Application of SPAR-H Based Bayesian Network Methodology to a Typical FBR Control Room Human Action
摘要
Standardized Plant Analysis Risk Human Reliability Analysis (SPAR-H) is one of the quantification tools to generate human error probability. The performance shaping factors used in SPAR-H method are based on expert judgment. However when there are uncertainties involved in selecting performance shaping factors, Bayesian network offers a flexible framework to model uncertainties. This is demonstrated with a typical Fast Breeder Reactor control room human action. The effect of uncertainty in Performance Shaping Factors on human error probability is demonstrated for this example.