Supply Chain Synchronization Through Deep Reinforcement Learning

Abstract

Synchronized supply chains can mitigate a cascading rise-and-fall inventory dynamic and prevent cycles of over and under-production. This paper demonstrated that a deep reinforcement learning agent could only perform adaptive coordination along the whole supply chain if end-to-end information transparency is ensured. Operational and strategic disruptions caused by the COVID-19 pandemic and the post-pandemic recovery can become a necessary kick-starter for required changes in information transparency and global coordination. This paper explores the capabilities of deep reinforcement learning agents to synchronize commodity flows and support operational continuity in the stochastic and nonstationary environment if end-to-end visibility is provided. The paper concludes that the proposed solution can perform adaptive control in complex systems and have potential in supply chain management and logistics. Among discovered benefits, it is essential to highlight that the proximal policy optimization is universal, task unspecific, and does not require prior knowledge about the system.