Data-Driven Models for Establishing Relationship between RPN and Reliability

Nowadays, the occurrence of natural hazards is increasingly frequent. Bridges, being the backbone of the societal economy by facilitating the transportation of goods, must be made more resilient to ensure they remain functional during any type of hazard. This work examines the risk profiles and reliability indicators of steel and concrete bridges through the computation of the risk priority number (RPN) and the evaluation of its correlation with structural reliability. Fatalities linked with different bridges were collected from public references, and the RPN was determined by considering severity (S), occurrence (O), and detectability (D) values. In the case of both bridges, reliability was calculated using two approaches, uncovering a contrary correlation between RPN and reliability, which was depicted through quadratic spline interpolation. Steel bridges displayed nonlinear variations, necessitating cubic spline interpolation for precise representation. Principal discoveries consist of a generally contrary correlation between RPN and reliability for both types of bridges, emphasizing the necessity for prioritized upkeep of high-risk structures. The study employs a dual-method approach, combining direct analytical techniques with statistical modeling, significantly enhancing reliability assessments’ robustness and precision. This investigation offers valuable perspectives for infrastructure governance, hazard evaluation, and safety strategizing, empowering informed decision-making for bridge examinations and upkeep maneuvers. The utilization of sophisticated interpolation methodologies guarantees accurate and pragmatic representations for managing infrastructure hazards, contributing to the fortification of bridge safety and reliability.