Hybrid metaheuristic solutions to inventory location routing problem

• The problem integrates location, allocation, inventory and routing decisions.
• We present a novel and low complexity method to calculate inventory cost.
• The problem is solved by decomposition into near exact subproblems.
• The metaheuristic is competitive when solving LRP and IRP benchmark problems.
• The metaheuristic can solve large-scale IRLP instances in reasonable time.

This paper considers a supply chain network with multiple depots and geographically dispersed customers, each of which faces non-constant demand over a discrete planning horizon. The goal is to determine a set of depots to open, the delivery quantities to customers per period and the sequence in which they are replenished by a vehicle fleet such that the total system-wide cost is minimized. To solve it, first we construct a mixed integer program, and then propose a hybrid metaheuristic consisting of initialization, intensification and post-optimization. Results show that the proposed heuristic is considerably efficient and effective for many classical instances.