A Decomposition-Based Deep Reinforcement Learning Framework for Omnichannel Replenishment and Fulfillment
Author(s)
Kolyaei, Maryam
Advisor(s)
Editor(s)
Collections
Supplementary to:
Abstract
Managing replenishment and fulfillment across multiple sales channels requires carefully balancing inventory flow to minimize stockouts and overstocks. Replenishment involves determining order quantity in cycles, while fulfillment focuses on meeting uncertain demand across channels with limited capacity, both posing significant challenges. We address this problem in large-scale networks and propose a decomposition-based Deep Reinforcement Learning (DRL) algorithm. Our framework enhances demand management, scalabilty, and, and efficiency, ensuring robust performance under varying demand conditions.
Sponsor
Date
2025-06
Extent
Resource Type
Text
Resource Subtype
Proceedings
Rights Statement
Unless otherwise noted, all materials are protected under U.S. Copyright Law and all rights are reserved