PROBABILISTIC SENSITIVITY ANALYSIS ILLUSTRATED FOR AN UNDERGROUND ROCK EXCAVATION

This paper presents an efficient way of approximating probabilistic sensitivity factors involving correlated random variables, and provides an illustration for the case of an underground rock excavation. Stress-controlled instability mechanisms are investigated, in which the jointed rock mass is regarded as an equivalent continuum. The proposed sensitivity factors are based on the Hasofer-Lind reliability index and the first-order reliability method (FORM), and can be extended to compute reliability sensitivities with respect to the mean value or standard deviation. Numerical differentiation and Monte Carlo simulations are carried out for verification. Comparisons between conventional deterministic sensitivity study and stochastic sensitivity analysis are made and discussed.