Influence of Probability Distribution of Shear Strength Parameters on Reliability-Based Rock Slope Analysis

This paper presents a study on the influence of the probability distribution curve of shear strength parameters on the reliability analysis of a rock slope. The findings of this investigation is applied a rock slope in Luoyang, China. The probability distribution of shear strength parameters (c and φ) may not follow the normal distribution assumed in most reliability-based analyses. There is a negative correlation between c and φ, which will affect significantly the results of the reliability analyses. In order to study the influence of a probability distribution curve, four different types of distribution curves are used in this study. It is concluded that the failure probability of a rock slope is sensitive to the probability distribution of the shear strength parameters. Based on this study, the failure probability can vary about ten times between two extreme cases where two different distribution curves are used. It is also found that the failure probability increases while the correlation coefficient between c and φ decreases. A sensitivity analysis of the probability distribution curve is suggested for future reliability analyses.