Quantitative Optimization and Assessment of Maintenance Rules in Nuclear Power Plants Using Probabilistic Safety Assessment

The traditional deterministic safety management systems adopted in early nuclear power plants (NPPs) often operate as inflexible frameworks, leading to overly conservative maintenance strategies and a lack of transparency in safety analysis logic.To address the limitations of deterministic approaches, this paper proposes a quantitative optimization and assessment framework for Maintenance Rules (MR) in Pressurized Water Reactors (PWRs) using Probabilistic Safety Assessment (PSA). The methodology integrates risk-informed decision-making to categorize Systems, Structures, and Components (SSCs) based on their risk significance using Fussell-Vesely (F-V) importance and Risk Achievement Worth (RAW). Furthermore, a dynamic baseline threshold approach is established to quantitatively determine plant-specific performance criteria, including the maximum allowed Unavailability (UA) and Maintenance Rule Functional Failures (MRFFs). A case study utilizing real operational data from the Residual Heat Removal System (RRA) and Component Cooling Water System (RRI) validates the proposed framework. The results demonstrate that the risk-informed MR framework effectively optimizes maintenance resource allocation and enhances plant economics without compromising core damage frequency (CDF) or large early release frequency (LERF). This quantitative approach provides a reproducible template for upgrading nuclear safety management systems in existing PWR fleets.