PROFAT: A user friendly system for probabilistic fault tree analysis

A methodology has been developed to conduct probabilistic fault tree analys is during risk assessment in chemical process industries. The methodology i s based on a system involving a) development of fault tree b) determination of minimum cutsets or shortest pathways leading from initiating events (ma lfunctioning) to the top event (the major accident), c) probability analysi s, and d) working out improvement index values. To this end techniques of B oolean algebra, structure moduling, analytical method of cutsets finding (t op-to-bottom algorithm), Monte-Carlo simulation, optimization technique and fuzzy probability set have been used. We named the methodology Analytical-Simulation Methodology (ASM) and we dev eloped a software package PROFAT (PRObabilistic fault tree analysis) to fac ilitate the use of the methodology in a rapid and effective, yet user-frien dly manner. PROFAT enables the user to find out, in an industry, i) initiat ing events which may eventually lead to a major accident, ii) shortest rout es (minimum cutsets) a series of initiating events may take place while aid ing each other in causing the accident, iii) the probabilities of occurrenc e of such initiating events and, finally, v) identifying initiating events with the greatest potential to cause the top event (major accident) so that accident prevention strategies and emergency preparedness plans can be foc used on them. The noteworthy attributes of the system are: resilience towards lack of pre cision in the basic data, swift processing with moderate requirements of co mputation capacity (sophistication of computers needed), ease of use, and d irectly utilizable output. The applicability of PROFAT has been demonstrate d with a case study of a sulfolane manufacturing unit.