TRANSFORMING RISK MANAGEMENT STRATEGIES IN OCCUPATIONAL HEALTH AND SAFETY MANAGEMENT SYSTEMS: FROM REACTIVE CONTROL TO PROACTIVE AI PREDICTION
DOI:
https://doi.org/10.32782/3041-2080/2026-6-28Keywords:
occupational safety management system, proactive strategy, situational awareness, artificial intelligence, risk assessmentAbstract
The article provides a systematic analysis of the evolution of risk management strategies in modern occupational health and safety management systems (hereinafter referred to as OHSMS). It highlights the pressing issue of stagnation in occupational injury rates caused by the exhaustion of traditional reactive approaches based on retrospective analysis of events and linear domino theory. It examines the methodological limitations of existing performance metrics, in particular “lagging indicators,” which are often unable to reflect complex nonlinear causeand- effect relationships in dynamic socio-technical systems. Based on a critical analysis of recent scientific publications (2016–2025), the urgent need to transition to a proactive risk management strategy (Safety-II) focused on ensuring system resilience and adaptability is justified. It has been proven that in conditions of increasing complexity of production processes and the influence of destabilizing factors of systemic crises (chronic stress, accumulated fatigue, changing environmental conditions, unhealthy psychological microclimate in the team of employees), the level of situational awareness of OHSMS specialists becomes a critical safety factor. A novel conceptual approach to the use of predictive artificial intelligence models, in particular ensemble machine learning methods (Random Forest), for the integration and identification of the hierarchy of influence of heterogeneous organizational and psychological risk factors is proposed. The architecture of “Human – Artificial Intelligence” is considered according to the “predictor-corrector” principle, where algorithms provide accurate risk prediction (level of situational awareness projection), and a specialist performs verification and makes management decisions. This synergistic approach allows for the mitigation of operators' cognitive limitations, overcoming the problem of information overload, and significantly improving the effectiveness of incident prevention at the production level.
References
Бочковський А.П., Сапожнікова Н.Ю. Теорія і практика оцінювання ризиків виникнення професійних небезпек. Зернові продукти і комбікорми. 2018. Т. 18 (№ 2). С. 4–10.
Haas E. J., Yorio P. L. Exploring the state of health and safety management system performance measurement in mining organizations. Safety Science. 2016. Vol. 83. P. 48–58. DOI: https://doi.org/10.1016/j.ssci.2015.11.009.
Milošević I., Stojanović A., Nikolić Đ., Mihajlović I., Brkić A., Perišić M., Spasojević-Brkić V. Occupational health and safety performance in a changing mining environment: Identification of critical factors. Safety Science. 2025. Vol. 184. Art. 106745. DOI: https://doi.org/10.1016/j.ssci.2024.106745.
Rao E. V. R., Pandey A. Industrial Accident Analysis and Predictive Models for Workplace Hazard Prevention. Disaster Advances. 2025. Vol. 18 (6). P. 58–64. DOI: https://doi.org/10.25303/186da058064.
Slater D. Human-AI Teaming based on Collaborative Predictor-Corrector Functions. ResearchGate. 2025. DOI: https://doi.org/10.13140/RG.2.2.11348.46720.
Salmon P. M., Cornelissen M., Trotter M. J. Systems-based accident analysis methods: A comparison of Accimap, HFACS, and STAMP. Safety Science. 2012. Vol. 50 (4). P. 1158–1170. DOI: https://doi.org/10.1016/j.ssci.2011.11.009.
Кружилко О. Є., Чеберячко С.І., Володченкова Н.В., Репін М.В., Майстренко В.В. Методика побудови математичної моделі прогнозування ризиків для стратегічного планування заходів із безпеки праці. Науковий Журнал Метінвест Політехніки. Серія: Технічні науки. 2025. № 5. С. 316–322. DOI: https://doi.org/10.32782/3041-2080/2025-5-36.
Laitinen H., Vuorinen M., Simola A., Yrjänheikki E. Observation-based proactive OHS outcome indicators – Validity of the Elmeri+ method. Safety Science. 2013. Vol. 54. P. 69–79. DOI: https://doi.org/10.1016/j.ssci.2012.11.005.
Grant E., Salmon P. M., Stevens N. J., Goode N., Read G. J. M. Back to the future: What do accident causation models tell us about accident prediction? Safety Science. 2018. Vol. 104. P. 99–109. DOI: https://doi.org/10.1016/j.ssci.2017.12.018.
Delikhoon M., Zarei E., Valdez Banda O., Faridan M., Habibi E. System thinking-based accident analysis models: A systematic review for sustainable safety management. Sustainability. 2022. Vol. 14, no. 10. Art. 5869. DOI: https://doi.org/10.3390/su14105869.
