System Reliability Assessment of Rock Slopes Under Seismic Loading Considering the Spatial Variability of Strength Parameters

A rock slope has multiple potential sliding planes and, thus, multiple failure modes. Currently, most existing studies consider only the reliability of the slope under a single failure plane, ignoring the system reliability under multiple failure planes. Furthermore, the influence of the spatial variability of parameters in the nonlinear Barton–Bandis failure criterion on slope reliability has been ignored. To address this research gap, an improved approach using the nonlinear Barton–Bandis failure criterion is used to analyze rock slope stability. On the basis of the concept of the local mean value from random field theory, a calculation approach to determine the local mean of parameters in the Barton–Bandis strength criterion is established, and Monte Carlo simulation is adopted to estimate the system failure probability of rock slopes. Two rock slopes are taken as examples to confirm the efficiency and accuracy of the proposed method. In addition, for rock slopes with multi-plane instability modes, the system failure probability is greater than that along a single structural plane. Using the failure probability of a single structural plane to evaluate the probability of stability of a slope can lead to an overestimation of its stability.