Probabilistic Methods for Stability Analysis of Rock Slopes: A Review

ABSTRACT: Stability analysis of rock slopes requires numerous reliable input parameters such as rock mass strength and joint conditions such as infillings, thickness, roughness, water pressure, dip, and dip direction. Generally, deterministic, and probabilistic methods are used to analyze the stability of rock slopes. The deterministic analysis utilizes the mean values of each input parameter for calculating the safety factor, and subsequently, the natural variability and uncertainty of these parameters are neglected. Probabilistic analysis, however, is based on calculating the probability of failure instead of the safety factor which is a more realistic and accurate approach for calculating the risk. Owing to the natural variability of input parameters, probabilistic methods have been proven to provide superior reliability estimates compared to deterministic methods. Several reliability analysis methods of rock slope stability have been presented for evaluating the slope reliability index. These include Monte Carlo simulation, Point Estimate Method, Response Surface Method, First- and Second-Order Reliability Methods, and the First-Order Second-Moment method. This paper presents an overview of the main probabilistic reliability-based methods available for the stability analysis of rock slopes and recommended allowable limits for practical applications. The review includes a systematic summary of each approach, formulation, and challenges associated with each method.