Allocating service parts in two-echelon networks at a utility company

We study a multi-item, two-echelon, continuous-review inventory problem at a Dutch utility company, Liander. We develop a model that optimizes the quantities of service parts and their allocation in the two-echelon network under an aggregate waiting time restriction. Specific aspects that we address are emergency shipments in case of stockout, and batching for regular replenishment orders at the central warehouse. We use column generation as a basic technique to solve this problem, and use various building blocks for single-item models as columns. Further, we study options to derive simple classification rules from the solution of our multi-item, two-echelon service part optimization problem using statistical techniques. Application of our models at Liander yields a solution that reduces costs by 15% and decreases the impact of waiting time for service parts by 52%.