Artificial Intelligence and Machine Learning for Job Hazard Analysis: A Survey

Job Hazard Analysis (JHA), sometimes referred to as Job Safety Analysis (JSA), is a fundamental process for identifying workplace hazards and specifying controls, yet traditional JHA methods are labor-intensive and prone to human oversight. In recent years, Artificial Intelligence (AI) and Machine Learning (ML) techniques have been applied to automate and enhance JHA. This survey provides a state-of-the-art review and research gap analysis of AI- and ML-enabled JHA research, synthesizing insights from ten peer-reviewed studies published over ten years (2015-2025). By capturing both opportunities and limitations, the survey provides a basis on howAI andMLestablishes a foundation for transforming JHA into a proactive, data-driven foundation for workplace safety. Our review reveals that AI- and ML-enabled JHA systems primarily automate one or more of the three core JHA columns: hazard identification, risk evaluation, and control recommendation. Techniques explored include rule-based Natural Language Processing (NLP) for extracting hazards from incident reports, ontology-based reasoning to structure safety knowledge, probabilistic Bayesian networks for risk scoring, large language models (LLMs) for generating contextual control measures, and spatial-temporal synchronization models for managing simultaneous operations.