Combination of local search and Constraint Logic Programming in the vehicle-fleet scheduling problem

Vehicle-fleet scheduling is one of the most commonly occurring problems of transport management. In this paper a new approach facing the problem is introduced. It draws upon the combination of the advantages of Constraint Logic Programming (CLP) with the benefits of local search in order to obtain satisfactory results with respect to execution time. CLP is used for generating an initial feasible solution and then for checking every intermediate solution obtained from local search so that it is in accordance with the constraints of the problem. Local search is implemented for the minimisation of the cost of the initial solution. Several local search methods were tested on various cases of the problem and the results obtained are discussed and compared with respect to the cost of the solution and the execution time.