Early-design improvement of human reliability in an experimental facility: A combined approach and application on SPES. (November 2019)

• Record Type: Journal Article
• Authors: Giardina, M.; Buffa, P.; Dang, V.; Greco, S.F.; Podofillini, L.; Prete, G.
• Is Part Of: Safety science. Volume 119(2019)
• Page Start: 300
• Page End: 314
• Publication Date: 2019-11
• Subjects: Hierarchical task analysis -- Human reliability assessment -- Human errors -- HEART -- SPAR-H -- CREAM; Industrial accidents -- Periodicals; Accident Prevention -- Periodicals; Safety -- Periodicals; Travail -- Accidents -- Périodiques; 363.11
• DOI: 10.1016/j.ssci.2018.08.008
• Languages: English
• ISSNs: 0925-7535

Abstract

Highlights: Human errors in Isotope Separation On-Line (ISOL) facility is performed. Integrated approach of Hierarchical Task Analysis and Human Reliability Analysis is used. A modified HEART approach is proposed. HEART, SPAR-H, and CREAM are compared to each other. Abstract: SPES (Selective Production of Exotic Species) is a second-generation Isotope Separation On-Line (ISOL) facility for advanced nuclear physics applications, currently under construction at INFN (National Institute of Nuclear Physics) of Legnaro, Italy. Despite the potentially important safety implications of human errors for ISOL facilities, only a limited number of studies addressing this issue have been performed worldwide. This paper tries to address this need by means of an integrated approach of Hierarchical Task Analysis (HTA) and three human error quantification methods: HEART (in an enhanced version), SPAR-H, and CREAM. The application of multiple Human Reliability Analysis (HRA) methods adds credibility to the analysis results and, for the present paper, this is required because the methods are applied to different performance conditions than typical, i.e. for operators in a nuclear power plant control room. The study emphasizes the potential range of task failure probabilities obtained in the analyses, rather than best-estimate values. Since the facility is being constructed, with elements such as the human-machine interface still undergoing design, some of the inputs typically required by HRA methods were not available; consequently, the produced HEP estimates may not fully represent future performance. The obtained results allowed informing some safety-enhancing recommendations that, coming at early stage, could be considered for implementation by the facility designers and managers.