Implications of an HRA framework for quantifying human acts of commission and dependency: Development of a methodology for conducting an integrated HRA/PRA

摘要

To support the development of a refined human reliability analysis (HRA) framework, to address identified HRA user needs and improve HRA modeling, unique aspects of human performance have been identified from an analysis of actual plant-specific events. Through the use of the refined framework, relationships between the following HRA, human factors and probabilistic risk assessment (PRA) elements were described: the PRA model, plant states, plant conditions, PRA basic events, unsafe human actions, error mechanisms, and performance shaping factors (PSFs). The event analyses performed in the context of the refined HRA framework, identified the need for new HRA methods that are capable of: evaluating a range of different error mechanisms (e.g., slips as well as mistakes); addressing errors of commission (EOCs) and dependencies between human actions; and incorporating the influence of plant conditions and multiple PSFs on human actions. This report discusses the results of the assessment of user needs, the refinement of the existing HRA framework, as well as, the current status on EOCs, and human dependencies.