Enhancing Slope Stability Assessment Under Seismic Conditions: A Comparative Analysis of Monte Carlo Simulation, Subset Simulation, and a Second-order Reliability Method for Probabilistic Analysis

Probabilistic approaches are widely used to address uncertainties in geotechnical parameters, particularly in studies related to slope stability. In this work, Monte Carlo simulation (MCS), subset simulation (SS), and the second-order reliability method (SORM) are employed via UPSS ADD-INs 3.0 within an Excel environment to perform a probabilistic assessment of a slope system. The analysis is carried out on a conceptual slope model with a height of 11.693 m and subjected to seismic loading conditions represented by horizontal seismic coefficients of kh = 0.12 and kh = 0.14. Different pore water pressure ratios, namely, ru = 0.0 and ru = 0.05, are considered on the basis of information reported in previous studies. To represent soil variability, lognormal random fields are adopted, and uncertainties in the soil parameters are modeled via Cholesky matrix decomposition. The reliability index and probability of failure ( $$\:{p}_{f}$$ ) are evaluated via the MCS, SORM, and SS to examine the influence of parameter uncertainty. The results indicate that subset simulation demonstrates better efficiency and reliability, particularly in estimating low probabilities of failure, than the MCS and SORM methods do.