A Genetic Algorithm Approach for Pavement Maintenance Program Optimization Usinf Safety Evaluation Data

Pavement management problems are specialized scheduling problems in which the number of solutions grows exponentially with the entity of the problem with the result that the number of solutions becomes unmanageable by conventional analytical optimization tools. In these conditions, there is a large number of possible solutions available, known as the “combinatorial explosion” of the feasible solution space. genetic algorithms (GAs) are becoming an increasingly popular way to search for huge solution spaces to find the best solutions. Pavement maintenance management is thus ideally suited for directed random search heuristics such as genetic algorithms. In this paper a procedure has been defined to make use of the available economic resources in the best way possible for pavement maintenance by using genetic algorithms. In particular, the optimization methodology has been applied to bituminous motorway pavements considering only traffic accident and pavement sideway force parameters. The objective has been to manage the multiyear pavement maintenance minimizing the predicted accident number and simultaneously maximizing the average Sideway Force Coefficient value for the whole motorway. The proposed procedure was applied to a motorway belonging to the road network of Eastern Sicily in Italy. The results indicate that the procedure represents an efficient approach to obtain an optimal solution out of the many possible ones in sufficiently short periods of time.