Development of risk importance measures for dynamic PRA based on risk triplet, 1; The Concept and measures of risk importance

Narukawa, Takafumi*; Takata, Takashi*; Zheng, X. ; Tamaki, Hitoshi ; Shibamoto, Yasuteru ; Maruyama, Yu ; Takada, Tsuyoshi  

Despite the advancements in dynamic probabilistic risk assessment (PRA) methods that account for the dynamics of event progression, establishing risk importance measures for these methods remains a significant research challenge. This study proposes novel risk importance measures from the perspective of the risk triplet: Timing-Based Worth (TBW) for the timing of scenario occurrence (scenario diversity), Frequency-Based Worth (FBW) for the frequency (probability) of scenarios, and Consequence-Based Worth (CBW) for the consequences of scenarios. To assess the effectiveness of these measures, a static PRA using the event tree method and a dynamic PRA using the continuous Markov chain Monte Carlo (CMMC) method are performed on a simplified reliability model. The results indicate that the proposed measures facilitate a comprehensive risk importance evaluation, incorporating resilience effects (the time margin) and consequence mitigation, alongside traditional frequency-based evaluations. This advancement is anticipated to improve the utilization of risk information derived from dynamic PRA.