Barrier-based safety programs are increasingly being used to manage high hazard facilities safely. Typical tools include bow tie diagrams which map all the prevention and mitigation barriers protecting against major process safety events. These make clear the most important barriers, the responsibility for the barrier, and then communicate this to all who need to know. This would work well if barrier performance was static, however the reality is that barrier performance degrades as soon as a facility is started-up, and each barrier degrades with its own characteristic time-constant. This makes establishing barrier status difficult and without this knowledge then risk management may fail.
Dynamic Barrier Management is a set of tools that aims to understand barrier status quantitatively at all times and to deploy this for maximum safety benefit in decision making. It does this by first establishing barrier importance by reference to a full QRA model for a similar facility and thereby to determine the effect of each barrier on the risk result. Where barriers are not explicitly part of the QRA model, then valuation is achieved by similarity. The barrier status is determined by harvesting a wide range of data already collected in SAP, Maximo, Synergi, the SCADA system and other (non-IT based) systems. This includes elements of what is termed Big Data analytics. Given the importance of each barrier and its current status, then this can be deployed for day-by-day decision making.
This paper describes the approach in detail and then shows how this is deployed for a case study. This system has the potential to make a step change in process safety.
See more of this Group/Topical: Global Congress on Process Safety