Probabilistic stability analyses of multi-stage soil slopes by bivariate random fields and finite element methods

This study aimed to investigate the stability of multi-stage drained and undrained slopes considering the spatial variability in soil properties. A bivariate random field was adopted to characterize the spatial randomness in soil strength parameters of each stage of soil slopes. The random field was then incorporated with the finite element method to obtain the factor of safety (FS) and evaluate the slope failure mechanism. Monte Carlo simulations were conducted for slopes with random soils, resulting in histograms of the FS. The results indicated that around 65%-75% cases of a random slope under the drained condition had an FS being less than that of a uniform slope, and this comparison value reached about 92.6% under the undrained condition. As such, the deterministic analysis of a multi-stage slope may result in an overestimated value of the FS. Moreover, the influence of cross-correlation between cohesive strength and friction angle on the stability of drained slopes was also investigated. The findings demonstrated that the failure probability of multi-stage slope increases as the correlation coefficient increases.