An ontological approach to hazard identification for safety-critical systems

Hazard identification is an essential and demanding task for the development of safety-critical systems (SCSs). Current practices suffer from one or several drawbacks: 1) a common hazard conceptualization is missing and thereby ambiguities may arise and, 2) there is still a need to formalize the experience of analysts and lessons learned from previous system development. It should be done in a structured way to facilitate future reuse and, 3) some hazard identification techniques require well-known system behaviors represented by models, such as automata and sequence diagrams, to identify hazards. However, such models are typically susceptible to changes or even not available in early stages of the development process. In this paper, we propose an ontological approach to support hazard identification in the early stages of the development of SCSs. The approach aims to improve the completeness of hazard identification results and to avoid ambiguities. A robotic strolling assistant system is used to evaluate the proposed approach.