Human reliability assessment for tank overfilling incident utilizing minimized human performance shaping factors

Abstract. Human errors are identified as significant contributors to process industry accidents. reliability analysis (HRA) has been conducted in previous studies improve human performance several industrial operations. However, error predictions greatly influenced by various performance-shaping factors (PSFs). Research also demonstrates that PSFs interdependent, which thereby complicates the modeling and analysis. Therefore, this study performs HRA, for a tank overfilling accident scenario resulted due failure. Fewer independent through careful classification were used estimate probability resulting from different human-triggered factors. For uses combination of Standardized Plant Analysis Risk Reliability (SPAR-H) Bayesian Belief Network (BBN). The failure distributions individual interconnecting tasks calculated using SPAR-H, probabilistic interdependence each task final was modeled BBN. From current analysis, divergent stream identification is determined key lead with 40% probability. This concludes BBN can be reliably employed Quantitative (QRA) framework examine estimation other human-related scenarios.