Probabilistic analysis of a spatially variable c’-ϕ’ slope

Y.H. Chok, M.B. Jaksa, D.V. Griffths, G.A. Fenton and W.S. Kaggwa

In practice, inherent soil variability is not commonly considered in routine slope stability analysis. This is due mainly to the fact that the effect of soil variability is complex and difficult to quantify. Furthermore, the majority of available slope stability analysis computer programs used in practice, which adopt conventional limit equilibrium methods, are unable to consider this aspect explicitly. To predict the stability of a slope more accurately, especially the marginally stable ones, the effect of soil variability needs to be accounted for. In this paper, an advanced probabilistic analysis method called the random finite element method (RFEM), developed by Griffiths and Fenton in the 1990s, is used to investigate the effect of soil variability on the reliability of a c’-ϕ’ soil slope. The results from the probabilistic study demonstrate that soil variability has a significant effect on the reliability of a slope. It is concluded that the deterministic factor of safety (FOS) is not a reliable measure of the true safety of a slope with spatially variable soils.