A fuzzy logic and probabilistic hybrid approach to quantify the uncertainty in layer of protection analysis

Layer of protection analysis (LOPA) is a widely used semi-quantitative risk assessment method. It provides a simplified and less precise method to assess the effectiveness of protection layers and the residual risk of an incident scenario. The outcome failure frequency and consequence of that residual risk are intended to be conservative by prudently selecting input data, given that design specification and component manufacturer's data are often overly optimistic. There are many influencing factors, including design deficiencies, lack of layer independence, availability, human factors, wear by testing and maintenance shortcomings, which are not quantified and are dependent on type of process and location. This makes the risk in LOPA usually overestimated. Therefore, to make decisions for a cost-effective system, different sources and types of uncertainty in the LOPA model need to be identified and quantified. In this study, a fuzzy logic and probabilistic hybrid approach was developed to determine the mean and to quantify the uncertainty of frequency of an initiating event and the probabilities of failure on demand (PFD) of independent protection layers (IPLs). It is based on the available data and expert judgment. The method was applied to a distillation system with a capacity to distill 40 tons of flammable n-hexane. The outcome risk of the new method has been proven to be more precise compared to results from the conventional LOPA approach.