Fault tree analysis of fuzzy mechanical systems

Abstract

The fault tree techniques involve the decomposition of a system into a logic diagram or fault tree in which certain primary events lead to a specified top event that signifies the total failure of the system. The fault trees can be used to find the minimal cut sets which, when coupled with the failure rates of the primary events, lead to an estimation of the reliability of the system. In many practical situations, conventional fault tree analysis is difficult because failure rates corresponding to the basic events are not clearly established. This paper presents a fuzzy fault tree technique for the reliability analysis of mechanical systems. The proposed technique uses fuzzy numbers to quantify the failure rates. Fuzzy arithmetic is then applied in the fault tree analysis to determine the probability of the top event and the corresponding system reliability as fuzzy numbers. The fuzzy approach is illustrated with the analysis of an overhead cam valve train.