In a one-of-a-kind and order-orient ed production corporation, job shop scheduling plays an important role in the prod uction planning system and production process control. Since resource selection in job shop sche...In a one-of-a-kind and order-orient ed production corporation, job shop scheduling plays an important role in the prod uction planning system and production process control. Since resource selection in job shop scheduling directly influences the qualities and due dates of produc ts and production cost, it is indispensable to take resource selection into acco unt during job shop scheduling. By analyzing the relative characteristics of res ources, an approach of fuzzy decision is proposed for resource selection. Finall y, issues in the application of the approach are discussed.展开更多
针对具有模糊加工时间和模糊交货期的作业车间调度问题,以最小化最大完工时间为目标,以近端策略优化(PPO)算法为基本优化框架,提出一种LSTM-PPO(proximal policy optimization with Long short-term memory)算法进行求解.首先,设计一种...针对具有模糊加工时间和模糊交货期的作业车间调度问题,以最小化最大完工时间为目标,以近端策略优化(PPO)算法为基本优化框架,提出一种LSTM-PPO(proximal policy optimization with Long short-term memory)算法进行求解.首先,设计一种新的状态特征对调度问题进行建模,并且依据建模后的状态特征直接对工件工序进行选取,更加贴近实际环境下的调度决策过程;其次,将长短期记忆(LSTM)网络应用于PPO算法的行动者-评论者框架中,以解决传统模型在问题规模发生变化时难以扩展的问题,使智能体能够在工件、工序、机器数目发生变化时,仍然能够获得最终的调度解.在所选取的模糊作业车间调度的问题集上,通过实验验证了该算法能够取得更好的性能.展开更多
文摘In a one-of-a-kind and order-orient ed production corporation, job shop scheduling plays an important role in the prod uction planning system and production process control. Since resource selection in job shop scheduling directly influences the qualities and due dates of produc ts and production cost, it is indispensable to take resource selection into acco unt during job shop scheduling. By analyzing the relative characteristics of res ources, an approach of fuzzy decision is proposed for resource selection. Finall y, issues in the application of the approach are discussed.
文摘针对具有模糊加工时间和模糊交货期的作业车间调度问题,以最小化最大完工时间为目标,以近端策略优化(PPO)算法为基本优化框架,提出一种LSTM-PPO(proximal policy optimization with Long short-term memory)算法进行求解.首先,设计一种新的状态特征对调度问题进行建模,并且依据建模后的状态特征直接对工件工序进行选取,更加贴近实际环境下的调度决策过程;其次,将长短期记忆(LSTM)网络应用于PPO算法的行动者-评论者框架中,以解决传统模型在问题规模发生变化时难以扩展的问题,使智能体能够在工件、工序、机器数目发生变化时,仍然能够获得最终的调度解.在所选取的模糊作业车间调度的问题集上,通过实验验证了该算法能够取得更好的性能.
