DocumentCode
2233454
Title
Reinforcement learning approach to re-entrant manufacturing system scheduling
Author
Liu, Chang-chun ; Jin, Hui-yu ; Tian, Yu ; Yu, Hai-Bin
Author_Institution
Shenyang Inst. of Autom., Chinese Acad. of Sci., Shenyang, China
Volume
3
fYear
2001
fDate
2001
Firstpage
280
Abstract
In this paper, we focus on the problem of optimally scheduling a closed re-entrant system with one type of parts and two service centers, each of which consisting of one machine. An algorithm based on reinforcement learning is proposed. The results of the experiments indicate that reinforcement learning can outperform some familiar heuristic methods and is closed to the workload balancing policy
Keywords
computer aided production planning; dynamic programming; learning (artificial intelligence); production control; dynamic programming; heuristic methods; reentrant manufacturing system; reinforcement learning; scheduling; temporal difference learning; workload balancing policy; Cities and towns; Dynamic programming; Heuristic algorithms; Job shop scheduling; Learning; Manufacturing automation; Manufacturing systems; Scheduling algorithm; Semiconductor device manufacture; Uncertainty;
fLanguage
English
Publisher
ieee
Conference_Titel
Info-tech and Info-net, 2001. Proceedings. ICII 2001 - Beijing. 2001 International Conferences on
Conference_Location
Beijing
Print_ISBN
0-7803-7010-4
Type
conf
DOI
10.1109/ICII.2001.983070
Filename
983070
Link To Document