As the nuclear industry advances toward deployment of advanced reactors, passive safety systems (PSS) are increasingly relied upon to reduce dependence on operator actions and external power while extending accident coping times. Despite these advantages, the reliability assessment of PSS remains challenging due to their strong dependence on complex thermal–hydraulic phenomena and associated uncertainties. This study, supported by the Electric Power Research Institute (EPRI), evaluates the global state of practice for PSS reliability analysis and examines its integration within a probabilistic risk assessment (PRA) framework. Existing methodologies for assessing PSS reliability are reviewed, with particular emphasis on the treatment of phenomenological variability and uncertainty inherent in natural circulation and other passive mechanisms. Building on this review, a consolidated reliability assessment methodology is developed that combines traditional fault tree analysis with systematic treatment of physical phenomena uncertainty. A Phenomena Identification and Ranking Table (PIRT) process is applied to identify and prioritize key thermal–hydraulic parameters influencing PSS performance, enabling focused modeling and highlighting knowledge gaps requiring further research. To address computational constraints, uncertainty propagation in a best‑estimate thermal–hydraulic code is performed using response surface techniques. The methodology is demonstrated through a pilot application to a simplified passive containment cooling system, yielding credible insights into system performance and reliability. The results underscore the importance of integrated, risk‑informed approaches to passive system reliability and support the continued advancement of passive safety concepts for next‑generation nuclear reactors.