A semi-quantitative reasoning methodology for filtering and ranking HAZOP results in HAZOPExpert

Hazard and Operability (HAZOP) analysis is the most widely used and recognized as the preferred Process Hazards Analysis (PHA) approach in the chemical process industry. Recently, a digraph-model based framework and an expert system called HAZOPExpert was developed for automating this analysis.(1) Upon testing the performance of the system on various industrial case studies, HAZOPExpert was found to successfully mimic the human expert's reasoning and identify the hazards. But, with the increasing complexity of the processes, the HAZOPExpert system generated a large number of consequences compared to those identified by a team of experts. This is mainly due to the strict qualitative reasoning approach implemented in the HAZOPExpert system. In order to filter and rank the consequences generated by the HAZOPExpert system, a semi-quantitative reasoning methodology is proposed using additional quantitative knowledge in the form of design and operating specifications of the process units, and process material property values. This filtering approach combines the qualitative digraph-based HAZOP models and the quantitative knowledge to eliminate the unrealizable consequences. Significant reduction in the number of consequences was obtained using this approach on an ethylene process plant HAZOP case study. (C) 1996 Elsevier Science Limited.