Fuzzy modelling of uncertain data in the layers of protection analysis

Abstract

Layers of Protection Analysis (LOPA) described in the IEC 61511 standard is one of the most popular methods that enables the adequacy of the layers of protection for an industrial process to be assessed. LOPA has found much broader application as a relatively simple risk assessment methodology. However, an important issue faced by risk analysts is how to deal with uncertainties that arise in the elements of the accident scenario like the severity of consequences, frequency data for initiating event, probability of failure of independent protection layers (IPL) and human reliability. (ICAT'14). This paper presents an approach of fuzzy LOPA which takes into account uncertain and/or vague data. A possibilistic representation of data by fuzzy intervals is proposed. Consequence reduced frequency is computed by extended multiplication using a-cuts method. The proposed approach is illustrated by an application to the protection of an operational system which is a heater.