Professional Documents
Culture Documents
stability analyses
B. Nilsen
Abstract The paper discusses alternative ap- Mots cls Stabilit de pente rocheuse 7 Analyse
proaches to stability analyses including the tradi- dterministe 7 Facteurs partiels 7 Analyse
tional deterministic principle, the so-called partial probabiliste 7 Facteur de scurit 7 Paramtres de
factor principle and the probabilistic approach calcul
which assesses the probability of failure rather than
the factor of safety. To illustrate the different prin-
ciples and approaches, stability analyses of road
cuts near Trondheim, central Norway, are used as
examples. It is concluded that although the tradi- Introduction
tional deterministic approach has the advantage of
being well established and easy to understand, to Traditionally, in the fields of geological engineering and
conform with new standards and guidelines it is rock mechanics, the deterministic principle of calculating
likely to be replaced by the partial factor principle. the stabilizing and driving forces to arrive at a factor of
To obtain the best possible basis for evaluation it is safety has been the predominant method of rock slope
useful to include a probabilistic analysis. stability analysis. In the last few years, however, alternative
methods have become more widespread. To a great extent,
Rsum Larticle prsente de nouvelles approches this is due to the introduction of new standards and
pour les analyses de stabilit comprenant lanalyse recommendations, such as Eurocode 7 (Comit Europen
dterministe traditionelle, la mthode dite des coef- de Normalisation 1994) and NS 3480 Geotechnical Plan-
ficients partiels et lapproche probabiliste qui ning (Norwegian Council for Building Standardization
value une probabilit de rupture plutt quun 1988) which require all stability analyses to be carried out
facteur de scurit. Afin dillustrer les diffrentes according to the so-called partial factor method, with
mthodes et approches, des analyses de stabilit partial factors for action and strength. In soil mechanics
relatives des dblais routiers prs de Trondheim, this principle was commonly adopted many years ago,
au centre de la Norvge, sont prsentes titre while in geological engineering/rock mechanics the calcu-
dexemples. Bien que lanalyse dterministe tradi- lation principles have been changed only to a minor
tionnelle ait lavantage dtre bien admise et facile extent.
comprendre, on conclut quelle sera probablement In addition, the probabilistic approach, with the calcula-
remplace par la mthode des coefficients partiels tion of the probability of failure instead of a factor of safety
afin de suivre les nouvelles rgles et recommanda- against failure, has become more common practice, as
tions. Pour obtenir la meilleure base possible de reflected by the many international conferences over the
diagnostic il est utile de considrer galement une last few years focusing on this issue (Li and Lo 1993;
approche probabiliste. Shackelford et al. 1996; Lee and Lee 1998).
The steps of the analysis that precede calculation, i.e. defi-
Key words Rock slope stability 7 Deterministic nition of potential stability problem and quantification of
analysis 7 Partial factors 7 Probabilistic analysis 7 input parameters, are crucial for the final result. The main
Factor of safety 7 Input parameters focus of this paper, however, is the advantages and disad-
vantages of the alternative methods, in particular the inter-
pretation of calculation results.
Table 1
Factor of safety calculation (figures for the probabilistic approach are mean values after truncation)
while for maximum water pressure but no earthquake it is Table 2 presents the results of the stability analysis using
1.26. the partial factor method for the same basic cases as for
the factor of safety approach. As can be seen, according to
this concept the stabilizing forces are greater than the
driving forces for all situations except for the worst case.
The partial factor method
Probabilistic analysis
According to the partial factor method, partial factors for For the road-cut example, the parameters related to geo-
actions and materials are to be applied instead of an metry and the unit weights are unambiguously defined and
overall safety factor. In principle, the calculation is carriedin the analysis represent constants. Other input parame-
out as follows: ters, i.e. water pressure, seismic acceleration and friction
angle, may, however, vary within wide limits. Thus, a prob-
FdpFk7gf
abilistic approach, with probability distributions assigned
MdpMk/gm to those parameters, has obvious advantages.
The following probabilistic analyses are based on the
where Fd is dimensioning action and Md is dimensioning
computer programs BestFit and 6RISK developed by the
strength; Fk is characteristic action and Mk is characteristic
Palisade Corporation (1996, 1997) and inspired by analyses
strength; gf is partial factor for action and gm is material
described by Hoek (1998). In the probabilistic analysis, the
factor. The design is considered to be satisfactory if:
considerations and quantifications of the respective varia-
Md 1 Fd bles are as follows:
In terms of slope stability, this means: if stabilizing forces
Water pressure (U)
are greater than driving forces.
The build-up of water pressure is assumed to be according
Guidelines for defining partial factors are described by the
to the triangular distribution described earlier. However,
Comit Europen de Normalisation (1994) and Norwegian
during the lifetime of the slope, it is unlikely that the water
Council for Building Standardization (1997). For the
pressure will equate to the maximum value given by the
example:
equation in the figure. The most common situation will be
W, U: gf p1.0 a practically dry slope, i.e. UF0. The most realistic proba-
Fa: gf p1.3 bilistic model of water pressure is believed to be a trun-
tan wa: gmp1.2 cated exponential function, with truncation represented by
Table 2
Stabilizing and driving forces for the road-cut example based on the partial factor method
0.0018 0.16
0.0016 a Water pressure (kN/m)
0.0014 0.14
0.0012 0.12
0.0010
Probability
0.10
0.0008
0.0006 0.08
0.0004
0.06
0.0002
0 0.04
0 1000 2000 3000 4000 5000
0.02
35
2
b Seismic acceleration (m/s ) 0
30 70 78 86 94 102 110 118 126 134 142
2
25 Normal stress (kN/m )
20
Fig. 3
15 Probability distribution of sn as calculated by 6RISK based on
10 2000 iterations
5
0
0 0.05 0.10 0.15 0.20 0.25
tion of wa with a standard deviation of 57 within the actual
0.08 normal stress range (between the worst case and best
0.07 c Active friction angle (deg) case in Table 1; see also Fig. 2c). Truncation is taken to be
0.06 to the likely highest and lowest realistic wa values 76 and
0.05 467 respectively.
0.04 Generally, most probabilistic analyses are based on
0.03 mutually independent input parameters. In this case,
0.02 however, two of the input parameters, wa and sn, are
0.01
distinctly interrelated and, as a consequence, a two-step
calculation procedure has to be followed. Step 1 of this
0
0 10 20 30 40 50 60 70 80 procedure is the calculation and definition of the sn distri-
Fig. 2
bution. The calculation is carried out by the program
Assumed probability distributions of a water pressure (U), b 6RISK (Palisade Corporation 1996). The result, based on
seismic acceleration (aa) and c active friction angle (wa) Latin Hypercube sampling (a technique giving comparable
results to the Monte Carlo technique but with fewer
samples) and 2000 iterations, is shown in Fig. 3. According
to BestFit, it can be described as a beta distribution with
the Umax value defined earlier and the mean value (prob- parameters a1p3.12/a2p1.0. Step 2 is an 6RISK calcula-
ably somewhat conservatively) defined as Umax/3 (see tion of the factor of safety (F). Here, wa and sn are treated
Fig. 2a). as interdependent variables with a dependency coefficient
of 0.9 (representing a negative correlation during
Seismic acceleration (a) sampling, i.e. high values of wa are selected for low values
Most realistically, earthquake activity is modelled by an of sn and vice versa). The result, based on Latin Hypercube
exponential distribution indicating that large earthquakes sampling and 2000 iterations, is shown in Fig. 4.
are very rare while small ones are common (Hoek 1998). In According to the probabilistic approach, as illustrated by
the present example, the maximum seismic acceleration Fig. 4, the probability of failure of the actual slope is:
during the lifetime of the slope is assumed to be
P (failure)pP (F~1.0)p0.015
amaxp0.1 g and the mean value amax/3. The resulting
exponential distribution of a is shown in Fig. 2b; in the This indicates that during the lifetime of the slope and for
calculation the distribution is truncated according to the assumed combinations of water pressure, seismic
amaxp0.1 g. acceleration and friction, the probability of failure is 1.5%.
Alternatively, the result may be interpreted as indicating
Active friction angle (wa) that for the specified conditions, two out of a hundred
The input parameter of active friction angle is generally slopes could be expected to fail.
associated with a considerable amount of uncertainty and, As can be seen from Fig. 4, the probability distribution
as discussed earlier, it is also a function of the normal closely resembles a normal distribution. The mean value of
stress. For the present example, based on the rock mass F is 1.75 and the standard deviation is 0.43 (for a standard
data and calculated normal stresses (Table 1), a mean wa distribution indicating that 68% of the 2000 calculated F-
value of 617 is assumed and a truncated normal distribu- values are between 1.32 and 2.18).
0.25
3480, are intended mainly for the design of buildings and
0.20 civil engineering works and make no clear distinction
0.15 between different categories of structures or short- and
long-term stability.
0.10
As a result of the basic differences, the partial factor
0.05 method normally tends to give a more conservative design
0 than when the approach uses the deterministic factor of
1.81
3.91
safety. As shown by the example calculations of a road cut
1.29
1.55
2.08
2.34
2.60
2.86
3.13
3.39
3.65
0.51
0.76
1.03
the four cases when the partial factor method is used (see
0.8 Tables 1 and 2).
0.6 It is important to be aware that, due to the uncertainty of
the input parameters, even Fp1.0 (the calculated stabil-
0.4 izing forces much greater than the destabilizing forces)
0.2 does not necessarily mean that the probability of sliding is
equal to zero. Hence, if not fully understood, the determin-
0 istic approach may sometimes give a false impression of
2.71
2.22
2.95
3.69
0.51
1.24
1.48
1.73
1.97
2.46
3.20
3.44
3.93
0.75
0.99
safety.
Factor of safety, F The final conclusion, i.e. the decision whether the calcu-
lated values represent a satisfactory level of safety or not, is
Fig. 4 not simple to make. In the case discussed in this paper, as
a Probability density function and b cumulative distribution
function of the factor of safety (F) for the road-cut example
earthquake activity is rare and simultaneous heavy rainfall
and maximum earthquake conditions are very unlikely, it
is realistic to assume the stability is satisfactory.
Probabilistic approach
To interpret the result of the probabilistic approach, i.e. to
Interpretation of results determine what probability of failure can be accepted, is
often difficult. Guidelines do, however, exist. For example,
The various calculation methods all give very distinct the Norwegian national guidelines for buildings and civil
results. However, there is a considerable degree of uncer- engineering works in potential slide areas are shown in
tainty connected with the interpretation of the calculated Table 3. According to the guidelines, a structure is consid-
values, i.e. what is the consequence of the calculated value ered to be safe when the probability of slide is lower than
in terms of stability and what value is required to be on the the respective limit given in the table.
safe side? For the road-cut example, with a P (F~1.0)p0.015 and an
assumed structure lifetime of 50 years, the annual proba-
Deterministic approach bility of slide would be 3!10 4. According to the criteria
According to its definition, a safety factor of F 1 1.0 means in Table 3, this would satisfy the stability requirement of a
stabilizing forces are greater than sliding forces and hence structure in safety class 2, including roads of the category
the slope should be stable. As there is always some degree discussed here.
of uncertainty connected to the input parameters, however,
this may not necessarily be the case. To take the uncer-
tainty into account and in order to allow for the different
stability requirements of different types of structures, the Table 3
following criteria for stability are often used: short-term Criteria regarding localisation of structures in potential slide
stability (e.g. temporary slopes in an open pit mine), areas according to Statens Bygningstekniske Etat (1995)
F61.3; long-term stability (e.g. permanent mine slopes or
road cuts), F61.5 Safety Consequence Max. annual
The factor of safety concept is easy to understand and has class of slide probability of slide
the advantage of having been standard procedure for a 1 Minor 10 2
long time. In comparison, the partial factor method may 2 Medium 10 3
be argued to give better control of the calculation by 3 Major ~10 4
assigning partial factors to actions and materials. The
Conclusions References
As illustrated by the calculation examples in this paper, Barton N, Bandis S (1990) Review of predictive capabilities of
rock slope stability analysis and particularly interpretation JRC-JCS model in engineering practice. Proceedings of Interna-
of the results is difficult even for simple slope geometry. In tional Conference on Rock Joints, Loen, Norway. AA Balkema,
Rotterdam, pp 603610
addition, different calculation principles give considerably Comit Europen de Normalisation (1994) Eurocode 7:
different results. To conform with new standards and geotechnical design part 1: general rules. European Prestan-
recommendations such as Eurocode 7, traditional deter- dard ENV 19971. Comit Europen de Normalisation, Brus-
ministic analysis based on the calculation of one factor of sels, 123 pp
safety is to be replaced by the partial factor method, Dahle A, Bungum H, Alsaker A (1992) Earthquake hazard at
involving partial factors on actions and material strength. strait crossings. In: Norwegian subsea tunnelling, 8. Norwegian
The principle of considering individual factors obviously Soil and Rock Engineering Association, Tapir, Trondheim,
pp 4346
has certain advantages. However, the available standards Hoek E (1998) Rock engineering. Chapter 8 factor of safety and
and recommendations concerning interpretation of the probability of failure. Course notes, Internet edition, http:/
results are at present largely limited to buildings and civil /www.rockeng.utoronto.ca/hoekcorner.htm, pp 105114
engineering projects. Therefore, in fields such as mining Hoek E, Bray JW (1991) Rock slope engineering, 3rd edn. Insti-
and dam foundation, for instance, it is likely that for some tute of Mineralogy and Metallurgy, London, 358 pp
time to come the traditional factor of safety method will Lee I-M, Lee M-J (1998) Optimization of rock slopes using relia-
continue to be the most commonly used. bility concepts. Proceedings NARMS 98, Cancun, Mexico. Int J
Rock Mech Min Sci 35 : 661
Due to the uncertainty of input parameters, including their Li KS, Lo S-CR (eds) (1993) Probabilistic methods in geotechnical
variability, the probabilistic approach for stability analysis engineering. Proceedings of International Conference,
has obvious advantages. In addition, safety criteria are Canberra. AA Balkema, Rotterdam, 333 pp
sometimes based on given probability limits. In important Nilsen B (1985) Shear strength of rock joints at low normal
rock slope stability analyses, the option to include the stresses a key parameter for evaluating rock slope stability.
probabilistic approach as a supplement to more routine Proceedings of International Symposium on Fundamentals of
deterministic analyses should always be considered. Rock Joints, Centek, pp 487494
Norwegian Council for Building Standardization (1988)
It is hoped to further investigate the applicability of proba- Norwegian Standard 3480 geotechnical planning (in Norwe-
bilistic methods and particularly the interpretation of gian). Norwegian Council for Building Standardization, Oslo,
calculation results during a research project to be initiated 11 pp
at the Norwegian University of Science and Technology. Norwegian Council for Building Standardization (1997)
National Application Document (NAD) for NS-ENV
Acknowledgements Professor Ugur Ozbay of the Colorado 19971 : 1997 Eurocode 7: geotechnical design. Norwegian
School of Mines, Mining Engineering Department, reviewed a Council for Building Standardization, Oslo, 14 pp
draft of this paper. His comments and suggestions are gratefully Palisade Corporation (1996) 6RISK users guide. Palisade
acknowledged. Corporation, New York, 306 pp
Palisade Corporation (1997) BestFit users guide. Palisade
Corporation, New York, 133 pp
Shackelford CD, Nelson PP, Roth MJ (eds) (1996) Uncer-
tainty in the geologic environment: from theory to practice.
Proceedings of American Society of Civil Engineers Confer-
ence, Uncertainty 96, Madison. American Society of Civil Engi-
neers, 12, 1453 pp
Statens Bygningstekniske Etat (1995) Technical regulations
related to the Planning and Building Law of June 14, 1985
chapter VII: personal and material safety (in Norwegian).
Statens Bygningstekniske Etat, Oslo, 12 pp