An application of fault tree analysis for computing the bounds on system failure probability through qualitative data in intuitionistic fuzzy environment

Abstract

AbstractFor the safety assessment of complex engineering systems, fault tree analysis (FTA) is a widely used method and can be performed only when all its basic events (BEs) have their corresponding failure data. However, in many applications, obtaining precise quantitative failure data of all the components of the system is not feasible due to changing environment, insufficient data or new components. The present research develops a novel FTA to identify the best estimate and bounds of system failure probability through qualitative data processing. The failure data of BEs are uncertain therefore expert elicitation method is used incorporating intuitionistic fuzzy (IF) set theory to evaluate the top event failure probabilities. The novelty of the proposed work comes with the use of linguistic values as failure possibilities of the system components and utilizing triangular intuitionistic fuzzy numbers (TIFNs) to quantify the linguistic terms into mathematical form to assess the best‐estimated value of the system failure probability, and also the lower bound and the upper bound of the system failure probability. To exemplify the appropriateness of the proposed study, the case studies Combustion Engineering Reactor Protection System (CERPS) and Emergency Core Cooling System (ECCS) are carried out and a comparison between computed results and outcomes from the previous methods is demonstrated. The results are exceptionally near to the outcomes of conventional FTA, which confirms that the proposed procedure is a real substitute in the IF environment when components have imprecise quantitative failure data.