Development of a predictive model for workers' involvement in workplace accidents in an underground coal mine

Abstract

Underground mines are dynamic and dangerous. These features of underground coal mines, coupled with the low level of mechanization, have made underground Indian coal mines accident-prone. The mine managers are much stressed about achieving high productivity with safety. It is a fact that human performance is the primary driving force for operating these mines safely, and the work-related factors significantly impact human performance and safety. With the help of demographic data and work-related characteristics, this study seeks to assess the chance of accidents. We achieve this goal using improved work compatibility and a binary logit model. This study employs a step-wise backward elimination technique to develop the logit model with significant work-related factors. When testing the model with available data, we obtained encouraging accuracy. This study employs data envelopment analysis to identify and prioritise work-related factors for the prevention of workplace accidents. Finally, we made some suggestions that may work for enhancing productivity and safety. © 2023, Indian Academy of Sciences.