A GRASP Algorithm Based on New Randomized Heuristic for Vehicle Routing Problem

This paper presents a novel GRASP algorithm based on a new randomized heuristic for solving the capacitated vehicle routing problem. This problem is characterized by a fleet of homogenous vehicle capacity that will start from one depot, to serve a number of customers with demands that are less than the vehicle capacity. The proposed method is based on a new constructive heuristic, and a simulated annealing procedure as an improvement phase. The new constructive heuristic uses four steps to generate feasible initial solutions, and the simulated annealing enhances these solutions found to reach the optimal one. We tested our algorithm on two sets of benchmark instances and the obtained results are very encouraging.