Sumit - S40808-022-01597-Y

Modeling Earth Systems and Environment (2023) 9:1445–1462
https://doi.org/10.1007/s40808-022-01597-y
REVIEW ARTICLE
Recent advances in 3D slope stability analysis: a detailed review

Sumit Kumar1 · Shiva Shankar Choudhary1 · Avijit Burman1
Received: 1 August 2022 / Accepted: 28 October 2022 / Published online: 23 November 2022
© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2022
Abstract
Regarding the geotechnical aspects, it is very important to correctly estimate a slope’s safety factor against failure. Slope
failures, in general, cost a lot of money, hinder transportation, and can even kill people in the worst cases. To successfully
prevent such failures, it is required to find proper understanding of the failure mechanisms such that proper analysis techniques
may be chosen to appropriately determine the stability of any soil/rock slope. Slope stability analyses in two dimensions
have been very popular among the practicing engineers because of the relative simplicity of the underlying concepts, as well
as because it also provides conservative estimates of the FOS against failure. However, in many situations where the slope
geometry and the loading conditions do not conform the plane strain idealization, a two-dimensional slope analysis can yield
results which are far from accurate. In such situations, the only viable option is to study the slope failure analysis in three
dimensions. Stability status of a slope in three dimensions is essential when the failure mass and slope geometry alter later-
ally, the soil/rock properties are anisotropic and not homogeneous, and the local surcharges are applied to the slope. Based
upon finite element analysis, limit analysis, and limit equilibrium, stability status in three dimensions has been determined
since the 1970s. Many of them are only applicable in specific circumstances. The present paper focuses on presenting an
overall picture of different aspects of slope stability analysis in a comprehensive manner.
Keywords Three dimension · Safety factor · Critical failure surface · Rock mass · Strength reduction method · Limit
analysis method
Introduction Leshchinsky 1987; Leshchinsky and Baker 1986; Leshchin-

sky et al. 1985; Cavounidis and Kalogeropoulos 1992). 3D
Regarding the geotechnical aspects, determining a slope’s slope failure analysis is needed when the failure mass and
safety factor (FOS) against failure is very important. Numer- slope geometry alter laterally, the soil/rock properties are
ous slope failures worldwide cause great financial loss, inter- anisotropic and not homogeneous, and the local surcharges
ruption of transportation, and may lead to loss of human are applied to the slope. Three different methodologies are
lives in extreme circumstances. Proper analysis and design of available to carry out 3D slope stability checks, which are
a slope are essential to prevent such failures. It was assumed the Limit Analysis Method (LAM), Finite Element Method
that the plane strain conditions existed in two dimensions (FEM), and Limit Equilibrium Method (LEM). When figur-
(2D), and the slope stability problem was first formulated ing out the 3D FOS, most researchers focused on FEM and
in 2D. Recently, many researchers have presented a slope LEM. For the first time, Anagnosti (1969) extended the Mor-
stability analysis in three-dimensional (3D) (Baker and genstern and Price (1965) formulation for slope analysis in
3D. Several investigators have proposed other 3D LEM for
stability analysis of a slope (Hovland 1977; Dennhardt and
* Avijit Burman Forster 1985; Leshchinsky and Huang 1992a, b; Ugai 1985;
[email protected]
Hungr 1987; Xing 1988; Ugai 1988; Jiang and Yamagami
Sumit Kumar 2004; Sun et al. 2012).
[email protected]
The LEM describes the FOS of slope as the ratio between
Shiva Shankar Choudhary resistive and driving moments. Three sections of LEM can
[email protected]
be used to check the slope’s stability: (1) determination of
1
Civil Engineering Department, National Institute direction of slide (DOS) and the FOS of any assumed failure
of Technology Patna, Patna, India
13
Vol.:(0123456789)

1446 Modeling Earth Systems and Environment (2023) 9:1445–1462
surface, (2) searching for the global minimum value of FOS meaning the slope cross-section always remains the same
amongst all investigated failure surfaces. In two dimensions, along the longitudinal direction. As a result, the stresses
the DOS of a slope is always along the direction of the given along the third direction can be ignored while formulating
slope. However, it is specifically necessary to determine the expression for determining FOS. However, this assumption
DOS of a slope while applying LEM in 3D slope analysis, loses its validity when the slope cross-section changes along
as it may not be the same as the overall fall direction of the the third direction. Also, for a finite slope, if corner stability
slope. The DOS of a 3D slope is different than the usual fall is to determined, two dimensional formulations would no-
direction when asymmetric loading/geometric profiles are longer be adequate. In such situations, three-dimensional
considered. A suitable search strategy, such as a grid-based slope stability analysis can only provide accurate description
search strategy or a metaheuristic optimization method like of the mechanisms of slope failure.
Gray Wolf Optimization (GWO), Particle Swarm Optimi-
zation (PSO), etc., must be adopted to estimate the global Determination of FOS in 3D
minimum FOS and the corresponding critical failure surface
(CFS) in 3D slope failure analysis. Three dimensional slope stability analysis is complicated as
Limit Analysis Method (LAM), on the other hand, defines it is required to account for both lateral (i.e. along the direc-
a collapse mechanism of any slope subjected to predefined tion of sliding of the slope) and the longitudinal/transverse
constraints and finds out FOS of the slope using upper and directions of the slope. Any LEM begins with an assumption
lower bounds theorem. The solution obtained from LAM of the type of failure mass. In case of 3D, the failure mass
can be construed as exact solution of the problem. Many of a soil can be a spheroid, ellipsoid, paraboloid etc. For a
researchers have applied LAM for developing graphical particular type of failure surface, it would then be necessary
solutions as stability charts. to formulate the expression of FOS. The two-dimensional
Clough and Woodward (1967) were the first to study FOS determination strategies following Bishop’s (1955a,
the FEM in geotechnical engineering. FEM is a preferred b), Morgenstern-Price (1965), Janbu’s (1965), and Spen-
technique to solve geotechnical problems because it can cer’s (1967) methods were extended to three dimensions by
model how materials behave when they experience nonlin- various researchers (Anagnosti 1969; Chen and Chameau
ear stresses and strains. A Strength Reduction Technique 1983; Hungr 1987). In 3D slope analysis, it is necessary to
(SRT) based on FEM can be used to efficiently compute the determine the DOS separately because it may vary depend-
FOS of a 3D slope. The approach aims to find out the state ing on the geometric and loading profile of the slope. In 3D
of instability of a slope by minimizing the soil/rock strength analysis, the failure mass of a soil is subdivided into several
parameters successively with the help of any relevant fac- columns, and the state of forces on such a column is shown
tor and designating the same as the FOS of the slope when in Fig. 1.
the slope becomes unstable. In 3D slope stability analysis,
FEM is considered more rigorous. Although these methods
take more time than others to complete, they are currently
most extensively used to check the 3d slope stability status
(Ugai and Leshchinsky 1995; Jeremić 2000; Griffiths and
Marquez 2007; Tan and Sarma 2008; Kelesoglu 2016). In
the following sections, detailed reviews of the works by sev-
eral researchers in the field of 3D slope analyses have been
presented.
Limit equilibrium methods (LEM) for 3D

slope analysis
Almost all 3D LEMs are expansions of 2D slice methods.

The stability check of any slope is describe with the help of
a metric called the safety factor, simply the ratio between
upward resisting and downward mobilizing force. If the
FOS > 1.0, the slope is considered as stable. For the first
time, Anagnosti (1969) showed a 3D slope failure analysis
formulation. Two-dimensional slope failure analysis is based
upon the assumption that plane strain conditions are valid, Fig. 1 Internal and external soil column forces
13
Modeling Earth Systems and Environment (2023) 9:1445–1462 1447
Where Pvk,l external force in vertical direction; Exk−1,l have an impact on the ratio between 3 and 2D safety factor.
and Exk,l are inter-column normal forces acting towards x Dennhardt and Forster (1985) provided a three-dimensional
axis; Eyk,l−1 and Eyk,l are inter-column normal forces that model for computing the FOS with ellipsoidal slip surfaces
acts along the y axis; Xxk−1,l and Xxk,l are the Shear forces and a symmetrical external force. To overcome the prob-
between vertical columns act in the x axis; Xyk,l−1 and Xyk,l lem’s indeterminacy, they considered normal stress distribu-
are Shear forces between vertical columns along the y axis; tion uniform on the failure surface. This method reported 3D
Hxk−1,l and Hxk,l are inter-column horizontal shear forces act- FOS values greater than the corresponding 2D FOS.
ing towards x axis between columns; Hyk,l−1 and Hyk,l are Leshchinsky et al. (1985) proposed a 3D approach for
inter-column horizontal shear forces acting towards y axis homogenous and symmetrical slopes based upon limiting
between columns; Sk,l and Nk,l are mobilized shear and total equations originally proposed by Kopacsy (1957) using
normal forces act at the column’s base (k, l). variational calculas (1957). They used a variational formu-
The FS against slope failure can be calculated using lation to get the lowest FOS by initializing three unknown
moment equilibrium methods as the ratio between shear functions, comprising the rotational surface equation in the
strength ( 𝜏f ) and shear stress (s) as shown in Eq. (1). form of a sphere or cylinder, a function relating to the shear
𝜏f stress distribution directions on the failure mass of the soil,
Fs = (1) and the normal stress. The authors found that the shallow
s slip surfaces were formed when the soil was less cohesive
The consideration dictates slope failure that the overall or the slope was steeper. They discovered that higher value
resisting moment is the same as the driving moment to main- of FOS for spherical slip surfaces compared to the cylindri-
tain the global moment equilibrium. Taking into account all cal failure surfaces. The failure mass was supposed to be a
of the columns enclosed by the spherical failure surface, cylindrical shape joined to the cap ends. For the value of
the global moment equilibrium for all columns and FS 3D FOS, the authors used conditions of force equilibrium
against rotation can be calculated as shown in Eqs. (2) and along the transverse and vertical axes and a half-sliding
(3) respectively. mass moment equilibrium around the axis of rotation. The
most notable difference was found between 3 and 2D FOS
∑ ∑ ck,l Ak,l + (Nk,l − uk,l Ak,l ) tan 𝜑k,l for cohesive slopes, according to their comparative analy-
M= Rk,l
∑ ∑F (2) sis. When the slope face was parallel to the shallow slip
− Wk,l Rk,l mz − Wk,l keq ek.l surface, they found identical FOS values for 3D and 2D
evaluations of cohesion less slopes. Baker and Leshchinsky
∑ � � � � (1987) extended Leshchinsky et al. (1985)’s 3D technique
Rk,l ck,l Ak,l + Nk,l − uk,l Ak,l tan 𝜑k,l
FOS = (3) to examine homogeneous, symmetrical conical slopes based
on a simplified approach by disregarding pore pressure and
∑ � �
Wk,l Rk,l mz + keq ek,l
external loads. The surface equation and also its central
Here, Rk,l is the geometric center distance of trial slip area point defined the shape and position of the failure surface.
to the axis of rotation for the k, l column, Ak,l is the col- Baker and Leshchinsky (1987) studies established that the
umn’s base area, Wk,l is the weight of the k, l column, keq is ratio between 3 and 2D FOS can be as high as 1.6. They
for earthquake loading, the horizontal pseudo-acceleration came to the conclusion that this ratio declines with increase
coefficient, ek.l is the horizontal driving force moment arm. in cohesiveness, and obtained values of 3D and 2D FOS are
The term mz = sin 𝛼k,l is introduced where 𝛼k,l represents the equivalent for entirely frictional soils. The authors further
dip angle at the column base. For all general moment equi- concluded that a two-dimensional investigation of conical
librium, safety factor (FOS) can be estimated by Eq. (3). heaps produces safer results based on the slope angle and
Anagnosti (1969) presented a 3D approach to deter- material type.
mine the FOS of general slip surfaces as an improvement Hungr (1987) presented a formulation for determining 3D
of Mongenstern and Price (1965) technique. A set of LEM FOS from the original work of two-dimensional Bishop’s
equations were developed, whereas the failure mass was not method Bishop (1955a, b). The failure surface in this sym-
constrained to any certain shape. The value of FOS increased metrical problem was considered to be a rotating circular
by 50% in 3D analysis, according to Anagnosti’s findings. surface. Following Bishop’s assumption, shear forces on
Hovland (1977) proposed a 3D slope analysis of cohesive the column’s vertical sides were neglected. To arrive at the
soil to expand the 2D ordinary slice method. The DOS was FOS equation, this method evaluates the equilibrium of all
believed to be vertically aligned. Hovland’s method dis- vertical forces on any column and the total moment equilib-
regarded pore-water pressure as well as all forces on the rium of slipping soil mass around the rotation axis. Hungr
column faces. His research also revealed that the degree of (1987) re-examined the works of Chen and Chameau (1983)
cohesion, angle of friction, and slip surface geometry all to deduce that 3D to 2D safety factor ratio is always more
13

than unity in usual circumstances. Ugai (1988) suggested remains constant, the role of 3D analysis becomes more
a series of 3D approaches based on the development of 2D essential.
slice methods, such as Fellenius, simplified Janbu, simpli- Leshchinsky and Huang (1992b) suggested a generalized
fied Bishop, and Spencer, to attain higher values of 3D FOS 3D method based on rigorous LEM and variational analy-
as compared to the corresponding 2D factor. Hungr et al. sis using Leshchinsky and Huang’s (1992a) 2D method.
(1989) developed three-dimensional approaches after modi- Continued log-spiral or a generic symmetrical surface was
fying Janbu’s (1954) and Bishop’s (1955a, b) two dimen- considered to be failure surface shape. Resolving a set of
sional slope stability formulations. A similar set of assump- linear equations as well as three nonlinear equations in an
tions was adopted by author in his extension of Bishop’s iterative procedure, this approach was able to converge on
approach. In this approach, inter-column shear forces were an FOS. According to the authors, the provided convergence
neglected, while all other shear forces acted on the failure criteria could have a considerable impact. Their findings
surface which was considered to be parallel to the plane revealed that in the back-analysis of slopes, ignorance of the
of symmetry. Assumptions were made regarding the slip end effect can lead to overestimated FOS and soil strength
surface symmetry and its nature of rotation. A benchmark parameters.
axis of rotation was considered for the non-rotational failure Lam and Fredlund (1993) suggested a three-dimensional
surfaces based on Fredlund and Krahn’s (1977) proposal. approach based on Fredlund and Krahn’s (1977) General
For the determination of FOS, it was necessary to satisfy Limit Equilibrium Methodology. Rotational slip surface with
the vertical force and the total moment equilibrium of each only one direction of movement was considered. Also, the
column and the sliding body, respectively. Bishop’s method angle of the inter-column forces that were created was cal-
is found to yield smaller 3D FOS results than rigorous culated using several different inter-column force functions.
3D methods. In bilinear slip surfaces, more conservative These definitions of inter-column forces were the same as
results obtained by extended Janbu’s approach as compare Morgenstern and Price's (1965) function, which comprised
to extended Bishop’s method. five correlations between the shear and normal forces that
Xing (1988) presented a simple three-dimensional act between columns. They agreed to exclude three of the
approach for determining the FOS of symmetric concave five inter-column forces because they play such a minor role
slopes with an elliptical surface of rotation. The failure sur- on ordinary slopes. They also developed two distinct FOS
face comprised of four geometrical features, along with its equations to determine the problem solutions: (a) the force
central location and three semi-radii. All inter-column forces equilibrium and (b) the moment equilibrium. Author dis-
were ignored acting normal to the sliding direction as the covered that 3D FOS value is significantly more than the
sliding movement took place parallel to the vertical plane. 2D factor and is essentially unaffected by the type of inter-
Xing (1988) derived the expression of 3D FOS by satisfy- column force functions.
ing force equilibrium in all three direction in addition to the Chen et al. (2003) developed a simplified 3D method by
rotational equilibrium of the sliding mass. An iterative tech- extending Spencer’s (1967) methodology. Rotational slip
nique was adopted to satisfy the equality of both moment surfaces were considered and numerous assumptions were
and force equilibrium FOS within a specified tolerance limit. made to establish the equilibriums of forces and moments.
For homogenous slopes, Xing (1988) assured that 3D FOS The authors decided that two of the four vertical inter-col-
of soil slope was higher than the 2D FOS by a maximum umn shear forces could be disregarded. Overall moment
deviation of 4.32 percent, which could increase to 10 percent and force equilibriums were used to find the 3D FOS value.
in the case of a weak layer. Comparative investigations by Chen et al. (2003) found that
Cavounidis and Kalogeropoulos (1992) suggested a 3D in symmetrical circumstances, 3D FOS was almost similar to
method to investigate vertical cuttings in cohesive soils other 3D techniques, while in an asymmetric case, 3D FOS
following the works of Azzouz and Baligh’s (1978). The was found higher values.
failure surface was supposed to be a combined central cylin- Jiang and Yamagami (2004) suggested a 3D technique for
drical slip surface with conical ends, with the rotation axis studying conical heap stability checks by combining vari-
specified on the cut crest. The FOS equation is based upon ational analysis based on Spencer’s (1967) technique. The
moment equilibrium, defined as the ratio between the sliding method utilized assumptions similar to Spencer’s method,
body’s overturning and resisting moment. They also arrived and DOS was supposed to be along the lateral length of
at the conclusion that the ratio between 3 and 2D FOS is the 3D slope. The directions and angle of application of
always not less than unity. The authors also discovered that all column forces acting on the column face were suitably
extending the length of the cylindrical component while assumed. The authors developed two separate equations
retaining the length of the conical ends constant reduces for FOS, satisfying the failure surface’s overall moment
the gap between 3 and 2D FOS values. When the length of and horizontal force equilibrium. The convergence of two
the conical ends is increased while the slip surface length
13
resulting plots from two different FOS equations was used Direction of slide (DOS)
to calculate the overall FOS.
Sun et al. (2012) presented a 3D technique for the system In 3D slope failure analysis, it is also mandatory to find the
of forces as an extension of Morgenstern Prices method (), DOS of a soil/rock slope with respect to its neutral axis as
which followed Zheng’s procedure (2009). As a result, no shown in Fig. 2, as reported by Kalatehjari et al. (2014).
internal force assumptions were made, and the failure mass For a symmetric slope with respect to geometry and loading
was not subdivided into several columns, yet all equilib- conditions, the neutral axis always will be located mid-way
riums were established. For the interpolation computation along the slope towards the DOS. However, when the cross-
of soil mass, a triangular mesh was utilized extending over section of the slope changes along the longitudinal direction,
the horizontal projection of the failure surface. A differ- the neutral axis will no longer be located along the mid-way
ent triangular type mesh was chosen on the failure mass to of the slope (Yamagami and Jiang 1996; Huang and Tsai
compute the corresponding integrals. The normal stress on 2000). It would then become essential to uniquely deter-
the failure mass was simulated using an advanced interpo- mine the orientation of the neutral axis to correctly identify
lation technique involving the triangular mesh, in addition the DOS. In 2D analysis, there is no need to work on DOS,
to Zheng’s (2009) nonlinear system. The solution was pro- but in 3D failure analysis of a slope, it is necessary to find
posed in the form of an optimization problem involving the DOS. If the geometric and loading conditions are symmet-
FOS and the other five unknowns. Solving the optimization ric, the DOS of a 3D slope will coincide with that of the
problem yielded the least value of FOS and other associated 2D model. However, if these conditions are not met, as in
unknowns. Except one case that involved a critical section, the case of a slope for which the section changes along the
Sun et al. (2012) found that 3D FOS had higher values than lateral directions or there is partial loading on some part of
2D factors in a comparative study. the 3D slope domain, or It is essential to find out the FOS of
a trial surface near the corner, the DOS will be different from
the DOS of a symmetric slope. Because the trial mass will
move in the direction of least resistance, the sliding direction
Fig. 2 Three ideas about the DOS: a failure mass of soil pass in the same sequence, b failure mass of soil pass toward each other, and c each
layer of soil moves away from the next. Source: Kalatehjari et al. (2014)
13

Yamagami and Jiang (1996) developed the initial

approach for calculating the DOS for slopes with angle
of inclination up to 45 degrees. This formulation was an
improvement over the Janbu’s (1954) approach, which could
take into account general slopes and slip surfaces. Huang
and Tsai (2000) incorporated the direction of slide in their
3D analysis but did not propose a method for calculating it.
Additionally, Chen et al. (2003) showed that this method
produced incorrect findings for slopes with numerous lat-
eral asymmetries since it neglected to account for horizontal
forces. Three alternative formulae were devised to deter-
mine the DOS and its accompanying FOS value. Huang
et al. (2002a) suggested a technique to find the DOS using
the directional factors of safety (DFOS) equations for a fail-
ure mass split into a number of columns. They presented
Fig. 3 Nature of symmetrical failure surface (Huang et al. 2002a, b)
a three-dimensional version of Janbu’s (1973) technique,
which used moment and force equilibrium to examine the
stability of a sliding mass with any arbitrary shape. Their
proposed method is applicable for asymmetric slope fail-
ures. Cheng and Yip (2007) developed simplified Bishop’s
(1955a, b), simplified Janbu’s (1954), and Morgenstern and
Price’s (1965) approach to present a series of 3D methods.
They adopted Morgenstern and Price’s (1965) method to
design a 3D non-symmetric problem, that could also be
reduced to simplified versions of Bishop’s and Janbu’s
approaches. They came up with a complicated way to figure
out the DOS and analyze the stability checks of a general
slope. This method used a number of connected iterative
processes to figure out the unknowns and angles to figure out
the DOS. Kalatehjari et al. (2014) provided an enhanced 3D
technique of stability status of a slope and the exclusive DOS
for ordinary slopes. This approach is based upon LEM, an
ingenious technique capable of calculating the single DOS
and the associated FOS for any general shape of the slope
with all possible external forces. Table 1 presents a summary
on the findings of many researchers encompassing various
Fig. 4 Nature of corner failure from 3D slope analysis. Source: aspects of determining FOS of a 3D slope.
Huang and Tsai (2000)
Critical failure surface in 3D
will also affect the slope’s safety factor. Various researchers Apart from determining the FOS using any standard tech-
have attempted to figure out how to compute the DOS in a nique, it would be further necessary to find out the position
non-symmetric problem so that 3D methods can be used in of critical failure surface (CFS). The CFS refers to the sur-
a wider range of situations (Huang et al. 2002a, b; Cheng face with global minimum FOS. To find CFS in LEM, many
and Yip 2007; Kalatehjari et al. 2014). Figure 3 shows the possible failure surfaces are investigated, and FOS values
nature of failure of a 3D symmetric slope problem analyzed are determined for each of them. The surface designated as
by Huang et al. 2002a, b. It can be seen that for a symmetric the CFS has the lowest FOS value. The accurate determi-
slope for both geometry and loading, the failure surface is nation of the location of the CFS is a very complex, time-
also symmetrical about the neutral axis. Figure 4 depicts the consuming and computationally demanding process. A grid
corner failure of a 3D failure mass, as reported by Huang based strategy can be utilized to perform the task of locating
and Tsai (2000). For the 3D soil mass shown in Fig. 4, the the CFS. In this process, a box like grid is set up containing
rules of plane strain condition do not apply and, hence 3D the locations of centers of failure surfaces, failure surfaces
slope analysis can only reveal the desired nature of failure. are generated from these points with varying radiuses, and
13
Table 1 Review and characteristics of LEM based on F0S and DOS

Methodology References DOS Application Assumed shape of failure surface
consid-
ered
Extension of Morgenstern Price’s Anagnosti (1969) × Generalized Generalized

method (1965) Sun et al. (2012) × Generalized Generalized
Extension of ordinary method of Hovland (1977) × Symmetrical Wedge/cone shaped
slice
Extension of Spencer’s method Chen and Chameau (1983) × Symmetrical Cylindrical center part along with
(1967) ellipsoidal ends
Chen et al. (2003) × Generalized Generalized rotational
Xing (1988) × Symmetrical Symmetrical elliptical slip surface
depicting a circular vertical cut
Limit equilibrium and Variational Leshchinsky et al. (1985) × Symmetrical Spherical and cylindrical
method Leshchinsky and Baker, (1986) × Symmetrical Cylindrical with curved ends
Baker and Leshchinsky (1987) × Conical heaps Cylindrical end caps/2D log-spiral
expansion
Limit Equilibrium Method Dennhardt and Forster (1985) × Symmetrical Ellipsoidal
Expansion of Bishop’s modified Hungr (1987) × Symmetrical Circular central section for rotation
method
Expansion of Fellenius, simplified Ugai (1988) × Symmetrical For any method
Bishop’s, simplified Janbu’s, and
Spencer’s method
Expansion of Bishop’s, and Janbu’s, Hungr et al. (1989) × Symmetrical Symmetrical rotational
simplified methods
Limit equilibrium and Variational Leshchinsky and Huang (1992a, b) × Symmetrical Symmetrical/log-spiral extensions
method
Expansion of Azzouz and Baligh’s Cavounidis and Kalogeropoulos × Symmetrical Central cylindrical slip surface with
(1978) method (1992) conical ends
Extension of 2D limit equilibrium Lam and Fredlund (1993) × Symmetrical Generalized rotational surfaces
formulation
Janbu’s Simplified method Yamagami and Jiang (1996) ✓ Generalized Generalized Shape
Extension of Bishop’s simplified Huang and Tsai (2000) ✓ Asymmetric slope Semi spherical
method
Extension of Janbu’s (1973) Huang et al. (2002a) ✓ Asymmetric slope Arbitrary shape
Extension of Spencer’s method Jiang and Yamagami (2004) × Symmetrical Rotational surfaces
based on variational analysis
Further development of Bishop’s, Cheng and Yip (2007) ✓ Generalized Spherical
Janbu’s simplified, and Morgen-
stern–Price’s methods
Limit equilibrium method Kalatehjari et al. (2014) ✓ Complex slopes Ellipsoidal/spherical, or cylindrical
finally FOS values are determined. Otherwise, the problem problems. Ugai (1985) provided a 3D approach for sym-
can be set-up as a minimization problem, where different metrical cohesive vertical cuts by applying variational cal-
optimization techniques like PSO, and GWO are used to find culus and limit equilibrium equations. The specific form of
out the least value of the FOS (i.e., objective function). The the 3D CFS is discovered to be cylinders with curved end
variables used in these optimization techniques are usually caps. The exact 3D slope FOS values, which are determined
the coordinates of the center of the failure surfaces and their as functions of slope length to vertical slope height ratios,
corresponding radiuses. In the following section, the works are higher than the 2D FOS suggested by previous research-
related to determination of CFS have been briefly presented. ers. Leshchinsky et al. (1985) developed a 3D approach for
Variational calculus usually aims to find out the extremum slopes that are the same on both sides, based on LEM and
values of a function. Since, the problem of finding out CFS Kopacsy's variational analysis (1957) method. The author
can be expressed essentially as a minimization problem, used variational limit equilibrium analysis to get the lowest
variational calculus can be effectively used to solve such FOS and the corresponding CFS at the same time. However,
13

their variational analysis investigations were limited to three dimensions. Using 3D limit equilibrium global optimi-
homogeneous problems. zation analysis, they studied different cases. Cheng’s (2003)
Yamagami and Jiang (1997) used the 3D simplified Jan- 2D simulated annealing method was extended to global 3D
bu’s method, originally suggested by Ugai (1988) and Ugai optimization technique to find 3D failure surfaces for vari-
and Hosobori (1988), with dynamic programming (DP) and ous problems.
generating random numbers. They implemented this analy- Ahangar-Asr et al. (2012) developed a computer program
sis into the 3D-DYRANUT computer program to find the employing a genetic algorithm (GA) known as Optimum3D
location, shape of 3D CFS and corresponding FOS values to compute the least FOS and the position of the failure sur-
for the slope of pore pressure and layered soils. A DP algo- face that corresponds to it for any slope that has an arbitrary
rithm alone cannot perform a 3D slope stability study. This profile for the initial slip surface.
study offered an innovative idea in which random number Hajiazizi and Tavana (2013) determined critical 3D
generation was used to generate states, reducing a 3D DP non-spherical slip surfaces for earth slopes by optimization
problem into a 2D problem and allowing DP to be used for techniques while applying LEM for 3D slope analysis. The
a 3D analysis. authors used the Three-Dimensional Alternating Variable
Huang et al. (2002a, b) presented a 3D version of Janbu’s Local Gradient (TDAVLG) technique to find the critical slip
(1973) technique that satisfied moment and force equilib- surface of non-spherical types. The proposed method is easy
rium in two directions to examine the stability status of a to understand but takes time to find the least FOS and the
slope with any arbitrary configuration. In addition, they also corresponding failure surface.
proposed a search procedure for obtaining the CFS in 3D. To Reid et al. (2015) proposed a simple grid-based box
do this, they considered appropriate failure surface shapes, search analysis technique in Scoops3D, where spherical
such as quadrilateral, log spiral, or planar surfaces; the appli- trial surfaces are produced at each search lattice node. The
cation of efficient optimization methods; and the provision spherical failure surface’s least FOS is presented as the CFS
of sufficient geological and geometrical data. following the completion of the search for all lattice points.
Jiang and Yamagami (2004) established a search strategy The Scoops3D computer program also offers a coarse-to-fine
for locating the 3D CFS by combining dynamic program- search variant to minimize overall calculation time while
ming with an expanded Spencer’s (1967) column method searching for the center of CFS and least FOS.
that fulfills both force and moment equilibrium. A search Kalatehjari et al. (2015) employed PSO techniques in 3D
was carried out on a 3D problem, i.e., (conical heaps) and slope stability analysis to find the CFS shape and failure
compared the minimum FOS associated with CFS. The direction and then build the FOS utilizing the limit equilib-
extended Spencer method was well-matched with variational rium approach. In Matlab, a program was made to calculate
approaches. the FOS and 3D form of critical slip surfaces. Finally, a 3D
Xie et al. (2003) introduced a new 3D deterministic slope model was made and tested in the lab with a vertical
grid-based geographic information system (GIS) model load. The authors re-analyzed PSO data and concluded that
developed using a column-based 3D LEM based stability the provided codes were quite useful in locating soil slope
analysis model. Traditionally, a deterministic technique was failure surfaces. Moresey et al. (2021) employed Gray wolf
adequate for both homogeneous and heterogeneous slopes. optimization techniques in a 3D slope stability checks to find
Monte Carlo Simulation (MCS) was employed to compute the FOS and associated CFS utilizing the limit equilibrium
the CFS of a 3D slope. All of the calculations are carried out approach. A program was written for spherical slip surface
by 3DSLOPEGIS, a computer program that includes a GIS in Matlab to calculate the FOS using force and moment
component for performing spatial analysis. equilibrium and compute the 3D form of the failure surface.
Mowen (2004) used MCS to develop a GIS-based 3D The results showed that GWO is more effective and use-
search technique for determining the three dimensional CFS, ful than other optimization methods for analyzing 3D soil
assuming the initial slip curve as a part of an ellipsoid. The slope CFS. This is because the FOS is lower and the failure
author extended the 2D slope analysis to 3D to determine shape is closer to the natural slope shape. Table 2 presents
the CFS, and studied various practical situations including an overview of the contributions of various researchers for
known minimum FOS and CFS in 2D slope stability checks. finding out CFS.
When comparing the cross-sections of the 3D CFS to the
2D result, there are no obvious differences. However, the
3D FOS for normal slope problems was found to be more Limit analysis method (LAM)
than 2D FOS.
Cheng et al. (2005) presented a practical method for A LAM was developed using the associated flow rules
employing NURBS (Non-Uniform Rational B-Splines) sur- and virtual work. Applying this method, valid results can
faces and ellipsoidal surfaces to simulate a sliding surface in be framed using upper and lower bounds by minimizing
13
Table 2 Review and characteristics of various optimization method and computer program for CFS
Optimization References Failure surface Remarks
method/computer
program
Box search Reid et al. (2015) Spherical slip surface Minimize computational time using coarse to fine search
MCS Xie et al. (2003) and Mowen (2004) Lower part of ellipsoid Minimum FOS obtained after about 100 trials
NURBS Cheng et al. (2005) Ellipsoid Insensitive to the initial solution
Metaheuristic optimization
GWO Moresey et al. (2021) Spherical slip surface Higher computational efficiency, fast, accurate, very simple
to use, more efficient to solve any engineering problems
PSO Kalatehjari et al. (2015) Ellipsoid Well suited for complex and multimodal problems, Higher
efficiency, Simple to operate
GA Ahangar-Asr et al. (2012) Rotational ellipsoid Effective and efficient technique
TDAVLG Hajiazizi and Tavana (2013) Non-spherical slip Easy to understand, high precision, high efficiency
3D-DYRANUT Yamagami and Jiang (1997) Generalized Shape Applicable either inhomogeneous cases or unsymmetrical
slope, obtained more efficient computation procedure for
large search regions
Dynamic Jiang and Yamagami (2004) Rotational surfaces High computational efficiency, high precision
program with
extended Spen-
cer’s method
the effects of the nature of application of the forces on are not always in equilibrium, and the failure mechanism
the columns in LEM. Several 3D upper-bound solutions is not always the same as the actual failure mechanism.
with collapse mechanisms based on translation or rotation The best (least) upper bound value can be discovered by
were shown. Giger and Krizek (1975) used upper bound analyzing several mechanisms. According to the lower
limit analysis theorem to investigate the stability check of bound theorem, if an equilibrium stress distribution can be
a vertical corner cut subjected to local surcharge loading. determined for the entire body that balances a number of
Leshchinsky et al. (1985) suggested a 3D strategy combin- external loads on the stress boundary and isn't higher than
ing variational principle and limit equilibrium, which was the material's failure criterion, the external loads aren't
an expansion of Baker and Garber’s (1978) rigorous 2D more than the actual collapse loads. It should be empha-
methodology. Michalowski (1989) presented a 3D result sized that the strain and displacements are not taken into
based on the mechanisms of translational collapse compris- account in the lower bound theory, and that the stress state
ing numerous stiff blocks. Michalowski’s (1985) findings does not always conform to the stress state at collapse.
were further improved by Farzaneh and Askari (2003) using The best (highest) lower bound value can be determined
transverse surfaces for every block for non-homogenous 3D by analyzing various permissible conditions of stress. A
soil slopes. Buhan and Garnier (1998) studied the upper- summary of some of the studies on the stability of 3D
bound load capacity of slope-adjacent square foundations slopes has been provided.
using rotating and punching collapse mechanisms. Farzaneh et al. (2008) carried out stability analysis of
A limited number of studies based on the 3D approach convex 3D slopes based on the upper bound theorem of limit
have been conducted to find the soil slope stability using analysis. This method benefits from the ability to calculate
the limit analysis method. Because it is difficult to create both the 3D FOS and the load-bearing capacity. After ana-
three-dimensional failure mechanisms in frictional soils, lyzing the obtained results of foundation bearing capacity
such studies are uncommon. Limit analysis models soil as tests, they concluded that those situated near straight slopes
a plastic substance that follows a flow law. This soil ide- have a greater capacity than those located adjacent to convex
alization allows for development of lower and upper plas- slopes.
tic bounding theorems (Drucker and Prager 1952; Chen Michalowski and Drescher (2009) performed a 3D limit
1975). The external loads induced are not smaller than analysis of curvilinear cone failure surface defined by log-
the actual collapse loads if all the external loads works spirals considering rotational mechanism. In the case of
are involved in the failure mechanism and the work car- cohesive soil, the failure surface takes the form of a torus.
ried out by the external loads per unit of displacement For a reasonable range of values, the critical height of slopes
equates the work done by the internal stresses. It should with finite width is found, and the findings are provided in
be emphasized that the internal stresses and external loads the form of graphs and tables.
13

Ganjian et al. (2010) suggested a 3D technique for deter- 2005). Over the conventional LEM-based slope analysis
mining the seismic slope stability under local loading based methodology, FEM-based slope stability solutions offer a
on the upper bound limit analysis technique. The seismic few distinct advantages. In LEM, it is required to assume a
stability factors were determined using the energy dissipa- pre-defined failure surface shape (i.e., spheroid, log-spiral,
tion approach and a 3D rotational collapse mechanism. On ellipsoid, paraboloid, etc.), and the FOS against slope fail-
locally loaded slopes, the effects of the dilatancy angle were ure is determined for the assumed failure surface shape. In
examined. reality, the assumption related to the shape of the failure
Nadukuru et al. (2011) studied the stability of a 3D surface may not exactly tally with the in-situ conditions. On
slope under the action of quasi-static distributed force. The the other hand, Strength Reduction Technique (SRT) based
developed charts for calculating the FOS were found to be on FEM tries to cause instability of the slope domain by
extremely useful. If the mechanism has limited width or gradually reducing the strength parameters (i.e., cohesion
constrained by local geology, the analysis is found to be and angle of internal friction) of soil by a suitable factor,
extremely accurate. They also devised a method for calcu- and designates the same as the FOS of the slope when fail-
lating the critical acceleration coefficient and displacements ure is initiated. The method is versatile in the sense that the
as a result of seismic excitation. The analysis, however, had instability of the slope is caused naturally by stress failure.
limitations as the procedure was only applicable to slopes Griffiths and Lane (1999) have used SRT based formulation
having inclination angle less than 4 50. Nadukuru and Mich- to solve a wide variety of slope stability problems. In com-
alowski (2013) used planar strain estimates and quasi-static parison to LEM, FEM provides information on deformed
load for seismic excitation. Based on the kinematic theorem failure mass, developed stresses and strains at different loca-
of limit analysis, a three-dimensional analysis of slopes is tions that are not available from LEM based formulations
performed. A 3D rotating mechanism with a failure sur- Lenchman and Griffiths (2000).
face at the slope toe was devised. A critical acceleration �
for 3D slopes is also determined, and a sliding block analy- c
(4)
�
cf =
sis is performed to solve for seismic slope displacements. SRF
Table 3 presents an overview of the contributions of various
researchers based on LAM.
( � )
� 𝜑
𝜑f = tan−1 (5)
SRF
here, SRF is strength reduction factor; cf and 𝜑f is shear

′ ′
3D slope stability analysis using finite
element method strength parameters.
Figures 5 and 6 show the deformed mesh of a 3D slope
The FEM is often used to model a domain with known mate- made up of undrained cohesive soil, earlier analyzed by
rial properties subjected to external loadings to find the Griffiths and Marquez (2007). Figure 5 presents the failure
resulting displacement, field stress and strain. For slope sta- pattern of a symmetrical slope with respect to geometry,
bility issues, several researchers have attempted to develop loading and support conditions, whereas Fig. 6 describes the
FEM-based solutions (Griffiths and Lane 1999; Chen et al. asymmetrical failure pattern of the same slope. The nature
Table 3 Review and characteristics of LAM based slope

References Methodology Application Important Discussion
Farzaneh et al. (2008) Upper bound theorem of Limit Convex slope Examined 3D FOS and Bearing
analysis capacity
Michalowski and Drescher (2009) Kinematic principle of Limit Excavation slopes Examine the critical height of slope
analysis for a finite width
Ganjian et al. (2010) Upper bound theorem of Limit Homogenous slope with local Examine the effect of dilatancy
analysis loading angle with varying local load on
a slope
Nadukuru et al. (2011) Kinematic principle of limit Analyses of existing slope Investigate the critical acceleration
analysis and find solutions for slope dis-
placements produced by seismic
shaking
Nadukuru and Michalowski kinematic theorem of limit Slope subjected to seismic load Examined the critical acceleration
(2013) analysis and sliding block analysis to solve
seismic slope displacements
13
introduced a 3D finite element limit equilibrium method

(known as 3DFLEM) that was developed on the basis of
principle of strength reduction and the single direction
of sliding. The mean value theorem of integrals is used
to determine the FOS corresponding to a sliding body of
any shape. They compared the 3D shear strength reduction
(SSR) approach with the 3D rigid limit equilibrium method.
The authors found that both the methods had good agree-
ment on FOS and critical slide surface.
Zhang et al. (2013) combined elastoplastic finite differ-
ence method and a strength reduction method to study how
Fig. 5 Symmetrical failure surface from FEM based SRT analysis. complicated shapes affect the stability of three-dimensional
Source: Griffiths and Marquez (2007) slopes. The analysis results were beneficial for preparing for
landslide hazards or designing infrastructures in a safe and
cost-effective manner.
Kelesoglu (2016) used the SRT and the finite difference
method (FDM) to figure out how each factor, like the slope’s
curvature, the pile contribution, and the slope’s local load-
ing, affected the slope. Concave slopes were discovered to
be more stable than straight slopes. When compared to a
straight slope, FOS values are found to be higher by 15–25%
for steep concave curves and up to 10% for smooth con-
cave slopes. When local loading exists on a slope due to
mobilization shear strains induced by a surcharge, the pile
row arrangement must be shifted upward toward the load to
preserve local and global stability. Table 3 summarizes the
Fig. 6 Asymmetric failure surface from FEM based SRT analysis.
Source: Griffiths and Marquez (2007) main contributions in the field of slope analysis using finite
element based solutions.
Tiwari et al. (2015) studied the stability of large land-
of the deformed mesh is indicative of the zone that can be slides using Specfem 3D Slope, an open-source spectral
regarded as the failure surface. element method (SEM) tool. In contrast to FEM, the SEM
Jeremić (2000) presented the latest technique for mod- technique is extremely advantageous since it greatly mini-
eling 3D slope stability issues. The failure mechanisms of mizes the enormous computational burden by adopting
slopes are defined by localized, continuous deformation, p-refinement techniques rather than h-refinement techniques.
which is described by a p-version of the FEM. It is shown As a result, SEM proved to be a very precise and strong
how, using a relatively small number of finite elements, the approach for elastoplastic modeling, deserving of its own
new method can be used to describe sharp deformation gra- computing technique for simulating large-scale landslides.
dients caused by shear localization during slope failures. It Table 4 presents an overview of the contributions of various
has been demonstrated that a convex curvature of the slope researchers based on FEM.
out of the plane can significantly reduce the safety factor.
Griffiths and Marquez (2007) performed a Mohr–Cou-
lomb failure criterion for a 3D analysis of elastic-perfectly Stability chart for 3D slope analysis
plastic soils using the finite element (FE) program, assuming
no dilatation. In the algorithm’s stiffness matrix creation and An effective method for the preliminary evaluation of slope
stress redistribution stages, the authors used 20-node hexa- safety can be obtained through the use of stability charts for
hedral elements with reduced integration. Using an elasto- slopes. Stability chart is a graphical tool used for plotting
plastic FE method, the authors presented the findings from the values of dimensionless stability number of a slope that
various 3D slope examples. They proved that the conver- can be readily used to determine the FOS without carrying
gence of the 3D safety factor with the 2D result increased. actual analysis. Stability charts avoids the need for perform-
Tan and Sarma (2008) used finite element method (FEM) ing iterative calculations to obtain the minimum FOS.
to evaluate a new technique based on the limit equilibrium Michalowski (2010) presented a kinematic limit approach
method for both homogeneous and non-homogeneous for 3D slope stability analysis. The approach produces an
slopes using non-associated flow rules. Li and Shao (2011) upper bound for the critical/limiting slope elevation, yielding
13

Table 4 Review and characteristics of FEM based slope

Jeremić (2000) p-version of the FEM Simple slope Examined the factor of safety value
decreased when slope is curved out
of plane in a convex way
Griffiths and Marquez (2007) Elastoplastic finite element method Homogenous and weak layer slope Examined the convergence of the 3D
safety factor with the 2D result
Tan and Sharma (2008) Finite element analysis Homogenous and non-homogenous Examined the normal and shear
slope stress distribution on slip surface
Li and Shao (2011) Finite element limit equilibrium Homogenous slope Comparing the 3D Rigid limit equi-
method (3DFLEM) librium technique to the 3D shear
strength reduction method
Zhang et al. (2013) Elastoplastic finite difference Boundary surface Examined the effect of turning arcs,
method turning corners and turning forms
Kelesoglu (2016) Strength reduction method Slope curvature, pile contribution, Examined the effect of plane curva-
and local structural loading ture of slope, contribution of the
pile and effect of local surcharge
act at crest
Tiwari et al. (2015) spectral element method (SEM) Large scale landslide Reduces computational load using
p-refinement instead of h-refine-
ment
a safety factor upper bound. To create slope stability charts, Sun et al. (2017) present 3D stability charts to evaluate
the author developed a 3D failure mechanism. The charts the FOS of convex and concave slopes with homogeneous
can be used without iterations to read the slope safety fac- soil. The Author employed a strength-reduction technique to
tor. Stability charts are useful when the failure mechanism carry out stability analysis of a slope to simplify the com-
is limited, such as on an excavation slope. puting procedure. Just like Kelesoglu (2016) and Sun et al.
Li et al. (2010) developed upper and lower bound limit (2017) also found the concave slopes as comparatively more
analysis (LA) based finite element procedures to provide stable than straight embankments, whereas convex slopes
chart solutions. These charts are useful for preliminary were observed to be relatively less stable. Engineers can
design to study for short term and long term stability of 3D use dimensionless 3D graphs to determine the early stabil-
slope. The author found that for the purpose of preliminary ity of excavated or manufactured slopes. Table 5 presents an
design of the slope where L/H > 5, 2D solutions can be used overview of the contributions of various researchers to the
in place of 3D solutions when applying the lower bound 3D stability chart.
analysis.
Michalowski and Martel (2011) investigated 3D steep
slopes using the kinematic limit analysis theorem. The
failure surface had the resemblance of a curvilinear cone 3D analysis of rock slopes
passing through the toe of the slope, typical of steep slopes,
satisfying a rotational failure mechanism. Stability charts The stability of rock slopes is still a very important topic
were designed to determine the safety factor using a quasi- in rock engineering due to the fact that the rock failure can
static technique, and the charts were found to be quite use- result in significant economic loss as well as serious casual-
ful because they do not require any iteration procedures to ties (Shen and Karakus 2014; Yang and Pan 2015; Yang and
estimate the FOS. Long 2016). As a consequence of this, one of the primary
Gao et al. (2013) developed stability charts for 3D responsibilities of a geotechnical engineer is the evaluation
homogeneous soil slopes under loading, which are static of the stability of rock slopes.
and pseudo-static situations over various parameters. Liu et al. (2013) introduced a FEM based multigrid
Based on Michalowski and Drescher’s (2009) 3D kinemati- method for calculating FOS of the slope. This method effi-
cally acceptable rotating failure mechanism, an analytical ciently combines nonlinear limit equilibrium and finite ele-
technique for analyzing the stability of uniform slopes in ment approaches for determining the stability of any surface
cohesive and frictional soils is developed. According to the or sliding block. The Drucker-Prager yield criterion-based
authors, the 3D rotational failure mechanisms provide the ideal elastic plastic incremental analytical approach was
most accurate upper bound estimate. used in the nonlinear FE computation. The FOS values from
13
Table 5 Review and characteristics of 3D slope using stability chart

References Methodology Application Important discussion
Michalowski (2010) Kinematic principle of limit analysis Loaded slope When the failure mechanism is con-
strained, such excavation slopes,
stability charts are beneficial
Li et al. (2010) finite element upper and lower Undrained and drained condition where L/H > 5, 2D solutions can be
bound limit analysis used in place of 3D solutions
Michalowski and Martel (2011) Kinematic principle of limit analysis Homogenous slope Developed stability chart and this
chart is useful for excavation slope
and physically constrained slope
Gao et al. (2013) Upper bound limit analysis Homogenous slope 3D rotational failure mechanisms
provide the best upper bound
estimate
Sun et al. (2017) Strength-reduction technique (SRT) Convex and Concave slopes with To assess the early stability of
homogenous soil excavated or constructed slopes,
engineers can employ dimension-
less 3D graphs
finite element analysis were higher and more accurate than to give FOS solutions that matched the analytical solutions.
those established using the rigid-body LEM. The authors showed that the FOS achieved in 3D and 2D
Shen and Karakus (2014) proposed a nonlinear SSR scenarios differed significantly.
method that incorporated Hoek–Brown (HB) criterion in the Yang et al. (2021) proposed a 3D nonlinear strength
FLAC3D platform for studying 3D rock slopes. The authors reduction numerical manifold method (3D-NSRNMM) that
conducted extensive case studies to check the behavior of utilized nonlinear Generalized Hoek–Brown strength crite-
the boundary conditions and the convergence criterion. The rion (GHB) to perform 3D rock slope analysis. The authors
authors found that when the slope angle is less than 50°, the suggested two approaches to transform the GHB parameters
boundary condition will substantially impact the value of the into their comparable instantaneous and average MC param-
safety factor for the given 3D slope. eters. The study established a significant effect of the bound-
Yang and Pan (2015) used a kinematical limit analysis ary conditions on the estimated FOS values. The 2D FOS
technique to determine the stability of three-dimensional was found more conservative compared to their 3D values.
rock slopes based on nonlinear Hoek–Brown criteria sub- Instantaneous equivalent MC parameters yielded FOS values
jected to seismic forces. The authors extended Michalowski almost similar to those found using average equivalent MC
and Drescher (2009) 3D static stability analysis with linear parameters. Table 6 presents an overview of the contribu-
MC yield criterion to nonlinear criterion under earthquake tions of various researchers on the 3D Rock slope.
loadings. The suggested upper bound LA method is an
excellent tool for evaluating the three-dimensional stability
of rock slopes using a nonlinear yield criterion. Application of artificial intelligence
Yang and Long (2016) examined a two-stage slope’s and machine learning in 3D slope analysis
seismic and static stability using the nonlinear HB crite-
rion and the upper bound theorem, and the stability factors Recently, artificial intelligence (AI) and machine learning
were determined using a thorough methodology. When the (ML) techniques have evolved dramatically with the advent
two-stage slope was reduced to one step, and the nonlinear of modern day computers. These techniques are being used
HB criterion was reduced to the linear MC criterion, useful in almost every area of science and technology. The topic
solutions were discovered. of slope stability analysis is no exception. The AI/ML tech-
XU and Yang (2018) used the limit analysis method to niques have been used in 2D slope stability analysis previ-
do a seismic rock slope analysis considering the material ously many researchers. Extreme gradient boosting method
response governed by Hoek–Brown criterion. The optimal (Wang et al. 2020; Zhou et al. 2019), relevance vector
FOS of a 3D rock slope is then obtained using an optimi- machine Zhao et al. (2012), Gaussian process regression
zation method based on the strength reduction approach. Zhu et al. (2021), extreme learning machine (Bardhan et al.
To estimate the FOS of a 3D rock slope, the equivalent 2022; Huang et al. 2017), artificial neural networks Ray et al.
Mohr–Coulomb strength terms derived from generalized (2020), hybrid machine learning Bardhan and Samui (2022)
tangent methodology are investigated. The authors used the are some of the ML method which have been used success-
generalized tangent method with Mohr–Coulomb parameters fully to present regression based results of 2D slope stability
13

Table 6 Review and characteristics of 3D Rock slope

Liu et al. (2013) Nonlinear finite element analysis and LEM Rock mass Compared stability factor of rigid-body limit equilibrium
method to finite element analysis
Shen and Karakus (2014) Shear strength reduction (SRT) Rock mass Examined the effect of boundary condition and conver-
gence criteria on FOS
Yang and Pan (2015) kinematical method of limit analysis Rock slope Compared stability factors to Michalowski and Drescher
(2009)
Yang and Long (2016) HB criterion and the upper bound theorem Rock slope Examined nonlinear HB criterion reduced to the linear MC
criterion
XU and Yang (2018) Limit analysis Rock slope Compared 3D FOS to 2D
Yang et al. (2021) 3D nonlinear strength reduction numerical Rock mass Examined the effect of boundary condition and compared
manifold method (3D-NSRNMM) 3D FOS of Rock slope to 2D FOS
solution of various problems. The scope of the current paper the sliding mass. Consequently, the 3D safety factor may
deals mainly with the topic of 3D slope stability solutions. be conservatively underestimated, while the back calculated
Regrettably it is seen that there is very scant application of mobilized shear strength may be exaggerated.
ML techniques in the area of 3D slope stability analysis. Arellano and Stark (2000) introduced a latest method for
After thorough search, we found one such paper that dis- adding the shear resistance on both sides of the sliding mass
cusses the application of artificial neural network to simulate into existing three-dimensional (3D) software. For a given
stability chart solution of 3D slopes Meng et al. (2021). This slope angle, the 3D/2D safety factors ratios goes up as width
is definitely an area which demands further investigation. to height ratios go down. The ratio of 3D/2D FOS value
increases when the slope inclination reduces for a particu-
lar width/height ratio. The 3D FOS value is more affected
Softwares for 3D slope stability analysis than the corresponding 2D value by the variation in shear
strength between the upper and lower layers.
To explore the stability of soils, several researchers worked Shen and Karakus (2014) introduced a nonlinear SSR
on various numerical modeling and optimization tools. technique that can model the nonlinear response of a rock
There are few softwares available in the market capable of mass based on Hoek–Brown (HB) yield and analyze 3D
carrying out 3D slope stability analysis. Commercially avail- slope stability in FLAC3D environment. The investiga-
able softwares usually perform analysis and design of varie- tors found that the FOS value depends on convergence and
ties of problems include three dimensional slope analysis. boundary condition criteria. The boundary conditions are
One of the popular software is the computer program called seen to severely influence the FOS value of a slope with a
STAB3D developed by Baligh and Azzouz (1975). Chen low angle of inclination than a steeper slope.
(1981) developed two separate computer programs, LEMIX Reid et al. (2015) developed a search analysis technique
and BLOCK 3, to evaluate rotational and translational slides, in Scoops3D-i software. The program evaluates the stability
respectively, based on the concept of LEM. FESPON, a of numerous possible failure surfaces within a user-defined
three-dimensional finite element computer program, was size range. The Scoops3D software also offers a coarse-to-
also developed to analyze rotating slides by Chen (1981). fine search variant to reduce overall run time while searching
Gens et al. (1988) provided a 3D technique for soils that for the center of CFS and least FOS. Table 7 presents an
are homogenous, isotropic, or entirely cohesive. Analytical overview of the contributions of various researchers using
solutions were developed considering general shape of the the software.
slip surface as a cylindrical section connected by splines
to planar or curved surfaces. They developed the computer
program DEEPCYL for planar ends, which calculates the Conclusions
3D FOS and critical slip surface given slope angle, slide
length, and (if necessary) depth factor D data. In addition, From a geotechnical perspective, the issue of establishing
they developed a computer program, F3SLOP, for different a slope’s factor of safety (FOS) against failure is essential.
ends, to calculate 3D FOS for a potential slide of length (L), Numerous slope failures throughout the world result in
slope angle (i), and height (H). significant economic damages, transport disruptions, and
Stark and Eid (1998) discovered that the 3D slope stabil- loss of human lives. Proper study and design of a slope are
ity programs (CLARA 2.31) ignores the shear resistance on essential in preventing such failures. Different approaches
13
Table 7 Review and characteristics of 3D slope based on computer algorithm

Software References Methodology Application Important discussion
LEMIX, BLOCK 3 and Chen (1981) Limit Equilibrium Method Cohesive and Compared 2D to 3D interslice
FESPON Cohesion angles, Comparison of LEMIX
less slope results with ordinary method
of column
DEEPCYL and DEEPCYL Gens et al. (1988) Swedish circle Simple slopes 3–30% increase in FOS in 3D as
compare to 2D
CLARA 2.31 Stark and Eid (1998) Extension of Bishop’s sim- Case histories Performs both 2D and 3D slope
plified, Janbu simplified stability analysis
method, Los Angeles method
and Method of slices
CLARA 2.31 Arellano and Stark (2000) Extension of Bishop’s simpli- Case history Comparison of 2D and 3D slope
fied, Janbu’s simplified analyses for different failure
method mass geometries
FLAC3D Shen and Karakus (2014) Shear strength reduction (SSR) Case studies Convergence criteria play
important role in determination
of 3d FOS
Scoops3D-i Reid et al. (2015) LEM Homogenous, Performs both 2D and 3D slope
Layered soil stability analysis
and conical
heaps
to slope-stability analysis have been established earlier. stability of excavated or constructed slopes, engineers can
Each has advantages and disadvantages in comparison to employ dimensionless 3D stability graphs.
the others. In this study, literature reviews based on three FEM based strength reduction technique (SRT) has been
different methods i.e., Finite Element Method, Limit Equi- found to be an extremely efficient tool in dealing with com-
librium Method, Limit Analysis Method and have been plex 3D slope stability problem. If not only reports the mini-
carried out to present an overview on different aspects mum FOS with sufficient precision but also can describe
of three dimensional slope stability analysis. The major- deformed mesh of the slope on the verge of failure along
ity of 3D slope stability LEMs are extended formulations with the associated stress states. It can efficiently simulate
from that of 2D LEMs. Even though the assumptions of the loadings such as pore pressure, earthquake etc.
the 3D slope stability formulations are the same as the Relatively less works have been reported on 3D rock
2D approaches, it is discovered that the 3D FOS is higher slope stability analysis incorporating appropriate materials
than the 2D FOS. LEM based formulations have been pro- model such as Hoek–Brown strength criteria, though there
posed for complex slope stability problems involving cor- are abundant presence of similar works in 2D. Therefore,
ner stability as well as for 3D soil domains where the cross 3D rock slope stability analysis can be construed as an unex-
section varies in the lateral direction. While carrying out plored field of research which demands attention from the
3D slope analysis using LEM, it is important to correctly researchers.
ascertain the direction of slide (DOS) of the columns pre- Currently, few commercially available softwares can per-
sent in the failure mass. Various optimization techniques form 3D slope analysis. Some of these softwares worthy
are used to obtain the CFS of a 3D slope. Among those of mention are STAB3D, LEMIX, BLOCK 3, FESPON,
optimization techniques, GWO and PSO gave higher com- DEEPCYL, CLARA 2.31, FLAC3D and Scoops3D-i.
putational efficiency and, are found to be fast, accurate, Among this software, Scoops3D-i is downloadable and easy
and easy to work with. These techniques are applied to to execute. Scoops3D-i can be applied to homogenous, non-
solve complex and multimodal problems. homogeneous, conical heaps and layered soil.
Very few research has been conducted on seismic slope Overall, it has been discovered that the delayed and/or
stability in three dimensions. LAM uses upper and lower time-dependent behavior of slopes has received very lit-
bound theorems to determine the limiting value of FOS tle attention. As a result, additional research effort needs
based on predefined failure mechanisms. In the majority of to be done to gain an understanding of the time-dependent
approaches employing LAM, charts have been developed to behavior of slopes. It would also be essential to investigate
compute the FOS in seismic loading scenarios, which has the 3D slope analysis based on the models developed for
shown to be highly useful as it avoids the need for perform- unsaturated soil, as little or no work is currently available.
ing complex slope stability calculations. To assess the early Applying ML techniques to the field of three-dimensional
13

slope stability analysis would be crucial because there is Chen RH (1981) Three-dimensional slope stability analysis. Joint
either no work in this area or very little of it. Highway Research Project. Report JHRP-81-17. Eng. Experiment
station, Purdue, West Lafayette, IN, USA
Cheng YM (2003) Location of critical failure surface and some further
studies on slope stability analysis. Comput Geotech 30(3):255–
Author’s contribution SK: Formal analysis, writing – original draft; 267. https://doi.org/10.1016/S0266-352X(03)00012-0
SSC: Writing – review and editing; AB: Conceptualization, Investiga- Cheng YM, Liu HT, Wei WB, Au SK (2005) Location of critical three-
tion, Methodology writing. dimensional non-spherical failure surface by NURBS functions
and ellipsoid with applications to highway slopes. Comput Geo-
Funding No external funding was required to perform the work pre- tech 32(6):387–399. https://doi.org/10.1016/j.compgeo.2005.07.
sented in this paper. 004
Cheng YM, Yip CJ (2007) Three-dimensional asymmetrical slope sta-
Data availability Data will be made available on request to the cor- bility analysis extension of bishop’s, Janbu’s, and Morgenstern—
responding author. Price’s Techniques. 133(December):1544–1555
Clough RW, Woodward RJ (1967) Analysis of embankment stresses
Declarations and deformations. J Soil Mech Found Div 93(4):529–549
De Buhan P, Garnier D (1998) Three dimensional bearing capacity
Conflict of interest The authors declare that the work presented in this analysis of a foundation near a slope. Soils Found 38(3):153–163
paper has no conflict of interest with any other works. Dennhardt M, Forster W (1985) Problems of three-dimensional slope
stability. Int Conf Soil Mech Found Eng 11:427–431
Drucker DC, Prager W (1952) Soil mechanics and plastic analysis or
limit design. Q Appl Math 10(2):157–165
References Farzaneh O, Askari F (2003) Three-dimensional analysis of nonho-
mogeneous slopes. J Geotech Geoenviron Eng 129(2):137–145
Ahangar-Asr A, Toufigh MM, Salajegheh A (2012) Determination of Farzaneh O, Askari F, Ganjian N (2008) Three-dimensional stability
the most probable slip surface in 3D slopes considering the effect analysis of convex slopes in plan view. J Geotech Geoenviron
of earthquake force direction. Comput Geosci 45:119–130. https:// Eng 134(8):1192–1200. https://doi.org/10.1061/( asce)1090-
doi.org/10.1016/j.cageo.2011.10.024 0241(2008)134:8(1192)
Anagnosti P (1969) Three-dimensional stability of fill dams. In: Pro- Fredlund DG, Krahn J (1977) Comparison of slope stability methods
ceeding of 7th international conference on soil mechanics and of analysis. Can Geotech J 14(3):429–439
foundation engineering, Mexico, pp 275–280 Ganjian N, Askari F, Farzaneh O (2010) Influences of nonassociated
Arellano D, Stark TD (2000) Importance of three-dimensional slope flow rules on three-dimensional seismic stability of loaded slopes.
stability analyses in practice. In: Slope stability 2000, pp 18–32 J Cent South Univ Technol 17(3):603–611
Azzouz AS, Baligh MM (1978) Three-dimensional stability of slopes. Gao Y, Zhang F, Lei GH, Li D, Wu Y, Zhang N (2013) Stability charts
Massachusetts Inst. of Technical report for 3D failures of homogeneous slopes. J Geotech Geoenviron
Baker R, Garber M (1978) Theoretical analysis of the stability of Eng 139(9):1528–1538. https://doi.org/10.1061/(asce)gt.1943-
slopes. Geotechnique 28(4):395–411 5606.0000866
Baker R, Leshchinsky D (1987) Stability analysis of conical heaps. Gens A, Hutchinson JN, Cavounidis S (1988) Three-dimensional analy-
Soils Found 27(4):99–110 sis of slides in cohesive soils. Geotechnique 38(1):1–23
Baligh MM, Azzouz AS (1975) End effects on stability of cohesive Giger MW, Krizek RJ (1975) Stability analysis of vertical cut with
slopes. ASCE J Geotech Eng Div 101(11):1105–1117. https://d oi. variable corner angle. Soils Found 15(2):63–71
org/10.3313/jls1964.23.16 Griffiths DV, Lane PA (1999) Slope stability analysis by finite ele-
Bardhan A, Dhilipkumar B, Mulagala L, Samui P (2022) Application ments. Geotechnique 49(3):387–403
of bi-directional extreme learning machine in predicting stability Griffiths DV, Marquez RM (2007) Three-dimensional slope stabil-
of slope of railway embankment in seismic condition. In: Stabil- ity analysis by elasto-plastic finite elements. Geotechnique
ity of slopes and underground excavations. Springer, London, pp 57(6):537–546. https://doi.org/10.1680/geot.2007.57.6.537
11–22 Hajiazizi M, Tavana H (2013) Determining three-dimensional non-
Bardhan A, Samui P (2022) Probabilistic slope stability analysis of spherical critical slip surface in earth slopes using an optimization
Heavy-haul freight corridor using a hybrid machine learning para- method. Eng Geol 153:114–124. https://d oi.o rg/1 0.1 016/j.e nggeo.
digm. Transp Geotech 37:100815 2012.11.014
Bishop AW (1955a) The analysis of stability of slopes. Géotech 5:7 Hovland HJ (1977) Three-dimensional slope stability analysis method.
Bishop AW (1955b) The use of the slip circle in the stability analysis J Geotech Eng Div 103(9):971–986
of slopes. First Technical Session: General Theory of Stability Huang C, Tsai C, Chen Y (2002b) Generalized method for three-
of Slopes dimensional slope stability analysis. J Geotech Geoenviron Eng
Cavounidis S, Kalogeropoulos H (1992) End effects on the stability of 128(10):836–848. https://d oi.o rg/1 0.1 061/( asce)1 090-0 241(2002)
cuts in normally consolidated clays. RIG 2:92 128:10(836)
Chen RH, Chameaut J (1983) Three-dimensional limit equilibrium Huang F, Huang J, Jiang S, Zhou C (2017) Landslide displacement
analysis of slopes. Geotechnique 33:31–40 prediction based on multivariate chaotic model and extreme learn-
Chen Z, Mi H, Zhang F, Wang X (2003) NOTE/NOTE a simplified ing machine. Eng Geol 218:173–186
method for 3D slope stability analysis. Can Geotech J 683:675– Huang C, Tsai C (2000) New method for 3D and a symmetrical stabil-
683. https://doi.org/10.1139/T03-002 ity analysis, October, 917–927
Chen J, Yin JH, Lee CF (2005) A three-dimensional upper-bound Huang C, Tsai C, Chen Y (2002a) Generalized method for three-dimen-
approach to slope stability analysis based on RFEM. Geotech- sional slope stability analysis, October, 836–848
nique 55(7):549–556 Hungr O (1987) An extension of Bishop’s simplified method of
Chen WF (1975) Limit analysis and soil plasticity. In: Developments slope stability analysis to three dimensions. Geotechnique
in geotechnical engineering. Elsevier, Amsterdam 37(1):113–117
13
Hungr O, Salgado FM, Byrne PM (1989) Evaluation of a three- Eng 137(2):183–189. https://d oi.o rg/1 0.1 061/( asce)g t.1 943-5 606.
dimensional method of slope stability analysis. Can Geotech J 0000412
26(4):679–686 Mongenstern NR, Price VE (1965) The analysis of stability of general
Janbu N (1954) Application of composite slip surface for stability slide surfaces. Geotechnique 15:79–93
analysis. Proc Eur Conf Stab Earth Slopes Sweden 3:43–49 Morsey NA, Gad S, Bakr MA (2021) 3-D critical slip surface based on
Janbu N (1973) Slope stability computations. Wiley, London artificial intelligent. Int J Geomate 21(85):99–106
Jeremić B (2000) Finite element methods for 3D slope stability analy- Mowen X (2004) A simple Monte Carlo method for locating the
sis. In: Proceedings of sessions of geo-denver 2000—slope sta- three-dimensional critical slip surface of a slope. Acta Geol Sin
bility 2000, GSP 101, 289, pp 224–238. https://doi.org/10.1061/ Engl Ed 78(6):1258–1266. https://doi.org/10.1111/j.1755-6724.
40512(289)17 2004.tb00783.x
Jiang JC, Yamagami T (2004) Three-dimensional slope stability analy- Nadukuru SS, Martel T, Michalowski RL (2011) 3D analysis of steep
sis using an extended Spencer method. Soils Found 44(4):127–135 slopes subjected to seismic excitation. In: Geo-frontiers 2011:
Kalatehjari R, Rashid ASA, Hajihassani M, Kholghifard M, Ali N advances in geotechnical engineering, pp 3546–3555
(2014) Determining the unique direction of sliding in three- Nadukuru SS, Michalowski RL (2013) Three-dimensional displace-
dimensional slope stability analysis. Eng Geol 182:97–108 ment analysis of slopes subjected to seismic loads. Can Geotech
Kalatehjari R, Arefnia A, Rashid ASA, Ali N, Hajihassani M (2015) J 50(6):650–661
Determination of three-dimensional shape of failure in soil slopes. Ray A, Kumar V, Kumar A, Rai R, Khandelwal M, Singh TN (2020)
Can Geotech J 52:1–66 Stability prediction of Himalayan residual soil slope using arti-
Kelesoglu MK (2016) The evaluation of three-dimensional effects on ficial neural network. Nat Hazards 103(3):3523–3540
slope stability by the strength reduction method. KSCE J Civ Eng Reid ME, Christian SB, Brien DL, Henderson S (2015) Scoops3D—
20(1):229–242. https://doi.org/10.1007/s12205-015-0686-4 software to analyze three-dimensional slope stability through-
Kopacsy J (1957) Three dimensional stress distribution and slip surface out a digital landscape. US Geological Survey Techniques and
in earth work at rupture. In: Proceedings of the 4th international Methods, Book, 14
conference on soil mechanics and foundation engineering, pp Shen J, Karakus M (2014) Three-dimensional numerical analysis
339–342 for rock slope stability using shear strength reduction method.
Lam L, Fredlund DG (1993) A general limit equilibrium model Can Geotech J 51(2):164–172. https:// d oi. o rg/ 1 0. 1 139/
for three-dimensional slope stability analysis. Can Geotech J cgj-2013-0191
30(6):905–919 Spencer E (1967) A method of analysis for stability of embankment
Leshchinsky D, Baker R (1986) Three-dimensional slope stability: end using parallel inters slice force. Geotechnique 17:11–26
effects. Soils Found 26(4):98–110 Stark TD, Eid HT (1998) Performance of three-dimensional slope
Leshchinsky D, Huang CC (1992a) Generalized three-dimensional stability methods in practice. J Geotech Geoenviron Eng
slope-stability analysis. J Geotech Eng 118(11):1748–1764 124(11):1049–1060. https:// d oi. o rg/ 1 0. 1 061/ ( asce) 1 090-
Leshchinsky D, Huang CC (1992b) Generalized slope stability analy- 0241(1998)124:11(1049)
sis: Interpretation, modification, and comparison. J Geotech Sun G, Zheng H, Jiang W (2012) A global procedure of evaluating
Eng 118(10):1559–1576. https://doi.org/10.1061/(ASCE)0733- stability of three-dimensional slopes. Nat Hazards 61:1083–1098.
9410(1992)118:10(1559) https://doi.org/10.1007/s11069-011-9963-9
Leshchinsky D, Baker R, Silver ML (1985) Three dimensional Sun C, Chai J, Xu Z, Qin Y (2017) 3D stability charts for convex and
analysis of slope stability. Int J Numer Anal Methods Geomech concave slopes in plan view with homogeneous soil based on the
9(3):199–223 strength-reduction method. Int J Geomech 17(1987):1–11. https://
Li H, Shao L (2011) Three-dimensional finite element limit equilibrium doi.org/10.1061/(ASCE)GM.1943-5622.0000809
method for slope stability analysis based on the unique sliding Tan D, Sarma SK (2008) Finite element verification of an enhanced
direction. Slope Stab Earth Retain Walls 216:48–55. https://doi. limit equilibrium method for slope analysis. Géotechnique
org/10.1061/47627(406)7 58(6):481–487. https://doi.org/10.1680/geot.2007.00084
Li AJ, Merifield RS, Lyamin AV (2010) Three-dimensional stability Tiwari RC, Bhandary NP, Yatabe R (2015) 3D SEM approach to
charts for slopes based on limit analysis methods. Can Geotech J evaluate the stability of large-scale landslides in Nepal Hima-
47(12):1316–1334. https://doi.org/10.1139/T10-030 laya. Geotech Geol Eng 33(4):773–793. https://doi.org/10.1007/
Liu Y, He Z, Li B, Yang Q (2013) Slope stability analysis based on s10706-015-9858-8
a multigrid method using a nonlinear 3D finite element model. Ugai K (1985) Three-dimensional stability analysis of vertical cohesive
Front Arch Civ Eng China 7(1):24–31. https://doi.org/10.1007/ slopes. Soils Found 25(3):41–48
s11709-013-0190-1 Ugai K, Hosobori K (1988) Extension of simplified Bishop method,
Meng J, Mattsson H, Laue J (2021) Three-dimensional slope stabil- simplified Janbu method and Spencer’s method to three dimen-
ity predictions using artificial neural networks. Int J Numer Anal sions. Doboku Gakkai Ronbunshu 1988(394):21–26
Methods Geomech 45(13):1988–2000 Ugai K, Leshchinsky DOV (1995) Three-dimensional limit equilibrium
Michalowski RL (1985) Limit analysis of quasi-static pyramidal and finite element analyses: a comparison of results. Soils Found
indentation of rock. Int J Rock Mech Min Sci Geomech Abstracts 35(4):1–7
22(1):31–38 Ugai K (1988a) Three-dimensional slope stability analysis by slice
Michalowski RL (1989) Three-dimensional analysis of locally loaded methods. In: Proceedings of the 6th international conference on
slopes. Geotechnique 39(1):27–38 numerical methods in geomechanics, Innsbruck, Austria
Michalowski RL (2010) Limit analysis and stability charts for 3D slope Wang L, Wu C, Tang L, Zhang W, Lacasse S, Liu H, Gao L (2020) Effi-
failures. J Geotech Geoenviron Eng 136(4):583–593. https://doi. cient reliability analysis of earth dam slope stability using extreme
org/10.1061/(asce)gt.1943-5606.0000251 gradient boosting method. Acta Geotech 15(11):3135–3150
Michalowski RL, Drescher A (2009) Three-dimensional stability of Xie M, Esaki T, Zhou G, Mitani Y (2003) Three-dimensional stability
slopes and excavations. Géotechnique 59(10):839–850 evaluation of landslides and a sliding process simulation using a
Michalowski RL, Martel T (2011) Stability charts for 3D failures of new geographic information systems component. Environ Geol
steep slopes subjected to seismic excitation. J Geotech Geoenviron 43(5):503–512. https://doi.org/10.1007/s00254-002-0655-3
13

Xing Z (1988) Three-dimensional stability analysis of concave slopes Zhao H, Yin S, Ru Z (2012) Relevance vector machine applied to
in plan view. J Geotech Eng 114(6):658–671 slope stability analysis. Int J Numer Anal Methods Geomech
Xu J, Yang X (2018) Seismic stability analysis and charts of a 3D 36(5):643–652
rock slope in Hoek–Brown media. Int J Rock Mech Min Sci Zhou J, Li E, Yang S, Wang M, Shi X, Yao S, Mitri HS (2019) Slope
112(August):64–76. https://d oi.o rg/1 0.1 016/j.i jrmms.2 018.1 0.0 05 stability prediction for circular mode failure using gradient boost-
Yamagami T, Jiang JC (1996) Determination of the sliding direction ing machine approach based on an updated database of case his-
in three-dimensional slope stability analysis. In: Proceedings of tories. Saf Sci 118:505–518
2nd international conference on soft soil engineering, vol 1, pp Zhu B, Hiraishi T, Pei H, Yang Q (2021) Efficient reliability analysis
567–572 of slopes integrating the random field method and a Gaussian pro-
Yamagami T, Jiang J (1997) A search for the critical slip surface in cess regression-based surrogate model. Int J Numer Anal Methods
three-dimensional slope stability analysis. Soils Found 37(3):1–16 Geomech 45(4):478–501
Yang XL, Long ZX (2016) Seismic and static 3D stability of two-stage
rock slope based on Hoek–Brown failure criterion. Can Geotech J Publisher's Note Springer Nature remains neutral with regard to
53(3):551–558. https://doi.org/10.1139/cgj-2015-0147 jurisdictional claims in published maps and institutional affiliations.
Yang Y, Xia Y, Zheng H, Liu Z (2021) Investigation of rock slope sta-
bility using a 3D nonlinear strength-reduction numerical manifold Springer Nature or its licensor (e.g. a society or other partner) holds
method. Eng Geol 292(November 2021):106285. https://doi.org/ exclusive rights to this article under a publishing agreement with the
10.1016/j.enggeo.2021.106285 author(s) or other rightsholder(s); author self-archiving of the accepted
Yang XL, Pan QJ (2015) Three dimensional seismic and static stabil- manuscript version of this article is solely governed by the terms of
ity of rock slopes. Geomech Eng 8(1):97–111. https://doi.org/10. such publishing agreement and applicable law.
12989/gae.2015.8.1.097
Zhang Y, Chen G, Zheng L, Li Y, Zhuang X (2013) Effects of
geometries on three-dimensional slope stability. Can Geotech J
50(3):233–249. https://doi.org/10.1139/cgj-2012-0279
13

Sumit - S40808-022-01597-Y

Uploaded by

Copyright:

Available Formats

Sumit - S40808-022-01597-Y

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Sumit - S40808-022-01597-Y

Uploaded by

Copyright:

Available Formats

Modeling Earth Systems and Environment (2023) 9:1445–1462

Recent advances in 3D slope stability analysis: a detailed review

Introduction Leshchinsky 1987; Leshchinsky and Baker 1986; Leshchin-

Limit equilibrium methods (LEM) for 3D

Almost all 3D LEMs are expansions of 2D slice methods.

Yamagami and Jiang (1996) developed the initial

Table 1 Review and characteristics of LEM based on F0S and DOS

Extension of Morgenstern Price’s Anagnosti (1969) × Generalized Generalized

here, SRF is strength reduction factor; cf and 𝜑f is shear

Table 3 Review and characteristics of LAM based slope

introduced a 3D finite element limit equilibrium method

Table 4 Review and characteristics of FEM based slope

Table 5 Review and characteristics of 3D slope using stability chart

Table 6 Review and characteristics of 3D Rock slope

Table 7 Review and characteristics of 3D slope based on computer algorithm

You might also like

Sumit - S40808-022-01597-Y

Uploaded by

Copyright:

Available Formats

Sumit - S40808-022-01597-Y

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Sumit - S40808-022-01597-Y

Uploaded by

Copyright:

Available Formats

Modeling Earth Systems and Environment (2023) 9:1445–1462

Recent advances in 3D slope stability analysis: a detailed review

Introduction Leshchinsky 1987; Leshchinsky and Baker 1986; Leshchin-

Limit equilibrium methods (LEM) for 3D

Almost all 3D LEMs are expansions of 2D slice methods.

Yamagami and Jiang (1996) developed the initial

Table 1 Review and characteristics of LEM based on F0S and DOS

Extension of Morgenstern Price’s Anagnosti (1969) × Generalized Generalized

here, SRF is strength reduction factor; cf and 𝜑f is shear

Table 3 Review and characteristics of LAM based slope

introduced a 3D finite element limit equilibrium method

Table 4 Review and characteristics of FEM based slope

Table 5 Review and characteristics of 3D slope using stability chart

Table 6 Review and characteristics of 3D Rock slope

Table 7 Review and characteristics of 3D slope based on computer algorithm

You might also like

Table 1 Review and characteristics of LEM based on F0S and DOS

Table 3 Review and characteristics of LAM based slope

Table 4 Review and characteristics of FEM based slope

Table 5 Review and characteristics of 3D slope using stability chart

Table 6 Review and characteristics of 3D Rock slope

Table 7 Review and characteristics of 3D slope based on computer algorithm