Probabilistic prediction of rockburst hazard using Monte Carlo simulation and MAIRCA approach

Abstract

The prediction of rockburst hazard is of great significance for the safe exploitation of deep mineral resources. To predict the probability of rockburst hazard reliably, a methodology that integrated the Monte Carlo simulation (MCS) and Multi-Atributive Ideal-Real Comparative Analysis (MAIRCA) approach was proposed in this paper. First, considering the heterogeneity and anisotropy of rock mass, the uniform, normal and triangular distributions were adopted to describe initial indicator information by introducing an uncertainty coefficient. Then, the MCS was used to randomly generate indicator values based on the probability distributions. Subsequently, the maximum deviation method was used to calculate the indicator weights, which can avoid the influence of personal subjectivity. After that, the MAIRCA approach was adopted to determine the rockburst hazard level of each sample, and the probability of rockburst hazard was obtained according to the law of large numbers. Finally, the proposed methodology was applied to predict the rockburst hazard in the Sanshandao gold mine, Laizhou city, Shandong Province, China. In addition, the effectiveness was demonstrated through sensitivity and comparison analyses. Results indicate that the rockburst hazard level is consistent with field conditions, and the proposed methodology is reliable for the probabilistic prediction of rockburst hazard.