Vehicle routing with source selection, integration sourcing in fleet deployment

We analyse the benefits and limitations of the integration of sourcing decisions into the operational route compilation task of a road-haulage company. A trucking company has to supply several customer sides. The demanded quantities are given. The trucking company has to decide which truck serves which customer location(s) (routing decisions). In contrast to previously reported fleet deployment problems the trucking company can select from several loading positions for each individual transport request (sourcing decisions). We propose a mathematical model for the integrated sourcing and vehicle routing decision problem. For this purpose, we merge a network flow model and a vehicle routing model. The first mentioned model represents the sourcing decision problem and the second model represents the fleet deployment (routing) decisions. We propose a matheuristic approach to solve the proposed integrated model. This matheuristic combines an algorithm for solving the network flow problem part and a metaheuristic that searches for least distance vehicle routes. Both algorithms interchange information through an adaptable distance matrix that is accessed by both algorithms. We use the proposed model-based approach to evaluate the benefits from integrating sourcing decisions in fleet deployment tasks and execute comprehensive computational experiments.