Near-miss management systems and observability-in-depth: Handling safety incidents and accident precursors in light of safety principles

Abstract

Accident precursors and near-miss management systems (NMS) are important safety tools in industries with major accident hazards, such as the airline industry, the nuclear industry, and the chemical industry, and they are increasingly adopted in other sectors such as manufacturing, construction, and health care. The aim of a NMS is to “harvest value” from near-miss data by assessing and prioritizing their risk implications, identifying their failure generating mechanisms, and guiding interventions and safety improvements and awareness. Recognizing that learning from near-misses is less costly than learning from accidents, the main value of a NMS is in the learning loop it provides within and across organizations, in focusing safety resources on addressing unsafe acts, reducing unsafe conditions and procedures, and improving design and operational safety issues. The present study first provides an updated review and synthesis of key ideas and challenges of NMS. It then proposes and examines important synergies between fundamental safety principles adopted in risk management, including defense- and observability- in depth, and NMS. Safety principles offer a new lens by which to view NMS. One important result is that near-miss data can be classified and interpreted in light of safety principles violated, and that safety interventions can be particularly effective when organized around such findings, the objectives being to (re-)establish and strengthen compliance with safety principles through workforce training, system redesign, and/or improved operational procedures. Finally, it is argued that NMS is one of the pillars of the implementation of observability-in-depth, and that the boundaries with the two other pillars (fault detection/online monitoring, and inspection) are likely to be blurred in the future, and that the next generation NMS (2.0) will likely integrate data from multiple sources to improve the efficacy of precursor identification, prioritization, and safety interventions, and ultimately accident prevention.