期刊
COMPUTERS & OPERATIONS RESEARCH
卷 39, 期 7, 页码 1315-1324出版社
PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.cor.2011.07.019
关键词
Scheduling; Unrelated parallel machines; Reinforcement learning; Tardiness
类别
资金
- Ministry of Education [10YJC630405]
- Natural Science Foundation of Guangdong Province, China [9451170003003938]
- Natural Science Foundation of China [50375082, 70771058, 60834004, 71002037]
- 863 Program of China [2008AA04Z102]
- Natural Science Foundation for Young Scholars of Dongguan University of Technology [2010ZQ18]
We address an unrelated parallel machine scheduling problem with R-learning, an average-reward reinforcement learning (RL) method. Different types of jobs dynamically arrive in independent Poisson processes. Thus the arrival time and the due date of each job are stochastic. We convert the scheduling problems into RL problems by constructing elaborate state features, actions, and the reward function. The state features and actions are defined fully utilizing prior domain knowledge. Minimizing the reward per decision time step is equivalent to minimizing the schedule objective, i.e. mean weighted tardiness. We apply an on-line R-learning algorithm with function approximation to solve the RL problems. Computational experiments demonstrate that R-learning learns an optimal or near-optimal policy in a dynamic environment from experience and outperforms four effective heuristic priority rules (i.e. WSPT, WMDD, ATC and WCOVERT) in all test problems. (C) 2011 Elsevier Ltd. 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