Inteligencia artificial en el análisis predictivo de accidentes laborales como estrategia para fortalecer el Sistema de Gestión de Seguridad y Salud en el Trabajo: una revisión de alcance

Abstract

Occupational Health and Safety (OHS) management has traditionally focused on reactive actions, which limits the ability to anticipate workplace accidents. In this context, artificial intelligence emerges as an alternative that can strengthen prevention through the use of predictive analysis models. Accordingly, the purpose of this study was to analyze the use of these tools in predicting workplace accidents as a strategy to support and improve the Occupational Health and Safety Management System, based on a review of scientific evidence published between 2019 and 2025. A scoping review was conducted through a systematic search of information in international academic databases such as Scopus and ScienceDirect, complemented by academic search engines like Google Scholar and university institutional repositories. The search included research articles, review papers, technical documents, and relevant regulations in both Spanish and English. The collected information was organized according to application sectors, techniques used, and reported results. The findings show an increase in scientific production on the use of artificial intelligence in OHS, with a predominance of machine learning techniques such as Random Forest and Support Vector Machines. These tools contribute to risk anticipation and decision-making processes; however, they also present limitations related to data quality, technological requirements, and ethical considerations. Overall, their application can support the strengthening of OHS management systems, provided that appropriate technical and organizational conditions are in place.