期刊
DECISION SUPPORT SYSTEMS
卷 45, 期 4, 页码 949-959出版社
ELSEVIER
DOI: 10.1016/j.dss.2008.03.007
关键词
Supply chain; Ordering policy; Multi-agent systems; Beer game; Reinforcement learning
A major challenge in supply chain ordering management is the coordination of ordering policies adopted by each level of the chain, so as to minimize inventory costs. This paper describes a new approach to decide on ordering policies of supply chain members in an integrated manner. In the first step supply chain ordering management has been considered as a multi-agent system and formulated as a reinforcement learning (RL) model. In the final step a Q-learning algorithm is proposed to solve the RL model. Results show that the reinforcement learning ordering mechanism (RLOM) is better than two other known algorithms. (C) 2008 Elsevier B.V. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据