HFACS-based FAHP implementation to identify critical factors influencing human error occurrence in nuclear plant control room

Abstract

Human factor is an inevitable element of safety-critical control room operations in nuclear power plants. Identifying human factors influencing operator error occurrence in such critical application is important to mitigate human errors. Conventional methods employed for human factor identification are often static, unable to deal with data and model uncertainty, and to consider independencies among failure modes. This study employs the integration of the soft computing technique, viz. fuzzy analytic hierarchy process (FAHP), into the human factors analysis and classifying system (HFACS) framework to identify the critical human factors that contribute to human errors in the nuclear control room application. Integration of FAHP improves the HFACS framework by providing an analytical foundation and group decision-making ability in order to ensure quantitative assessment of nuclear accidents. The proposed model has two phases. The first phase is study of 18 human performance-related events that occurred in Indian nuclear power plants, utilizing HFACS to analyze and determine the human and organizational factors (HOFs) responsible for such events. Existing HFACS can only be used to detect a wide range of human factors. The proposed model further explores the underlying causes of such a wide range of factors. The hierarchy of HOFs identified as adverse mental states and organizational process factors that contributed most failures causing accident in the first phase provides inputs for the critical human factor identification to the second phase, which is a quantitative analysis using FAHP. In the second phase, more than 40 adverse mental state and organizational process factors are identified from literature survey and ten subfactors are screened and selected by human reliability assessment and control room operation experts. The study was conducted by administering a questionnaire that is comprised of the screened factors. Data have been collected from 88 senior reactor operators in an operating nuclear power plant. The critical factors contributing to human errors identified from FAHP are attention, perception and memory under cognitive factor and decision making, training and communications under organizational factor. The study results reveal that the implementation of HFACS-based FAHP methodology enabled in determining intrinsic human factors that contribute to nuclear power plant control room operator performance.