Suggestion of specific performance shaping factor update for the human reliability analysis framework of the TRIGA research reactor

Abstract

Based on the human reliability analysis (HRA) framework of the TRIGA research reactor, performance shaping factor (PSF) estimation is an important step when considering the effects of specific operating or working cultures in determining human error probabilities (HEPs). This study aims to suggest a method to develop specific PSFs for the HRA framework of the TRIGA research reactor through a TRR-1/M1 case study. The PSF survey of the HRA framework was developed based on the EMBRACE method to consider the negative impacts of each PSF compared to the normal situation of all four cognitive activities in the errors of omission and errors of commission modes. Given the varied experiences of experts, expert elicitation was employed to categorize high-performing, low-performing, and informative experts to ensure reliable data for PSF analysis. For low-performing and informative experts, the survey results of PSFs were improved by additional surveys to support an appropriate dataset for analyzing the PSFs of the HRA framework. The impact of PSFs within the HRA framework was estimated using the updated normal distribution of the posterior PSFs based on the classical model. The success likelihood index method successfully integrated all subjective expert judgments into a cohesive representation and offered a better systematic consensus model to generalize HEPs in the form of a normal distribution based on a large group of experts.