Solving the joint production distribution problem by the variable neighborhood search algorithm

This paper addressed the dynamic integrated production routing problem where the objective is to minimise the total cost including production, inventory and distribution. Since this problem is NP-hard, we developed a variable neighbourhood search algorithm (VNS) to solve it. Therefore, two VNS algorithms were proposed: the first one is used to determine the produced quantity of each day and the second aims to determine the routing map. Insert and swap moves were used in the local search step. The obtained results showed the effectiveness of our proposed approach according to previous works, including greedy heuristic and memetic algorithm, in the literature.