This paper develops a Genetic-Algorithm-based procedure for solving multi-objective project level pavement maintenance and rehabilitation programming problems. A two-objective optimization model which considered maximum pavement performance and minimum action costs as functions is put forward. It was found that the robust search characteristic and multi-solution handling capability of genetic- algorithms were suitable for multi-objective optimization analysis. Formulation and development of the solution algorithm were described and demonstrated with a numerical example in which a hypothetical project level pavement maintenance and rehabilitation analysis was performed for two-objective optimization. From the result calculated by the computer program, chromosome 31020212322222300100 represents the following 20years maintenance strategies: Overlay in year 1, 9, and 5; Crack sealing in year 2, 7, and 18; Do nothing in year 3, 5, 16, 17, 19, and 20; and Pothole patching in year 4, 6, 8, 10, 11, 12, 13, and 14. Based on the computing results, the Pareto optimal solutions of the two objective optimization functions are obtained. The optimal solutions of this two – objective optimization model can provide the decision makers the maintenance and rehabilitation planning with maximum pavement performance and minimum action costs.
