Epistemic uncertainty in fault tree analysis approached by the evidence theory

Process plants may be subjected to dangerous events. Different methodologies are nowadays employedto identify failure events, that can lead to severe accidents, and to assess the relative probability ofoccurrence. As for rare events reliability data are generally poor, leading to a partial or incompleteknowledge of the process, the classical probabilistic approach can not be successfully used. Such anuncertainty, called epistemic uncertainty, can be treated by means of different methodologies, alternativeto the probabilistic one. In this work, the Evidence Theory or DempstereShafer theory (DST) is proposedto deal with this kind of uncertainty. In particular, the classical Fault Tree Analysis (FTA) is consideredwhen input data are supplied by experts in an interval form. The practical problem of informationacquisition from experts is discussed and two realistic scenarios are proposed. A methodology topropagate such an uncertainty through the fault tree up to the Top Event (TE) and to determine the beliefmeasures is supplied. The analysis is illustrated by means of two simple series/parallel systems. Anapplication to a real industrial safety system is finally performed and discussed.