考虑劣化维护的单机调度深度强化学习模型和算法

陈勇,杜习之,姜一炜,易文超,裴植,纪祖臻

Deep reinforcement learning models and algorithms for single-machine scheduling considering deteriorated maintenance

Yong CHEN,Xizhi DU,Yiwei JIANG,Wenchao YI,Zhi PEI,Zuzhen JI

表 9 DRL方法的成本优化效果

Tab.9 Cost optimization effect of DRL methods

规模 A2C DQN PPO $\Delta {\mathrm{mean}} $/% $\Delta \min $/% $\Delta {\mathrm{mean}} $/% $\Delta \min $/% $\Delta {\mathrm{mean}} $/% $\Delta \min $/% 10 −0.23 0.00 −12.31 0.00 −0.78 0.00 20 7.25 1.01 5.21 0.33 8.01 0.44 30 9.53 3.25 7.98 2.76 11.88 3.25 50 10.00 −0.27 −46.41 −9.44 12.73 0.98 80 9.57 −0.59 6.25 −0.76 12.38 1.03 100 8.33 −1.27 6.96 0.87 12.43 1.53 150 12.00 2.90 8.33 3.61 12.83 3.45