An integration of intelligent approaches and economic criteria for predictive analytics of occupational accidents

Occupational accidents are a significant concern, resulting in human suffering, economic crises, and social issues. Despite ongoing efforts to comprehend their causes and predict their occurrences, the use of machine learning models in this domain remains limited. This study aims to address this gap by investigating intelligent approaches that incorporate economic criteria to predict occupational accidents. Four machine learning algorithms, Random Forest (RF), Support Vector Machine (SVM), Multivariate Adaptive Regression Spline (MARS), and M5 Tree Model (M5), were employed to predict occupational accidents, considering three economic criteria: basic income (BI), inflation index (II), and price index (PI). The study focuses on identifying the most suitable model for predicting the frequency of occupational accidents (FOA) and determining the economic criteria with the greatest influence. The results reveal that the RF model accurately predicts accidents across all income levels. Additionally, among the economic criteria, II had the most significant impact on accidents. The findings suggest that a reduction in FOA is unlikely in the coming years due to the increasing growth of II and PI, coupled with a slight annual increase in BI. Implementing appropriate countermeasures to enhance workers' economic welfare, particularly for low-income employees, is crucial for reducing occupational accidents. This research underscores the potential of machine learning models in predicting and preventing occupational accidents while highlighting the critical role of economic factors. It contributes valuable insights for scholars, practitioners, and policymakers to develop effective strategies and interventions to improve workplace safety and workers' economic well-being.