Certified Inventory Control of Critical Resources

TL;DR

This paper introduces a data-driven inventory control policy that guarantees service levels with minimal assumptions, using online learning and finite-sample inference, validated on synthetic and real data.

Contribution

It presents a novel certified inventory control policy that ensures service levels under minimal demand assumptions, combining online learning with finite-sample inference.

Findings

Guarantees service levels with minimal demand assumptions

Validates approach with synthetic and real-world data

Provides finite-sample inference method

Abstract

Inventory control is subject to service-level requirements, in which sufficient stock levels must be maintained despite an unknown demand. We propose a data-driven order policy that certifies any prescribed service level under minimal assumptions on the unknown demand process. The policy achieves this using any online learning method along with integral action. We further propose an inference method that is valid in finite samples. The properties and theoretical guarantees of the method are illustrated using both synthetic and real-world data.