Incorporating Consequence Mitigation into LOPA

Layer of Protection Analysis (LOPA) is a proven tool for risk assessment in the process industries. Much of LOPA's utility comes from its ease of use, which is a result of the simplifications that are employed to calculate results. One critical simplification is the assumption that if a protection layer is effective, the consequence will not occur. In other works, the protection layer is preventive. This allows the analyst to only consider the frequency of a singular consequence event using only probabilistic means. In reality, there are many safeguards to do not prevent the consequence, but instead reduce the magnitude, or mitigate the consequence. From a risk analysis standpoint, this results in two different consequences that must be addressed, a larger one if the protection layer fails, and a smaller one if the protection layer is successful. Unfortunately, LOPA, as it is currently practiced, cannot address this situation, and mitigative safeguards are either ignored, or given too much credit (i.e., treated as preventive). This paper will discuss how to extend LOPA to consider consequence mitigation protection layers, by extending the analysis to consider multiple different consequences that are possible in a single given scenario. This type of analysis has seen more attention as the result of increased use of bowtie analysis for visualization of risk scenarios, and is being facilitated by the adoption of Unified Hazard Assessment data structure and tools that allow PHA, LOPA, Hazard Registers, and Bowtie diagrams to be all be viewed and edited using a single tool and single data file.