To effectively tackle system reliability problems of rock slopes involving multiple correlated potential failure modes at small probability levels

this paper aims to propose a subset simulation-based approach for system reliability analysis of rock slopes. The expressions for system failure probability calculation of typical slope systems are derived. To account for the correlations among multiple failure modes

the “max” and “min” functions are utilized to construct the system performance functions. A benchmark rock slope with two sliding blocks and the left abutment slope of Jinping I slope with multiple correlated failure modes are investigated to demonstrate the effectiveness of the proposed approach. The results indicate that the proposed approach is much more efficient than the Monte-Carlo simulation (MCS) and more accurate than the first order approximation

N-dimensional equivalent method and Ditlevsen’s bounds method when solving the system reliability problems of rock slopes with small probabilities. In summary

it provides an effective means for tackling system reliability problems of complex rock slopes at small probability levels. In addition

the correlations among different failure modes can be effectively quantified with the aid of the “max” and “min” functions

which greatly facilitates the system reliability analysis of rock slopes.