A human reliability analysis methodology based on an extended Phoenix method for severe accidents in nuclear power plants: Qualitative analysis framework. (October 2021)

• Record Type: Journal Article
• Authors: Chen, Shuai; Zhang, Li; Qing, Tao
• Is Part Of: Reliability engineering & system safety. Volume 214(2021)
• Publication Date: 2021-10
• Subjects: Human reliability analysis -- Severe accident -- Emergency response personnel -- Phoenix methodology -- Nuclear power plant; Reliability (Engineering) -- Periodicals; System safety -- Periodicals; Industrial safety -- Periodicals; Fiabilité -- Périodiques; Sécurité des systèmes -- Périodiques; Sécurité du travail -- Périodiques; 620.00452
• DOI: 10.1016/j.ress.2021.107750
• Languages: English
• ISSNs: 0951-8320

Abstract

Highlights: A qualitative framework to analysis human reliability in severe accident is proposed. The human error modes specific to emergency response personnel are identified. The performance influencing factors of emergency response personnel are explained. The characteristics of human response in severe accident are delineated. Human-related issues in past severe accidents and in drills are concluded. Abstract: The mitigation of the Fukushima accident exposed the detrimental action of human errors once again. However, deficiencies in existing human reliability analysis (HRA) methods prevent the effective prediction and analysis of potential human errors during the mitigation of severe accidents following core damage. To address this problem, this study presents a qualitative analysis framework for emergency response personnel (ERP) reliability analysis under severe accident conditions. This framework is developed using the Phoenix HRA method in conjunction with analyzing the ERP response characteristics and human-related issues from previous accidents, literature surveys, drill observations and interviews, as well as an assessment of the applicability of existing HRA methods. The analysis framework includes the establishment of an ERP response tree and a human failure fault tree (FT), as well as the identification and compilation of ERP cognitive functions, human error modes (HEMs), performance-influencing factors (PIFs), and the construction of Bayesian belief networks, to provide methodological guidance for the prevention of human errors under severe accident conditions.