A Multi-Criteria Decision Intelligence Framework to Predict Fire Danger Ratings in Underground Engineering Structures

Abstract

A wide variety of natural catastrophes are induced by coal mining, with fire hazard being one of the most significant threats to underground engineering structures. In recent years, there has been an alarming rise in mine fire accidents due to the abundance of coal deposits around the world. Underground fires and explosions have continuously been the primary reason for a significant proportion of deaths and the destruction of infrastructure over the last few decades. Underground mining fires deplete natural coal resources, have an adverse impact on the environment by releasing hazardous chemicals and greenhouse gases into the atmosphere, and cause subsidence due to coal depletion during the combustion process. This study aims to predict fire danger rating of underground mining production processes by using the application of state-of-the-art unsupervised and supervised machine learning techniques. The developed k-nearest-neighbors-based isometric feature mapping and fuzzy c-means clustering algorithm has shown its dependability and superiority with a higher accuracy and has been advantageous to the monitoring and prevention of fire danger in underground mining production processes. The proposed multi-criteria decision intelligence framework permits early fire detection, providing the emergency response team extra time to respond the critical situations in order to prevent the fire from spreading, hence promoting sustainable, green, climate-smart, environmentally friendly and safe mining engineering operations.