
p-y curve and its factors

The definition of the lateral load-deflection responses of soil‒pile interaction, that is, p-y curves,
are the key to the p-y method. In the p-y curve, p stands for the lateral soil resistance per unit
length of the pile, and y is the lateral deflection. The p-y curve is usually nonlinear and is not like
the linear spring in the Winker beam-on-elastic-foundation model. The shape of the p-y curve at
a certain depth of the pile depends on the following factors:
• Soil properties
• Loading type (static, cyclic, or dynamic)
• Pile diameter and cross-sectional shape

• Interface properties between soils and piles
• Depth below the ground surface
• Pile construction method
• Group interaction effects
Although the factors of the p-y curves are listed here, their individual impacts are not well
quantified. The p-y curve is usually empirically back-calculated from static load tests or using
correlation based on soil properties in p-y software. To determine p-y curves in soft clays, Mat-
lock’s method (1970) is usually employed. This method is based on reference data and two
lateral load tests on 324 mm diameter steel pipe piles in soils with su = 38kPa and su = 15kPa
. Matlock’s p-y curve for static loading in soft clays is shown in eq. 7.39 and Figure 7.30. The
ultimate soil resistance per unit length of pile, Pult , is smaller than that calculated with eq. 7.40.
1
p ( y )
3
= 0.5 | | (7.39)
pult ( y50 )
( y av' g J )
Pult = | 3 + z + z | su d (7.40 a)
( su d )
Pult = 9 su d (7.40 b)
Where:
y av' g = average effective unit weight from the ground surface to the depth is considered.
𝐽 = 0.5 for soft clay, 0.25 for medium clay.
d = pile diameter.
z = depth below the ground surface.
The deflection, y50 , at which 0.5 Pult occurs is:
y50 = 2.5e 50 d (7.41)
ε 50 = strain at one-half (50%) of the ultimate compressive strength of the clay.

Typical values of ε 50 are shown in Table 7.6.
Example 7.5: p-y curve by Matlock’s method

o A 400 mm diameter pile is embedded into a saturated medium clay that has an average unit
weight of 15 kN m3 and an undrained shear strength of 30 kPa. The groundwater table is near
the ground surface. Use the Matlock method to develop the static curve at a depth of 5 m.
Solution
From Table 7.6, e 50 = 0.01.
y ' = 15 - 9.8 = 5.2 kN m3

y50 = 2.5e 50 d = 2.5 ( 0.01) ( 400mm ) = 10mm
( y av' g J )
Pult = | 3 + z + z | su d
( su d )
( 5.2 0.25 )
= |3+ x5+ x 5 | x 30 x 0.4
( 30 0.4 )
= 84 kN
m
p/pult
y/y50
Figure 7.30 Matlock’s p-y curve for static loading of a single pile in soft clays
Table 7.6 Typical ε 50 values in clays
Consistency of clay su (kPa ) ε 50
Very soft <12 0.02

Soft 12–24 0.02
Medium 24–48 0.01
Stiff 48–96 0.006
Very stiff 96–192 0.005
Hard >192 0.004
Source: Reese et al. (2006).

Alternatively:
Pult = 9 su d
= 108 kN
m
Thus, use Pult = 84 kN .

m
Curve equation:
1
P ( y )
3
= 0.5 | | < 1;
Pult ( y50 )
1
( y)
3
P = 42 | |
( 10 )
7.6 Pile load test

The methods of pile load capacity determination in Section 7.4 involved some assumptions,
local empirical experience, and uncertainties that make predictions that do not match the real
behavior of installed piles. Thus, in most large projects, a specific number of load tests must be
conducted on piles. The primary reason is the unreliability of prediction methods. The vertical
and lateral load-bearing capacity of a pile can be tested in the field by applying the loads to the
test pile directly. Loads higher than design values, for instance, 200% of design values, may be
applied. The detailed configuration and interpretation of the pile load test will be presented in
this section.
7.6.1 Configuration of pile load test

A pile load test consists of the tested pile, reaction piles, reaction beams, hydraulic jacks, and
monitoring systems, as shown in Figure 7.31. The function of the reaction piles and beams is
to provide the required reaction force to load the tested pile; thus, the capacity of the reaction
system should be greater than that of the tested pile. Loadings and displacements were measured
during the test via load cells, LVDTs, strain gauges, etc.
There are two types of static load tests: stress-controlled tests and strain-controlled tests.
The former applies a predetermined load (the independent variable) and measured settlements
(the dependent variable). The latter increases the load to maintain a constant rate of penetra-
tion (usually vary from 0.25 to 2.5 mm min ). ASTM D1143 describes both procedures for
compression piles, and ASTM D3689 covers tests on uplift (tensile) piles. There are two types
of stress-controlled tests, that is, the maintained load test (ASTM D1143 Procedure B) and
the quick test (ASTM D1143 Procedure A). As mentioned in ASTM D1143, test loads should
be applied following one of the procedures described in the following for each test method.
If feasible, the maximum applied load should reach a failure that reflects the total axial static
compressive load capacity of the pile. Both methods give a load–settlement plot, and details
are introduced here.
Figure 7.31 Configuration of a pile load test.
Maintain load test

In the maintain load test, the load increment is approximately 25% of the design load, and each
load is maintained for two hours or until the rate of settlement is less than 0.25 mm per hour. The
loading process is continued until 200% of the design load (for a proof test) or failure is reached.
For the proof test, the maximum load is held for a longer period for creep effect observation.
Then, the pile is progressively unloaded.
Quick test
The quick test is faster than the maintain load test. The load increment is approximately 5% of the
design allowable load, and each load is held for a predetermined time interval regardless of the
rate of pile displacement at the end of that interval. ASTM permits intervals of 4 to 15 minutes.
The test will continue until approximately 200% of the design load or failure is reached. The
quick test generally requires 4 to 6 hours to complete, while the maintain load test requires sev-
eral days to complete. Thus, the quick test is commonly adopted to shorten the total test duration.
7.6.2 Interpretation of test results

After the pile load test, a load‒settlement curve (Figure 7.32) will be obtained to determine the
ultimate load capacity of the pile. Several determination methods for the ultimate load capacity
are suggested and introduced in the following.
1. Davisson’s method:
It is a widely used method to determine the load capacity of a pile, such as in the AASHTO
and international building code (IBC).
d QL
S = 4+ + (unit: mm) (7.42)
120 AE
Where:
S = settlement at the pile head.

d = pile diameter (mm).
Q = the applied load (kN).
L = pile length (mm).
A = pile cross section area ( mm 2 ).
E = the Young’s modulus of the pile ( kN mm 2 ).
In eq. 7.42, the 4 mm (approximately 1.5 in) refers to the pile head settlement contribute
from shaft displacement at the ultimate load condition, and d 120 (approximately 0.01d)
refers to the pile head settlement contribute by base displacement at the ultimate load
condition. The sum of these two terms is the offset, and QL AE is the elastic deformation
of the pile. Thus, the ultimate load capacity is the intersection of the settlement curve and the
linear line in eq. 7.42, as illustrated in Figure 7.32.
Qult
Q
d
d AE
PL AE
S
d
Q
L
A
E
fc
Figure 7.32 Load‒settlement curve of a pile load test.

2. Fuller and Hoy method:

The ultimate load is the load where the slope = 0.140 mm/kN on the plot of the load‒settle-
ment curve.
3. Terzaghi’s method:
The ultimate load is the load where the settlement equals 10% of the pile diameter.
Example 7.6: Interpretation of pile load test results

• A pile load test was performed on a pile 30 m in length and 0.6 m in diameter. The test result
is shown in the figure that follows. Determine the ultimate load capacity of the pile.
Solution
Solutions with different methods are shown in Table EX 7.6 and Figure EX 7.6a and b.
Table EX 7.6
Method Pu (kN)
Davisson 5827
Fuller and Hoy 5739
Terzaghi 5994
Figure EX 7.6a Load‒settlement curve in the pile load test.

Figure EX 7.6b Ultimate load capacity determination.
7.6.3 Correlation of pile load test results with other in situ tests
The pile load test gives engineers detailed information on the pile‒soil interaction behaviors,
and the most important among them is the ultimate load capacity of the pile on-site. The authors
collected more than 200 pile load test data points. The ultimate load capacities determined by
Davisson’s offset method were extracted from the data. In addition to the pile load test, other
in situ test data were also collected, such as data for the standard penetration test and cone
penetration test. The ultimate pile load capacity is thus correlated to the SPT N value and cone
resistance qc , which are very common in situ test results. Figure 7.33 shows the correlation
between ultimate load capacity and SPT-N and qc . The SPT-N values used in Figure 7.33a
were the average between 10 pile diameters above and 4 pile diameters below the pile tip. The
length-to-diameter ratio, L / d , of the pile is used as another factor in the figure. It is clear that
a higher average SPT-N and a larger L / d yield a higher ultimate load capacity. However, some
scatters are also observed in the data. In Figure 7.33b, the average cone tip resistances, qc , avg ,
are calculated along the pile length. In the figure, it is also clear that a higher qc , avg and a larger
L / d yield a higher ultimate load capacity.
7.7 Settlement analysis of single piles

The total settlement of a pile under a vertical working load, Qw, is given by:
Se = Se (1) + Se ( 2 ) + Se (3) (7.43)
Where:
qult N(L/d)
qult
N(L/d)
(a)
qult qc,avg L/d

qult
qc,avg L/d
(b)
Figure 7.33 Correlation between ultimate load capacity and in situ test data: (a) SPT-N and (b) qc .
Se(1) = elastic settlement of the pile.

Se( 2 ) = settlement of the pile caused by the load at the pile tip.
Se(3) = settlement of the pile caused by the load transmitted along the pile shaft.
(Qwp + E Qws ) L
Se(1) = (7.44)
AP E p
Where:
Qwp = load carried at the pile tip under working load conditions.
Qws = load carried by frictional resistance under working load conditions.
ξ varies between 0.5 and 0.67 and will depend on the nature of the distribution of the unit fric-
tion (skin) resistance along the pile shaft.
Qwp d
Se( 2 ) = (1 - u s2 ) I wp (7.45)
AP Es
Where:
d = width or diameter of the pile.

Es = modulus of elasticity of soil at or below the pile tip.
µ s = Poisson’s ratio of soil (refer to Table 2.4).
I wp = influence factor ≈ 0.85.
Qws d
Se(3) = ( ) (1 - u s2 ) I ws (7.46)
UL Es
Where:
U = perimeter of the pile.

L= embedment length of the pile.
I ws = influence factor = 2 + 0.35 L d (Vesic, 1977).
7.8 Negative skin friction

When piles are driven through soft layers (such as fill layers or soft clay) into firmer layers, the
piles are subjected to extra loads caused by negative skin friction in addition to the applied load
on the pile head. Negative skin friction is a downward drag acting on a pile due to the downward
movement, that is, settlement, of the surrounding compressible soil relative to the pile. Figure 7.34
shows the pile passing through a recently constructed cohesive soil fill. The soil below the fill
is completely consolidated under its overburden pressure. The determination of negative skin
friction is still being studied, and some models have been proposed. Figure 7.35 shows a neutral
plane model that assumes that a natural plane exists at a certain depth where the negative and
positive friction forces are identical.
Filled-up
soil
Figure 7.34 Illustration of negative skin friction.
Figure 7.35 Neutral plane model for negative skin friction: (a) pile settlement, (b) load trans-
fer, and (c) axial compression load in the pile.
Source: Redrawn from Timothy et al. (2014).
The negative skin friction force transfer to the pile depends on the pile material, the types
of soils, and the amount and rate of relative movement between the soil and the pile. The soil
actively settles and drags the pile downward; thus, a small relative movement on the order of 10
mm is necessary to fully trigger negative skin friction. A simple estimation of the extra load on
a pile due to negative skin friction, which is related to the undrained shear strength and adhesion
factor of the soil, is expressed as:
Qns = As x su x a (7.47)
Where:
Qns = load due to negative skin friction.

As = circumferential area of the pile.
su = average undrained shear strength along the pile shaft.
a = adhesion factor.
Possible measures used to reduce the negative friction include using a smaller diameter for
the pile shaft compared to the pile tip, driving the pile inside a casing and filling a viscous mate-
rial between them, and coating the piles with bitumen. Bitumen coating has been proven to be an
effective measure to reduce negative skin friction based on a full-scale test reported by Bjerrum
et al. (1969). The negative skin friction measured at one year after pile driving on the reference
pile (without bitumen coating) is approximately 3,000 kN, whereas that on the pile with bitumen
coating is only 150 kN. Notably, the bitumen coating should be well protected during the pile
driving process to ensure a reduction in negative skin friction.
7.9 Group pile effect

In most cases, piles are used in groups to transmit the structural load to the soil. In such a
situation, a pile cap is constructed over group piles, as illustrated in Figure 7.36. Determining
the load-bearing capacity of group piles is extremely complicated and has not yet been fully
resolved. When piles are placed close to each other, a reasonable assumption is that the stresses
transmitted by the piles to the soil will overlap and reduce the load-bearing capacity of the piles,
as shown in Figure 7.37. Significant settlement or complete general shear failure of the group
piles can occur even when pile load tests made on a single pile have demonstrated satisfactory
performance.
A classic case history of group pile failure was reported by Terzaghi and Peck (1967). The
foundation of the Charity Hospital at New Orleans consisted of approximately 20,000 timber
piles that were 7.9 m long. The pile penetrated the soft clay and terminated in a 1.8 m thick
dense sand that was underlain by 50 m or more soft soil. A pile load test was performed prior to
construction and showed a 6.5 mm settlement under a 300 kN load. However, after the building
was constructed, which is under a working load of 150 kN, the group piles settled more than
300 mm within two years. Thus, the group pile load capacity cannot be estimated by summing
the load capacity of a single pile.
Ideally, the piles in a group should be spaced so that the load-bearing capacity of the group is
not less than the sum of the bearing capacity of the individual piles. In practice, the minimum
center-to-center pile spacing, D, is 2.5d and in ordinary situations is actually approximately 3 to
3.5d. The minimum spacing, which considers different soil types and pile types, is listed in Table 7.7
Table 7.7 Minimum spacing for group piles
Pile length (m) Friction piles in sand Friction piles in clay End-bearing piles
Less than 12 m 3d 4d 3d
12 to 24 m 4d 5d 4d
Greater than 24 m 5d 6d 5d
Note: d is the pile diameter or the largest side. The pile spacing is measured at the pile cutoff level, unless
a racking pile is used, in which case the spacing is measured at an elevation 3 m below the pile cutoff level.
Figure 7.36 Photo of group piles.
Figure 7.37 Influence of the group pile on load transfer.

Den Norske Pelekomite (1973). It is clear from the table that longer piles or piles in clays gener-
ally need more spacing to avoid reducing the group pile capacity.
An easy method to estimate the group pile capacity is the “efficiency formula.” The efficiency
of the load-bearing capacity of a group pile can be calculated with eq. 7.48. The efficiency factor
is hardly empirical; thus, great care should be taken when using the efficiency formula.
Qg (u )
n= (7.48)
EQ u
Where:
η = the group efficiency.

Qg (u ) = ultimate load-bearing capacity of the group pile.
Qu = ultimate load-bearing capacity of each pile without the group effect.
In general, for a group pile that is clay, η is less than unity, approximately 0.7, as sug-
gested by Tomlinson (1969), for a spacing of two pile diameters. For wider spacing, the piles
in clay fail individually, but the efficiency ratio reaches unity at a spacing of approximately
eight pile diameters. Group piles in sand tend to compact the soil around the pile; thus, the
ultimate failure load of pile groups will tend to increase. Figure 7.38 shows the variation in
group efficiency for a 3 × 3 group pile in sand. For loose and medium sands, the magnitude
of the group efficiency can be larger than unity primarily due to the densification of sand
surrounding the pile. The efficiency ratio is less than unity when pile groups are driven into
dense sand.
Problems
7.1 Explain the following terms (with plots if necessary):
a. t-z curve
b. p-y curve
c. pile load test
d. replacement pile
7.2 A 600 mm diameter closed-end steel pipe pile is driven to a depth of 17 m into the following
soil profile. The groundwater table is at a depth of 6.0 m. The modulus of elasticity of the
well-graded sand is 35 MPa, and Poisson’s ratio is 0.3. Assume that s = 2 o ' . Compute the
ultimate load capacity. 3
Depth (m) Soil layer Unit weight (kN/m3) Friction angle (°)
Moisture Saturated
0–2.0 Silty sand 18.0 19.0 28

2.0–7.0 Fine to medium sand 18.5 19.5 31
7.0–16.0 Silty sand – 19.1 29
16.0–18.0 Well-graded sand – 19.8 35
d/D
Figure 7.38 Variation in the efficiency of pile groups in sand.

Source: Based on Kishida and Meyerhof (1965).
7.3 An 800 mm diameter concrete pile is driven to a depth of 20 m into the following soil
profile. The groundwater table is at the ground surface. Compute the ultimate load capacity
with a , B , and y methods .
Depth (m) Soil Saturated unit Undrained shear Friction

layer weight (kN/m 3) strength (kPa) angle (°)
0–3.0 Silty clay 17.0 12 28

3.0–10.0 Clayey silt 17.5 20 29
10.0–14.0 Clay 17.8 50 30
14.0–25.0 Clay 18.1 80 31
7.4 A static pile load test is conducted on a rectangular pile with dimensions of 1.0 m × 49.0 m
(diameter ×length). The load‒settlement curve is shown in Figure P 7.4. Please determine
the ultimate load capacity by Davisson’s method.
7.5 As problem. 7.4, use other methods, such as Terzaghi’s method or Fuller and Hoy method,
to determine the ultimate pile load capacity.
7.6 A pile with diameter of 1.2 m and length of 6.0 m is embedded in a clay layer with su =
30 kPa. Determine the ultimate lateral load capacity of the pile if the pile head is fixed.
Figure P 7.4 Pile load test result.
7.7 As problem 7.6, determine the ultimate lateral load capacity of the pile if the pile head is
free and the height of the lateral load is 1.0 m.
7.8 As problem 7.6 and 7.7, calculate the maximum moment for both cases.
7.9 As problem 7.6, the same pile is embedded in a sandy layer with unit weight y sat = 20 kN m3
and friction angle of 32. Determine the ultimate lateral load capacity and maximum
moment of the pile if the pile head is fixed.
7.10 As problem 7.9, determine the ultimate lateral load capacity and maximum moment of the
pile if the pile head is free and the height of the lateral load is 1.0 m.
7.11 A large sign is to be supported on a single post connected to a 1.2 m diameter pile drilled
shaft foundation. The soil is a stiff clay with su = 80 kPa. The design wind load is 600
kN, and the center of the sign is 20 m above the ground. Using a factor of safety of 3.0,
compute the minimum required depth of embedment, L, and the maximum moment in the
drilled shaft, M max .
7.12 A 600 mm diameter pile is embedded into a saturated medium clay that has an average unit
weight of 18 kN m3 and an undrained shear strength of 50 kPa. The groundwater table
is near the ground surface. Use the Matlock method to develop the static curve at a depth
of 10 m.
7.13 A driven concrete pile with diameter of 1.0 m and length of 25 m is embedded
in saturated sand. Given the unit weight of sand y sat = 20 kN m3 , drained friction
angle o ' = 33 , and the frictional angle of sand-pile interface s = 25 , estimate the
ultimate frictional resistance of the pile. (Assume the effective earth pressure coef-
ficient is 1.2.)
7.14 As problem 7.13, estimate the ultimate end-bearing capacity of the pile.
7.15 As problem 7.13. A working load on the pile is 3,500 kN. Skin resistance carries 2,100
kN of the load, and end-bearing carries the rest. Estimate the total settlement of the pile.
(Assume the modulus of the elasticity of pile E p = 21x 106 kPa and a uniform distribu-
tion of skin friction along the pile.)
References
Bjerrum, L., Johannessen, I.J. and Eide, O. (1969), Reduction of skin friction on steel piles to rock. Pro-
ceeding of the 7th International Conference. in Soil Mechanics and Foundation Engineering, Mexico,
Vol. 2, pp. 27–34.
Broms, B.B. (1964a). The lateral resistance of piles in cohesive soils. Journal of the Soil Mechanics &
Foundation Divisions, ASCE, Vol. 90, No. 2, pp. 27–63.
Broms, B.B. (1964b). The lateral resistance of piles in cohesionless soils. Journal of the Soil Mechanics &
Foundation Divisions, ASCE, Vol. 90, No. 3, pp. 123–156.
Broms, B.B. (1965), Design of laterally loaded piles. Journal of SMFED, ASCE, No. SM3, pp. 79–99.
Burland, J.B. (1973), Shaft friction piles in clay—a simple fundamental approach. Ground Engineering,
Vol. 6, No. 3.
Coyle, H.M. and Reese, L.C. (1966), Load transfer for axially loaded pile in clay. Journal of the Soil Me-
chanics and Foundations Division, ASCE, Vol. 92, pp. 1–26.
Das, B.M. (2016), Principles of Foundation Engineering, SI, 8th edition, Boston: Cengage Learning.
De Mello, V.F.B. (1969), Foundations of buildings in clay. State-of-the art report. In Proceeding of the 7th
International Conference. Soil Mechanics and Foundation Engineering, Mexico City, SOA Volume,
pp. 49–136.
De Nicola, A.D. and Randolph, M.F. (1993), Tensile and compressive shaft capacity of piles in sand. Jour-
nal of the Geotechnical Engineering, ASCE, Vol. 19, No. 12, pp. 1952–1973.
Den Norske Pelekomite (1973), Veledning ved pelefundementering. Oslo: Norwegian Geotechnical Insti-
tute, Veiledning No. 1.
FHWA-HRT-04-043 (2006), A Laboratory and Field Study of Composite Piles for Bridge Substructures,
McLean: Office of Infrastructure R&D Federal Highway Administration.
Kishida, H. (1967), Ultimate bearing capacity of piles driven into loose sand. Soils and Foundations, Vol.
7, No. 3, pp. 20–29.
Kishida, H. and Meyerhof, G.G. (1965), Bearing capacity of pile groups under eccentric loads in sand.
In 2, Proceeding of the Fifth International Conference, Soil Mechanic and Foundation Engineering,
Montreal pp. 270–274.
Kraft, L.M., Ray, R.P. and Kagawa, T. (1981), Theoretical t-z curves. Journal of the Geotechnical Engi-
neering Division, ASCE, Vol. 107, No. 11, pp. 1543–1561.
Kulhawy, F.H. (1991), Drilled Shaft Foundations, Foundation Engineering Handbook, 2nd edition, Chap.
14. Ed. H.-Y. Fang, New York: Van Nostrand Reinhold.
Kulhawy, F.H., Trautmann, C.H., Beech, J.F., O’Rourke, T.D., McGuire, W., Wood, W.A. and Capano,
C. (1983), Transmission line structure foundations for uplift-compression loading, Rep. No. EL-2870,
Electric Power Research Institute, Palo Alto, CA.
Ladanyi, B. (1963), Evaluation of pressuremeter tests in granular soils. In Proceeding 2nd Panamerican
Conference Soil Mechanic and Foundation Engineering, Sao Paulo, 1, pp. 3–20.
Matlock, H. (1970), Correlation for design of laterally loaded piles in soft clay. In The Second Annual
Offshore Technology Conference, Houston, Texas: Onepetro.
Meyerhof, G.G. (1956), Penetration test and bearing capacity of cohesionless soils. Journal of the Soil
Mechanics and Foundation Division, ASCE, Vol. 82, No. SM1, pp. 1–19.
Meyerhof, G.G. (1976), Bearing capacity and settlement of pile foundations. Journal of Geotechnical and
Geoenvironmental Engineering Division, ASCE, Vol. 102, No. GT3, pp. 195–228.
Mohan, D., Jain, D.S. and Kumar, V. (1963), Load bearing capacity of piles. Geotechnique, London, Vol.
13, No. 1, pp. 76–86.
NAVFAC. (1982), Foundations and Earth Structures Design Manual 7.2, Department of the Navy Naval
Facilities Engineering Command, Alexandria, VA.
Nishida, Y. (1964), A basic calculation on the failure zone and the initial pore pressure around a driven pile.
In Proceeding of the First Regional Asian Conference on Soil Mechanics and Foundation Engineering,
the International Society of Soil Mechanics and Foundation Engineering, Tokyo, 1, pp. 217–219.
Poulos, H.G. and Davis, E.H. (1980), Pile Foundation Analysis and Design, New York: Wiley, c1980.
Reese, L.C., Cox, W.R. and Koop, F.D. (1974), Analysis of laterally loaded piles in sand. Offshore Tech-
nology Conference, Houston, Texas, May.
Reese, L.C., Isenhower, W.M. and Wang, S.-T. (2006), Analysis and Design of Shallow and Deep Founda-
tions, Hoboken, New Jersey: John Wiley and Sons.
Skempton, A.W. (1951), The Bearing Capacity of Clays, Vol. 1, London, Building Research Congress,
pp. 180–189.
Soderberg, L.O. (1962), Consolidation theory applied to foundation pile time effects. Geotechnique, Lon-
don, Vol. 11, No. 3, pp. 217–225.
Sowers, G.F., Martin, C.B., Wilson, L.L. and Fausold, M. (1961), The bearing capacity of friction pile
groups in homogeneous clay from model studies. Proceedings of the 5th International Conference on
Soil Mechanics and Foundation Engineering, Vol. 2, pp. 155–159.
Terzaghi, K. and Peck, R.B. (1967), Soil Mechanics in Engineering Practice, 2nd edition, New York: John
Wiley and Sons, p. 729.
Terzaghi, K., Peck, R.B. and Mesri, G. (1996), Soil Mechanics in Engineering Practice, 3rd edition, New York:
John Wiley and Sons, p. 549.
Timothy, C., Siegel, P.E., G.E., D.GE with Dan Brown and Associates PC (2014), Designing piles for
drag force. 43rd Annual Midwest Geotechnical Conference, October 1–October 3, 2014, Bloomington,
Minnesota: United State Departments of Transportation Federal Highway Administration.
Tomlinson, M.J. (1969), Foundation Design and Construction, 2nd edition, New York: Wiley, p. 785.
Vesic, A.S. (1967a). A study of bearing capacity of deep foundations. Final Report, Project B-189, Georgia
Institute of Technology, Atlanta, pp. 231–236.
Vesic, A.S. (1967b). Model testing of deep foundations in sand and scaling laws. In Proceedings North
American Conference on Deep Foundations, Panel Discussions, Session II, Mexico City.
Vesic, A.S. (1977), Design of pile foundations. National Cooperative Highway Research Program, Synthe-
sis of Practice No. 42, Transportation Research Board, Washington, DC, p. 68.
Zhang, Y., Andersen, K.H. and Tedesco, G. (2016), Ultimate bearing capacity of laterally loaded piles in
clay-Some practical considerations. Marine Structures, Vol. 50, pp. 260–275.
Chapter 8
Slope stability
8.1 Introduction
Predicting the stability of soil and rock slopes is a classic problem for geotechnical engineers and
plays an important role in designing embankments, dams, roads, tunnels, and other engineering
structures. Many researchers have focused on assessing the stability of slopes (Taylor, 1948;
Morgenstern, 1963; Fredlund and Krahn, 1977; Hoek and Bray, 1981; Goodman and Kieffer,
2000). However, the problem still presents a significant challenge to designers. Gravitational
and seepage forces cause instability in natural and man-made slopes. The latter frequently occur
in practice and may arise as a result of the construction of embankments, canals, excavations,
and earth dams. Slope stability is usually analyzed using limit equilibrium procedures, where a
failure surface is assumed before a factor of safety (Fs) calculation. For engineering designs, a
primary concern is whether the slope is safe at a given slope angle. This chapter introduces some
basic knowledge of slope stability problems for undergraduate students. The contents include
the slope failure modes, stability assessments, stability charts, and stabilization.
Generally, current design practices for slope stability rely largely on the value for Fs obtained
from the conventional limit equilibrium method or finite element method. The limit equilib-
rium method is used widely because it is relatively simple and easy to use. Conventional slope
stability analyses investigate the equilibrium of a mass of soil bounded below by an assumed
potential slip surface and above by the surface of the slope. Notably, simplifications are neces-
sary in all design methods. The assumed failure mechanisms are more or less crude approxima-
tions of the actual failure mechanism. Certain assumptions are also made regarding the slope
geometry and the loads acting on the slope. However, engineers sometimes misuse the limit
equilibrium method or finite element method because there is a lack of knowledge about method
backgrounds. In general, assumptions are a requirement because otherwise, the design methods
would be overwhelmingly complex and nearly impossible to use rationally. In a robust design
method, the necessary assumptions have very little influence on the end result.
8.2 Failure modes

Based on the geological structure and the stress state of soil and rock masses, certain slope fail-
ure modes appear to be more likely than others. Figure 8.1 displays some common slope failure
modes. A rockfall (Figure 8.1a) refers to quantities of rock falling freely from a cliff face. This
term is also used for the collapse of rock from the roof or walls of mines or quarry workings.
Figure 8.1b is a slope rotational shear failure mode. For the rotational shear failure mode, the
individual particles in a soil or rock mass should be very small compared to the size of the slope,
and these particles are not interlocked as a result of their shape (Hoek and Bray, 1981).
DOI: 10.1201/9781003350019-8
Slope stability 357
Figure 8.1 Different slope failure modes: (a) rockfall, (b) rotational failure, (c) toppling,
(d) planer slides.
358 Slope stability
Rotational shear failure generally occurs in soil slopes. The issue of whether a rock mass
can be considered heavily fractured is thus mostly a matter of scale. Rotational shear failure in
a large-scale slope would probably involve failure primarily along preexisting discontinuities,
with some portions of the failure surface possibly going through intact rock. Additionally, rota-
tion and translation of individual blocks in the rock mass would help create a failure surface.
The resulting failure surface would follow a curved path. The failure surfaces are drawn to be
relatively deep, but they could also be shallower.
Figure 8.1c shows the slope toppling failure mode. This type of failure is characterized by
a successive breakdown of the rock slope. Failure could initiate by crushing of the slope toe,
which in turn causes load transfer to adjacent areas that may fail. Obviously, the orientations
of discontinuities and in situ stresses in relation to the rock strength are important factors gov-
erning this failure mode. The presence of discontinuities in the rock mass can result in several
secondary modes of failure, as shown in Figure 8.2.
Figure 8.2 Common classes of toppling failures: (a) block toppling of columns of rock contain-
ing widely spaced orthogonal joints, (b) flexural toppling of slabs of rock dipping
steeply into the face, and (c) block flexure toppling characterized by pseudocon-
tinuous flexure of long columns through accumulated motions along numerous
cross joints.
Source: Redrawn from Goodman and Bray (1976).
Slope stability 359
Figure 8.1d shows the plane shear failure mode, which could be a single discontinuity (plane
failure), two discontinuities intersecting each other (wedge failure), or a combination of several
discontinuities connected together (irregular failure surface). A common feature of most failure
modes is the formation of a tension crack at the slope crest. For failure to occur, release surfaces
must be present to define the rock block moving in the lateral direction.
A more detailed description of the governing failure mechanisms for typically occurring
slope failures is given as well as those based on field observations. Notably, a real failure surface
would be a combination of different failure modes. For example, there may be a combination
of plane failures and rotational shear failures, with or without tension cracks at the slope crest.
8.3 Slope stability analyses

As mentioned previously, two-dimensional limit equilibrium analysis is the most popular
approach to evaluate slope stability. It is known that two-dimensional solutions utilized in design
will obtain a conservative evaluation for slope failure. With advances in computer hardware,
evaluating slope stability using finite element method numerical analysis has become more
prevalent and has gradually been widely adopted by engineers. The displacement finite ele-
ment technique is widely used for predicting the load‒deformation response and hence collapse
of geotechnical structures. In contrast, very few numerical analyses have been performed to
evaluate slope stability based on limit theorems. Limit analysis methods that produce solutions
in terms of limit load can be based on either the kinematically admissible velocity field (upper
bound) or the statically admissible stress field (lower bound). Due to the difficulty in computing
the stress field for a slope, most studies are based on the upper bound method. In recent years,
finite element upper bound (Lyamin and Sloan, 2002a; Krabbenhoft et al., 2005) and lower
bound (Lyamin and Sloan, 2002b) limit analysis methods have become popular approaches
in slope analyses. One of the merits of using the finite element method or finite element limit
analysis method is that the techniques can generate failure mechanisms automatically.
8.3.1 Factor of safety in the limit equilibrium method

For the simplest form of limit equilibrium analysis, only the equilibrium of moment is satisfied.
The sum of the forces acting to induce the sliding of parts of the slope is compared with the sum
of the forces available to resist failure. The ratio between these two sums is defined as the factor
of safety Fs, which is:
R
Fs = (8.1)
D
where:
R = resistance moment.
D = driving moment.
This simple definition of the safety factor can be interpreted in many ways. It can be expressed
in terms of loads, forces, moments, etc. The merit of a safety factor is that the stability of a slope
can be quantified by a number. According to eq. 8.1, a safety factor less than 1 indicates that fail-
ure is possible. If there are several potential failure modes or different failure surfaces that have
360 Slope stability
a calculated safety factor less than 1, the slope can fail. By using the limit equilibrium method,
the slip surface needs to be assumed prior to any safety factor calculation. Thus, the obtained Fs
would be unreasonable if the assumed failure mechanism is far from the real one.
8.3.2 Basic limit equilibrium methods

For the general concept of the limit equilibrium method, soil masses above the potential
failure surface are considered as a combination of vertical slices. Therefore, this approach
is called the “procedure of slices.” The actual number of slices and their width can be deter-
mined by the soil profiles, slope geometry, and assumed shape of the potential failure surface.
Importantly, the assumed failure surface could significantly influence the final obtained mag-
nitude of Fs.
Various approaches have been proposed based on the limit equilibrium concept. For the
assumed potential failure surface, some methods require circular failure shapes arbitrarily. Oth-
ers can work with noncircular shapes. In fact, it is very difficult to predict whether the failure
surface is circular or noncircular. As discussed in Section 8.2, an actual failure surface can be a
combination of different failure modes. Therefore, a reasonable assumption of a potential failure
surface relies highly on the judgment made by experienced engineers without monitoring data.
This issue is also a disadvantage of using the limit equilibrium method.
In general, the equilibrium of moments about the center of the circle for the entire soil mass
above the assumed failure surface is considered. A circular failure surface is shown in Fig-
ure 8.3, where the overturning moment can be expressed as:
M
d
= E Wi b i (8.2)
where Wi is the weight of the ith slice and b i is the horizontal distance between the center of
the circle and the center of the slice. Theoretically, b i should be the distance between the center
of the circle and the center of gravity of the slice. However, the difference can be neglected if
sufficient numbers of slices are used.
Figure 8.3 General configuration of slices for a circular failure surface.

Slope stability 361
In eq. 8.2, the moment arm b i can be expressed by eq. 8.3, and thus, the driving moment is
described by eq. 8.4.
bi = r sin a i (8.3)
M d = r E Wi sin a i (8.4)
The resisting moment is from the shear stress ( τ ) along the failure surface, which is on the
base of each slice. Based on the Mohr‒Coulomb failure criterion, the shear strength for each
slice ( s i ) can be expressed as:
si = c + a tan o (8.5)
The resisting moment will be:
M r = r E si A i (8.6)
where ∆li is the thickness of the ith slice. The factor of safety of the slope can be expressed as:
r E si A i E s Ai i
Fs = = (8.7)
r E Wi sin a i EW sin a
i i
Notably, eq. 8.7 is valid only for a circular failure surface.
8.3.3 Infinite slope

For cohesionless soils (c = 0), only the friction angle plays a role in the safety calculation. In
general, the failure surface depth will not significantly influence the factor of safety. It is pos-
sible to form a small failure depth, thus meeting the requirements for an infinite slope. This
condition would appear when there is a stronger soil or rock formation at shallow depths. In
Figure 8.4, the sliding soil mass has weight W = y Lz cosB , and thus, the shear force (V) and
normal force (N) can be expressed by eqs. 8.8 and 9, respectively.
V = W sin B = y Lz cos B sin B (8.8)

N = W cos B = y Lz cos 2 B (8.9)
Figure 8.4 Slice with forces for an infinite slope.

362 Slope stability
From eqs. 8.8 and 8.9, the driving shear stress and normal stress can be calculated as shown
in eqs. 8.10 and 8.11, respectively.
Sd = y z cos B sin B (8.10)
o = y z cos B2
(8.11)
Based on the Mohr‒Coulomb failure criterion, the resisting shear strength at failure is
expressed as eq. 8.12. Therefore, Fs can be calculated using eq. 8.13. For cohesionless soils
(c = 0), Fs will be expressed simply as the ratio of tano / tanB .
S = c + y z cos 2 B tan o (8.12)
c + y z cos 2 B tan o
Fs = (8.13)
y z cos B sin B
Figure 8.5 shows the seepage effects when the water table is at the slope surface. For
this case, the sliding soil mass has weight W = y sat Lz cosB . The pore water pressure acts
on the bottom of the typical section. By considering the effective stress theorem, eq. 8.11
based on the effective stress can be expressed by eq. 8.14. In addition, the resisting shear
strength at failure, including soil cohesion, is shown in eq. 8.15. However, groundwater
contributes to the saturated unit weight, which can increase the driving shear stress, as
shown in eq. 8.16.
o = o - uw = y sat z cos 2 B - y w z cos 2 B = y 'z cos 2 B (8.14)
S f = c + y 'z cos 2 B tan o (8.15)

Sd = y sat z cos B sin B (8.16)
Figure 8.5 Seepage consideration for an infinite slope.

Slope stability 363
Figure 8.6 Slice with forces adopted in the ordinary method of slices.
From eqs. 8.15 and 8.16, the factor of safety can be derived as follows:
c + y z cos 2 B tan o
Fs = (8.17)
y z cos B sin B
Example 8.1
For the infinite slope for cohesionless slope ( B = 25º) shown in Figure EX 8.1, compute Fs
when y = 19 kN m3 and φ = 33º.
Figure EX 8.1 Infinite slope information.
Solution
Fs = tano / tanB = tan 33º/tan 25º = 1.393
364 Slope stability
8.3.4 Finite slope
Ordinary method of slices

The forces on the sides of slices are ignored for the ordinary method. The force conditions
considered by the ordinary method of slices are shown in Figure EX 8.1. The normal force can
be expressed by eq. 8.18. Bishop (1955) indicated that the normal force would exist only if the
resultant force of the forces on the side of the slice acted in a direction parallel to the base of
the slice. This condition is almost impossible. Moreover, interslice forces should be zero for all
forces on the slices to achieve equilibrium.
N = W cos a (8.18)
The normal stress on the base of a slice can be calculated by dividing the normal force by the
area of the slice base ( 1x A i ), as expressed by eq. 8.19.
W cos a
o= (8.19)
A i
From eqs. 8.18 and 8.19, the factor of safety of the slope can be expressed by eq. 8.20.
Fs =
E ( cA + W cos a tan o )
i
(8.20)
EW sin a i i
Effective stress strength parameters should be adopted, as shown in eq. 8.25, if the soil mass
is in the drained condition.
Fs =
E ( c'A + W cos a tan o ')
i
(8.21)
EW sin a i i
The weight of the slice can be expressed in eq. 8.22, and thus, the factor of safety is shown
in eq. 8.23. Note that eq. 8.23 will lead to a significantly low effective stress (o '=y h cos 2 a - u),
even negative, when the pore water pressure is larger or the failure surface is steep.
W = y hb = y hA cos a (8.22)
Fs =
E ||c ' A + (W cos a - uA cos a ) tan o '||
i i i
2
(8.23)
EW sin a i i
Based on the ordinary method of slices, a better expression for the factor of safety can be
obtained by considering the effective slice weight, W ′ , which is expressed as:
W ' = W - ub (8.24)
Slope stability 365
From eqs. 8.18, 8.21, and 8.24, the factor of safety derived from moment equilibrium is
shown in eq. 8.25. This equation does not calculate negative effective stresses if the pore water
pressure is less than the total vertical pressure.
Fs =
E |c| 'A + (o - u ) tan o '|| (8.25)
EW sin a
Example 8.2
For the plane strain slope shown in Figure EX 8.2, compute Fs using the ordinary method of slices.
Note that for a real-world problem, you would have to choose a failure circle and the number of
slices and then measure all the parameters you are given for slice information. For the plane strain
slope shown in the following, the factor of safety is computed using the ordinary method of slices.
Slice information for example 8.2:
Slice Ave. h (m) Ave. h w (m) α (degree)
1 0.20 0.20 -24

2 0.60 0.60 -14
3 1.35 1.35 -11
4 2.40 2.40 -3
5 3.40 3.20 0
6 4.35 3.60 3.5
7 5.25 3.80 11.5
8 5.60 3.80 14
9 5.25 3.70 24
10 4.75 3.40 29
11 4.20 3.10 32.5
12 3.50 2.60 38.5
13 2.50 1.70 46
14 1.25 0.60 57
*h w is the water height in a slice.
Figure EX 8.2 Slope information.

366 Slope stability
Solution
The only trick here is to correctly calculate the total weight of each slice, given that part of every
slice is dry and a part is saturated. Namely:
(
Wi = b hwi y sat + ( hi - hwi ) y dry )
b
Additionally, note that tl i = and ui = hwiy w .
cos a i
Solution information for example 8.1:
Slice h hw α b Sin α Cos α c'xtl u = y wh w u × tl W sin α c' + c' (Wcos

α –ul)tan q '
1 0.2 0.2 −24 1 −0.41 0.91 16 2 2 −1.63 16.97

2 0.6 0.6 −14 1 −0.24 0.97 15 6 6 −2.90 17.49
3 1.35 1.35 −11 1 −0.19 0.98 15 13 13 −5.15 20.02
4 2.4 2.4 −3 1 −0.5 1.00 15 24 24 −2.51 23.89
5 3.4 3.2 0 1 0.00 1.00 15 31 31 0.00 28.09
6 4.35 3.6 5.5 1 0.10 1.00 15 35 35 8.11 32.81
7 5.25 3.8 11.5 1 0.20 0.98 15 37 38 20.01 37.26
8 5.6 3.8 14 1 0.24 0.97 15 37 38 25.70 39.00
9 5.25 3.7 24 1 0.41 0.91 16 36 40 40.69 35.22
10 4.75 3.4 29 1 0.48 0.87 17 33 38 43.96 32.14
11 4.2 3.1 32.5 1 0.54 0.84 18 30 36 43.24 29.37
12 3.5 2.6 38.5 1 0.62 0.78 19 26 33 41.78 26.42
13 2.5 1.7 46 1 0.72 0.69 22 17 24 34.13 24.85
14 1.25 0.6 57 1 0.84 0.54 28 6 11 19.22 28.15
Total 264.65 391.66
Fs = 391.66/264.65 = 1.480
Simplified Bishop procedure

For the simplified Bishop method, all forces applied to a slice are displayed in Figure 8.7. The
normal interaction forces between adjacent slices are assumed to be collinear. In addition, no
shear stresses between slices are assumed. The force equilibrium equation in the vertical direc-
tion can be written as:
N cos a + S sin a - W = 0 (8.26)
From eq. 8.5, the shear force expressed by the shear strength and factor of safety can be writ-
ten as eq. 8.27.
si tl
S= (8.27)
Fs
Slope stability 367
Figure 8.7 Slice with forces based on the simplified Bishop method.
By considering effective stresses, we can write:
1
S= |c'A + ( N - u A ) tan o '|| (8.28)
Fs |
From eqs. 8.26 and 8.28, the normal force, N , can be presented by eq. 8.29 based on the
Mohr‒Coulomb failure criterion.
W - (1/ Fs ) ( c'A-u A tan o ' ) sin a

N= (8.29)
cos a +
( sin a tan o ')
Fs
The effective normal stress on the base of the slice is:
N
o' = -u (8.30)
A
By combining eqs. 8.29 and 8.33, the factor of safety can be written as:
| |
| ' '|
| c A cos a + (W - u A cos a ) tan o |
E | ( sin a tan o ') |
| cos a + |
| Fs |
Fs = (8.31)
E W sin a
368 Slope stability
In eq. 8.31, Fs appears on both sides. To use the simplified Bishop procedure, the Fs on the
right-hand side should be assumed for the trial. The iterative process can converge on Fs. Com-
monly, a computer program is used to address problems.
Inclusion of additional known forces

In general, there are additional known driving and resisting forces. For example, in Figure 8.8,
the pseudostatic method is used to simulate seismic loading. Moreover, additional forces could
result from slope reinforcement. Their effects must be included in the equilibrium equation
to compute the factor of safety. The simplified Bishop procedure is employed for illustration
to include the additional forces. For the overall moment equilibrium about the center of a cir-
cle, the two components are the soil weight and shear strength. Therefore, equilibrium can be
expressed in eq. 8.32. Counterclockwise and clockwise moments are considered positive and
negative, respectively.
si A i
y E Fs
-y EW sin a
i i =0 (8.32)
As shown in Figure 8.8, if there are seismic forces ( kWi ) and forces from soil reinforcement
( Ti ), the equilibrium equation can be expressed as:
si A i
y E Fs
-y EW sin a - E kW d + E T h
i i i i i j =0 (8.33)
where k is the seismic coefficient, d i is the vertical distance between the center of gravity
of the slice and the center of the circle, and h j is the moment arm of the reinforcement force
Figure 8.8 Slope with additional known seismic and reinforcement forces.
Slope stability 369
Figure 8.9 Individual slice with additional known forces.
about the center of the circle. Notably, the reinforcement force is not horizontal, as shown in
Figure 8.9. The magnitude of k is generally determined by design codes.
In eq. 8.33, the forces represented by the last two summations involve only known quanti-
ties, and thus, it is convenient to replace them by a single moment, M n . An additional moment
because of a force, P , is from the known forces. Eq. 8.33 can be represented by eq. 8.34.
si Ali
y E Fs
-y EW sin ai i + Mn = 0 (8.34)
The factor of safety that satisfies moment equilibrium is shown in eq. 8.35.
Fs =
E s A i i
(8.35)
Mn
E Wi sin a i -
r
Based on the effective stresses, the factor of safety can be expressed as eq. 8.36 when the
summation of each slice is considered.
Fs =
E |c| ' + (o - u ) tan o '|| A (8.36)
Mn
EW sin a - r
In Figure 8.9, the known forces are included in each slice. They are a seismic force ( kW ),
a water force ( P ), and a reinforcement force (T ). The force P is applied perpendicular to the
top of the slice, and the reinforcement force is inclined at an angle, ψ, from the horizontal. To
determine the normal force in eq. 8.36, the summations of vertical forces in Figure 8.10 can be
presented in eq. 8.37.
N cos a + S sin a - W - P cos B + T sinY = 0 (8.37)

370 Slope stability
Figure 8.10 Taylor’s chart.

Source: Redrawn form Taylor (1937).
where β is the inclination of the top of the slice. Notably, the seismic force in Figure 8.9 is
horizontal, and therefore, it makes no contribution to the vertical force equilibrium. In fact, the
vertical force can be taken into account if needed. To simplify the contribution of the known
forces in eq. 8.37, Fv represent the vertical components of all known forces, as shown in equa-
tion 8.38. The only exception is the slice weight.
Fv = - P cos B + T sinY (8.38)
Therefore, the summation of vertical forces in eq. 8.37 can be expressed as:
N cos a + S sin a - W + Fv = 0 (8.39)
Based on the Mohr‒Coulomb failure criterion and eq. 8.28, the normal force can be solved
and presented as:
( 1 )
W - Fv | | ( c'A-u A tan o ' ) sin a
N= ( Fs )
(8.40)
cos a +
( sin a tan o ')
Fs
Slope stability 371
The normal force in eq. 8.29 combines with eq. 8.36 for the factor of safety, and then the fac-
tor of safety can be expressed as:
| |
| ' '|
| c A cos a + (W - Fv - u A cos a ) tan o |
E | ( sin a tan o ') |
| cos a + |
| Fs |
Fs = (8.41)
Mn
E W sin a -
r
A summary of the very basic limit equilibrium methods for analyzing slopes is shown in
Table 8.1, where the required inputs and their limitations are included. For strength parameter
selections (drained or undrained condition), hydraulic conductivity can be considered. Practi-
cally, rainfall-induced slope failures generally take a few days to allow water infiltration into a
certain depth (failure surface) of soil. For the greatest magnitude of clay hydraulic conductivity,
rainfall will most likely not reach this depth easily. Therefore, drained parameters are suggested
for slope stability analyses under rainfall.
Table 8.1 Available limit equilibrium method for homogeneous slope stability evaluations
Analysis method Application(s) Limitations
Cohesionless soils (sands, c = 0 )
Infinite slope Moist ( γ m ) , submerged ( y '), plane failure, same F s for moist and
seepage ( γ sat ) , homogeeous submerged sands
Cohesive soils (clays, silts, sand mixed with clays/silts, c > 0 )

Infinite slope Moist ( γ m ) , submerged ( y '),
plane failure, limited to very long
seepage ( γ sat ) , homogeneousslopes
Wedge (Culmann) Moist ( γ m ) , submerged ( y '),
plane failure, acceptable for near-
homogenous vertical slopes, iterate for Fs if o = 0
Taylor’s charts Moist ( γ m ) , submerged ( y '),
circular failure, iterate for F s if, o = 0
homogenous, rapid drawdown nonrigorous solution, short-term
total stress analysis with cu - ou
(or c - o )
Ordinary method Moist ( γ m ) , submerged ( y '), circular failure, nonrigorous solution,
of slices seepage ( γ sat ) , nonhomogenous long- or short-term analysis *
(Fellenius) profile, external loads
Simplified method Moist (γ m ) , submerged (y ') , circular failure, iterate for F s if o = 0 ,
of slices (Bishop) Seepage (γ sat ) , nonhomogenous semirigorous solution, long- or
profile, external loads short-term analysis *
Bishop- Moist (γ m ) , submerged (y ') , Circular failure, long-term effective
Morgenstern and seepage (γ sat ) , homogenous, stress analysis with c' - o '
Barnes tables simplified method of slices
Cousins charts Moist (γ m ) , submerged (y ') , Circular failure, long-term effective
seepage (γ sat ) , homogenous, stress analysis with c' - o '
simplified method of slices
Morgenstern rapid Seepage (γ sat ) , homogenous Short-term effective stress drawdown
drawdown analysis with c' - o '
*For slice methods: Long-term effective stress analysis with c' - o ' and u ≥ 0 (smallest F s for cuts in satu-
rated clay); short-term total stress analysis using cu - ou or c - o and u = 0 (smallest F s for embankments
on saturated clay).
372 Slope stability
8.4 Slope stability charts

Stability charts for soil slopes were first produced by Taylor (1937), and they continue to be used
extensively as design tools and draw the attention of many investigators (Morgenstern, 1963; Zan-
bak, 1983; Michalowski, 2002; Qian et al., 2015). The presented chart solutions cover a range of
slope and soil conditions. They provide an easy and simple way of obtaining the factor of safety.
They are still useful for many conditions. Because stability charts are still produced based on the
methods in Section 8.3, only some of the most basic chart solutions will be introduced in this section.
Figure 8.10 shows the chart solutions proposed by Taylor (1937) for homogeneous cohesive
soil, such as clay. In this case, the soil has only cohesive strength (c), with φ = 0, which is the und-
rained condition (short-term). In Figure 8.10, d is the depth factor, and N0 is the stability number.
Notably, N0 is dimensionless. For a two-dimensional purely cohesive slope, the slope failure mode
is generally a deep circle when the slope is gentle. In other words, the slope failure surface will be
controlled by the bottom rigid layer, such as the rock formation. Therefore, the stability number
can be influenced by the depth factor. For steep slopes, toe failure is the common failure mode.
For drained conditions, c' = 0 should be adopted for pure sand, which rarely appears alone
in slopes. In practice, soil compaction will be generally applied when constructing fill slopes.
Because of overconsolidation from compaction, the soil could have both cohesion and friction
angles, known as cohesive-frictional ( c' - o ' ) soil. In addition, the soils in natural slopes are rarely
pure components of clay or sand. They will generally have slight overconsolidation, unsaturated
effects, or rock particles in slopes. Therefore, c' - o ' soil is also more appropriate for describ-
ing soil strength. For the preceding cases, Taylor’s chart considering uniform undrained shear
strength is no longer applicable. A number of charts have been published based on the effective
stress theorem, but they usually apply to very specific conditions. One of the more useful charts
was presented by Hoek and Bray (1981), who considered different groundwater conditions. Two
examples are shown in Figure 8.11, where the effects of tension cracks are included. Figure 8.12
Figure 8.11 Water table and tension crack assumptions for Hoek-Bray charts.
Source: Redrawn form Hoek and Bray (1981).
Slope stability 373
Figure 8.12 Hoek-Bray charts.

Source: Redrawn form Hoek and Bray (1981).
374 Slope stability
shows the stability charts relevant to Figure 8.11 by Hoek and Bray (1981). For a given slope, the
slope Fs can be assessed when the soil strength ( c ' and φ ' ), unit weight, and slope height (H) are
known. Using the saturated unit weight here is a conservative consideration.
In fact, the safety factor is highly relevant to the failure mechanism. As mentioned previously,
the deep circles are the main failure mode of gentle slopes for purely cohesive soils. However,
the general failure mode for c' - o ' soil is toe circles. For a given slope angle, the depth of the
failure surface decreases with increasing friction angle. Therefore, the bottom rigid layer does
not influence the factor of safety and the failure mode if it is deep enough.
Example 8.2
A 10 m high slope with a 20° slope angle has the parameters c ' = 2kPa, o ' = 25. , and
y sat = 16kN/m3. Under long-term conditions, the water table is at the ground surface for dis-
tances greater than 40 m behind the toe of the slope.
Solution
c' 2
Using Figure 8.13c, calculate = = 0.027 .
y H tan o ' (16)(10)(tan 25. )
For a slope angle of 20° , read off the chart.
Either c '/ y HFs = 0.009 or tan o '/ Fs = 0.318 is obtained.

Hence, Fs ≈ 1.41 .
In addition, three-dimensional boundary effects will affect failure surface development, as

shown in Figure 8.13. Generally, they make the failure surface shallower. The effects are more
obvious for purely cohesive soils (Lim et al., 2015) because the failure surface is deeper for
purely cohesive soils than for cohesive-frictional soils. This difference also means that three-
dimensional effects are relatively significant in purely cohesive soil slopes. Gens et al. (1988)
indicated that the difference in the slope stability assessment between two- and three-dimensional
B
b
Figure 8.13 Three-dimensional slope failure mechanism modified with a plane insert.
Slope stability 375
Figure 8.14 Stability charts considering seismic effects.

Source: Redrawn from Michalowski (2002).
analyses can range from 3% to 30% and an average of 13.9%. As indicated by Li et al. (2009),
the differences in the factor of safety between two-dimensional and three-dimensional analyses
can be up to several hundred percent. This phenomenon is observed much less in cohesive-
frictional soil slopes, particularly for B / H ≥ 5. Therefore, engineers should be aware of this
issue when assessing three-dimensional slope stability.
Michalowski (2002) adopted the pseudostatic method to account for seismic force and then
produced stability charts for cohesive-frictional soil slopes without a groundwater table. The
chart solutions are displayed in Figure 8.14 for various horizontal seismic coefficients ( kh ). In
fact, the pseudostatic method is a simplified approach that would not be appropriate for slopes
with a high groundwater table because an earthquake can generate high excess pore water pres-
sure. Notably, the range of the predicted failure surface increases with increasing kh when
the Mohr‒Coulomb failure criterion is applied to the slope material (Li et al., 2020). Many
commercial limit equilibrium software programs should set up searching boundaries. To obtain
reasonable critical failure mechanisms, it is important to adjust the search boundary properly.
8.5 Stabilization methods

For slope stability problems, an appropriate stabilization method is determined by considering
potential failure mechanisms, as shown in Figure 5.20. Specifically, the slope reinforcement
should be installed over the range of potential failure surfaces. The failure surface development
makes it more difficult to pass through the slope reinforcement. Therefore, the factor of safety
increases after slope reinforcement. In Chapter 5, many reinforced slope stabilization methods,
such as anchor and geosynthetics in Figures 5.25b and 5.28, respectively, have been well intro-
duced. In this section, only two commonly used methods will be briefly included.
376 Slope stability
Figure 8.15 is a configuration of the drilled shaft. The retaining wall can have a certain space
(Sp) between two piles. Notably, the space between two piles is also helpful for drainage. The
soil arching effects shown in Figure 8.16 can be utilized to increase the factor of safety. After
inserting the piles, the horizontal soil stress is significantly different. The horizontal stresses will
be concentrated at both side piles after installation. Of course, the distance between the two piles
increases, and then the soil arching effects decrease. The drilled shaft behaves like a cantilever
wall. In general, the pile toe will be installed into the bottom rigid layer to prevent sliding.
Therefore, the drilled shaft alone would not be suitable for slopes with a very thick soil layer.
Figure 8.15 Pile retaining wall.
Figure 8.16 Soil arching effect (top view).

Slope stability 377
Figure 8.17 Potential failure mechanisms with multiple drilled shaft.
In some cases, a two-row pile could be adopted for a slope, as shown in Figure 8.17, where
three types of failure surfaces can be obtained based on Li et al. (2020). Type A failure surfaces
are restricted by piles. Type B and Type C failure surfaces pass through a single-row and a two-
row pile, respectively. For Type A, Figure 8.18 shows that the slope can be divided into three
parts between shafts. Thus, each part can be assessed individually, implying that Fs increases
because of the reduction in the slope height. In addition, Li et al. (2022) proposed slope stabil-
ity charts (Figure 8.18), which are suitable for Type A and Type B failure surfaces. However,
their solutions are not applicable to slopes with Type C failure surfaces. Theoretically, the most
effective location of drilled shafts is near the middle of the slope, because the failure surface
development could be suppressed.
378 Slope stability
Figure 8.18 Stability chart for the results obtained under various S p/D ratios (dry slopes).
Figure 8.19 shows the use of a drainage well to collect and lower the groundwater table. For
some potential large-scale landslides, commonly used reinforced methods would not be very
effective because the reinforcements have difficulty reaching ranges beyond the potential slid-
ing slope masses. Thus, the general slope reinforcement scale is much smaller than the potential
sliding mass, so drainage wells could be more useful than general slope reinforcements. In
particular, the groundwater table can rise quickly during rainfall, and thus, the slope is critical.
The horizontal drainage pipe is useful to slow down the rise in water pressure. This feature
contributes directly to the factor of safety because of the increasing soil effective strength. We
investigated a range of soil slope parameters, which showed that H is the main controlling fac-
tor for Fs when H ≥ 20 m and there is a fairly high water level. This issue is commonly faced
by geotechnical engineers in Taiwan, where there are a large number of colluvium slopes. This
Slope stability 379
Figure 8.19 Configuration of drainage well.
feature occurs because changes in natural soil strength have only limited effects on Fs. For
some large-scale landslides, an Fs of 1.5, the general static design requirement, is difficult to
satisfy. Therefore, maintaining a relatively low water level by the drainage well is an advisable
approach.
In addition to the stabilization methods, long-term monitoring on the slope is required. For
example, an inclinometer can be used to measure soil movements and identify the moving
direction, as shown in Figure 8.20. Figure 8.20 also shows various monitoring techniques which
have been frequently applied recently. In general, they are used to measure groundwater table,
surface or underground movements.
380 Slope stability
Figure 8.20 Slope monitoring techniques.

For the past few decades, many researchers have dedicated their time and efforts to soil mechan-
ics, and therefore, we can obtain better understandings of soil strength and behavior. They also
enhance knowledge of slope stability problems. Thus far, our lessons are still learned from expe-
rience. Perhaps some of these experiences are bad. Investigations of past slope failures help us
understand potential factors that can cause slope instability. However, it is believed that we have
not captured the entire phenomenon. In particular, global warming brings more extreme weather
conditions, which will create major challenges for civil engineers in the near future. Reducing
the losses of lives and property requires advances in this area.
Problem
8.1 For the slope shown in Figure P8.1, determine the factor of safety.
Figure P8.1
Slope stability 381
8.2 Figure P8.2 shows a 10 m high slope in saturated clay. For the soil, assume y = 19 kN/m3,
o ' = 20 , and c' = 20 kN/m 2 . Determine Fs using the ordinary method of slices.
Figure P8.2
8.3 Refer to problem 8.2. For the same slope shown in Figure P8.2, compute the factor of
safety using the simplified Bishop method of slices.
8.4 Figure P8.3 shows a slope with dimensions similar to those in Figure P8.2. However, there
is steady-state seepage. The phreatic line is shown in Figure P8.3. The other parameters
remain the same. Determine Fs using the ordinary method of slices.
Figure P8.3
382 Slope stability
8.5 A 15 m high slope with a face angle of 40° is to be excavated in soil with density
y = 16 kN/m3 , a cohesive strength of 40 kPa, and a friction angle of 30°. Find the factor of
safety of the slope, assuming that it is a saturated slope subjected to heavy surface recharge.
8.6 A slope with B = 60o is to be constructed with a soil that has o ' = 20o and c' = 25 kPa.
The unit weight of the compacted soil will be 19 kN/m3.
a. Find the critical height of the slope using Michalowski’s solution.
b. If the height of the slope is 15 m, determine the factor of safety using Michalowski’s
solution.
References
Bishop, A.W. (1955), The use of the slip circle in the stability analysis of slopes. Geotechnique, Vol. 5,
No. 1, pp. 7–17.
Fredlund, D.G. and Krahn, J. (1977), Comparison of slope stability methods of analysis. Canadian Geo-
technical Journal, Vol. 14, No. 3, pp. 429–439.
Gens, A., Hutchinson, J.N. and Cavounidis, S. (1988), Three-dimensional analysis of slides in cohesive
soils. Geotechnique, Vol. 38, No. 1, pp. 1–23.
Goodman, R.E. and Bray, J.W. (1976), Toppling of rock slopes. Proceedings of the Specialty Conference
on Rock Engineering for Foundations and Slopes, ASCE, Vol. 2, pp. 201–234.
Goodman, R.E. and Kieffer, D.S. (2000), Behavior of rock in slope. Journal of Geotechnical and Geoen-
vironmental Engineering, Vol. 126, No. 8, pp. 675–684.
Hoek, E. and Bray, J.W. (1981), Rock Slope Engineering, London: Institute of Mining and Metallurgy.
Krabbenhoft, K., Lyamin, A.V., Hjiaj, M. and Sloan, S.W. (2005), A new discontinuous upper bound limit
analysis formulation. International Journal for Numerical Methods in Engineering, Vol. 63, No. 7,
pp. 1069–1088.
Li, A.J., Jange, A., Lin, H.D. and Huang, F.K. (2020), Investigation of slurry supported trench stability under
seismic condition in purely cohesive soils. Journal of GeoEngineering, Vol. 15, No. 4, pp. 195–203.
Li, A.J., Lin, H.D. and Wang, W.C. (2022), Investigations of slope reinforcement with drilled shafts in col-
luvium soils. Geomechanics and Engineering, Vol. 31, No. 1, pp. 71–86.
Li, A.J., Merifield, R.S. and Lyamin, A.V. (2009), Limit analysis solutions for three dimensional undrained
slopes. Computers and Geotechnics, Vol. 36, No. 8, pp. 1330–1351.
Lim, K., Li, A.J. and Lyamin, A.V. (2015), Three-dimensional slope stability assessment of two-layered
undrained clay. Computers and Geotechnics, Vol. 70, pp. 1–17.
Lyamin, A.V. and Sloan, S.W. (2002a), Lower bound limit analysis using non-linear programing. Interna-
tional Journal for Numerical Methods in Engineering, Vol. 55, No. 5, pp. 573–611.
Lyamin, A.V. and Sloan, S.W. (2002b), Upper bound limit analysis using linear finite elements and non-
linear programing. International Journal for Numerical Methods in Engineering, Vol. 26, No. 2, pp.
181–216.
Michalowski, R.L. (2002), Stability charts for uniform slopes. Journal of Geotechnical and Geoenviron-
mental Engineering, Vol. 128, No. 4, pp. 351–355.
Morgenstern, N. (1963), Stability charts for each slopes during rapid drawdown. Geotechnique, Vol. 13,
pp. 123–131.
Qian, Z.G., Li, A.J., Merifield, R.S. and Lyamin, A.V. (2015), Slope stability charts for two-layered purely
cohesive soils based on finite-element limit analysis methods. International Journal of Geomechanics,
Vol. 15, No. 3.
Taylor, D.W. (1937), Stability of earth slopes. Journal of the Boston Society of Civil Engineering, Vol. 24,
pp. 197–246.
Taylor, D.W. (1948), Fundamentals of Soil Mechanics, New York: John Wily and Sons, pp. 406–479.
Zanbak, C. (1983), Design charts for rock slopes susceptible to toppling. Journal of Geotechnical Engi-
neering, ASCE, Vol. 109, No. 8, pp. 1039–1062.
Solutions
Chapter 1
1.1
Table S1.1
Test Purpose
Soil physical properties Obtain grain distribution, moisture content, unit weight, liquid
tests limit, and plastic limit of soil and classify soil type based on
these properties.
Triaxial unconsolidated Obtain undrained shear strength (s u) of clayey soil.
undrained test (UU)
Triaxial consolidated Obtain total strength parameter (c, ϕ) and effective strength
undrained test (CU) parameter (c', ϕ').
Oedometer test Obtain compression index and swell index (C c, C s).
Direct shear test Obtain (c', ϕ ') of sandy soil.
Permeability test Obtain the value of hydraulic conductivity (k) of soil.
Dynamic triaxial test Obtain stress–strain relationship, stress path, and liquefaction
resistance strength.
1.3
Ans:
Clayey soil.
1.5
Ans:
Table S1.5
Depth (m) N1 N2 N3 N-value
1.5 2 3 2 5
3.0 2 2 4 6
4.5 1 3 3 6
6.0 2 3 4 7
7.5 3 4 4 8
384 Solutions
1.7
Ans:
d 02 - di2 50.82 - 34.92
Ar = (split spoon) = = = 111.9%
di2 34.92
d 02 - di2 76.22 - (76.2 - 3.1) 2

Ar (Thin-walled tube) = = = 8.7%
di2 (76.2 - 3.1) 2
Chapter 3
3.1
c = c' = 0
ϕ = ϕ' = 34˚, Nc = 52.64, Nq = 36.5, Nγ = 38.04 (see Table 3.1)
a. 3m wide strip footing
Eq. 3.5 is used.
qu = (1)(0)(52.64) + (1)(18.5)(36.5) + (0.5)(18.5)(3)(38.04) = 1,730.86 kN/m2
b. 3 m × 3 m square footing
Eq. 3.8 is used.
qu = (1.3)(0)(52.64) + (1)(18.5)(36.5) + (0.4)(18.5)(3)(38.04) = 1,519.74 kN/m2
c. Circular footing of 3 m in diameter
Eq. 3.9 is used.
qu = (1.3)(0)(52.64) + (1)(18.5)(36.5) + (0.3)(18.5)(3)(38.04) = 1,308.62 kN/m2
3.3
Eq. 3.10 is used.
c = c'= 0
ϕ = ϕ' = 34˚, Nc = 42.16, Nq = 29.44, Nr = 31.15
Solutions 385
a. 3m wide strip footing
B = 3 m, L = ∞ , Df = 1 m
Fcs = 1, Fqs = Frs = 1
Fcd = 1.125, Fqd = Frd = 1.063
Fci = 1, Fqi = Fri = 1
q = 18.5 kN/m2
1
qu = cNcFcsFcdFci+qNqFqsFqdFqi + γBNrFrsFrdFqi
2
= (0)(42.16)(1)(1.125)(1) + (18.5)(29.44)(1)(1.063)(1) + (0.5)(18.5)(3)(31.15)(1)(1.063)(1)
= 1,497.8 kN/m2
b/c. 3 m × 3 m square footing; circular footing of 3 m in diameter
B = 3 m, L = 3 m, Df = 1 m
Fcs = 1.707, Fqs = Frs = 1.354
Fcd = 1.125, Fqd = Frd = 1.062
Fci = 1, Fqi = Fri = 1
1
qu = cNcFcsFcdFci+qNqFqsFqdFqi + γBNrFrsFrdFqi
2
= (0)(42.16)(1.707)(1.125)(1) + (18.5)(29.44)(1.354)(1.063)(1) + (0.5)(18.5)(3)(31.15)
(1.354)(1.063)(1)
= 2,028 kN/m2
3.5
a. c = 0, o = 34 , y = 18.5 kN/m3

N q = tan 2 ( 45 + o 2 ) eT tan o = tan 2 ( 45 + 34 2 ) eT tan 34 = 29.44
( )
N c = cot o N q - 1 = cot 34 ( 29.44 - 1) = 42.16
( )
Ny = tan 1.4o N q - 1 = tan 1.4 x 34 ( 29.44 - 1) = 31.15
Fcs = 1 + 0.2 ( B L ) tan 2 ( 45 + o 2 ) = 1 + 0.2 ( 3 4 ) tan 2 ( 45 + 34 2 ) = 1.531
Fqs = Fy s = 1 + 0.1( B L ) tan 2 ( 45 + o 2 ) = 1 + 0.1( 3 4 ) tan 2 ( 45 + 34 2 ) = 1.265
( )
Fcd = 1 + 0 2 D f B tan ( 45 + o 2 ) = 1 + 0.2 (1 3) tan ( 45 + 34 2 ) = 1.125
( )
Fqd = Fy d = 1 + 0.1 D f B tan ( 45 + o 2 ) = 1 + 0.1(1 3) tan ( 45 + 34 2 ) = 1.063
( )
2
Fci = Fqi = 1 - B 90 =1
Fy i = 1 - ( B o ) = 1
2
qu = cN c Fcs Fcd Fci + qN q Fqs Fqd Fqi + 0.5y BNy Fy s Fy d Fy i

= 0 +18.5 x 1x 29.44 x 1.265 x 1.063x 1 + 0.5 x 18.5 x 3 x 31.15 x 1.265x 1.063 x 1
= 1894.75 kN
N/m 2
qu ( net ) = 1894.75 - 18.5 x 1 = 1876.25 kN/m 2
386 Solutions
b. qb ( net ) = 6000 / (3 x 4) - 18 x 1 = 482

=FS 1876.
= 25 / 482 3.893
3.7
B = 3 m, L = 4 m, Q = 6,000 kN, ML = 1,200 kN-m, eL = 1,200/6,000 = 0.2 < 4/6

qb,max/min = [(6,000)/(3 × 4)](1 ± (6)(0.2)/3) = 700 kN/m2/300 kN/m2
Using eq. 3.22:
B' = min(3 − 2 × 0.2, 4) = 2.6 m, L' = max(3 − 2 × 0.2, 4) = 4 m

c = c' = 0
ϕ = ϕ' = 34˚, Nq = 29.44, Nc = 42.16, Nγ = 31.15
Fcs = 1 + (0.2)(2.6/4)tan2(45 + 34/2) = 1.46
Fqs = Fγs = 1 + (0.1)(2.6/4)tan2(45 + 34/2) = 1.23
Fcd = 1 + (0.2)(1/3) tan(45 + 34/2) = 1.125
Fqd = Fγd = 1 + (0.1)(1/3) tan(45 + 34/2) = 1.063
qu = (0)(42.16)(1.46)(1.125)(1) + (1)(18.5)(29.44)(1.23)(1.063)(1) + (1/2)(18.5)(2.6)(31.15)
(1.23)(1.063)(1) = 1,691.63 kN/m2
3.9
Eq. 3.24 is used.
B = 3 m, L = 4 m, H = 5–1 = 4 m, Df = 1 m
m = L/B = 4/3, n = H/(B/2) = 8/3
If = 0.862 (refer to Table 3.4)
1 - 2us
I s = F1 + F2 = 0.398
1- u s
( 1 - ( 0.3)2 )
Si( rigid ) = 0.93Si( center ) = ( 0.93) ( 250 ) ( 4 ) ( 3/2 ) | | ( 0.398 ) ( 0.862 ) = 0.0363 m
| 12000 |
( )
= 36.3 mm
3.11
o 0' = (18 - 9.81) x (1 + 4 / 2 ) = 24.57 kN/m 2 < pc' = 80 kN/m 2

200x 3x x4
Ao ave' = = 80
(3 + 2) x ( 4 + 2)
o 0' + Ao ave' = 104.57 > pc' = 80
Solutions 387
Cs H c p' C H o ' + Ao ave'

Sc = log c + c c log 0
1 + e0 o 0' 1 + e0 pc
0.05 x 4 80 0.25 x 4 104.57
= logg + log = 121.6 mm
1 + 0.8 24.57 1 + 0.8 80
3.13
1. Settlement controlled.
Use eq. 3.24.

1 - u s2
Si = qb ( net ) (a B' ) Is I f
Es
μs = 0.3
For the center of the footing, α = 4:
m = L/B = 4/3 = 1.33, n = H/(B/2) = 4/(3/2) = 2.67 => F1 = 0.36, F2 = 0.0669

Is = 0.36 + [(1 − 2(0.3))/(1 − 0.3)](0.0669) = 0.398
B/L = 3/4 = 0.75, Df/B = 1/3 = 0.33 => If = 0.862
Si (rigid) = (0.93)qall(net)(4)(3/2)[(1 − 0.32)/12,000](0.398)(0.862) = 0.04 m = 40 mm
qall(net) = 275.53 kN/m2
2. Bearing capacity controlled.
c = c' = 0 kN/m2, ϕ = ϕ' = 34˚, γ = γ' = 18.5 kN/m2, Nq = 29.44, Nc = 42.16, Nγ = 31.15
Fcs = 1 + (0.2)(3/4)tan2(45 + 34/2) = 1.531
Fqs = Fγs = 1 + (0.1)(3/3)tan2(45 + 34/2) = 1.265
Fcd = 1 + (0.2)(1/3) tan(45 + 24/2) = 1.125
Fqd = Fγd = 1 + (0.1)(1/3) tan(45 + 24/2) = 1.063
Fci = Fqi = 1
Fγi = 1
qu = (0)(42.16)(1.531)(1.125)(1) + (1)(18.5)(29.44)(1.265)(1.063)(1) + (1/2)(18.5)(3)
(31.15)(1.265)(1.063)(1) = 1894.75 kN/m2
qu (net) = 1894.75 − 18.5(1) = 1876.25 kN/m2
qall(net) = 1876.25/2 = 938.125 kN/m2
=> Considering conditions 1 and 2, choose the smaller one.
qall(net) = 275.53 kN/m2 (The allowable settlement governs!)
Chapter 4
4.1
For o ' = 32o , K 0 = 1 - sin o ' = 0.47

At z = 0 m , o h' = 0
388 Solutions
At z = 2.5 m , o h' = 16 x 2.5 x 0.47 = 18.8 kN/m 2

At z = 9.5 m , o h' = (16 x 2.5 + 7 x 20 - 9.81x 7) x 0.47 = 52.3 kN/m 2
uw = 9.81x 7 = 68.7 kN/m 2
Total lateral force = 512.8 kN/m 2 , z = 2.885 m
4.3
a. K a = =
1 , zc 40
= / 17 2.353 m
b. Assume that tension cracks are formed and no water is in the tension cracks.
At zc = 2.353 m, o a = 0 ,
At z = 4 m − , o a = 4 x 17 - 2 x 20 = 28 kN/m 2
At z = 4 m + , o a = 4 x 17 - 2 x 30 = 8 kN/m 2
At z = 9 m, o a = 4 x 17 + 5 x18 - 2 x 30 = 158 - 60 = 98 kN/m 2
c.
Pa1 = 28 x 1.647 / 2 = 23.06 kN/m
Pa 2 = 8 x 5 = 40 kN/m
Pa3 = (98 - 8) x 5 / 2 = 225 kN/m
P = Pa1 + Pa 2 + Pa3 = 288.06 kN/m
z = 2.09 m
4.5
o ' = 34o , K a = 0.283

At z = 0 m, o a' = 0 kPa
At z = 2 m, o a' = 9.06 kPa , uw = 0 kPa
At z = 6 m , o a' = 20.59 kPa uw = 9.81x 4 = 39.24 kPa
Pa1 = 9.06 x 2 / 2 = 9.06 kN/m
Pa 2 = 9.06 x 4 = 36.22 kN/m
Pa 3 = (20.59 - 9.06)) x 4 / 2 = 23.07 kN/m
PW = 9.81x 4x 4 / 2 = 78.48 kN/m
P = Pa1 + Pa 2 + Pa 3 + PW = 146.83 kN/m
z = 1.70 m
4.7
For clay, o = 0 , K a = 1
a.
= zc 40
= / 16 2.5 m
Solutions 389
b. Assume that tension cracks are formed and water is in the tension cracks.
At z = 0 m, a a = 0 kPa , uw = 0
At z = 2.5 m, a a = 2.5 x 16 - 2 x 20 = 0kPa , uw = 9.81x 2.5 = 24.53kPa
At z = 4 m − , a a = 1.5 x 16 = 24 kPa
At z = 4 m + , a a = 4 x 16 - 2 x 25 = 64 - 50 = 14 kPa
At z = 8 m, a a = 4 x 16 + 4 x18 - 2 x 25 = 136 - 50 = 86 kPa
c.
Pa1 = 24 x 1.5 / 2 = 18 kN/m
Pa2 = 14 x 4 = 56 kN/m
Pa 3 = (86 - 14) x 4 / 2 = 144 kN//m
P = Pa1 + Pa 2 + Pa 3 = 218 kN/m
Pa1 x (1.5 / 3 + 4) + Pa 2 x 4 / 2 + Pa3 x 4 / 3
z= = 1.77 m
P
4.9
o1' = 30o o2' = =

32o , K a1 0.=
333, K a 2 0.307
a. At z = 0 m, a a' = 0
At z = 2 m − , a a' = y 1' x z x K a1 = 6.13 kPa
At z = 2 m + , o a' = y 1' x z x K a 2 = 5.64 kPa
At z = 5 m , a a' = 15.95 kPa , uw = 9.81x 5 = 49.05 kPa
b. P = 6.13 x 2 / 2 = 6.13 kN/m

a1
Pa 2 = 5.64 x 3 = 16.92 kN/m
Pa3 = (15.95 - 5.64)) x 3 / 2 = 15.47 kN/m
Pw = 49.05 x 5 / 2 = 122.63 kN/m
P = Pa1 + Pa 2 + Pa 3 + Pw = 161.15
1 kN/m
z = 1.66 m
4.11
For sand, o ' = 32o , K a,sand = 0.307
For clay, o = 0 , K a,clay = 1
At z = 0 m, a a' = 0kPa , uw = 0 kPa

At z = 3 m − , a a' = y 1' x z x K a ,sand = 9.39 kPa , uw = 9.81x 3 = 29.43 kPa
At z = 3 m + , a a = y 1 x z x K a ,clay - 2 su = 20 kPa
At z = 7 m , a a = 60 + 4 x 16 - 40 = 84 kPa
390 Solutions
Pa1 = 9.39 x 3 / 2 = 14.09 kN/m

Pa2 = 20 x 4 = 80 kN/m
Pa 3 = (84 - 20) x 4 / 2 = 120 kN/m
Pw = 29.43 x 3 / 2 = 44.15 kN/m
P = Pa1 + Pa 2 + Pa 3 + Pw = 266.22 kN/m
z = 2.75 m
4.13
c' = 0 kPa , o '=34o , y =16 kN/m 2 , y sat = 20 kN/m 2
When s = o ' / 2 = 17o , 0 =0o , B =

= 0o , K a 0=
.26 , K ah 0.25
z = 0 m, a h' = 0 kPa
z = 2 m, a h' = 2 x 16 x 0.25 = 8 kPa
z = 6 m, a h' = || 2 x 16 + 4 x ( 20 - 9.81) || x 0.25 = 18.19 kPa , uw = 4 x 9.81 = 39.24 kPa
Pah = 0.5 x 2 x 8 + 8 x 4 + 0.5x 4 x (18.19 - 8) + 0.5x 4 x 39.24 = 138.9 kN/m
z = 1.68 m
When s = 2o ' / 3 = 22.67o , 0 =0o , B =

= 0o , K a 0=
.25 , K ah 0.23
z = 0 m, a h' = 0 kPa
z = 2 m, a h' = 2 x 16 x 0.23 = 7.36 kPa
z = 6 m, a h' = || 2 x 16 + 4 x ( 20 - 9.81) || x 0.23 = 16.73 kPa
uw = 4 x 9.81 = 39.24 kPa
Pah = 0.5 x 2 x 7.36 + 7.36 x 4 + 0.5x 4 x (16.73 - 7.36) + 0.5 x 4 x 39.24 = 134 kN
N/m
z = 1.66 m
4.15
ou = 0o , K
= p1 K=
p2 1
z = 0 m , a p = 0+2 x 20 = 40 kPa
z = 4 m - , a p = (4 x 17 x 1) + 2 x 20 x 1 = 108 kPa
z = 4 m + , a p = (4 x 17 x 1) + 2 x 30 x 1 = 128 kPa
z = 9 m , a p = [4 x 17 + 5 x 18] x 1 + 2 x 30x 1 = 218 kPa
Pp = 40 x 4 + (108 - 40) x 4 / 2 + 128 x 5 + (218 - 128) x 5 / 2 = 1161kN/m
|160 x 7 + 136 x (5+4/3) + ( 640 x 2.5 ) + ( 225 x 5 / 3) ||
z=| = 3.41 m
1161
4.17

K p = 3.537
Solutions 391
z = 0 m , a 'p = 0 kPa
z = 2 m , a 'p = (2 x 16 x 3.537) = 113.2 kPa
z = 6 m , a 'p = [2 x 16 + 4 x (20 - 9.81)] x 3.537 = 257.4 kPa
uw = 4 x 9.81 = 39.24 kPa
Pp = 0.5 x 2 x 113.2 + 113.2 x 4 + 0.5x 4 x (257.4 - 113.2) + 0.5 x 4 x 39.24 = 932 7 kN/m
z = 2.1 m
4.19
When s = o ' / 2 = 17o , 0 =0o , B=

= 0o K p 6.=
767 , K ph 6.471
z = 0 m , a h' = 0
z = 2 m , a h' = 2 x 16 x 6.471 = 207.1 kPa
z = 6 m , a h' = [2 x 16 + 4 x (20 - 9.81)] x 6.471 = 470.8 kPa
Pph = 2 x 207.1 / 2 + 207.1x 4 + 4(470.8 - 207.1) / 2 + 4(4 x 9.81) / 2
= 1641.4 kN/m
|( 207.1x 4.67 ) + ( 828.4 x 2 ) + ( 527.4 x 4 / 3) + (78.48 x 4 / 3)||
z=|
1641.4
= 2.09 m
When s = 2o ' / 3 = 22.67o , 0 =0o , B = 0o
=K p 8.=
946 , K ph 8.255
z = 0 m , a h' = 0 kPa
z = 2 m , a h' = 2 x 16 x 8.255 = 264.2 kPa
z = 6 m , a h' = [2 x 16 + 4 x (20 - 9.81)] x 8.255 = 598.5 kPa
Pph = 2 x 264.2 / 2 + 264.2 x 4 + 4(598.5 - 264.2) / 2 + 4(4 x 9.81) / 2
= 2068.1 kN/m
|( 264.2 x 4.67 ) + (1056.8 x 2 ) + ( 668.6 x 4 / 3) + (78.48 x 4 / 3)||
z=|
2068.11
= 2.1 m
4.21
K 0 = 1 - sino ' = 0.47

z = 0 to 9.5 m
a h 0 = 30 x 0.47 = 14.1kPa
z = 0 m , a h0' = 0 kPa
392 Solutions
z = 2.5 m , a h' 0 = 2.5 x 16x 0.47 = 18.8 kPa

' = [2.5 x 16 + 7 x (20 - 9.81)] x 0.47 = 52.3 kPa
z = 9.5 m , a h0
uw = 7 x 9.81 = 68.67 kPa
P0 = 0.5 x 2.5x 18.8 + 7(18.8 + 52.3) / 2 + 7 x 68.67 / 2 + 14.1x 9.5 = 646.7 kN//m
|( 23.5 x 7.83) + (131.6 x 3.5 ) + (117.34 x 7 / 3) + (240.35 x 7 / 3) + (133.955 x 9.5 / 2||
z=|
646.7
= 3.27 m
4.23
su1 = 20 kN/m 2 , y 1 = 17 kN/m 2 ; su2 = 30 kN/m 2 , y 2 = 18 kN/m 2

ou = 0o , K p = 1
z = 0 m, a p = 2 x 20 x 1 + 40 = 40 + 40 = 80 kPa
z = 4 m - , a p = (4 x 17 x 1) + 1x 40 + 2 x 20x 1 = 68 + 40 + 40 = 148 kPa
z = 4 m + , a p = (4 x 17 x 1) + 1x 40 + 2 x 30x 1 = 68 + 60 + 40 = 168 kPa
z = 9 m , a p = [4 x 17 + 5x 18] x 1 + 60 + 40 = 158 + 60 + 40 = 258 kPa
Pp = 0.5 x 4 x (80 + 148) + 0.5 x 5 x (258 + 168) = 1521 kN/m
z = 3.7 m
Chapter 5
5.1
( 5 (
B = tan -1 | | = 68.2

( (12 - 8) / 2 )
→ B < o'
In this condition, the construction of earth retaining structures is necessary.
5.3
ANS: a: (1); b: (1)

The figure depicts a reinforced wall with wraparound facing. The wall facing is vegetated
for natural appearance. Although the reinforced wall with vegetation on the facing looks
like a natural slope, if you look closer, you still can observe the wraparound configura-
tion and reinforcement layers from the wall facing. The reinforced wall is classified as
an internally stabilized fill wall. The stability of internally stabilized walls is provided by
internal resistance through the reinforcement tensile force mobilized by soil–reinforcement
interaction.
Solutions 393
5.5
Calculate Rankine’s active earth pressure coefficient
1
Ka =
3
Horizontal force components
1 1
Pa = x 18 x x 52 = 75.0 kN/m
2 3
Pahh = Pa = 75.0 kN/m
Vertical force components

W1 = 0.5 x (5 - 0.3) x 24 = 56.4 kN/m
W2 = 2.5 x (5 - 0.3) x 18 = 211.5 kN/m
W3 = 0.3 x 3.5x 24 = 25.2 kN/m
Pav = 0 kN/m
EW + P i av = V = 293.1 kN/m
Moment i = Wi x Moment Arm

M1 = 56.4 x 0.75 = 42.3 kN . m/m
M 2 = 211.5 x 2.25 = 475.9 kN . m/m
M 3 = 25.2 x 1.75 = 44.1 kN . m/m
EM R = 562.3 kN . m/m
5
M D = 75 x = 125.0 kN . m/m
3
Calculate the factor of safety against overturning.
Fo =
EM R
=
562.3
= 4.50 > 2 ok
EM D
125.0
Calculate the factor of safety against sliding.
293.1x tan(20o)
Fs = = 1.42 < 1.5 NG!
75.0
5.7
Calculate Coulomb’s active earth pressure coefficient
2
a = B = 10o,0 = 0o, s = o ' = 20o
3
K a = 0.43
394 Solutions
Horizontal force components

1
Pa = x18x 0.2973 x 5.12 = 100.66 kN/m
m
2
Pah = Pa cos(10) = 99.13 kN/m
Vertical force components
W1 = 0.5 x (5 - 0.3) x 24 = 56.4 kN/m
W2 = 0.5 x 2.5 x 0.1x 18 = 2.25 kN/m
W3 = 2.5 x (5 - 0.3) x 18 = 211.5 kN/m
W4 = 0.3 x 3 5 x 24 = 25.2 kN/m
Pav = 100.66 x sin(10o) = 17.48 kN/m
EW + P i av = V = 312.83 kN/m
Moment i = Wi x Moment Arm
M1 = 56.4 x 0.75 = 42.33 kN . m/m
M 2 = 2.25 x 2.67 = 6.0 kN . m/m
M 3 = 211.5 x 2.25 = 475.9 kN . m/m
M 4 = 25.2 x 1.75 = 44.1 kN . m/m
M av = 17.48 x 3.5 = 61.18 kN . m/m
EM R = 629.448 kN . m/m
5.1
M D = 99.13 x = 168.52 kN . m/m
3
Calculate the factor of safety against overturning.
Fo =
EM R
=
629.48
= 3.74 > 2 ok
EM D
168.52
Calculate the factor of safety against sliding.
312.83 x tan(20o)
Fs = = 1.15 < 1.5 NG!
99.13
5.9
Answer to part A
Calculate maximum reinforcement load at the top and bottom layers.
Top layer:
Kr
= 1 for geosynthetics
Ka
Kr
oh = x K a x (y z + q ) - 2c K a = - 4.3 kPa
Ka
Tmaax = 0 kN/m
Solutions 395
Due to the soil a tension (negative lateral earth pressure) occurs at the top of the wall; therefore,
no reinforcement tensile force is mobilized (Tmax = 0 kN/m).
Bottom layer:
a h = 36 kPa
Tmax = a h x Sv = 36 x 0.5 = 18 kN/m
Calculate long-term reinforcement tensile strength.
Tult 150
Tal = = = 33.33 kN/m
RFCR x RFD x RFID 2 x 1.5 x 1.5
Calculate Fbr for all reinforcement layers.
Table S5.9a
No. of layer z (m) σ’ v (kPa) σ h (kPa) Tmax (kN/m) F br
16 (top layer) 0.25 4.20 −4.31 0 –

15 0.75 12.60 −1.62 0 –
14 1.25 21.00 1.06 0.53 62.57
13 1.75 29.40 3.75 1.88 17.74
12 2.25 37.80 6.44 3.22 10.34
11 2.75 46.20 9.13 4.57 7.29
10 3.25 54.60 11.82 5.91 5.63
9 3.75 63.00 14.51 7.25 4.59
8 4.25 71.40 17.20 8.60 3.87
7 4.75 79.80 19.89 9.94 3.35
6 5.25 88.20 22.57 11.29 2.95
5 5.75 96.60 25.26 12.63 2.64
4 6.25 105.00 27.95 13.98 2.38
3 6.75 113.40 30.64 15.32 2.17
2 7.25 121.80 33.33 16.67 2.00
1 (bottom layer) 7.75 130.20 36.02 18.01 1.85
The bottom layer is the most critical for reinforcement breakage because the lateral earth
pressure increases linearly with depth.
Bottom layer:
Tal 33.33
Fbr
= = = 1.85 > 1.5 ok
Tmax 18.01
Answer to part B
Calculate pullout resistance at the top and bottom layers.
Top layer:
Pr = C x F * x a x a v' x Le = 2 x 0.4 x 0.8 x 4.2 x 4.36 = 0.31 kN/m

where
C=2
2
F * = tan 31 = 0.40
3
396 Solutions
a = 0.8
(8 - 0.25)
Le = 4.5 - = 0.12 m
31
tan(45 + )
2
Bottom layer:
Pr = 2 x 0.4 x 0.8 x 130.2 x 4.36 = 363.19 kN/m
Calculate Fpo for all reinforcement layers.
Table S5.9b
No. of layer z (m) Le (m) σ' v (kPa) P r (kN/m) Tmax (kN/m) F po
16 (top layer) 0.25 0.12 4.20 0.31 0 –

15 0.75 0.40 12.60 3.21 0 –
14 1.25 0.68 21.00 9.15 0.53 17.20
13 1.75 0.96 29.40 18.14 1.88 9.66
12 2.25 1.25 37.80 30.16 3.22 9.36
11 2.75 1.53 46.20 45.23 4.57 9.91
10 3.25 1.81 54.60 63.34 5.91 10.72
9 3.75 2.10 63.00 84.49 7.25 11.65
8 4.25 2.38 71.40 108.68 8.60 12.64
7 4.75 2.66 79.80 135.91 9.94 13.67
6 5.25 2.94 88.20 166.19 11.29 14.72
5 5.75 3.23 96.60 199.51 12.63 15.79
4 6.25 3.51 105.00 235.87 13.98 16.88
3 6.75 3.79 113.40 275.27 15.32 17.97
2 7.25 4.08 121.80 317.71 16.67 19.06
1 (bottom layer) 7.75 4.36 130.20 363.19 18.01 20.17
The top layer is not the most critical for reinforcement pullout because of no reinforcement
tensile force mobilized at the top of the wall.
As shown in the table, the 12th layer is the most critical for reinforcement pullout due to the
combined effect of the short reinforcement embedment length and low overburden pressure.
The 12th layer:
Pr 30.16
Fpo
= = = 9.36 > 1.5 ok
Tmax 3.22
5.11
Calculate long-term reinforcement tensile strength.

120
Tal = = 26.7 kN/m
2.5 x 1.5 x 1.2
T
Fbr = al
Tmax
Tal 26.7
Tmax
= = = 17.8 kN/m
Fbr 1.5
Solutions 397
Kr
ah = x K a x (y z + q ) - 2c K a = 5.27 z - 5.65
Ka
Determine the vertical spacing of the reinforcement.
Tmax = a h x Sv
Tmax 17.8
Sv = =
ah 5.27 z - 5.65
The bottom layer is the most critical for reinforcement breakage.
at z = 6 m: Sv = 0.68 m
select Sv = 0.6 m for design
Determine the length of the reinforcement.
2
2 x x 0.8 x a v' x Le
P C x F * x a x a v' x Le 0.916x a v' x Le
Fpo = r = = 3 =
Tmax a h x Sv a h x 0.7 ah
1.5
xa h
(H - z)
Le = 0.916 = L-
a v' o'
tan(45 + )
2
1.638x a h (H - z)
L= +
a v' o'
tan(45 + )
2
1.638x (0.32 x 16.5x z - 2 x 5 x 0.32 ) (6 - z ) -16.5 z 2 +114.283z - 16.373
L= + =
16.5x z 1.767 29.156z
The top layer is not the most critical for reinforcement pullout because of no reinforcement
tensile force mobilized at the top of the wall. In the calculation, several reinforcement layers are
evaluated to identify the most critical layer.
at z = 0.3 m: L = 1.87 m
at z = 0.9 m: L = 2.79 m
at z = 1.5 m: L = 2.69 m
at z = 2.1 m; L = 2.46 m
select L = 2.8 m for design
5.13
Calculate Rankine’s active and passive earth pressure coefficients.

30
K ah1 = tan 2 (45 - ) = 0.333
2
20
K ah 2 = tan 2 (45 - ) = 0.490
2
20
K ph 2 = tan 2 (45 + ) = 2.040
2
398 Solutions
Calculate lateral earth forces.
1
Pau1 = x18x 32 x 0.333 = 26.973 kN/m
2
( )
Pau2 = 18 x 3 x 0.49 - 2 x 10 x 0.49 x ( H 0 +3)
= 12.46H 0 + 37.38
Pau 3 =
1
2
( ( ))
(18x 3 + 20( H 0 + 3)) x 0.49 - 2 x100 x 0.49 - 18 x 3 x 0.49 - 2 x 10 x 0.49 x ( H 0 +3)
= 4.9 H 0 2 + 29.4H 0 + 44.1

Ppu1 = 2 x 10 x 2.04 H 0
= 28.57H 0
1
Ppu2 = ( 20 x H 0 x 2.04 ) H 0
2
= 20.4 H 02
Determine the wall embedment length to satisfy the required FS.

Take moment on point O.
Mr
FS = > 1.5
Md
=
Pp1 ( 12 H 0 ) + Pp 2 ( 13 H 0 )
Pa1 (4 + H 0 ) + Pa 2 ( 12 H 0 + 3) + Pa 3 ( 13 H 0 + 3)
H0 > 9 m
Hp = 1.2H0 = 10.8 m
Compare the forces above and below point O to ensure the wall embedment length is sufficient.
Ppu1 =257.13 kN/m

Ppu2 =1652.40 kN/m
Pau1 = 26.973 kN/m
Pau2 = 149.52 kN/m
Pau 3 =705.6 kN/m
a al1 = 74.2 kPa

a al 2 = 91.84 kPa
a pl1 = 628.32 kPa
a pl 2 = 701.76 kPa
Solutions 399
Ppl = 1197.08 kN/m

Pal = 149.44 kN/m
Ppl - Pal = 1047.65 kN/m
R = Ppu - Pau = 1027.4 kN/m
Ppl – Pal > R ok
5.15
Answer to part A
Assume the wall length is less than the depth of groundwater level.
Calculate Rankine’s active and passive earth pressure coefficients.
( 35 )
K ah = K a = tan 2 | 45 - | = 0.271
( 2 )
( 35 )
K ph = K p = tan 2 | 45 + | = 3.69
( 2 )
Calculate lateral earth forces.
a a = K a y ( H e + H p ) = 32.25 + 4.607H p
a p = K a y H p = 3.69 x 17 x H p = 62.73H p
Pa =
1
2
(
x (32.25 + 4.607 H p ) x H p + H e)
= 2.304H p 2 +32.253H p +112.8775 kN/m
1
Pp = x 62.73H p x H p
2
= 31.365H 2p kN/m
Determine the wall embedment length to satisfy the required FS.
M r Pp x L p
FS = = > 1.5
Md Pa x La
(2 )
Pp x | H p + H e - H s |
= ( 3 ) > 1.5
\2 |
( )
Pa x | H p + H e - H s |
|3 |
Hp ≥ 2.7 m
The wall length is 9.7 m, which is still above the groundwater level (Hw = 10 m). The assump-
tion is valid.
400 Solutions
Answer to part B
Summary of resultant forces of the active and passive earth pressures.
a a = 44.69 kPa
a p = 169.371 kPa
Pa = 216.74 kN/m
Pp = 228.65 kN/m
Determine the anchor load.

Take moment on point O, and solve equation to obtain T.
( 7 + 2.7 ) ( 228.65 )
| 216.74x 3 | - | 1.5 x 0.9 |
T =( ) ( )
cos 15x (7 + 2.7 - 1)
= 67.06 kN/m
Solve T again based on the horizontal force equilibrium.
228.65
216.74 -
T= 1.5 = 66.57 kN/m
cos15
For the anchor load with anchor horizontal spacing of 3 m:
FT = 67.06 x 3 = 201.19 kN
Answer to part C
The length of the tie rod should extend beyond the potential failure surface to achieve adequate
anchorage.
o'
( )
L cos a > H e + H p - H s - L sin a tan(45 -
2
)
For simplicity, ignore the Lsinα in the calculation.
L ≥ 4.68 m
Chapter 6
6.1
a. Load factor method

=
Coulomb Theory: K a 0=
.3, K p 6.13
K a , h = K a cos s = 0.3x cos(20.1) = 0.2817

K p , h = K p cos s = 6.13 x cos(20.1) = 5.7566
Solutions 401
At GL-5.5 m:
a a' , h = a v' x K a , h = 88x 0.2817 = 24.79 kPa
At GL-9.0 m:
a a' , h = a v' x K a , h = 144 x 0.2817 = 40.56 kPa
At GL-17.0 m:
a a' , h = a v', a x K a , h = 225.52x 0.2817 = 63.53 kPa

a 'p , h = a v', p x K p , h = 81.52 x 5.76 = 469.28 kPa
Pp , h L p 16581.23 + 2772.96
Fb = = =3.1
Pa , h La 151.84 + 64.394 + 2433.6 + 811.6 + 2772.96
b. Strength reduction method

By trial and error, try Fb = 2.05.
( tan o ) -1 ( tan(30) )
or = tan -1 | | = tan | | = 15.73o
F
( b ) ( 2.05 )
=
Coulomb Theory: K a 0.=
518, K p 2.226
K a , h = K a cos s = 0.518x cos(10.54) = 0.509

K p , h = K p cos s = 2.226 x cos(10..54) = 2.188
At GL-5.5 m:
a a' , h = a v' x K a , h = 88 . 0.509 = 44.79 kPa
At GL-9.0 m.
a a' , h = a v' x K a , h = 144 x 0.509 = 73.296 kPa
At GL-17.0 m.
a a' , h = a v', a x K a , h = 225.52 x 0.509 = 114.79 kPa

a 'p , h = a v', p x K p , h = 81.52x 2.188 = 178.366 kPa
Pp , h L p
≈ 1.0
Pa , h La
Hence, Fb = 2.05.
402 Solutions
6.3

=
.3, K p 6.13
K a , h = K a cos s = 0.3x cos(20.1) = 0.28

K p , h = K p cos s = 6.13 x cos(20.1) = 5..76
At GL-5.5 m:
a a' , h = a v' x K a , h = 88 x 0.2817 = 24.79 kPa
At GL-9.0 m:
a a' , h = a v' x K a , h = 144 x 0.2817 = 40.56 kPa
At GL-(9 + Hp):
( )
a a' , h = a v', a x K a , h = 144 +10.19 H p x 0.2817 = 40.56 + 2.87 H p
a 'p , h = a v', p x K p , h = 10.19 H p x 5.76 = 58.69 H p
Pp , h L p
=Fb = 1.20
Pa , h La
By trial and error, it can be obtained that Hp = 2.454 m.
b. Strength factor method

( tan o ) -1 ( tan(30) )
Fb = 1.2, or = tan -1 | | = tan | | = 25.693o
F
( b ) ( 1.2 )
=
.35, K p 4.3
K a , h = K a cos s = 0.35 x cos(17.2) = 0.334

K p , h = K p cos s = 4.3x cos(17.2) = 4.107
At GL-5.5 m:
s a' , h = s v' x K a , h = 88 x 0.334 = 29.39 kPa
At GL-9.0 m:
a a' , h = a v' x K a , h = 144 x 0.334 = 48.096 kPa
At the wall bottom:
( )
a a' , h = a v', a x K a , h = 144 +10.19 H p x 0.334 = 48.096 + 3.4H p
Solutions 403
a 'p , h = a v', p x K p , h = 10.19 H p x 4.107 = 41.85H p

Pp , h L p
≈ 1.0
Pa , h La
By trial-and-error method, Hp = 3.06 m.
6.5
According to Rankine’s earth pressure theory, o = 0o , K a = K p = 1.0 .

Active earth pressure
At GL-5 m:
a v = (2 x 16 + 3 x17) = 83 kPa, a a = 83x 1 - 2 x 20 = 43 kPa ;
At GL = –9 m:
In clay 2 layer: a v = (2 x 16 + 7 x17) = 151 kPa, a a = 151x 1 - 2 x 20 = 111 kPa ;
In clay 3 layer: a v = (2 x 16 + 7 x17) = 151 kPa, a a = 151x 1 - 2 x 30 = 91 kPa ;
At GL-20 m:
a v = (2 x 16 + 7 x17 + 11x 18) = 349 kPa, a a = 349 x 1 - 2 x 30 = 289 kPa ;
Passive earth pressure

At GL-9 m:
A v = 0 kPa, A a = 0 + 2 x 30 = 60 kPa
At GL-20 m:
A v = (11X 18) = 198 kPa, A a = 198X 1 + 2X 30 = 258 kPa
4
Pa1 = 43 X 4 = 172 kN, L a1 = =2m
2
1 2
Pa 2 = (111 - 43) X 4 X = 136 kN, L a 2 = 4 X = 2.67 m
2 3
11
Pa 3 = 91X 11 = 1001 kN, L a 3 = 4 + = 9.5 m
2
404 Solutions
1 2
Pa 4 = (289 - 91) X111X = 1089 kN, L a 4 = 4 + 11X = 11.33 m
2 3
11
Pp1 = 60 X 11 = 660 kN, L p1 = 4 + = 9.5 m
2
1 2
Pp 2 = (258 - 60) X 11X = 1089 kN, L p 2 = 4 + 11X = 11.33 m
2 3
Mr = eP L p p = 660 X 9.5 + 1089 X 11.33 = 18612 kN-m/m
Md = eP L a a = 172 X 2 + 136X 2.67 + 1001X 9.5 + 1089 X 11.33 = 22558 kN-m/m
Mr
Fb = = 0.83
Md


At GL-5 m:
20
a v = (2 X 16 + 3 X17) = 83 kPa, a a = 83X 1 - 2 X = 30.37 kPa
0.76
At GL-9 m:
20
In clay 2 layer: a v = (2 x 16 + 7 x17) = 151 kPa, a a = 151x 1 - 2x = 98.37 kPa
0.76
30
In clay 3 layer: a v = (2 x 16 + 7 x17) = 151 kPa, a a = 151x 1 - 2x = 72.05 kPa
0.76
At GL-20 m:
30
a v = (2 x 16 + 7 x17 + 11x 18) = 349 kPa, a a = 349 x 1 - 2 x = 270.05 kPa
0.76

At GL-9 m:
30
a v = 0 kPa, a a = 0 + 2 x = 78.95 kPa
0.76
At GL-20 m:
30
o v = (11x 18) = 198 kPa, o a = 198x 1 + 2x = 276.95 kPa
0.76
Solutions 405
4
Pa1 = 30.37 x 4 = 121.48 kN, L a1 = =2m
2
1 2
Pa 2 = (98.37-30.37) x 4 x = 136 kN, L a 2 = 4 x = 2.67 m
2 3
11
Pa 3 = 72.05 x 11 = 792.55 kN, L a 3 = 4 + = 9.5 m
2
1 2
Pa 4 = (270.05-72.05) x 11x = 1089 kN, L a 4 = 4 + 11x = 11.33 m
2 3
11
Pp1 = 78.95 x 11 = 868.45 kN, L p1 = 4 + = 9.5 m
2
1 2
Pp 2 = (276.95-78.95) x 11x = 1089 kN, L p 2 = 4 + 11x = 11.33 m
2 3
Mr =E Pp L p = 868.45 x 9.5 + 1089x 11.33 = 20592 kN-m/m
M = E P L = 121.48x 2 +136 x 2.67 + 792.55x 9.5 + 1089x 11.33 = 20477 kN-m/m

d a a /
E P L / E P L =20592/20477 ~ 1.0 . Hence, F = 0.76.

p p a a b
6.7

Rankine’s earth pressure theory, o = 0o , K a = K p = 1.0 .

At GL-6 m:
a v = (2 x 18 + 4 x19) = 112 kPa, a a = 112 x 1 - 2 x 25 = 62 kPa
At GL-10 m:
In clay 2 layer: a v = (2 x 18 + 8 x19) = 188 kPa, a a = 188 x 1 - 2 x 25 = 138 kPa

In clay 3 layer: a v = (2 x 18 + 8 x19) = 188 kPa, a a = 188x 1 - 2 x 30 = 128 kPa
At GL-21 m:
a v = (2 x 18 + 8 x19 + 18 x11) = 386 kPa, a a = 386x 1 - 2 x 30 = 326 kPa

At GL-9 m:
a v = 0 kPa, a a = 0 + 2 x 30 = 60 kPa
At GL-20 m:
a v = (11x 18) = 198 kPa, a a = 198x 1 + 2x 30 = 258 kPa

406 Solutions
4
Pa1 = 62 x 4 = 248 kN, L a1 = =2m
2
1 2
Pa 2 = (138-62) x 4 x = 152 kN, L a 2 = 4 x = 2.67 m
2 3
11
Pa 3 = 128 x 11 = 1408 kN, L a 3 = 4 + = 9.5 m
2
1 2
Pa 4 = (326-128)) x 11x = 1089 kN, L a 4 = 4 + 11x = 11.33 m
2 3
11
Pp1 = 60 x 11 = 660 kN, L p1 = 4 + = 9.5 m
2
1 2
Pp 2 = (258 - 60) x 11x = 1089 kN, L p 2 = 4 + 11x = 11.33 m
2 3
Mr = EP L p p = 660 x 9.5 + 1089x 11.33 = 18612 kN-m/m
Md = EP L a a = 248 x 2 +1522 x 2.67 + 1408x 9.5 + 1089 x 11.33 = 26619 kN-m/m
Mr
Fb = = 0.7
Md

su1/Fb = 12/0.6 = 20 kPa, su2/Fb = 25/0.6 = 41.7 kPa, su3/Fb = 30/0.6 = 50 kPa

At GL-6 m:
a v = (2 x 18 + 4 x19) = 112 kPa , a a = 112 x 1 - 2 x 41.7 = 28.6 kPa
At GL-10 m:
In clay 2 layer: a v = (2 x 18 + 8 x19) = 188 kPa, a a = 188x 1 - 2 x 41.7 = 104.6 kPa

In clay 3 layer: a v = (2 x 18 + 8 x19) = 188 kPa, a a = 188 x 1 - 2 x 50 = 88 kPa
At GL-21 m:
a v = (2 x 18 + 8 x19 + 18 x11) = 386 kPa, a a = 386 x 1 - 2 x 50 = 286 kPa

At GL-9 m:
a v = 0 kPa, a a = 0 + 2 x 50 = 100 kPa

Solutions 407
At GL-20 m:
a v = (11x 18) = 198 kPa, a a = 198x 1 + 2x 50 = 298 kPa

4
Pa1 = 28.6 x 4 = 114.4 kN, L a1 = =2m
2
1 2
Pa2 = (104.6-28.6) x 4x = 152 kN N, L a2 = 4 x = 2.67 m
2 3
11
Pa 3 = 88 x 11 = 968 kN, L a 3 = 4 + = 9.5 m
2
1 2
Pa 4 = (286-88) x 11x = 1089 kN, L a 4 = 4 + 11x = 11.33 m
2 3
11
Pp1 = 100 x 11 = 1100 kN, L p1 = 4 + = 9.5 m
2
1 2
Pp 2 = (298 - 100) x 11x = 1089 kN, L p 2 = 4 + 11x = 11.33 m
2 3
Mr =E P L = 1100x 9.5 + 1089x11.33 = 22788 kN-m/m
p p
M = E P L = 114.4 x 2 + 152 x 2.67 + 968 x 9.5 + 1089 x 11.33 = 22169 kN-m/m

d a a
E P L / E P L ~ 1.0 . Therefore, F = 0.6.

p p a a b
6.9

At GL-5 m:
a a = 83 - 2 x 10 = 63kN/m 2
At GL-13 m (layer 2):
a a = 219 - 2 x 20 = 179kN/m 2
a a = 219 - 2x 30 = 159kN/m 2
At GL-20 m:
a a = 345 - 2 x 30 = 285kN/m 2

At GL-9 m:
a p = 0 + 2x 30 = 60kN/m 2
408 Solutions
At GL-20 m:
a p = 198 + 2 x 30 = 258kN/m 2
16 ( 2 )
M d = 504 x 4 + 464 x + 1113 x 11.5 + 441x | 7 x + 8 | = 22, 876kN-m/m
3 ( 3 )
( 11 ) ( 2 )
M r = 660 x | + 4 | + 1089 x |11x + 4 | = 18, 612kN-m/m
(2 ) ( 3 )
Mr
Fb
= = 0.81
Md

su1/Fb = 10/0.75 = 13.33 kPa, su2/Fb = 20/0.75 = 26.67 kPa, su3/Fb = 30/0.75 = 40 kPa

At GL-5 m:
a a = 83 - 2x 26.67 = 29.66kN/m 2
a a = 219 - 2 x 26.7 = 165.66kN/m 2
a a = 219 - 2 x 40 = 139kN/m 2
At GL-20 m:
a a = 345 - 2 x 40 = 265kN/m 2

At GL-9 m:
a p = 0 + 2 x 40 = 80kN/m 2
Solutions 409
At GL-20 m:
a p = 198 + 2 x 40 = 278kN/m 2
Mr =E P L = 880x 9.5 + 1089 x11.33 = 20698 kN-m/m

p p
M = E P L = 237.28x 4 + 5444 x 5.3 + 973x 11.5 + 441x 12.7 = 20622 kN-m/m

d a a
E P L / E P L ~ 1.0
p p a a
Therefore, Fb = 0.7.
c. Slip circle method

Assuming the failure surface passes the wall toe, the radius of the failure surface is equal to 15 m.
15 8 8
a1 = tan -1 = 1.31 rad =75.1o , a 2 = cos -1 = 1 rad =57.8o , a 3 = sin -1 = 0.56 rad =32.2o
4 15 15
The weight of the soil column is W = 2 x 15 x 16 + 7 x 15 x 17 = 2265 kN/m
T / 2.4 T / 3.1 T / 5.58
M
| 15x 30 x (15x d0 ) + | 15 x 30x (15 x d0 ) + | 15 x 20x (15x
x d0 )
Fb = r = 0 0 0
= 1.06
Md 2265x15 / 2
d. Terzaghi’s method
D = 16 m. According to Terzaghi’s method, the radius of the failure surface is
B/ 2 =20/ 2 =14.14<D .
Qu 5.7 su2 B / 2 5.7 x 25 x 20 / 2

Fb = = =
W - su1 H e (y H e + qs ) B/ 2 - su1 H e (16 x 2 + 17 x 7 ) 20 / 2 - (10 x 2 + 20 x 7 )
= 1.02
e. Bjerrum and Eide’s method

The radius of the failure surface is:
B 20
= =14.14 m
2 2
L / B = o, H e / B = 9 / 20, then get the N c = 5.9 from the figure.
N c x Su 5.9 x su3 5.9x 30
Fb = = = = 1.17
y x H e +qs y x He 2 x 16 + 7 x17
410 Solutions
6.11
Draw a flow net.
B = 60 m, H p = 8 m, AH w = 9 m, N d = 18, N f = 6
Terzaghi’s method:
( 0.22 + 0.36 ) AH w ( 0.22 + 0.36 ) 9

iavg = 2 = 2 = 0.33
Hp 8
The uplift force:
1 2 1
U= H P iavg y w = x82 x 0.33 x 9.81 = 103.6
2 2
The weight of soil column:
1 2 1
W' = H P (y sat - y w ) = x82 x (20 - 9.81) = 326.1
2 2
W ' 326.1
Fs
= = = 3.15
U 103.6
Harzar’s method:
L = 1.84 m, Ah = 9 / 18 = 0.5 , i=
exit(max) 0=
.5 / 1.84 0.272
y' 10
icr = = = 1.22
y w 9.81
icr 1.22
=Fs = = 4.49
imax(exit) 0.272
6.13
uw = 9.81x 25 = 245.25 kPa

a v = (11 + 5) x 18 = 288 kPa
av 288
Fup = = = 1.17
uw 245.25
6.15
a. Draw the apparent earth pressure

10x 2 + 20 x 7 16 x 2 + 17 x 7
su , avg = = 17.8kPa , y avg = = 16.8kPa
2+7 2+7
Solutions 411
y H e 16.8 x 9
= = 8.5 > 4 (soft to medium clay)
su 17.8
y H e 16.8 x 9
Nb = = = 5.04 < 5.14 (m = 1.0)
sb 30
( 4s ) ( 4 x 17.8 )
K a + |1 - m u | + |1 - | + 0.53
( y He ) ( 16.8 x 9 )
pa ,1 = K a y H e = 0.53 x 16.8 x 9 = 80.14kPa
pa ,2 = K a y H e = 0.30 x 16.8 x 9 = 45.36kPa
Choose pa = 80.14kPa
b. The horizontal strut spacing, L = 5 m

With the half method:
Q1 = (2 + 1.5 + 1.25) x 0.5 x pa x L
Q1 = (2 + 1.5 + 1.25) x 0.5 x 80.136 x 5 = 951.615kPa
Q2 = (1.5 + 1.5) x pa x L
Q2 = (1.5 + 1.5) x 80.136 x 5 = 1202.04kPa
b.
Active earth pressure distribution is assumed: K=
a K=
p 1.
2 su 2 x 10
Depth of tension crack: zc = = = 1.25 m
y sat 16
Active side:
At GL −1.25m, a a = 0kPa
At GL − 2m(upper), a a = a v K a - 2 su K a = (2 x 16) - 2(10) = 12kPa
At GL − 2m(lower), a a = a v K a - 2 su K a = (2 x 16) - 2(20) = -8kPa ~ 0kPa
At GL − 9m(upper), a a = a v K a - 2 su K a = 32 + (7 x 17) - 2(20) = 111kPa
At GL − 9m(lower), a a = a v K a - 2 su K a = 32 + (7 x 17) - 2(30) = 91kPa
At GL − 20m, a a = a v K a - 2 su K a = 151 + (11x 18) - 2(30) = 289kPa

412 Solutions
Passive side:
At GL − 9m, a p = a v K a + 2 su K a = 0 + 2(30) = 60kPa

At GL − 20m, a p = a v K a + 2 su K a = (11x 18) + 2(30) = 258kPa
Location of the assumed support:

Pa la
l= - s = 8.9 m
Pp
The maximum bending moment will occur at the location of the shear force equal to 0.
The maximum bending moment is equal to 508.85 kN-m
M x,max 508.85
The section modulus, S = = = 2.99 x 10-3 m3 .
a all 170x1000
The size and type of the sheet pile can then be selected according to the product manual.
c. The strut load at the second level of the strut Q 2 = 3117.79kN.

The horizontal strut spacing L = 5 m. The uniform pressure acting
on the wale:
=p 3117.
= 79 / L 3117.79 / 5 = 623.558 kN/ m
If the simply supported model is adopted:
1 2 1
M max = pL = x 623.558 x 52 = 1948.62kN -m
8 8
M max 1948.62
The section modulus: S = = = 1.15 x 10-2 m3
a all 170 x1000
If the fixed end beam model is employed:

1 1
M max = pL2 = x 623.558 x 52 = 1299.08kN -m
12 12
M max 1299.08
The section modulus: S = = = 7.64 x 10-3 m3
a all 170 x 1000
6.17
Calculate the system stiffness as:
EI 65000
Sw = = = 80
y w havg
4
9.81x 34
The factor of safety against basal heave using Terzaghi’s method Fb = 1.22.
According to Figure 6.22, we can find s hm / H e = 1.0% .
The maximum wall deflection δhm = 0.01 × 9,000 = 90 mm.
The maximum surface settlement dvm ≈ 0.75δhm = 68 mm.
Solutions 413
Let d represent the distance away from the wall.

According to Clough and O’Rourke’s method:
d = 0, δv = 68 mm
d = 0.75He = 6.75 m, δv = 68 mm
d = 2He = 18 m, δv = 0
Chapter 7
7.1
a. t-z curve
t: mobilized side friction resistance
z: displacement at a depth of pile
The t-z method is a widely used static analysis for developing pile load‒settlement response at
the pile head and vertical load distribution along the pile length.
b. p-y method
p: stands for the lateral soil resistance per unit length of the pile
y: the lateral deflection
It’s a method for determination of later load capacity of piles. It considers the soil‒pile interaction by
using a series of nonlinear springs, which is similar to the method introduced for the mat foundation.
c. In situ pile load test

In order to obtain or estimate the load–deformation relationship of a single pile in actual use, to obtain
a judgment for judging the supporting force of the foundation pile or the integrity of the pile body.
d. Replacement pile
Remove soils by boring first, and then fill the borehole (either with casing or without casing)
with concrete.
7.3
Known: d = 0.8 m
1
Ab = T d 2 = 0.5 m 2
4
Qbu = 9 su Ab = 9 x 80 x 0.5 = 361.73 kN
a method:
Qsu = a x su x As
Qsu (total ) = 88.67 + 323.55 + 341.63 + 651.11 = 1404.96 kN
Qu = Qbu + Qsu (total ) = 361.73 + 1404.96 = 1766.69 kN......Ans
414 Solutions
B method:
Qsu (total ) = E a ' K s tan s As = 700.72 kN
Qu = Qbu + Qsu (total ) = 361.73 + 700.72 = 1062.45 kN......Ans
y method:
d = 20 m , y = 0.173
10.79 x 3 + 48.49 x 7 + 91.38x 4 + 1332.23x 6
a 'v, avg = = 76.54 kPa
20
12 x 3 + 20 x 7 + 50 x 4 + 80 x 6
Cu , avg = = 42.8 kPa
20
Qsu = y (a 'v, avg + 2 x Cu , avg ) As = 0.173(76.54 + 2 x 42.8) x 0.8T x 20 = 1409..96 kN
Qu = Qbu + Qsu = 361.73 + 1409.96 = 1771.69 kN......Ans
7.5
Terzaghi’s method: Pult at s10% diameter
d =1 m
d=
10% .1 m 100 mm
0=
: Pult ~ 4400 tf
7.7
use table (7.32)
| e L e L e e L |
Pult = 9 x su x d 2 | (4 x ( ) 2 + 2 x ( ) 2 + 4 x ( )( ) + 6 x ( ) + 4.5)0.5 - (2 x ( ) + ( ) + 1.5) |
| d d d d d d d |
| 1 2 6 2 1 6 1 1 6 |
= 9 x 30x 1.22 |(4 x ( ) +2 x ( ) +4 x ( )( )+6 x ( )+4.5)0.5 - (2x ( )+(( )+1.5) |
| 1..2 1.2 1.2 1.2 1.2 1.2 1.2 |
= 279.3 kN
7.9
( o') ( 32o )
K P = tan 2 | 45 + | = tan 2 | 45 + = 3.25
( 2) ( 2 |)
y ' = y sat - y w = 20 - 9.88 = 10.2 kN/m3
3
Pult = x K p x y ' x d x L2
2
3
= x 3.75 x 10.2 x 1.2 x 62
2
= 2478.6 kN
M max = K P x y ' x d x L3
= 3.25x 10.2 x 1.22 x 63
= 8592.5 kN . m
Solutions 415
7.11
Assumption: Rigid free-head pile
Pu 600
f = = = 0.694 m
9su d 9 x 80 x 1.2
Take moment at pile tip,
1
Pu ( L + e) - (9 su d )( L - 1.5d )2 + 9 su dz 2 = 0
2
L = 1.5 x 1.2 + 0.694 + 2 z
1
z= ( L - 2.494)
2
Substitute z to the equaation,
-252L2 + 1257.36 L + 11719.92 = 0
: L = 9.756 m
Take moment at max. depth (1.5d + f )
1
M = Pu (e + 1.5d + f ) - (9 su d ) f 2 = 13.288 MN . m
design 2
M max
FS = =3
M
design
: M max = 3 x 13.288 = 39.865 MN . m
7.13
consider the sand arch effect, use L'= 15D =15 m
Z = 15 (m) , f s = Ks x tan(s ) x a 'v,15D

= 1.2 x tan(225o) x (20 - 9.81) x 15
= 85.53 kN/m
85.53
Qs = E f s As = x T x 1x 15 + 85.53 x T x 1x 10 = 4702.26 kN
2
7.15
-
Ap = x 12 = 0.785 m 2 E = 0.67 u s = 0.3 (from table 2.4) Iwp =0.85
4
L 25
U=- L=25 m I ws = 2 + 0.35 = 2 + 0.35 = 3.7
75
d 1
416 Solutions
total settlement Se(total) = Se(1) +Se(2) +Se(3)

(Qwp + E Q ws ) x L (1400 + 0.67 x 2100) x 25
Se(1) =Elastiic settlement = =
Ap x E p 0.78544x 21x 106
= 4.255 x 10-3 m = 4.255 mm
(Qwp ) x d (1400) x1
Se(2) = By the load at the tip = x (1- u s 2 ) x I wp = x (1 - 0.32 ) x 0.85
Ap x Es 0.7854x 35000
=3.9394x 10-2 m = 39.394 mm
( Q ) d
Se(3) = By the load transmitted along the pile shaft = | wss | x x (1- u s 2 ) x I ws
( U x L ) Es
( 2100 ) 1
=| | x x (1 - 0.32 ) x 3.75
( - x 25 ) 35000
07 x 10-3 m = 2.607 mm
=2.60
Se(total) = 4.255 + 39.394 + 2.607 = 46.256 mm
Chapter 8
8.1
tan 0 ' tan 30

Fs = = = 1.732
tan a 1
3
8.3
Table S8.3a
1 2 3 4 5 6 7 8 9 10
Slice no. b (m) Weight (kN/m) α sin α cos α c'b Wsinα Wtan φ (8) + (9)
1 1.2 68.4 60 0.87 0.50 24 59.24 24.90 84.13

2 2.5 370.5 48 0.74 0.67 50 275.34 134.85 410.19
3 2.5 498.75 35 0.57 0.82 50 286.07 181.53 467.60
4 2.5 484.5 23 0.39 0.92 50 189.31 176.34 365.65
5 2.5 356.25 15 0.26 0.97 50 92.20 129.66 221.87
6 2.5 209 4 0.07 1.00 50 14.58 76.07 90.65
7 2 95 −5 −0.09 1.00 40 −8.28 34.58 26.30
8 2 68.4 −13 −0.22 0.97 40 −15.39 24.90 9.51
9 2 26.6 −20 −0.34 0.94 40 −9.10 9.68 0.58
− − − − − − Total 883.97 − −
Solutions 417
Table S8.3b
1 1st try 2nd try 3rd try 4th try 5th try
Fs = 1.5 Fs = 1.3 Fs = 1.28 Fs = 1.27 Fs = 1.26
Slice no. mα (10)/m α mα (10)/m α mα (10)/m α mα (10)/m α mα (10)/m α
1 1.32 63.74 1.45 58.18 1.46 57.59 1.47 57.29 1.48 56.99
2 1.41 291.49 1.52 269.72 1.53 267.39 1.54 266.20 1.55 265.01
3 1.44 323.78 1.54 303.57 1.55 301.37 1.56 300.25 1.56 299.12
4 1.42 256.84 1.50 243.60 1.51 242.14 1.51 241.39 1.52 240.64
5 1.38 160.64 1.44 153.54 1.45 152.75 1.46 152.35 1.46 151.94
6 1.29 70.45 1.33 68.10 1.34 67.83 1.34 67.69 1.34 67.56
7 1.18 22.27 1.21 21.75 1.21 21.69 1.21 21.66 1.22 21.63
8 1.07 8.91 1.08 8.79 1.08 8.78 1.08 8.77 1.08 8.77
9 0.95 0.61 0.96 0.61 0.96 0.61 0.96 0.61 0.96 0.61
Total 1,198.74 Total 1,127.86 Total 1,120.14 Total 1,116.22 Total 1,112.28
1. First trial and error.
1
(E )c ' b +W tan 0 '
ma 1198.74
Fs = = = 1.36
(EW sin a ) 883.97
2. Second trial and error.
1
(E )(c'b +W tan 0 ')
ma 1127.86
Fs = = 1.28
(EW sin a ) 883.97
3. Third trial and error.
1
(E )(c ' b +W tan 0 ')
ma 1120.14
Fs = = = 1.27
(EW sin a ) 883.97
4. Fourth trial and error.
1
(E )(c ' b +W tan 0 ')
ma 1116.22
Fs = = = 1.26
(EW sin a ) 883.97
5. Fifth trial and error.
1
(E )(c ' b +W tan 0 ')
ma 1112.28
Fs = = = 1.26 (OK)
(EW sin a ) 883.97
Factory of safety = 1.26

418 Solutions
8.5
c 40
= = 0.29
y H tan 0 ' 16x 15 x tan 30
tan 0 '
From Figure 8.13, for the 40º, the corresponding value of = 0.95 .
Fs
Factory of safety = 0.61
Index
A cohesionless soil, 61
adhesion between wall and soil, see cohesion, properties, 62–66, 71–72
between the wall and soil cohesive soil, 61
allowable bearing capacity, 103–104, 118–119 properties, 66–68, 86–87
angular distortion, 120–121 combined footing, 90–91
assumed support method, 270, 273, 275 compensated foundation, mat, 123–125
Atterberg limit compression index, 61, 75
liquid index, 67 cone penetration test, 18, 21
liquid limit (or LL), 17–18, 46 cone tip resistance, 65
plasticity index (or PI), 46, 206 correlations, 66–67, 72
plastic limit (or PL), 17, 18 consolidation, 12, 43, 47, 116–117
continuous footing, 90–91
B creep, 210, 342
basal heave analysis method, see base shear failure critical depth, 313
analysis method cross wall (or CW), 280–283, 289
basal heave, see base shear failure
base shear failure, 84–85, 219, 245–249, 251–253, D
255, 257–258, 264, 276, 278, 289 deflection, 242, 264, 275–281, 289
base shear failure analysis method deformation, 52–53, 61, 68, 75, 80, 178, 203,
Bjerrum and Eide’s, 249–252, 258–259, 289 206–209, 217–219, 242, 266, 275–276,
earth pressure equilibrium, load factor, 246, 281–283
247, 253, 255, 259 depth factor, bearing capacity, 100
earth pressure equilibrium, strength factor, 247, diaphragm wall, 181, 189, 217, 241, 285, 289
257, 259 displacement, 79–83, 87, 134, 157, 169, 189, 207,
slip circle, 84, 252–253, 258–259, 289 219, 266, 284–286, 289
Terzaghi’s, 247–249, 252, 258–259 drainage, 191, 206
beam on elastic foundation, 131, 337, 338 drained
bottom-up construction method, 244 analysis, 51, 53, 61, 98–99, 101–102, 140
braced excavation, 242–244 behavior, 51–53, 55, 61, 86
building protection, 185 material, 53, 86
buttress wall (or BW), 280–283, 289 shear strength, 60, 61, 87
C E
cantilevered wall, 217, 221–223 earth pressure
characteristic soil parameters, 81–86 apparent, 265–266, 269, 270–273, 289
coefficient of vertical subgrade reaction Caquot-Kerisel, 160–162, 169, 170
coefficient, 125–128 Coulomb, active, 152–154, 158, 159, 167,
cohesion 169, 170
effective, 18, 61, 99, 140, 253, 313, 319 Coulomb, passive, 154–156, 158, 160, 162,
between the wall and soil, 193, 259, 168–170
316–318, 348 earthquake, 166–169
420 Index
Rankine, active, 140, 141, 145, 147, 149, I

154, 169 immediate settlement, 111–116
Rankine, passive, 144, 151, 154, 160, 162, 170 inclination factor, bearing capacity, 100
Rankine, sloping ground, 145 inclinometer, 284–285, 289, 379
at rest, 61, 75, 134–137, 164, 169 index
earthquake, 166–169, 192, 219 compression, 17–18, 47
earth retaining wall swelling, 47
classification, 176–178
selection, 184–187 L
eccentrically loaded foundation, 106–111 lateral load capacity
effective stress analysis, 53–55, 86–87, 98–99, nonrigid pile, 337
101–102, 140, 144, 149, 247, 265 rigid pile, 330
parameters, 61 limit state design method, 80–81
load transfer mechanism, 311
F local shear failure, 92–94
factor of safety long-term
excavation, basal heave (see excavation, base analysis, 58, 98–99, 101–102, 371
shear failure) behavior, 53, 142
excavation, base shear failure, 246–249, 252, durability, 210–211
253, 255, 257, 258, 264, 278, 289 load condition, 71, 79
excavation, sand boiling, 262 performance, 83
excavation, upheaval failure, 260 strength, 57, 60–61, 210–213
pile, 1, 6–7, 21, 41, 48
retaining wall, 3–5, 192 M
retaining wall, bearing capacity, 197–199 mat foundation, 90–91, 123–128
retaining wall, facing connection, 212–213 mechanically stabilized earth wall (or MSE wall),
retaining wall, overturning, 194 178, 203
retaining wall, reinforcement breakage, modulus, 61, 68–74
210–211 Mohr-Coulomb
retaining wall, reinforcement pullout, failure criterion, 61, 193, 361, 362, 367, 370,
211–212 375
retaining wall, sliding, 195 failure line, 58
shallow foundation, bearing capacity, 3, 21, theory, 193
31–32, 41–43, 47–48, 103
slope, 356, 359, 361, 363–369, 371, 372, N
374–376, 378 negative skin friction, 347
failure modes, 191, 206, 218 neutral plane, 347
fixed earth support method, 219–223 nongravity wall, 181, 217
flow net, 253, 262, 263
free earth support method, 219–221, 223–225 O
friction angle overconsolidation, 372
drained, 61–64, 265, 267, 372, 374
effective (see drained) P
soil-reinforcement interface, 207–208, 212 penetration depth, 33, 62, 247, 249, 252–253, 255,
soil-wall interface, 135, 169, 192–193, 253, 267 262, 314
undrained, 51 performance-based design method, 81–83
full casing pile, 309–310 permeability, 17–18, 32, 34–35, 43, 49, 191, 206
piled raft foundation, 128–130
G pile installation
general shear failure, 92–94 bored pile, 307
geosynthetic-reinforced soil wall (or GRS wall), driven pile, 301
178, 203 pile load capacity
geosynthetics, 203, 205, 209 end-bearing resistance, 312
gravity, and semi-gravity wall, 178, 189 frictional resistance, 315
ground improvement, see soil improvement uplift, 321
group pile, 348–350 pile load test, 341, 342–343
group efficiency, 351 interpretation of test results, 342
Index 421
maintain load test, 342 strut load, 245, 264, 268, 269, 272,
quick test, 342 287, 289
plate loading test, 122–123 support wall, 217, 221–227
poisson’s ratio, 28, 71
preconsolidation stress, 117 T
punching shear failure, 92–94 tilt angle, 285
p-y method, 338 tiltmeter, 285–286, 289
tolerable settlement, shallow foundation,
R 119–121
Rankine active zone, 101 top-down construction, 242, 244–245, 289
Rankine passive zone, 95 total stress analysis, 51, 53–55, 86, 98, 102, 266,
reinforced wall, 181, 203 371
reverse circulation drilling (or RCD), 308 triaxial test, 34, 49
RQD, 16 t-z curve, 322
S U
safety factor, see factor of safety ultimate bearing capacity, 94–102
sand boiling, 219, 245, 261 ultimate strength design method, 79
settlement undrained
due to excavation, 242, 264, 275–281, analysis, 51, 53–54, 61, 142, 149, 247
284–286, 289 behavior, 51–54, 59, 61
due to foundation, 32, 41–43, 46–48, 111–121, material, 53, 55, 86
186, 191–192, 199, 207 shear strength, 18, 55, 60, 61, 66–68, 72, 81,
due to pile, 345 84–87, 98–99, 102, 142, 144, 249–250,
various definitions, 126, 128, 130 253, 257, 266–268, 278, 372
shape factor, bearing capacity, 100 undrained, 18, 54, 61, 71
shear strength, 61
cohesionless soil, 62–66 V
cohesive soil, 66–68 vane shear test, 6, 18, 24–26, 43, 49
sheet pile, 181, 217, 241, 262, 272, 274–275, 278 correlations, 67–68
short-term vertical stress, 116
analysis, 58, 98–99, 102, 371
behavior, 52 W
load condition, 79 wale, 241–242, 269, 272
performance, 83 water table, effect on bearing capacity,
strength, 60 104–106
soil improvement, 253, 279–280, 289 Winkler foundation, 125–128
soil spring constant, see coefficient of vertical working stress design method, 78–79
subgrade reaction
soldier pile, 181, 217, 241, 278 Y
stability number, 372 Young’s modulus
standard penetration test, 19–20 effective (or drained), 53, 61
correlations, 64–65, 68, 71–73 initial, 18, 28
N value, 19–20, 62–64 secant, 71, 112, 115
stress path tangent, 71, 72
effective, 57–58, 60 unloading/reloading (or elastic), 51, 61,
total, 57–58, 60 75, 249

Fundamentals of Foundation Engineering (2023)

Uploaded by

Copyright:

Available Formats

Fundamentals of Foundation Engineering (2023)

Uploaded by

Document Information

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

Fundamentals of Foundation Engineering (2023)

Uploaded by

Copyright:

Available Formats

Fundamentals of Foundation Engineering

Chang-Yu Ou, Kuo-Hsin Yang, Fuchen

ISBN: 978-1-032-39498-5 (hbk)

1 Geotechnical site investigation 1

1.10 Reports and interpretation 41

2 Principles of foundation design 51

3.4.1 Ultimate bearing capacity 91

4 Lateral earth pressure 134

5 Earth retaining structures 175

6.6 Control of excavation-induced movement 279

7 Pile foundations 295

8 Slope stability 356

8.3.3 Infinite slope 361

Chang-Yu Ou is a chair professor of the Department of Civil and Construction Engineering

Geotechnical site investigation

1.2 Plan for subsurface exploration

1.2.1 Investigation methods

1.2.2 Site investigation layout

Building Shallow foundation ○ – – – ○ ○ ○ – – – ○ – – – ○ ○

Legend: ○, often used; △, sometimes used; –, rarely

1.2.3 Depth of subsurface exploration

can be determined by the following rules:

as q equals estimated net stress on the foundation.

1.3 Borehole exploration

1.3.1 Wash boring method

1.3.2 Rotary drilling method

Figure 1.5 Fishtail drill bit.

Figure 1.6 Schematic of wash boring operation.

Figure 1.7 Diamond drill bits of the drill bit.

The same system to wash boring in Figure1.6

Diamond casing shoe

Figure 1.8 Typical rig with wireline coring equipment.

1.4 Soil sampling methods

1.4.1 Thin-walled tube sampler

1.4.2 Split-spoon sampler

1.4.3 Piston sampler

1.4.4 Core sampler

Figure 1.11 Sampling pr ocess by piston sampler.

Figure 1.12 Triple-tube core barrel.

Table 1.2 Common borehole sizes

Core barrel type Hole diameter Core diameter Core sample

Table 1.3 RQD classification index

RQD (%) Rock mass quality

<25% Very poor

1.5 Sampling disturbance

di = inside diameter of the sampler tube.

An example is shown to calculate the area ratio in Figure 1.12:

Table 1.4 Test for disturbed samples and undisturbed samples

Disturbed samples Undisturbed samples

Sieve analysis Shear strength

Table 1.5 Common laboratory tests for soils

Laboratory tests Soil property

General physical property tests

Water (moisture) content test ω, γ sat , γ d

Constant head permeability test k (for coarse-grained soil)

Oedometer test - c', OCR, C c , C ur

Shear strength tests

Direct shear test c′, 0 '

1.6 Laboratory test

Symbols used in Table 1.5