Machine learning-based predicted mean vote modeling for mine worker well-being

With the depletion of surface mineral reserves, miners are confronted with ever hotter and typically more humid working conditions in deeper mines, which threatens their well-being, safety, and productivity. This study examines thermal comfort parameters affecting workers in subterranean environments. The approach employs the predicted mean vote model based on the ASHRAE (American Society of Heating, Refrigerating, and Air-Conditioning Engineers) thermal sensation scale and employs the classic machine learning techniques of logistic regression, support vector machine (SVM), k-nearest neighbor (KNN), and decision trees. The decision tree model attained the highest level of accuracy (100%), followed closely by KNN (95.08%), SVM (93.85%), and logistic regression (78.69%). The findings suggest that maintaining thermally optimal conditions is crucial to boost miners’ well-being. This can be achieved by enhancing ventilation, providing appropriate clothing, and ensuring regular rest breaks. These actions are essential for minimizing health risks and enhancing productivity.