This paper presents a novel approach to enhancing Safety Management through occurrence reporting in Maintenance, Repair and Overhaul (MRO) environments by developing a complex emergent safety causation model and predictive analytics. Traditional safety models fall short in capturing the non-linear and emergent nature of safety incidents. Through a thorough literature review, key principles and gaps in current safety causation models were identified. This led into the formation of the Cooking Cauldron Model (CCM) in understanding incidents better. Moreover, recognizing the limitations in reactive and proactive approaches, a predictive approach was attempted. By applying Six Sigma DMADOV methodology, this study leverages Artificial Intelligence (AI), Machine Learning (ML) and Natural Language Processing (NLP) techniques to extract predictive trends from historical occurrence reports. The resulting models not only identifies high-risk areas ahead of aircraft layovers but also introduces a paradigm shift toward predictability, laying the groundwork for a future-focused safety culture in aviation.
Grech et al. (Thu,) studied this question.