Marine Nuclear Transportation: Probabilistic Safety Assessment of Radioactive Materials Based on Accident Data and Event Tree Analyses

This research study presents the development and validation of a probabilistic safety assessment (PSA) methodology specifically tailored for the marine transportation of radioactive materials (RAMs). The research addresses key limitations in existing PSA applications, particularly concerning data scarcity, associated accident scenarios, and adaptability of probabilistic models across different transport modes. A modular and scalable PSA methodology is proposed, integrating accident frequency estimation, probabilistic modeling, and event tree analysis. At the core of this methodology is a Python-based data analysis tool designed to process generic accident datasets, offering flexibility for application across various accident types. The tool’s practical application is demonstrated through a case study in established PSA models. Given the rarity of nuclear-specific marine incidents, the methodology leverages generic marine accident data from 2011 to 2021 to estimate accident frequencies and assess low-probability, high-consequence events. The resulting probabilistic models and event tree structure provide a robust basis for risk quantification, offering critical insights into the safety implications of RAM marine transport and informing future regulatory and operational strategies.