现代制造工程Issue(7)：20-30,19,12.DOI:10.16731/j.cnki.1671-3133.2025.07.003
面向多目标动态作业车间调度的强化学习决策算法研究
Research on reinforcement learning decision algorithm for multi-objective dynamic job shop scheduling
摘要
Abstract
To address the multi-objective dynamic job shop scheduling problem and meet the real-time scheduling needs of manu-facturing workshops in environments with variable scales,a method combining Proximal Policy Optimization(PPO)with GoogLeNet,named GLN-PPO,is proposed.This method constructs the state space of the scheduling problem using multidimen-sional matrices,designs an action space based on various priority rules,and devises a multi-objective reward function.To verify the effectiveness of the proposed algorithm,it is trained and tested in three environments:a static public environment based on common benchmark problems,a static real environment based on actual cases,and a dynamic real environment.Experimental re-sults show that compared to genetic algorithms,GLN-PPO can provide high-quality scheduling results,meet the real-time sched-uling requirements of enterprises,and adapt flexibly to environments with variable scales.关键词
深度强化学习/作业车间调度/GoogLeNet/近端策略优化Key words
deep reinforcement learning/job shop scheduling/GoogLeNet/Proximal Policy Optimization(PPO)分类
信息技术与安全科学引用本文复制引用
张宁宁,万卫兵,张梦晓,赵宇明..面向多目标动态作业车间调度的强化学习决策算法研究[J].现代制造工程,2025,(7):20-30,19,12.基金项目
科技部科技创新2030——"新一代人工智能"重大项目(2020AAA0109300) （2020AAA0109300）
