Models and algorithms for the cyclic inventory routing problem

The key to efficient supply chain management is the integration of different stages in the chain. A good example for this is 'Vendor Managed Inventory' (VMI), in which a distributor is given the freedom by its customers to determine how often, at what time and in what quantities he will replenish them. This freedom gives the distributor the opportunity to design more efficient vehicle routes. The cost minimisation problem arising under VMI, integrating the decisions on quantities to be delivered and vehicle routes to be made, is known as the 'Inventory Routing Problem' (IRP). This dissertation studies the IRP under the assumption of stable customer consumption patterns, for which a cyclic approach is appropriate. A novel solution approach is presented that aims at finding a three-way cost trade-off between fixed vehicle costs, transportation costs, and inventory costs, while taking into account various side-constraints encountered in practice. The excellent performance of this new approach is illustrated by (i) evaluation on a large set of test instances with varying characteristics, and by (ii) comparison to other existing methods. Further, the approach is also applied to a real-life case. In this case study, the robustness of the approach w.r.t. variations in the customer consumption patterns can be observed, illustrating once more the practical relevance of this dissertation.

Birger Raa
  • El-Houssaine Aghezzaf
Dec. 4, 2006, 1:42 p.m.
Research group