A model-driven approach for incorporating human reliability analysis in early emergency operating procedure development

Current Probabilistic Risk Assessment (PRA) methods analyze operator actions in accident scenarios using Human Reliability Analysis (HRA) methods after Emergency Operating Procedures (EOPs) and complex system design are largely complete. This paper proposes the early Model-based HRA (eMHRA) method that couples PRA, HRA, and EOP development together and shifts analysis earlier into the complex system design process. By moving the development of these related and important steps in complex system design earlier in the design process, significant modifications to the complex system can be made much more inexpensively and consume much less time to address critical issues found in PRA, HRA, and EOP development. Further, EOP developers can benefit from rapid and early feedback from the HRA and PRA information. A software tool was developed to implement the eMHRA method presented in this paper and is demonstrated in the paper. A case study is presented of a subsystem of a generic Pressurized Water Reactor (PWR) civilian nuclear power plant. The case study shows that HRA and EOP insights can be incorporated into PRA models early in the design process to better inform system designers of potential high likelihood failure events in operator actions. The eMHRA method presented in this paper provides a new tool for risk analysts to better predict and understand failure scenario outcomes early in the design process. With this information, engineers will be better able to develop new EOPs and operator interfaces to reduce failure likelihood in due to missed operator recovery actions.