DEVELOPMENT OF AN ARTIFICIAL INTELLIGENCE-BASED PREDICTIVE RISK ASSESSMENT FRAMEWORK FOR INDUSTRIAL CONSTRUCTION SAFETY USING NEAR-MISS AND INCIDENT DATA

This paper proposes an AI-based predictive risk assessment framework for industrial construction safety using near-miss reports, incident records, safety observations, and project characteristics. The methodology integrates data preprocessing, feature engineering, machine learning, probability estimation, explainable AI, and decision support. The framework aims to support proactive safety management by identifying recurring hazards, estimating future incident likelihood, and prioritising preventive interventions. Its effectiveness will depend on data quality, consistent reporting, appropriate validation, and the interpretation of predictions within established safety-management practices. Future research should implement the proposed framework using real industrial construction datasets, compare the predictive performance of candidate algorithms, and validate the resulting model across independent projects. Further prospective evaluation should determine whether model-informed interventions produce measurable improvements in workplace safety.

1.                Baker, H., Hallowell, M. R., & Tixier, A.-J.-P. (2020a). Automatically learning construction injury precursors from text. Automation in Construction, 118, 103145. https://doi.org/10.1016/j.autcon.2020.103145 2.                Baker, H., Hallowell, M. R., & Tixier, A.-J.-P. (2020b). AI-based prediction of independent construction safety outcomes from universal attributes. Automation in Construction, 118, 103146. https://doi.org/10.1016/j.autcon.2020.103146 3.                Tixier, A.-J.-P., Hallowell, M. R., Rajagopalan, B., & Bowman, D. (2016). Application of machine learning to construction injury prediction. Automation in Construction, 69, 102–114. https://doi.org/10.1016/j.autcon.2016.05.016 4.                Zongzhi Wu, Rujun Wang, 2014. Concern with the safety management of oil and gas pipelines--Status. Chinese Safety News 6, p. 1.N. Craciun, The Cost Assessment for Protection and Prevention of Activities Carried at Height, (2015) 5.                Wu ZongzhiˈZhang Shengzhu, Yu Lijinˈ2014. A method for long-distance oil and gas transmission pipeline route selection based on loss of life. China Safety Science Journal 24, p. 71.A. Targoutzidis, E. Koukoulaki, K. Schmitz-Felten, M. Kuhl, O. Karen, E. Hengel, K. Rijken, R. Van den Broek, R. Klüser, The business case for safety and health at work: (2014). 6.                Young, Do Jo, Bum, Jong Ahn, 2005. A method of quantitative risk assessment for transmission pipeline carrying natural gas, Journal of Hazardous Materials 123, p. 1S. Sullivan, J Occup Environ Med.,46(6 suppl), S56, (2004) 7.                Han, Z.Y, Weng, W. G, 2011. Comparison study on qualitative and quantitative risk assessment methods for urban natural gas pipeline network, Journal of Hazardous Materials 18, p. 509. 8.                Roberts, T., Buckland, I., Beckett, H., 2001. Consequences of jet-fire interaction with vessels containing pressurized, reactive chemicals, Hazardous: Analyzing the past, planning the future 148, p. 147.A. Burdorf, Scand J Work Environ Health., 33(3), 161, (2007) 9.                P. Lanoie, S. Tavenas, Saf Sci., 24(3), 181, (1996) 10.             Gomez-Mares, M., Munoz, M., Casal, J., 2009. Axial temperature distribution in vertical jet fires, Journal of Hazardous Materials 172, p. 54.T. Mahboob, B. Tariq, S. Anwar S., M. Khanum, Int. J of Comp. App., 121, 20 (2015)