Maintenance Cost Optimization for Bridge Structures Using System Reliability Analysis and Genetic Algorithms

Abstract

Maintenance cost optimization and performance prediction of bridge structures have become important challenges in bridge management systems. The performance of bridge structures should be carefully monitored, especially in severe climatic conditions. The objective of this study is to develop a rational method that predicts the most cost-effective intervention schedule for bridges, where the structural safety is maintained with the minimum possible lifecycle cost. The framework functions through (1) a central database that contains the asset inventory along with the maintenance actions list, (2) a biquadratic system reliability–based deterioration model, (3) an intervention effect model that simulates the effect of undertaking various intervention scenarios on the bridge superstructure performance, (4) a financial model that computes the lifecycle costs throughout the planning horizon, and (5) an optimization model that utilizes a genetic algorithms engine to compare the different intervention scenarios and selects the most cost-effective one. This method is applied to a simply supported bridge superstructure case study, designed in accordance with Canadian highway bridge design standards. The results indicate that undertaking less costly minor repair actions may considerably reduce the lifecycle costs as a result of decreasing the number of costly major interventions. The optimum scenario resulted in an equivalent uniform annual cost of US$8,277 per year, which shows 4.5 times cost saving as compared with the conventional scenario where only major repairs are performed. This innovative combination of reliability analysis, nonlinear finite-element modeling, and genetic algorithms optimization supports asset managers in long-term planning and ensures undertaking rational and objective decisions.