Machine Learning and Deep Learning in Chemical Health and Safety: A Systematic Review of Techniques and Applications

This comprehensive systematic review categorizes more than 100 papers on ML/DL applications in chemical health and safety, covering supervised and unsupervised learning, deep learning architectures, and their applications to hazardous property prediction, process fault detection, and consequence modeling. For MOC, this review provides an essential reference for understanding which AI techniques are applicable to various aspects of change risk assessment. Process changes may alter chemical interactions, introduce new fault modes, or change consequence scenarios; the ML/DL methods surveyed here—including QSPR models for chemical property prediction and neural networks for fault detection—can be directly applied to evaluate safety implications of proposed modifications in chemical process environments.