Lifetime multi-objective optimization of maintenance of existing steel structures

Abstract

In this paper, the lifetime performance of deteriorating structures, defined by their time dependent condition index and reliability index, is analyzed. The effect of preventive and essential maintenance actions on performance and cost in predicted, and the optimal times of application of preventive and essential maintenance actions are found. Due to significant uncertainty in the initial performance, effects of deterioration and of maintenance actions, as well as, times of application and cost of maintenance actions, the analysis is performed in a probabilistic framework. The reduction in performance due to deterioration is simulated using an extension of the model proposed by Frangopol (1998). The probabilistic condition index, reliability index, and cumulative cost profiles are computed using Latin Hypercube simulation. Optimization of times of application is performed using genetic algorithms. Results show the significant importance of preventive maintenance actions in reducing the lifetime cost of existing structures, but also their fundamental role of essential maintenance action in keeping structures safe and serviceable during the entire lifetime.