Validation of system safety hazard analysis in safety-critical industries: An interview study with industry practitioners

While many hazard analysis techniques exist, little empirical research has been dedicated to their use in in-dustrial contexts, in particular concerning how practitioners validate hazard analyses. This raises questions about the accuracy, comprehensiveness, and credibility of safety analyses, and how practitioners consider this issue in relation to the overall system safety work. Acquiring qualitative evidence regarding the validation of hazard analysis among practitioners is important to support evidence-based safety practices. This paper qualitatively investigates the state of practice in hazard analysis and its validation for system safety among practitioners. Twenty semi-structured interviews were conducted with practitioners in safety-critical industries in North America. Feedback from practitioners indicates that only a limited number of hazard analysis methods are applied in industry, which are mainly based upon linear accident theory. It is also found that almost all prac-titioners perform some form of validation as they believe this type of safety work improves safety. Experts Re-views and benchmark exercises are the only methods reported for validating hazard analysis. In addition, practitioners highlighted several weaknesses of the current hazard analysis and hazard analysis validation practices, of which subjectivity is seen as the most important one. The authors discuss this in context of the emerging academic consensus that hazard analysis is inherently subjective, but that it can nevertheless be very useful especially when it relies on strong evidence. Also, several opportunities for organizations, regulatory bodies, and academic institutions are identified to improve the current state of the practice in both hazard analysis and hazard analysis validation.