%0 Journal Article
%T 一种基于强化学习的作业车间动态调度方法
%A 魏英姿
%A 赵明扬
%J 自动化学报
%P 765-771
%D 2005
%X ？Productionschedulingiscriticaltomanufacturingsystem.Dispatchingrulesareusuallyapplieddynamicallytoschedulethejobinadynamicjob-shop.Existingschedulingapproachessel-domaddressmachineselectionintheschedulingprocess.Compositerules,consideringbothmachineselectionandjobselection,areproposedinthispaper.Thedynamicsystemistrainedtoenhanceitslearningandadaptivecapabilitybyareinforcementlearning(RL)algorithm.Wedefinetheconceptionofpressuretodescribethesystemfeature.Designingarewardfunctionshouldbeguidedbytheschedulinggoaltoaccuratelyrecordthelearningprogress.CompetitiveresultswiththeRL-basedapproachshowthatitcanbeusedasreal-timeschedulingtechnology.
%K Reinforcementlearning
%K compositerules
%K meantardiness
%K dynamicjob-shopscheduling
%U http://www.aas.net.cn/CN/abstract/abstract15981.shtml