Hybrid solution methods for inventory routing problems

Abstract

This research project addresses the inventory routing problem (IRP) and its practical variants. This problem integrates essential activities in the supply chain management, namely inventory control and vehicle routing. In the IRP, a supplier is responsible for delivering products to its customers and controlling their stocks in a given planning horizon, which results in a challenging optimization problem. Because the IRP has received an increasing attention only in the last years, there is still a considerable lack of research regarding solution methods to solve it. Thus, the objective of this research project is to propose a hybrid method combining metaheuristic algorithms with a branch-price-and-cut method. Practical variants of the IRP will be also considered aiming to make the addressed problem as realistic as possible, so the proposed methods can be an interesting contribution to the literature and useful for decision makers in practice. To solve the basic variant of this problem, we have already developed two metaheuristic algorithms which can find good feasible solutions within relatively short running times for large instances. We are currently developing an exact method based on a branch-price-and-cut algorithm, which it is expected to provide optimal and near-to-optimal solutions for medium and large instances as well as good lower bounds for the optimal costs, which represents an advantage in relation to heuristic algorithms. By combining these approaches, it will be possible to exploit the advantages of both classes of methods to yield an enhanced solution method, which can be able to solve reasonably large problem instances in acceptable running times for the IRP and its variants. This research will be conducted under the supervision of Professors Raf Jans and Jean-François Cordeau, both full Professors of the Department of Logistics and Operations Management, HEC-Montréal, who are experts in these topics. (AU)