The fault tree analysis is an important technique in the context of reliability and risk analysis providing support for decision making and helping managers in assuring that a given system and their components will perform accordingly, given a set of operational and environmental conditions. Nowadays, one of the main problems emphasized in the fault tree related literature is concerned with procedures that contemplate computational efficiency as well as precision in the estimation of the fault tree top event probability. This paper suggests a new approach to tackle this inference problem, which mainly consists of removal of the fault tree redundancies with subsequent application of recursive algorithms. It is also shown the improved precision of the proposed approach when compared to traditional techniques. The proposed approach is then illustrated by means of an example application.
reliability; fault trees; redundancies
