A Neural Network Approach for Health-Based Capability Classification of Mine Workers in Hazardous Environments

Mine workers are frequently exposed to hazardous environments, including extreme temperatures, dust, and toxic gases, which significantly impact their health and performance. This system leverages artificial intelligence and machine learning techniques to monitor and evaluate workers’ health in real time. The proposed model utilizes key health and behavioural parameters such as age, Body Mass Index (BMI), oxygen saturation (SpO₂), alcohol consumption, and job satisfaction. A neural network-based classification algorithm processes this data to categorize workers into three capability levels: Best, Moderate, and Limited. These classifications assist supervisors in making informed decisions regarding task allocation, identifying workers who may require medical attention, and ensuring optimal workforce management. The system features a user-friendly interface that enables easy data entry and provides real-time predictions and recommendations, making it accessible to non-technical users. By enabling proactive health monitoring and data-driven decision-making, the system reduces health risks, enhances operational efficiency, and promotes a safer and more sustainable mining environment. Furthermore, its scalable design allows integration with larger datasets and supports future expansion, ensuring long-term applicability and effectiveness.