Human Factors Correlation Analysis for Nuclear Micro-Reactors Based on Literature Review and Evidence Theory

Abstract

Nuclear micro-reactors have drawn increasing attention in the nuclear industry since they can be potentially applied for a variety of purposes such as desalination and hydrogen production. Nuclear micro-reactors have unique characteristics (e.g. wider use of automation) compared with currently operating large nuclear power plants. These characteristics inevitably influence operator performance. To perform human reliability analysis and system design optimization, it is necessary to have a thorough investigation of such effects. To this end, in this paper we propose a method based on literature review and evidence theory to analyze human factor correlations for nuclear micro-reactors. A literature review is performed to identify studies related to characteristic human factors for micro-reactors. We divide human factors into independent variables that are used to characterize microreactors (e.g., level of automation) and dependent variables (e.g., situation awareness) that are influenced by the independent variables. The correlations between the independent and dependent variables are extracted from the reviewed studies. We should note that there may be conflicts between the correlations extracted from different studies. To address this problem, we use evidence theory to aggregate the potentially conflicting information to draw final conclusions about the correlations. The results obtained in this study demonstrate the high complexity of the relations between independent and dependent variables, including the uncertainties in the relations, the non-monotonic relations, and the inconsistent effects of one independent variable on different dependent variables. The potential applications of the results in human reliability analysis for nuclear micro-reactors and system design optimization are discussed.