This study proposes a dynamic human reliability assessment (HRA) method that explicitly accounts for environmental stressors such as radiation exposure and high temperature–high humidity conditions during severe nuclear accident scenarios. Conventional HRA methods, including THERP and IDHEAS, typically apply static performance shaping factors (PSFs), which are insufficient to represent time-dependent and location-specific stress variations. To address this limitation, a dynamic PSF model was developed by incorporating physically based indicators, particularly the Wet Bulb Globe Temperature (WBGT), which reflects heat stress and the effects of protective clothing.
A representative accident mitigation scenario involving alternative water injection using portable equipment was analyzed across multiple work environments, including the main control room, outdoor areas, and inside the reactor building. Baseline human error probabilities (HEPs) were first calculated using conventional methods, and then dynamically updated based on temporal stress accumulation and residual effects during rest periods.
The results indicate that while short-duration tasks show comparable or slightly lower HEP values, long-duration tasks and those influenced by residual stress exhibit significantly increased HEP, up to five times higher than static estimates. Overall, the dynamic HRA approach yields higher and more realistic HEP values compared to conventional methods.
This study demonstrates that incorporating time-dependent environmental stress enables more accurate evaluation of human reliability under severe accident conditions.