Operator response modeling and human error probability in TRIGA Mark II research reactor probabilistic safety assessment

Abstract

This study presents the quantitative part of the analysis of human actions (HAs) in human reliability analysis (HRA) for a research reactor level 1 probabilistic safety assessment (PSA). In responding to the abnormal conditions, type C HAs which are identified in the previous study are elucidated furthermore for their cognitive response element, prior to the quantification of their human error probabilities (HEPs). The HEPs are afterward quantified by mean of Standardized Plant Analysis Risk-Human Reliability Analysis (SPAR-H) method. For the type A and B HAs, their HEPs are quantified by using the Technique for Human Error Rate Prediction (THERP) method. Incorporation of the human failure events (HFEs) and their respective HEPs into the PSA models further allow the evaluation of human contribution to the risk associated with the operation of the research reactor. In this study, it is quantitatively shown that human attribute has major contribution to the risk and with the introduction of nominal level of procedure-performance shaping factor (PSF), risk can be significantly reduced.