开环可重入排队网络的递阶增强型学习调度

Hierarchical Reinforcement Learning Scheduling ofOpen Re-entrant Queueing Network

对平均费用型马氏决策过程 ,研究了一种递阶增强型学习算法 ;并将算法应用于一个两台机器组成的开环可重入生产系统 ,计算机仿真结果表明 。

In this paper, a hierarchical reinforcement learning algorithm is investigated for Markov decision process with average cost. And it is applied to an open re-entrant manufacturing system composed of two machines as an example. Computer simulation results demonstrate that the algorithm outperforms some well-known heuristic scheduling policies.