摘要
提出一个基于神经元动态规划解决可重入生产系统调度问题的仿真框架.根据可重入生产系统的特点建立状态集,并将调度问题表示成相应的马尔可夫决策过程.选择合理的性能指标,采用神经元动态规划产生每一步的调度,并在仿真中优化策略.仿真算例验证了该方法的有效性,三种调度策略的结果比较表明了神经元动态规划方法的优越性.本仿真框架还可拓展至其他类型的生产调度问题.
A simulation framework of scheduling re-entrant lines with Neuro-Dynamic Programming （NDP） is presented. The state set is constructed based on the characteristics of re-entrant lines, and the scheduling problem is described as the corresponding Markov decision process. Proper performance index is selected, and each scheduling decision is produced and improved in simulation with NDP. An example is given to illustrate the validity of the method. The comparison of three scheduling policies indicates the superiority of NDP in scheduling re-ehtrant lines. The presented simulation framework can be extended to scheduling of other kinds of production systems.
出处
《信息与控制》
CSCD
北大核心
2007年第2期218-223,共6页
Information and Control
关键词
仿真框架
可重入生产系统
神经元动态规划
调度
simulation framework
re-entrant line
neuro-dynamic programming
scheduling
