(32ci) Learning Smarter Lessons from Incidents with Bowtie and Consupmap Analyses

Too many process safety incidents are continuing to occur across the world. Voluminous amounts of information has been produced on how to prevent future occurrences. There are incident investigation reports, incident meta-analyses, industry and regulator guidance materials and academic publications. There is too much information available for people to review, let alone comprehend how to action and the volumes are increasing exponentially. This raises the question: Is there a S.M.A.R.T.er (Specific, Measurable, Actionable, Realistic, Timebound), more user-centric way to capture and communicate learnings from incidents?

Bowtie Analyses are becoming an increasing populate method for visualising and communicating the control measures associated with preventing and mitigating unwanted events in hazardous systems. Since their inception in the 1970s, the bowtie framework has been applied in a variety of ways across many high consequence event scenarios. Research has identified that quality bowties - bowties that are easy to understand and practical to use - can be produced by applying a disciplined approach to the construction of the bowtie and the selection and articulation of the prevention and mitigation barriers or controls. This presentation will describe this quality bowtie approach then discuss subsequent research that investigated whether the same bowties can be used to capture learnings from incidents.

One of the limitations of the bowtie analysis approach is in its ability to visualise organisational factors that either support or erode barrier or control performance. To address this shortcoming further research explored adapting Rasmussen's Accimap framework to provide a structured means of visualising control support management systems referred to as ConSupMaps (Control Support Maps). The presentation will provide an introduction of ConSupMaps and how they can also be used to, complement bowtie work and support the visualisation and communication of learnings from incidents. The presentation will then conclude with a case study example showing how Bowtie Analysis and ConSupMaps can illustrate visually and quantitatively the incident lessons in a S.M.A.R.T.er manner.