Lebanese University

Faculty of Engineering
Department of Civil Engineering

Mass Movement and

slope Improvement
Dalia Abdel Massih, Ph.D.

1
 PART 1
Slope Stability Analysis –
Theoretical Part

2
 CONTENT
1. Introduction – Example of slope failures
2. Shape of failure surfaces
3. Causes of slope failure surface
4. Safety factor definition
5. Infinite Slope Analysis
6. Finite slope stability methods:
6.1. Limit equilibrium methods: Definition
6.1.1. Plane failure surfaces/Culman method
6.1.2. Rotational Failure surface
6.1.3.Slices methods:
6.1.3.1. Fellenius
6.1.3.2. Bishop
6.2. Limit analysis method: Definition
6.2.1. Planar failure surface
6.2.2. Logspiral failure surface 3
 1- Introduction
Example of slope failures

4
Levee failures can also be caused by massive slope failures, such as that
shown here, on the river side of the Marchland Levee in Louisiana, which
occured in 1983. 5
Failure due to an earthquake
6
Failure due to an earthquake
A large part of the village is destroyed 7
The structure is intact. There is no structural problem.
The problem is related to the soil slope instability. 8
Examples in Lebanon
Hammana

Fractured rocks
Block failures
Rock topples

9
 Lebanon !!!

Failures related to roads construction.

Side slopes are not well analyzed to be stable.
Examples: Highway of Metn

10
Mansourieh 1994

Failure occurred during

the spring season. The
whole slope region there
is critical.

11
2- Shape of failure surfaces

12
Rock failure vs slope failure

Rock failure Soil failure

pre-determined failure failure plane along line of
plane max stress

13
Rock failure vs slope failure
Rock failure
failure along pre-determined planes of weakness
Soil failure
failure along lines of max. stress
• frictionless, cohesive = rotational (j=0)
• frictional, incohesive = planar (c=0)

14
Type of soil failure surfaces
Cohesionless Soil c=0
Plane failure surface
Failure surface

q q0 tan d
Logspirale failure surface r  r0e
x O
q0 d Angle formed by the tangent
r0 q to the failure surface and the
rh Failure surface perpendicular to the radius
A r
Passing by the toe This is the angle formed by
the velocity of deformation
I and the tangent to the failure
d
B surface = Dilation angle
d  f (j )  linear function 15
Type of soil failure surfaces
Cohesion and friction Soil c, j soil
Logspirale failure surface
O
Failure surface
Passing by the toe when j is high
and b is high (greater than 53o)

Failure surface
Passing below the toe when j
is low c is high and b is small
(smaller than 53o)

16
Type of soil failure surfaces
Frictionless Soil j=0
Circular failure surface (The logspiral will become
circular because d will become equal to 0)
O
r
r
Failure surface
Passing by the toe when b is high
(greater than 53o)
b

Failure surface
Passing below the toe when b
is small (smaller than 53o)

17
Type of soil failure surfaces
Soil with a weak bedding plane
Planar failure surface along the interface of the
bedding plane

Failure surface
Along the bedding plane

Bedding plane

18
3- Causes of slope failures

19
Causes of slope failures
 Steepness of the Slope (Slope
geometry)
 Water and Drainage (ex: rain infiltration
with no drainage)
 Soil Composition (Nature of the soil)
 Remove of Vegetation (roots work as
reinforcements)
 Bedding Planes
 Joints & Fractures
 Sudden Shocks (ex: earthquakes)
20
4. Safety factor definitions

21
Basics of slope stability evaluation

• Safety Factor FS
• Shape and location of failure is not known
a priori but assumed (trial and error to find
minimum F)
• Static equilibrium (equilibrium of forces
and moments on a sliding mass)

22
Safety factor : 1st definition

Mobilizing Forces a function of:

1. Elevation Difference (Height)
2. Slope Angle
3. Weight of Material(s)
Resistance to Sliding is a function of:
1. Material Properties ( φ and c’)
2. Internal Stress (σ)
3. Pore Pressure Conditions
23
 Safety factor : 2 st definition
tf
FS 
t m or (d )
where
tf = shear strength = c+s.tan(j)
tm ( or d) = mobilized (or developed) shear resistance
obtained from the equilibrium of forces
c  s tan j
FS 
tm
c  s tan j c tan j
tm   s  cm  s tan j m
FS FS FS

Theoretically F = 1: Failure
24
F > 1: Safe
Design Factor of Safety (Practical)
 Minimum FS = 1.25 for highway side slopes
 Use FS = 1.3 to 1.5 for critical slopes such as
end slopes under abutments, slopes
containing footings, major retaining
structures
 Minimum FS = 1.3 for non-permanent slopes
and excavations (provisoires)
 Use FS=1.05 for pseudostatic analysis
(under earthquake conditions)
25
5. Infinite slope analysis

26
 Definition of infinite slope

 Slope that extends for a relatively long

distance and has consistent subsurface
profile (homogeneous) can be considered
as infinite slope

 Failure plane parallel to slope surface

27
Infinite slope analysis in dry Sands
S = Shear resistance force = t.A = c.A+s.A.tanj = c.A+N.tanj

28
Infinite slope analysis in c-j soil with water
S = Shear resistance force = t.A = c.A+s.A.tanj = c.A+N.tanj
b
l
cos b

l
(For the calculation of U see next slide)

29
Infinite slope analysis in c-j soil with water
Pore Water pressure calculation
AB is an equipotential line
hA  hB
uA uB
 zA   zB
w w
uA
 0  0  zB
w
z B  h cos b  cos b
 h cos2 b
Reference line
u A   w h cos2 b
b
l
cos b U A ( force)   w h
b
cos2 b
cos b
U A   w hb cos b
30
6. Finite slope stability
analysis

31
6.1. Limit Equilibrium Method - Basis
 Limit equilibrium of the mass in failure
 Predefined shape of the failure surface
 Assumption on the shape of the normal
stress distribution (s) along the failure
surface :
Different methods have been generated based
on the assumption on the shape of (s). Some
authors have considered (s) uniformly
distributed along the failure surface others
exponential, others sinusoidal…, )
32
6.1.1. Plane failure surfaces / Culman Method
Assumptions: plane failure + s uniformly distributed

33
6.1.1. Plane failure surfaces / Culman Method

34
6.1.1. Plane failure surfaces / Culman Method

35
6.1.1. Plane failure surfaces / Culman Method
Safety factor calculation
FS=tf/td

Minimize FS with respect to q

 Can be done using an optimization toolbox :

Matlab or Excel Solver
36
 6.1.2. Rotational failure surface

Limit Equilibrium:
1) Assume some circular (or other shape) failure surface
2) Calculate driving forces (moment about O)
3) Calculate resisting forces (moment about O)
Issues:
• Where is the center of mass?
• How does resistance vary along surface?
•How does normal stress vary along
surface?
•Water table and seepage forces?
•Soil layering?
•More complex geometry?

37
6.1.2. Rotational failure surface / Total
stress Analysis for saturated clay

38
6.1.2. Rotational failure surface / Total stress
Analysis for saturated clay

39
6.1.2. Rotational failure surface / Total
stress Analysis for saturated clay
Effect of Tension Crack
Reduces the angle of the
sliding sector
Height of tension crack:
For frictionless (j=0) soil
2c
hc 

Cohesive and frictional soil c = cohesive strength (kPa)
2c j  = unit weight of soil (kN/m3)
hc  tan( 45  ) f = friction angle
 2
40
6.1.2. Rotational failure surface / Total stress
Analysis for saturated clay
Effect of Tension Crack

41
6.1.3. Slices Methods
Assumption made on interslices forces instead of
the global normal stress distribution

42
 6.1.3. Slices Methods
• Assume some failure surface
• Discretize failure surface into smaller elements (slices) of width b. (DXi)
• Bottom of each slice passes through one type of material
• Curved bottom of each slice approximated as segment
• More slices = more refined solution
• Around 50 slices typically sufficient (less for hand solutions)
• Calculate factor of safety for each slice (strength/stress) and overall
factor of safety
• Find lowest FS for different failure surfaces

Side forces make the problem statically indeterminate 43

6.1.3. Slices Methods
Geometry and forces acting on one slice (slice
number i)

44
.

6.1.3. Slices Methods

Available equations and unknown
Equations Condition
A 2n force equilibrium in two directions for each slice
B n moment equilibrium for each slice
C n Mohr-Coulomb failure criterion

4n total number of equations

Unknowns Description
D 1 factor of safety
E n normal force at the base of each slice, Pi
F n location of normal forces at the base of slices
G n shear force at the base of each slice, Si
H n-1 interslice horizontal force, Ei
I n-1 interslice vertical force, Ti
L n-1 location of interslice forces (line of thrust) hi

6n-2 total number of unknowns

The difference between equations and unknowns gives the number

of assumptions to render the problem statically determinate:
6n  2  4n  2n  2 45
 6.1.3. Slices Methods
All methods make n assumptions on the locations of Pi.

Concerning the remaining n-2 assumptions, methods

differ greatly.

Usually, n-1 assumptions are made about interslice

forces and an extra-unknown to be determined together
with the factor of safety is introduced. (For the exact
method that have used all the equations of equilibrium)

46
6.1.3.1. Fellenius Method
 Developed by Wolmar Fellenius as a result of
slope failures in sensitive clays in Sweden

 First method of slices to be widely accepted and

used

 Reduces the force resolution of the slope to a

statically determinate structure

 Simplest method of slices; also produces the

lowest factor of safety (most conservative)
47
6.1.3.1. Fellenius Method
All interslice forces are
equal
DEi=0
DTi=0
 2n-2 assumptions
 2n-2 additional
equations DLi
n-2 - (2n-2) = -n
i.e. n eqs are not
satisfied

 SM for all the slices

48
6.1.3.1. Fellenius Method

DLi

Ti = Smobilized

49
6.1.3.2. Bishop Method
In Bishop’ simplified method, n-1 assumptions are made DTi=0
As one more assumption is made than required, one
equilibrium condition cannot be satisfied. Therefore, the
horizontal equilibrium of one slice cannot be satisfied with
the computed safety factor.
n
cDli cosi  Wi tan f  m i
i 1
FBishop  n
Wi sin i
i 1

tan f
mi  cosi  sin i
F 50
51
6.2. Limit Analysis Method – Upper Bound
method

D  c.vt .Area Where vt is the tangential component of the

deformation velocity v with respect to the
failure surface

52
3.2.1. Planar failure surface

Rigid
W

H v
j Rigid
90q

q
1
W  H 2 tan( 90  q )
2
The rate of work done by the gravity force is the
vertical component of the velocity multiplied by the
weight of the soil wedge:
W  H 2 tan( 90  q )  V cosf  90  q 
 1
2 53
3.2.1. Planar failure surface

Where v is the velocity of the rigid block.

The rate of energy dissipated along the discontinuity
surface

vcosj 
 H
Dc
sin q
2ccosj
D  W  H 
 cosq sinq  j 

54
3.2.1. Planar failure surface

The critical height is obtained by minimizing H with

respect to q

 j  4c    j 
q cr   H cr    tan  
4 2    4 2

The safety factor FS is included implicitly in this

equation by dividing c and tanj by FS. Then, FS is
minimized using a minimization toolbox (Matlab or
Excel Solver etc…)
55
3.2.2. Logspiral failure surface

r q   r0 expq  q 0  tan j 

56
3.2.2. Logspiral failure surface

rh  r q h   r0 expq h  q 0  tan f 

H  rh sinq h  r0 sinq 0
L  r0 cosq 0  rh cosq h

 sinq h expq h  q 0  tan f   sinq 0

H
r0

 cosq 0  cosq h expq h  q 0  tan f 

L
r0 57
3.2.2. Logspiral failure surface

58
 3.2.2. Logspiral failure surface

Consider next the triangular region O-A-B . The rate of work

done by the weight of the region is:
1  1 
W 2   Lr0 sinq 0  2r0 cosq 0  L 
2  3 

W 2  r0 f 2 q h ,q 0 
3

1 L L
f 2 q h ,q 0    2 cosq 0   sinq 0
6 r0  r0 

59
 3.2.2. Logspiral failure surface
A similar technique can be used for the triangular
area O-A-C

W3  r0 f 3 q h ,q 0 
 3

f 3 q h ,q 0   cos2 q h exp2q h  q 0  tan f 

1H
3 r0

60
3.2.2. Logspiral failure surface

Work of external forces

W  W1  W2  W3  r0  f1  f 2  f 3 
    3

Rate of internal energy dissipation

qh
rdq cr0 
2
D   cV cosj 
  exp2q h  q0  tan j   1
q0 cosj 2 tan j

61
3.2.2. Logspiral failure surface

Equating the work of external forces to the energy dissipated

along the failure surface gives:

D  W  H  f q h ,q 0 
  c


f q h ,q 0  
exp2q h  q 0  tan f  1 sinq h expq h  q 0  tan f  sinq 0 
2 tan f  f1  f 2  f 3 

62
3.2.2. Logspiral failure surface

Two problems can be solved:

1- Determine the critical slope height beyond which the
slope will fail:

Minimize H to find Hcr

The variables of minimizations are q h ,q 0 

2- For a given slope height determine the safety factor:

FS is included implicitly in the energy equation D  W by
dividing c and tanj by FS

Minimize FS implicitly under the constraint D  W

The variables of minimizations are q h ,q 0 

63