EURO 2024 Copenhagen
Abstract Submission

EURO-Online login

3177. The Optimal Policy for a Cyclic Capacitated Lost Sales Inventory

Invited abstract in session MC-50: Retail Inventory Management II, stream Retail Operations.

Monday, 12:30-14:00
Room: M2 (building: 101)

Authors (first author is the speaker)

1. Ludwig Brieditis
Mechanical Engineering Sciences, Lund University
2. Peter Berling
Industrial Economics and Technology Management, Norwegian University of Science and Technology

Abstract

Determining optimal replenishment of brick-and-mortar retail stores is a challenging problem. Based on experience from a large north European grocery store, these retailers typically face a non-stationary demand with a recurrent periodic pattern where excess demand is lost. Moreover, replenishment opportunities are limited by a weekly schedule with limited transport capacity allocated to each store.

The optimal replenishment decisions for a store in this setting is previously unknown. We show that a periodic modified base-stock policy is optimal, i.e. ordering so that the resulting inventory is as close as possible to the optimal base-stock level of the period by modeling the problem as a Markov Decision Process. It is straightforward to use the model to find the optimal base-stock levels. A numerical study shows the resulting cost increases when using solution methods that disregard some of the setting characteristics, e.g. non-stationary demand, limited capacity, lost sales.

Efficiently solving this problem is important not only to minimize the cost at the stores but also to evaluate various planning decisions, e.g. scheduling and routing, in the greater distribution network.

Keywords

Status: accepted

Back to the list of papers