Stability measures to re-optimise the inventory routing problem under parameters uncertainties
Khaled HADJ-HAMOU, Professor, DISP Laboratory
We consider the inventory routing problem (IRP) when facing a variety of uncertainties (demands, travel times). Re-optimisation, i.e. recalculating a new schedule when new data appears, is one way to cater for these uncertainties. But re-optimisation can yield very different solutions : stable solutions that deviate the least from the original ones are required. However, there is no consensus in the literature to define the concept of stability. We present a panel of stability measures from the operations management literature. These measures are adapted for the IRP and their mathematical formulations proposed. We show the need to define stability functions for the IRP and its variants, ensuring both external and internal stability for the routing and the inventory management components of the problem.