A reliability based analysis provides a useful means for modelling uncertainty in the ground by producing results in terms of probability of failure, or a reliability index. This can be more helpful in decision making than a single factor of safety. The present work focuses on the stability assessment of a rock slope, using two common reliability techniques, the Monte Carlo simulation and Point Estimate Method. It was found that Monte Carlo is more suitable and not computationally demanding when variability is small.
The stability of rock slopes is critical in civil engineering design as failure can cause catastrophic consequences, especially for the slope cuts in road construction. Therefore an assessment of the rock slope stability must be considered. There are mainly four primary modes of failure, against which stability must be examined. These include plane and circular failures, wedge failure and toppling failures. The significance of probabilistic analysis in rock slope stability has been recognised due to a number of uncertainties involved in the analysis, for example the shear strength properties or the orientation of rock discontinuities. These analyses have been used in many practical applications, such as slope stabilization schemes (e.g. McGuffey et al. 1980) and landslide hazards (e.g. Cruden 1997).
A further stage was established using advanced computational techniques involving complex reliability analysis in geotechnical engineering. Different methods for reliability analysis in geotechnical engineering can be found in Baecher&Christian (2003). Two of these methods used for reliability analysis are the Monte Carlo simulation and the Point Estimate Method (PEM). Monte Carlo simulation has been used extensively in geotechnical engineering, especially in combination with stochastic finite element modelling (e.g. Fenton et al. 2005, Hicks & Onisiphorou 2005, Onisiphorou 2000).
