Causes of Uncertainty in FTA Method and Use of Fuzzy Logic to Solve this Problem
Document Type : Original Article
Authors
1 Abadan Institute of Technology, Petroleum University of Technology, Abadan, Iran
2 Faculty of Environmental Studies, University Putra Malaysia, Malaysia
Abstract
Nowadays increasing the number of chemical industries leads to more industrial accidents. Due
to consequences of accidents imposed high costs to industry, society and environment. This
makes safety analysis more important. Therefore qualitative and quantitative hazard analyses are
essential for identification and quantification of hazards. Fault tree analysis (FTA) is an
established technique in hazard identification. People when using FTA often suffers from a lack
of detailed data on failure rates, uncertainties in available data, imprecision and vagueness. This
may lead to uncertainty in results and process risk level. Fuzzy logic deals with uncertainty and
imprecision, and is an efficient tool for solving problems where knowledge of uncertainty may
occur. In this paper traditional FTA was combined with fuzzy set theory. In fuzzy method, all
variables are replaced by fuzzy numbers in the process of fuzzification and subsequently fuzzy
probability of the top event is used for fault tree.
to consequences of accidents imposed high costs to industry, society and environment. This
makes safety analysis more important. Therefore qualitative and quantitative hazard analyses are
essential for identification and quantification of hazards. Fault tree analysis (FTA) is an
established technique in hazard identification. People when using FTA often suffers from a lack
of detailed data on failure rates, uncertainties in available data, imprecision and vagueness. This
may lead to uncertainty in results and process risk level. Fuzzy logic deals with uncertainty and
imprecision, and is an efficient tool for solving problems where knowledge of uncertainty may
occur. In this paper traditional FTA was combined with fuzzy set theory. In fuzzy method, all
variables are replaced by fuzzy numbers in the process of fuzzification and subsequently fuzzy
probability of the top event is used for fault tree.
Keywords
)