Utilizing Machine Learning Techniques for Safety Management and Severity of Accidents Forecasting in Construction Projects

Abstract ? Data-driven systems are increasingly being used to address the safety concerns of large-scale construction projects. The detailed investigation of these issues requires converting the recorded information into extensive data domains due to the complexity and heterogeneity of the data. This paper explores applying different machine learning models for safety assessment to predict potential scenarios and develop preventative strategies. This research demonstrates how to manage highly heterogeneous safety data by providing prediction models grounded on empirical data. Machine learning methods are used in this context to estimate the severity of potential accidents. It also outlines different strategies to prevent or mitigate safety failures. The machine learning models developed through this research can contribute to construction industry professionals in predicting future safety concerns. Hence, the study discusses a precautionary measure that should be implemented to ensure the safety of construction in large projects.