Online Safety Risk Management for Underground Mining and Construction Based on IoT and Bayesian Networks

Management of environmental hazards in underground mining and construction sites has always been a challenging task for project managers and site engineers. Poor ventilation, the production of hazardous gases, dust, and considerable heat and humidity are some of the inherent characteristics of these ecosystems. Most conventional underground risk management methods are static and overly simplistic, making it almost impossible to predict and control these complex hazards. This paper aims to develop an online safety risk management system for underground mining and construction environments that enables dynamic and remote monitoring, analysis, and control of safety risks in underground space. The proposed system benefits from an automated combination of Internet of Things (IoT) wireless sensors as an environmental perception layer and Bayesian networks (BNs) as a powerful risk modelling engine. Using an open-source dataset collected in a real underground coal mine, a proof-of-concept example is presented to demonstrate the applicability of the proposed system. The proposed system will enable real-time and remote monitoring of underground ecosystems and enhance worker safety upon implementation.