SIMULATION ANALYSIS OF SLOPE STABILITY: A CASE STUDY ON SLOPE FAILURE AT NEW LABORATORY OF FACULTY OF MECHANICAL ENGINEERING.UTM FELIX LING NGEE LEH A project report submitted in partial fulfillment of the requirement for the award of the degree of Master of Engineering (Civil-Geotechnics) Faculty of Civil Engineering Universiti Teknologi Malaysia OCTOBER 2003Universiti Teknologi Malaysia BORANG PENGESAHAN STATUS TESIS* JUDUL:_SIMULATION ANALYSIS OF SLOPE STABILITY: ‘A CASE STUDY ON SLOPE FAILURE AT NEW LABORATORY ‘OF FACULTY OF MECHANICAL ENGINEERING UTM. ‘SESI PENGASIAN: 2003/2004 1 Saya LIX. NGEE LEH (HURUF BESAR) ‘mengaku membenarkan tesis (PSM/Sarjana/Doktor Faisafah)* ini disimpan di Perpustakaan Universiti Teknologi Malaysia dengan syarat-syarat kegunaan seperti berikut: Tesis adalah hakmilik Universiti Teknologi Malaysia. Perpustakaan Universiti Teknologi Malaysia dibenarkan membuat salinan untuk tujuan pengajian sahaja. Perpustakaan dibenarkan membuat salinan tesis ini sebagai bahan pertukaran antara institusi pengajian tinggi. **Sila tandakan (V) (Mengandungi maklumat yang berdarjah keselamatan atau ‘epentingan Malaysia seperti yang termaktub di dalam AKTA RAHSIA RASMI 1972) (Mengandungi maklumat TERHAD yang telah ditentukan oleh organisasi/badan di mana penyelidikan dijalankan) EI ‘TIDAK TERHAD Me (TANDATANGAN PENULIS) (TANBAT'ANGAN PENYELIA) Alamat Tetap: 4G, Jalan il EN. \N KASS] ‘Sarawak, (Nama Penyelia) 8 OKTOBER 2003, Tarik 8 OKTOBER 2003 CATATAN TPotong yang dak borkenan ‘+ Sica teis int SULIT atau TERILAD, sit lampickan surat daripadapihak berkuasslorganisasi ‘perkenaan dengan menyatakan skal sebab dan empoh esis ini pets dikeaskan sebagai ‘SULIT stau TERHAD. ‘+ Tesi dimaksudkan scbagai tess bagi jazah Doktor Fasafth dan Sarjanasecara penyelidikan atau disertasi bagi pongaian secara kerja kursus dan penyeikan, ‘tau LaporanProjek Sarjana Moda (FSM),ii “I declare that this project report is my own work except for the quotations and summarizes which I have explained the source” Signature : pee th Name of Author: FELIX LING NGEE LEH Date F 8 October 2003To my be loved ones and those who love me iiiiv ACKNOWLEDGEMENTS I would like to express my sincere gratitude to my supervisor of master project, Mr. Azman Kassim for his guidance, inspiration, support and friendship. Lam very grateful to lecturers of the faculty, especially Madam Fauziah Kassim, Mr. Nazli Ali and Associate Professor Z’aba Ismail for their guidance and knowledge that went into this project report. 1 would like to thank my employee Kole} Universiti Teknologi Tun Hussein ‘Onn (KUiTTHO) who made it possible for me to study here. I would like to thank my parents, brothers, sisters and my friends for their love and support.ABSTRACT The soil movement on a failed slope had caused substantial failure of sheet pile wall positioned at about 6m from the new laboratory of Faculty of Mechanical ia (UTM). To facilitate investigating ‘causes of the failure, a computer simulation of slope stability using PLAXIS 7.2 is performed to simulate slope conditions before and after construction of the study area. The results of simulation analysis establishes the fact that global soil mass movement started from the elevation of the newly filled water tank to downhill direction toward to installed sheet pile generates a combination of mobilized shear force and lateral pressure larger than the capacity or strength of the sheet pile. The associated stresses also create the occurrence of the up heaving underneath the road Engineering, Universiti Teknologi Mal pavement adjacent to the new buildings. Furthermore, the simulation analysis deduces that the slope instability becomes greater as moisture or pore-water pressure in the slope increases or decreasing in soil’s shear strength.vi ABSTRAK Pergerakan bumi atas cerun telah menyebabkan kegagalan tembok cerucuk kkeping yang terletak kira-kira 6m daripada Bangunan Tambahan Fakulti Kejuruteraan Mekanikal (FKM), Universiti Teknologi Malaysia (UTM). Model simulasi dengan penggunaan perisian PLAXIS 7.2 telah dijalankan untuk menentukan kestabilan cerun sebelum dan selepas aktiviti pembinaan di kawasan tersebut. Model simulasi ini adalah bertujuan untuk mengenalpasti punca-punca dan sebab-sebab berlakunya kegagalan. Keputusan analisis simulasi menunjukkan bahawa pergerakan bumi ini adalah berbentuk serantau bermula daripada puncak bukit di mana lokasi tangki air yang baru dibina menuju ke arah kaki bukit di mana tembok cerucuk keping dibina. Daya ricih dan tekanan sisi yang diaruh oleh pergerakan ini adalah lebih besar daripada kekuatan cerncuk keping yang dibina dan telah menyebabkan bonggokkan tanah dibawah jalan raya sebelah bangunan baru tersebut. Keputusan analisis ini juga mendapati bahawa kestabilan cerun akan terjejas dengan kenaikan tekanan air atan dengan penurunan kekuatan ricih tanah.CHAPTER CHAPTER I CHAPTER TABLE OF CONTENTS ITEM DECLARATION DEDICATION ACKNOWLEDGEMENTS ABSTRACT ABSTRAK TABLE OF CONTENTS LIST OF FIGURES LIST OF TABLES LIST OF SYMBOLS LIST OF PHOTOS INTRODUCTION 1.1 Background of the Study 1.2. Problem Statement 1.3 Objective and Scope of the Study LITERATURE REVIEW 2.1, Introduction 2.2 Investigation of Slope Failure 2.2.1 Types of Slope FailureCHAPTER IL 2.2.2 Classification of Slope Failure Review of Slope Stability Analysis 23.1 History of Development 23.2. Types of Analysis 2.4 Review of Limit Equilibrium Theory in Slope Stability Analyses 24.1 Method of Slices Review of Finite Element Theory in Slope Stability Analysis 2.5.1 Enhanced Limit Methods 2.5.2 Direct Methods 2.5.3 Information Required for Finite Element Analysis 2.5.4 The Use of FE Software Package 2.6 Back Analysis 2.6.1 Application of Back Analysis in Slope Stability Analysis 2.6.2 Limitations of Back Analysis 2.7 Closing Remark fa & 2 RESEARCH METHODOLOGY 3.1 Introduction 3.2. Computer Modeling 3.2.1 PLAXIS 3.3. Parametric Study 3.3.1 Stability Analysis 3.3.2. Results of Analysis and Interpretations 3.3.3 Discussion of Parametric Study 3.4 Landslide Occurrence: A Case Study 3.4.1 General 3.42 Description of Study Area viii 15 17 7 19 23 34 36 39 41 42 46 49 50 53 54 54 55 55 61 67 67 67CHAPTER IV CHAPTER V REFERENCES 3.43 General Geology of the Study Area 3.44 Simulation Modelling of Slope ANALYSES AND DISCUSSIONS 4.1. Introduction 4.2. Results of Analysis and Interpretations: 4.2.1 Case I- Original/natural Condition 4.2.2 Case Il- After Cut to Platform Level 4.2.3 Case Ill- After Occurrence of Landslide 4.2.4 Back Analyses 42.5. Deformation of Sheet Pile Wall 4.3 Discussion of Case Study CONCLUSIONS AND RECOMMENDATIONS 5.1 General 5.2. Conclusions 5.3 Recommendations for Future Work 4 4 86 86 87 87 87 87 88 88 151 152 153FIGURE Figure 2.1 Figure 2.2 Figure 2.3 Figure 2.4 Figure 2.5 Figure 2.6 Figure 2.7 Figure 2.8 Figure 2.9 Figure 2.10 Figure 2.11 Figure 2.12 Figure 3.1 Figure 3.2 Figure 3.3 Figure 3.4 Figure 3.5 Figure 3.6 LIST OF FIGURES ITEM Types of slope failure Velocity of movement for slope failure forms Slopes, slices and general unknowns in limit equilibrium methods Conventional effective stress analysis applied to critical CD case for unloading problem ‘The forces acting on a slice: Fellenius analysis : Bishop simplified analysis Janbu’s correction factor for non-circular failure The forces acting on a sli surfaces A typical clement in the Morgenstern-Price method Finite Element approaches proposed in computing the factor of safety in a slope stability analysis Basic back analysis approaches applied for the slope forming materials obeying linear failure envelopes Value of back analysis Value of back analysis, continued Overview of research methodology The overview of parametric study Cross section of idealized slope for parametric study Location of study area Initial proposed platform level Extend of slides PAGE 10 14 20 22 29 30 33 35 48 SI 52 56 56 57 68 70 7Figure 3.7 Figure 3.8 Figure 3.9 Figure 3.10 Figure 4.1 Figure 4.2 Figure 4.3 Figure 4.4 Figure 4.5 Figure 4.6 Figure 4.7 Figure 4.8 Figure 4.9 Figure 4.10 Figure 4.11 Figure 4.12 Figure 4.13 Figure 4.14 Figure 4.15 New proposed platform level & location of installed sheet pile Schematic diagram of slope failures The slope cross sections The overview of case study Case I- Without water tank (Zero pore water pressure) Case I- Without water tank (Water table at 25m below crest of slope) Case I- Without water tank (Water table at 10m below crest of slope) Case I- Without water tank (Water table at Sm below crest of slope) Case I- With water tank- No displacements (Zero pore water pressure) Case I- With water tank — No displacements (Water table at 25m below crest of slope) Case I- With water tank ~ No displacements (Water table at 10m below crest of slope) Case I- With water tank — No displacements (Water table at 5m below crest of slope) Case I- With water tank- Displacements allowed (Zero pore water pressure) Case I- With water tank — Displacements allowed (Water table at 25m below crest of slope) Case I- With water tank — Displacements allowed (Water table at 10m below crest of slope) Case I- With water tank — Displacements allowed (Water table at Sm below crest of slope) Case II- Without water tank (Zero pore water pressure) Case II- Without water tank (Water table at 25m below crest of slope) Case I- Without water tank (Water table at 10m below crest of slope) 3 77 78 92 93 95 97 98 103 105 107Figure 4.16 Figure 4.17 Figure 4.18 Figure 4.19 Figure 4.20 Figure 4.21 Figure 4,22 Figure 4.23 Figure 4.24 Figure 4.25 Figure 4.26 Figure 4.27 Figure 4.28 Figure 4.29 Figure 4.30 Figure 4.31 Figure 4.32 Case II- Without water tank (Water table at Sm below crest of slope) Case Il- With water tank- No displacements (Zero pore water pressure) Case I1- With water tank — No displacements (Water table at 25m below crest of slope) Case Il- With water tank —_No displacements (Water table at 10m below crest of slope) Case Il- With water tank —No displacements (Water table at Sm below crest of slope) Case Il- With water tank- Displacements allowed (Zero pore water pressure) Case II- With water tank — Displacements allowed (Water table at 25m below crest of slope) Case Hl- With water tank — Displacements allowed (Water table at 10m below crest of slope) Case Il- With water tank ~ Displacements allowed (Water table at 5m below crest of slope) Case III- Without water tank (Zero pore water pressure) Case Il- Without water tank (Water table at 25m below crest of slope) Case III- Without water tank (Water table at 10m below crest of slope) Case IIl- Without water tank (Water table at Sm below crest of slope) Case III- With water tank- No displacements (Zero pore water pressure) Case Ill- With water tank — No displacements (Water table at 25m below crest of slope) Case III- With water tank — No displacements (Water table at 10m below crest of slope) Case IH With water tank — No displacements (Water table at Sm below crest of slope) xii 108 109 10 il 2 113 4 4s 116 1g 119 120 121 122 123 124 125Figure 4.33 Figure 4.34 Figure 4.35 Figure 4.36 Figure 4.37 Figure 4.38 Figure 4.39 Figure 4.40 Figure 4.41 Figure 4.42 Figure 4.43 Figure 4.44 Figure 4.45 Case IIl- With water tank- Displacements allowed (Zero pore water pressure) Case IIl- With water tank — Displacements allowed (Water table at 25m below crest of slope) Case III- With water tank — Displacements allowed (Water table at 10m below crest of slope) Case I- With water tank — Displacements allowed (Water table at Sm below crest of slope) ‘Combinations of back-calculated shear strength on failure plane for case without water tank when safety factor is equal to one Combinations of back-calculated shear strength on failure plane for case with water tank (No deformation is allowed for water tank) when safety factor is equal to one Combinations of back-calculated shear strength on failure plane for case with water tank (Deformation of water tank is allowed) when safety factor is equal to one Deformation of sheet pile wall — with water tank ~No displacements (zero pore water pressure) Deformation of sheet pile wall — with water tank — No displacements (water table at 25m below crest of slope) Deformation of sheet pile wall — with water tank —No displacements (water table at 10m below crest of slope) Deformation of sheet pile wall — with water tank — No displacements (water table at Sm below crest of slope) Deformation of sheet pile wall — with water tank — Displacements allowed (zero pore water pressure) Deformation of sheet pile wall — with water tank — Displacements allowed (water table at 25m below crest of slope) xiii 126 127 128 129 131 132 133 135 137 139 141 143 145,Figure 4.46 Figure 4.47 xiv Deformation of sheet pile wall — with water tank ~ Displacements allowed (water table at 10m below crest of slope) 147 Deformation of sheet pile wall - with water tank — Displacements allowed (water table at Sm below crest of slope) 149TABLE Table 2.1 Table 2.2 Table 2.3 Table 2.4 Table 2.5 Table 3.1 Table 3.2 Table 3.3 Table 3.4 Table 3.5 Table 3.6 Table 3.7 Table 3.8 Table 3.9 Table 3.10 LIST OF TABLES: 1TEM ‘Vames (1978) classification system A classification of slope failures Basic solution requirements satisfied by the various methods of analysis, List of commonly used method of slices: assumptions concerning interslice forces for different method of slices Characteristics of equilibrium methods of slope stability analysis Material properties of idealized slope Descriptions and assumptions for simulations of slope with zero pore water pressure Descriptions and assumptions for simulations of slope with piezometric line at three quarters of the slope height Changing of effective cohesion, c* Changing of effective friction angle, 6” Changing of unit weight, y Changing of poisson ratio, v Soil properties for simulation analysis Descriptions and assumptions for simulations of the slope stability analysis for original/natural slope Descriptions and assumptions for simulations of the slope stability analysis for slope after cut to platform level xv PAGE 12 13 16 24 25 58 59 62 63 65 19 81Table 3.11 Table 3.12 Table 3.13 Table 4.1 Table 4.2 Table 4.3, Table 4.4 Table 4.5 Descriptions and assumptions for simulations of the slope stability analysis for slope after occurrence of landsli Properties of the sheet pile section ESC-B8, Grade $275 Descriptions and assumptions for back analysis of soil strengths at failure Calculated safety factor and deformation of original/ ‘Natural slope Calculated safety factor and deformation of slope after cut to platform level Calculated safety factor and deformation of slope after occurrence of landslide Back-Calculated soil strength at failure Deformation of sheet pile wall xvi 82 3 85 1 104 nz 130SYMBOL kp SPREE You Yary Yoet LIST OF SYMBOLS PARAMETER Effective cohesion Modulus Young Hydraulic gradient Coefficient of permeability (or hydraulic conductivity) Coefficient of permeability (horizontal flow) Coefficient of permeability (vertical flow) Coefficient of Active earth pressure Coefficient of earth pressure at rest Coefficient of Passive earth pressure Effective friction angle Poisson ratio Bulk unit weight Dry unit weight Wet unit weight Shear stress Xvi UNITS KN/m? or kPa N/m? Degrees KN/m? kN? KN/m* Vm? or kPaLIST OF PHOTOS PHOTO ITEM PAGE 3a Physical damages of sheet pile wall — left side view 5 32 Physical damages of sheet pile wall — right side view 8CHAPTER I INTRODUCTION 1.1 Background of the Study Population growth and accompanying economic development have meant that civil engineering projects are increasingly being carried out in mountainous regions in tropical countries. Wesley (1994) found that probably highway construction is the most important among those hilly development, however hydroelectric construction, geothermal power plants, and recreational resorts also constitute major projects. ‘Over the last decade, we have had numerous landslides culminating in tragedies and it is no coincidence that they all happened in hilly regions, The fact is, slopes are inherently unstable. While erosion in the form of soil run-offs occurs daily, slopes can collapse and result in destructive landslides, Slope failures increase when a naturally occurring weak zone or what is called a “fault plane” in geological terminology is present. Landslides on November 20, 2002 at Taman Hillview, Ampang, occurred in such a high-risk zone. Actually, this kind of tragedy is not a new story in our country for instance the infamous Highland Tower Tragedy that happened in 1993 which is just at the side of the tragedy that happened. Landslides and other gravity-stimulated mass movements are important and costly problem, and they are a continual source of concern for geotechnical engineers and engineering geologists throughout the world, particularly in geologically ‘active?regions. They occur worldwide and are described as sudden, short-lived geomorphic events that involve the rapid-to-slow descent of soil or rock in sloping terrains. They ‘can occur on any terrain given the right conditions of soil, moisture and the angle of slope. Risks of landslides are enhanced in the tropics, where thick, loose residual soil, the result of deep weathering, can be easily eroded, Until and unless a holistic stance in tackling landslides is taken, lives and properties will be at stake and taxpayers’ money will be wasted on rescue missions. Although many mitigation works had been planned and designed prior to the construction of the project, there still exist many uncertainties associated with the ‘material, spanning from its complex origin. Hence, its come the importance to analyze the stability of the existing slope. However, most of the times it is too costly ‘or impossible to monitor the slope for the whole of its service life. 1.2 Problem Statement There are many circumstances in slopes, where the civil engineer must investigate the stability of slope by performing slope stability analysis. Over the years, much research work had been carried out by researchers namely, Wu and Kraft (1970), Comell (1971), Alonzo (1976), Tang et al. (1976), and Vanamreke (1977), regarding the reliability analyses of slopes. They found that the uncertainties ‘occur due to the variability of soil properties, systematic errors or model errors in measurement of properties, and model errors in analytical methods (Oka & Wu, 1990). It should be noted that all calculations for their work ‘were made for the critical slip surface. The search for the critical failure surface, carried out over a ‘number of admissible slip surfaces, is by and large still performed by repeated trials or by the grid search method (Nguyen, 1985), Existing methods of slope stability analysis using slices (Bishop 1955, Janbu 1957) are based on the limit equilibrium theorem. An implicit assumption inequilibrium analyses of slope stability is that the stress-strain behaviour of the soil is ductile, ic, that the soil does not have a brittle stress-strain curve (where the shearing resistance drops off after reaching a peak). This limitation results from the fact that the methods provide neither information regarding the magnitudes of the strains within the slope, nor any indication about how they may vary along the slip surface (Duncan, 1996). Besides it, the analysis only considered force and moment acting on the slices with total disregard to the deformation developed in the slices. ‘Thus, itis not possible to obtain reliable results from the analyses if solely based on the method of slices (Terado et al., 1999). Thus, in order to obtain a unique solution it is necessary to introduce extra conditions. Better analysis should therefore take into account the displacement and deformation of the slices, and also the stresses in the soil mass in determining the stability of slope. However, the problem arises in incorporating these extra conditions in the conventional slope stability analysis, In the other hand, the stability analyses are performed not only to provide a factor of safety once the soil properties are known, but also to establish field shear strengths from the study of failures. It is rational to carry out the study determining what actually happened after an unexpected instability has occurred. It is therefore necessary to do some analyses in reverse, which is usually termed as “back analysis” ‘The investigation is not mean to blame who or whom should be responsive to the failure but it collects valuable information that could be used in designing the remedial works as well as guidelines for further projects. The awareness of importance of back analysis has resulted in development of various methods in back analysis. However, the problem always arises in determining the suitable method of analysis and the way back analysis can be carried out.1.3. Objective and Scope of the Study ‘The objective of this study is to determine the stability of the slope before and after the construction of Bangunan Tambahan Fakulti Kejuruteraan Mekanikal, UTM. Stability analysis of slope is carried out based on the computer modelling using PLAXIS V7.2 (professional edition), a finite clement package. A real case study of slope failure was chosen in fulfilling the objective of the study. Shear strength reduction technique is chosen for the determination of safety factor for its formulation based on finite element, The critical failure surface is found automatically. The values of shear strength parameters (c, 4) at failure along the failure plane are back calculated with the factor of safety is assumed to be unity. The analysis is based on long-term condition. Drained analysis is used in this simulation modelling. For the simplicity of the analysis, unsaturated soil condition is out of the scope of study. Stresses of the soil mass along the critical slip surfaces as well as the displacement and deformation are determined using the theory of finite element.CHAPTER IL LITERATURE REVIEW 2.1 Introduction ‘The stability of a slope depends on its ability to sustain load increases or ‘changes in environmental conditions, which may affect the geomaterials _ mechanically or chemically (e.g. weathering) (Eberhardt, 2002). The first outward sign of slope instability is usually a tension crack in the ground behind the crest of the slope, sometimes accompanied by slumping of the soil in front of the crack (Sarsby, 2000), The rate of failure is generally relatively slow, developing over a period of hours to days. First part of this chapter reviews ity including classification system that available and investigation of slope inst factors that encourage instability. The understanding of slope instability has not come about in an ordered ‘manner. Ideas have been advanced from time to time; some in being accepted have passed into the body of knowledge, others, because they represent concepts too advanced for the times, have been rejected, only to be resurrected at a later date, or because of their patent absurdity have been consigned to the rubbish bin. The second part reviews some of the published landmarks in early slope stability analysis, many of which are valid today, and without which none of the concepts described here would have been possible.Over the year, limit equilibrium theory has been widely used in slope stability analysis, Equilibrium methods of slope stability analysis all involve assumptions because the number of equilibrium equations available is smaller than the number of unknowns involved (Duncan & Wright, 1980), Third part of literature review gives aan overview of limit equilibrium theory that commonly used, showing how these methods may be selected which avoid significant errors arising from the mechanics of the analysis, Increased computational power has enabled engincers to concentrate more on obtaining the right input for a slope stability analysis and on selecting the appropriate model. In the past 25 years, the finite element method has becn used to analyze a large number of dams, as well as other embankments and slopes. The principal requirement for achieving reasonably accurate and useful results from these analyses in suitable presentation of the stress-strain behaviour of the soils involved (Duncan, 1996). The fourth part of this chapter reviews the theory of FE in slope stability analysis as well as information required to formulate the problem and the use of software package. Owing to the complexities of the laboratory testing, which require costly laboratory equipment, expert test techniques and long test times, itis unpractical to rely solely on laboratory testing, In addition, the difficulties in undisturbed soil sampling also greatly limit applications of laboratory testing. Back analysis has been widely accepted for determination of the parameters for landslide control works because it avoids many of the problems associated with laboratory tests. The final part of literature review presents a detailed description of back analysis currently in use, with emphasis on the advantages and limitations of each procedure.2.2 Investigation of Slope Failure In the sense used herein, the term “failure” connotes an unacceptable difference between expected and observed performance (Leonards, 1982). Included are rupture or erosion of the ground, unexpected cracking that is unsightly, or reduces load-carrying capacity, or both, movements that impair the function of a structure, and monitoring systems that do not convey intended warnings in time. When a sudden rupture occurs, herein called “instability,” the result is usually spectacular and often incurs heavy property damage and possibly loss of life. Itis commonly believed that investigation of instabilities offers the ultimate opportunity to gain new insights, to evaluate analytical tools, and to improve the design process. This, however, is not always the case, for the following reasons (Leonards, 1982): i. The geometry of the structure, the loading conditions, the soil stratigraphy, and the ground-water regime prior to failure must be inferred from the original site investigation and extrapolation from adjacent ground conditions, from the plans and specifications, and from examination of the post-failure conditions. In many cases, this information is incomplete, and the very fact that adjacent material did not fail implies that conditions in the failure zone may have been different, ii, _Itis always difficult and frequently impossible to reconstruct the sequence of ‘events that led to failure; thus the factors which initiated the instability seldom can be positively identified, iii, _As the original conditions are irreparably altered (and cannot be fully Teconstructed), itis not possible to test different hypotheses of failure on the prototype,‘That is the why, though slope stability analysis is one of the oldest geotechnical engineering subjects, yet as we enter the 21 century it remains one of the most active areas of study. 2.2.1 Types of Slope Failure In spite of the wide variety in mass movement, there are three predominant types, which are translational slide, rotational failure and wedge failure as shown in Figure 2.1 (Sarsby, 2000). Modes of failure are usually rotational or composite (often with essentially circular slip surfaces); non-circular slip surfaces normally develop because of the influence of ground stratigraphy. Each type of failure has general characteristics, which are: * A translational slide is usually relatively shallow (typically 0.5-2m deep), with the failure surface being more or less parallel to the ground surface. This type of failure is frequently seen on the sides of newly formed cut slopes, where failure of the topsoil cover has occurred. Typically, after failure the slipped material accumulates as a soft heap at the toe of the slope © Rotational instability is usually a deep-seated failure mechanism, with the sliding surface often being more or less circular. There is a large volume of ‘material associated with the movement, and this type of failure is characterized visually by a steep rear scarp slope at the top, an upper plateau which tits backwards from the slope face and an outwards bulge at the toe (ie. the bottom of the slope). * Wedge failure represents an intermediate mechanism between the two Previous types. The failure surface is composed of one or more essentially straight lines, and for a single wedge failure the sliding surface is not parallel to the ground, The shape of the failure surface results from the presence of weak or hard strata orientated in unfavourable directions. A category of slope instability that is not discussed here, but which is relevant to the behaviour of slopes of tailings (waste) material, is a flow slide. The term “flow” is used when the slipping material becomes disaggregated and, although a9 flow can remain in contaet with the surface of the ground it travels over, this is by no means always the case. Earth flows consist of slow movement of softened or weathered debris (such as from the base of a broken up slide mass). Mudflows comprise liquidized clay debris and move very quickly. 2.2.2 Classification of Slope Failure Landslides, slips, slumps, mudflows, rock falls, are some of the terms which are normally used to describe movements of soils and rocks under the influence of gravity. In view of the vast range of different ways in which these movements can occur, a system of description and classification is required, so that all practitioners can describe it in consistent terms and be understood. Besides it, it is also important so the technical literature is consistent. There must be, at the very least, an agreed terminology. Such a consensus unfortunately does not exist There are too many classification systems that are available. These include the notable schemes by Sharpe (1938), Hutchinson (1968), Varnes (1978), and Hunt (1984), Varnes (1978) classification system, which takes in account types of movement and types of material, is illustrated in Table 2.1. Hunt (1984) gives quite a Good system (Table 2.2). The major factors in this are: movement form, failure surface form (are, planar, etc), mass coherency, constitution and failure cause. Besides it, Hunt (1984) also classifies slope failure forms in terms of the velocity of movement (Figure 2.2). Brombead (1986) found that a major source of difficulty with these schemes is the limited terminology which can be used to describe different types of mass movement: this gives a superficial degree of similarity to the various classification systems, but a descriptive term in one scheme may Tepresent something completely 4ifferent in another. Furthermore, mass movements come in such an enormous range of sizes, shapes and types that even if there was only one classification system, it would be often be difficult to decide precisely how to classify a particular mass movement.vss ag 7 ? Foture Surtoce (a) Translational (planar) failure Zp em AEE A Eo o ieneaorentita Fotwre Surroce (b) Deep-seated (rotational) failure(© Simple wedge failure (@) Multiple wedge failure Figure 2.1: Types of slope failure (Source: Sarsby, 2000)Table 2.1: Varnes (1978) Classification System (Source: Fell, 1994) TYPE OF MATERIAL ENGINEERING SOILS TYPE OF MOVEMENT Predominantly | Predominantly BEDROCK fine coarse Falls Earth fall Debris fall Rock fall ‘Topples Earth topple | Debris topple | Rock topple Rotational Earth slump |” Debris slump | ~ Rock slump "| Few |” Earth block | Debris block | Rock block 3 Units Slide slide slide @ | Translational | Many Earth Debris Rock Units Slide slide Slide Lateral Spreads Earth spread | Debris spread | Rock spread Flows Earth flow Debris flow Rock flow (soil creep) (deep creep) Complex Combination of two or more principal types of movementTable 2.2: A classifica mn of slope failures (Source: Hunt, 1984) Type Form Defi Falls Free fall Sudden dislodgment of single or multiple blocks of soil or rock which fallin free descent Topple ‘Overturning of @ rock Block about a pivot point located below its center of gravity ‘STides Rotational | Relatively slow movement of an essentially coherent block or slump (or blocks) of soil, rock, or soil-rock mixtures along some well-defined arc-shaped failure surface Planar or Slow to rapid movement of an essentially coherent block (or ‘translational | blocks) of soil or rock along some well-defined planar failure surface ‘Subclasses © Block glide | A single block moving along a planar surface + Wedges Block or blocks moving along intersecting planar surfaces. ‘A number of intact blocks moving as separate units with © Lateral 0 eventually yields the following force equilibrium equation: =“, ~ (2.24)Infinitesimal element of a larger, linearly bounded slice Eccentricities of these forces, dw, dN’, dP, Kk areles thanx so may be AS — nedeted tei Eccentricity of dPb is g: GN is gy dW is gy Line of thrust of P,, y=b@) Slip surface y = y(x) Sliding mass subdivided into a small number of large slices, Figure 2.8: A typical element in the Morgenstern-Price method (Source: Bromhead, 1992) 3334 2.5 Review of Finite Element Theory in Slope Stability Analysis Bishop (1952) noted that the stresses from a limit equilibrium method of analysis did not agree with the actual stresses within an earth structure. Other researchers have confirmed this observation both with experimental evidence and with numerical modelling. La Rochelle (1960) estimated the stress conditions in steep slopes using photoelastic tests on gelatine models. The results showed that stresses along a slip surface were over-stressed in the lower portion of the slip circle Brown and King (1966) produced critical slip surface from a finite element stress analysis of slopes using a linear elastic soil model. The critical slip surfaces were Produced by using the angle of obliquity, @, along the slip surface (i.e., @equal to (45° + g'/a)). Each critical slip surface represented a close approximation to an essentially circular shaped slip surface (Fredlund & Scoular, 1999), Clough and Woodward (1967) undertook a study to evaluate the effect of incremental loading with single step loading as it related to stresses and deformations. It was concluded that: 1) stresses and deformations in an embankment obtained from a direct application of the gravitational body forces on the complete structure were not completely accurate, and 2) changing Poisson’s ratio interferes with the relationship between stresses and displacements, requiring a new analysis for each case. It was concluded, “meaningful analysis can be made only if the stress distribution within the structure can be predicted reliably.” (Fredlund & Scoular, 1999). Hence, a number of finite element slope stability methods have been proposed and the methods can be categorized as “enhanced limit methods” or “direct methods”, as shown in Figure 2.9.7 oaeeanaee Finite Element Slope ‘Stability Methods [i | Direst eto’ (finite element analysis | only) Enchanced limit methods ] Ginte clement analysis wih a | | Lead increase ae ae limit equiium analysis) | L ] Definition of Factor of Safety _f 1 i ae strength Stross level Stone & Sires Kuthawy Zienkiewsczot af Adikari and Cummins 1969 | 1975 L | 1985 Figure 2.9: Finite element approaches proposed in computing the factor of safety in a slope stability analysis (Source: Fredlund & Scoular, 1999)36 2.5.1 Enhanced Limit Methods Kulhawy (1969) developed a computer program to obtain an independent assessment of the normal and shear stress distribution along an assumed slip surface. The normal and shear stresses from an elastic analysis were used to calculate an overall factor of safety. The formulation of Kulhawy (1969) was classified as an “Enhanced Limit Strength Method” and shown in equation 2.25. (Fredlund & Scoular, 1999) _ D(c'+o'tan g")AL Factor of safety, Fe ¥y TAL (2.25) where, ° = effective cohesion, a = effective normal stress, 6° =effective friction angle, T = shear stress, AL =size of each slices, ‘Wright (1969) compared the factors of safety calculated using the “enhanced limit strength” method with factors of safety calculated using Bishop’s Simplified method (1952). A slip surface was selected for comparative purposes that had a factor of safety of 1.0 when using the Bishop’s Simplified method. It was concluded that the factors of safety determined by the “enhanced limit strength” method (Kuthawy, 1969) were approximately 3% higher than those determined applying Bishop’s Simplified method. Wright et al, (1973), using the “enhanced limit strength” method, showed that: 1) along one third of the slip surface, the local factors of safety are less than the overall factor of safety, 2) the factors of safety calculated by the finite element method using linear elastic material properties ranged from 0% to 4.5% higher than those calculated using Bishop’s Simplified method, and 3) the factors of safety calculated by the finite element method using nonlinear elastic37 material properties increased with Poisson’s ratio and are 2% to 8% higher than those calculated using the Bishop’s Simplified method (Fredlund & Scoular, 1999), Resendiz (1974) agreed with the concept of using the finite element method to calculate the stability of a slope: however, disagreed with points No. 2 and No. 3 of the results of Wright et al. (1973) because the factor of safety differences were too small. Resendiz had developed a finite element method of slope stability analysis defined as an “enhanced limit stress-level” method in 1972. This method used the ‘maximum principal stress difference of the soil at failure to define the factor of safety. Analyses made using non-linear stress versus strain relationships led to factors of safety, which in all cases were higher (i.e., differences as large as 30%) than conventional factors of safety (¢.g., Ordinary method or Bishop’s Simplified method) (Fredlund & Scoular, 1999). Zienkiewicz et al. (1975) also proposed a finite element method of analysis to compute the factor of safety by using the principal stress difference in the soil at failure to define the factor of safety. The method is an “enhanced limit stress-level method” and the formulation is shown in equation 2.26. Both the Resendiz (1972) and Zienkiewicz et al. (1975) formulations are classified as “enhanced limit stress- level” methods (Fredlund & Scoular, 1999) Factor of safety , F: ~) (2.26) aa i } where, o;' = effective principal major stress, 63' = effective principal minor stress, AL = size of each slices, Naylor (1982) established two types of finite element slope stability methods, a “direct” and an “enhanced limit” method of analysis. The direct method used a finite element nodal formulation to define the slip surface and the factor of safety38 directly from the analysis. The proposed “direct” slope stability method defined the factor of safety either as the increased load necessary to cause failure, or as the reciprocal of the reduction in the strength properties required in order to achieve failure. These methods have also been studied by Martins et al. (1981) and Tan and Donald (1985) (Fredlund & Scoular, 1999). The “enhanced limit” slope stability methods are based on stresses calculated using a finite element analysis and combined with a limit equilibrium type of analysis along a prescribed slip surface, to define the factor of safety. The prescribed slip surface is the one defined by the lowest factor of safety and is found using a trial and error procedure. The stresses along the slip surface are computed using a finite element analysis and can either be used in a “strength” method or a “stress-level” method, Farias and Naylor (1996) stated that when using the “direct” finite element method it is, “not easy 10 obtain a safety factor accurate to within the confidence mits achievable by limit equilibrium methods”. The authors noted that: 1) a fine mesh is required, 2) a code capable of giving reliable results with the Mohr Coulomb elasto-plastic model for loading states close to failure is needed, and 3) itis usually mecessary (0 carry out a set of analyses with c’ and tang’ progressively reduced by a factor which will become the safety factor when failure is eventually reached. “Enhanced limit” methods require only one finite element analysis to calculate factors of safety for a slope with various combinations of c’ and tang’ (Fredlund & Scoular, 1999), Adikari and Cummins (1985) produced a finite element method that combine the “strength” and the “stress-level” methods as defined by Kulhawy (1969) and Zienkiewicz et al. (1975), respectively. The formulation is shown in equation 2.27. The Adikari and Cummins (1985) method produced factors of safety that were between the values obtained when applying the Kulhawy (1969) and the Zienkiewicz etal. (1975) methods. It was noted that for near-failure conditions (i.e., as defined by Bishop's Simplified method, 1955), the value ofthe factor of safety calculated by the Adikari and Cummins (1985) method approached 1.0, while the value of the factor of safety calculated by the Zienkiewicz et al. (1975) method remained high. The factor of safety by the Kulhawy (1969) method also approached unity with the factor of safety being dependent on the percentage of the strength mobilization in the39 component materials. The main difference in results appears related to using the stresses on the principal plane rather than on the plane. By definition, failure does not ‘occur on the plane of principal stress and therefore, the Zienkiewicz et al. (1975) method (or any stress-level method) is computing a factor of safety that must be higher than the factors of safety produced by a “strength” method (Fredlund & Scoular, 1999), Tie+ortan gar} [eros] e” Terconl Creer } Factor of safety, F. where, c= effective cohesion, o effective normal stress, ° effective friction angle, AL = size of each slices, or = effective principal major stress, ©;' = effective principal minor stress, 2.5.2 Direct Methods ‘An alternative tool for geotechnical stability analysis is the FEM-based strength reduction technique. This technique has been applied mostly for slope stability analysis, where it has been shown to produce results comparable to the methods of slices and other limit-equilibrium methods. For slope, the factor of safety, Fs is traditionally defined as the ratio of the actual soil shear strength to the minimum shear strength required to prevent failure (Bishop, 1955). Duncan (1996) describes Fs as the factor by which the soil shear ‘strength must be divided to bring a slope to the verge of failure. Hence, an obvious way of computing Fs with a finite element program is simply reduce the soil shear strength until collapse occurs. The resulting factor of safety is the ratio of the soil’sactual shear strength to the reduced shear strength at failure. This “shear strength reduction technique” was used as early as 1975 by Zienkiewicz ef al. (1975), and has since been applied by Naylor (1982), Donald & Giam (1988), Matsui & San (1992), Ugai (1989), Ugai & Leshchinsky (1995) and many others (Dawson, et al., 2000), Application of the technique has been limited in the past due to the long ‘computer run times required. However, with increasing speed of desktop computers, the technique is becoming an effective alternative to the method of slices, and is being used increasingly in engineering practice. Strength reduction is now a built-in feature of commercial geotechnical FEM codes such as PLAXIS (PLAXIS BV, 1998). The technique is known as Phi-C Reduction in PLAXIS. The total multiplier Ms is used to define the value of the soil strength parameters at a given stage in the analysis: LAN Pint ‘input Msp = EP inpwt Simp (2.28) "ANP reduced — Creduced where the strength parameters with the subscript ‘input’ refer to the properties entered in the material sets and parameters with the subscript ‘reduced’ refer to the reduced values used in the analysis. DMs/iis set to 1.0 at the start of a calculation to set all ‘material strengths to their unreduced values, The strength parameters are successively reduced automatically until failure of the structure occurs. At this point the factor of safety is given by: Locilabe serength fl See oenath ot fatarg 7 VeINe OF EMF at fitare (2.29) This approach resembles the method of calculation of safety factors conventionally adopted in slip-circle analyses. When using Phi-c reduction, stress-dependent stiffness behaviour and hardening effects of the soil are excluded.