Bayesian Inference of Structural Equation Modelling in Mine Accident and Safety Research – An Approach

Abstract

This paper proposed an approach of Bayesian inference in structural equation modelling (SEM) to evaluate the accident causation in underground coal mines in India. The statistics on accident events and reportable incidents has not shown the corresponding levels of improvement. In the area of major hazards control, the mining industry has emphasized mainly on past experiences and lessons learnt. However, the conventional risk management processes are not able to achieve the goal of zero accident potential (ZAP) due to a tonne of reasons. Bayesian inference SEM is necessary to develop the models and the coefficient of parameter estimation. The Markov Chain Monte Carlo sampling in the form Gibbs sampling was applied for sampling from the posterior distribution. The results revealed that all coefficients of SEM parameters are statistically significant. The Bayesian error statistics reveals that this model provides an approach to reduce accidents in underground coal mines of India.