시스템 특성함수 기반 평균보상 TD(λ) 학습을 통한 유한용량 Fab 스케줄링 근사화 = Capacitated Fab Scheduling Approximation using Average Reward TD(λ) Learning based on System Feature Functions

다국어 초록 (Multilingual Abstract)

In this paper, we propose a logical control-based actor-critic algorithm as an efficient approach for the approximation of the capacitated fab scheduling problem. We apply the average reward temporal-difference learning method for estimating the relative value functions of system states, while avoiding deadlock situation by Banker's algorithm. We consider the Intel mini-fab re-entrant line for the evaluation of the suggested algorithm and perform a numerical experiment by generating some sample system configurations randomly. We show that the suggested method has a prominent performance compared to other well-known heuristics.

목차 (Table of Contents)

• 1. 서론
• 2. 문제 정의 및 관련 연구 동향
• 3. NDP 모델링 및 근사화 알고리즘
• 4. 실험 설계 및 결과
• 5. 결론
• 참고문헌

참고문헌 (Reference)

1 Reveliotis,S.A, "The destabilizing effect of blocking due to finite buffering capacity in multi-class queueing networks" 45 (45): 585-588, 2000

2 Wein,L.M, "Scheduling semiconductor wafer fabrication" 1 (1): 115-130, 1988

3 Kumar,P.R, "Scheduling semiconductor manufacturing plants" 14 (14): 33-40, 1994

4 Kumar,P.R, "Scheduling manufacturing systems of re-entrant lines, In Stochastic Modeling and Analysis of Manufacturing Systems" Springer-Verlag 325-360, 1994

5 Choi,J.Y, "Relative value function approximation for the capacitated re-entrant line scheduling problem" 2 (2): 285-299, 2005

6 Sutton,R.S, "Reinforcement Learning(An Introduction)" MIT Press 1999

7 Reveliotis, S. A, "Real-time management of resource allocation systems" Springer 2005

8 Kumar,S, "Queueing network models in the design and analysis of semiconductor wafer fabs" 17 (17): 548-561, 2001

9 Jolliffe, I. T, "Principal Component Analysis" Springer-Verlag 2002

10 Bertsekas,D.P, "Neuro-Dynamic Programming" Athena Scientific 1996

11 Puterman, M. L, "Markov Decision Processes : Discrete Stochastic Dynamic Programming" Wiley 1994

12 Kumar,S, "Fluctuation smoothing policies are stable for stochastic re-entrant lines" 6 : 361-370, 1996

13 Tsitsiklis,J.N, "Feature-based methods for large scale dynamic programming" 22 : 59-94, 1996

14 Lu,S.H, "Efficient scheduling policies to reduce mean and variance of cycle-time in semiconductor manufacturing plants" 7 (7): 374-385, 1994

15 Lu,S.H, "Distributed scheduling based on due dates and buffer priorities" 36 (36): 1406-1416, 1991

16 Choi, J. Y, "Computationally efficient neuro-dynamic programming approximation method for the capacitated re-entrant line scheduling problem"

17 Tsitsiklis,J.N, "Average cost temporaldifference learning" 35 : 1799-1808, 1999

18 Rossetti,M.D, "A simulation-based approximate dynamic programming approach for the control of the intel mini-fab benchmark model" 2009