An Analysis of Multi-Agent Reinforcement Learning for Decentralized Inventory Control Systems

TL;DR

This paper explores multi-agent reinforcement learning for decentralized inventory control, demonstrating that agents trained with a centralized critic can achieve near-centralized performance while respecting real-world information constraints.

Contribution

It introduces a multi-agent reinforcement learning approach using proximal policy optimization with a centralized critic for decentralized inventory management.

Findings

Multi-agent PPO with centralized critic performs close to centralized solutions.

Decentralized agents outperform distributed model-based methods in most scenarios.

The approach respects real-world information constraints in supply chain networks.

Abstract

Most solutions to the inventory management problem assume a centralization of information that is incompatible with organisational constraints in real supply chain networks. The inventory management problem is a well-known planning problem in operations research, concerned with finding the optimal re-order policy for nodes in a supply chain. While many centralized solutions to the problem exist, they are not applicable to real-world supply chains made up of independent entities. The problem can however be naturally decomposed into sub-problems, each associated with an independent entity, turning it into a multi-agent system. Therefore, a decentralized data-driven solution to inventory management problems using multi-agent reinforcement learning is proposed where each entity is controlled by an agent. Three multi-agent variations of the proximal policy optimization algorithm are investigated through simulations of different supply chain networks and levels of uncertainty. The centralized training decentralized execution framework is deployed, which relies on offline centralization during simulation-based policy identification, but enables decentralization when the policies are deployed online to the real system. Results show that using multi-agent proximal policy optimization with a centralized critic leads to performance very close to that of a centralized data-driven solution and outperforms a distributed model-based solution in most cases while respecting the information constraints of the system.