A Hybrid Hopfield Neural Networks Based Simulation Approach for Optimisation of Manufacturing Group Scheduling

以混合式霍克菲爾神經網路爲基礎之模擬方法求解製造群組排程最佳化問題

10.29977/JCIIE.200707.0005

Thien-My Dao；Cherif Makrem；Seraphin C. Abou

電腦整合製造 ； 單元製造 ； 群組技術 ； 類神經網路 ； 禁忌搜尋法 ； 電腦模擬 ； 生產控制 ； computer integrated manufacturing ； cellular manufacturing ； group technology ； neural network ； tabu search ； computer simulation ； production control

工業工程學刊

24卷4期（2007 / 07 / 01）

300 - 308

英文

一個混合式神經網被發展出來，並運用模擬技術來求解複雜的群組排程問題。目標為給定一特定數量的工作數及機器數，在滿足各種不同的限制情況下，找出最佳化的績效指標。本論文所建議的求解方法係根基於作業研究和人工智慧。在本論文中探討及展現了如何結合霍克菲爾神經網路(HNN)和禁忌搜尋法來定義出最佳的作業群組，以加速途程表之產生。針對每一個作業，霍克菲爾神經網路被用來產生可行的製造方案，並使用禁忌搜尋法來找出最佳排程。所建議的方法，亦即混合合霍克菲爾神經網路(HHNN)以及模擬技術，可使得作業間可節省時間和工作量之不均衡最小化。所建議的方法之潛在能力已透過求解具有大規模的群組排程問題例子來表示出來。

A hybrid Neural Networks have been developed and the simulation techniques have been used to solve complex group scheduling problems. The objectives are to find the best performance criterion optimum for a given number of jobs and machines in order to satisfy different production constraints. The solution methods proposed in this paper, have their roots in both operational research and artificial intelligence. The use of combining Hopfield Neural Networks (HNN) and Tabu (local search) approach to define optimal groups of operations which will facilitate the generation of the route sheet has been discussed and presented in this paper. For each one of the operations, the (HNN) will be used to generate the feasible manufacturing alternatives, and the Tabu search technique will be applied to identify the best schedule. The proposed approach, Hybrid Hopfield Neural Networks (HHNN) and the simulation techniques, save time and minimise unbalanced workloads among the operations. The potential ability of the proposed approach is shown using a numerical case example of a big size of group scheduling problem.

1. Abou, S. C.,T. M. Dao(2001).A new neuron network based approach for optimisation of manufacturing group scheduling.Proceedings of the IIE Annual Conference, Industrial Engineering Research Conference,Dallas, Texas, USA:
2. Binato, S.,W. J. Hery,D. M. Loewenstern,M. G. C. Resende,C. C. Ribeiro,P. Hansen (eds.)(2002).Essays and Surveys in Metaheuristics.Kluwer Academic.
3. Cheng, R.,M. Gen,Y. Tsujimura(1999).A tutorial survey of job-shop scheduling problems using genetic algorithms, part II: hybrid genetic search strategies.Computers & Industrial Engineering,36,343-364.
4. Dao, T. M.,S. Rodner,P. S. Benoit,M. Galopin(1995).A knowledge based simulation system for solution identification in group scheduling.Proceedings of the 14th IASTED International Conference on Modelling Identification and Control,Innsbruck, Austria:
5. Grossberg, S.(1988).Non-linear neural networks: Principles, mechanisms, and architectures.Neural Networks,1,17-61.
6. Hanada, A.,K. Ohnishi(1993).Near optimal job shop scheduling using neural network parallel computing.Proceedings of the 19th Annual Conference of IEEE Industrial Electronics,Maui, HI, USA:
7. Hopfield, J. J.,D. W. Tank,(1985).Neural computation of decisions in optimization problems.Biological Cibernetics,52,141-152.
8. Le Cun, Y.,E. Bienenstock (eds.)(1986).Disordered Systems and Biological Organisation.Springer-Verlag:New-York.
9. Nowicki, E.,C. Smutnicki.A fast tabu search algorithm for the job-shop scheduling problem.Management science,42,797-813.
10. Sun, K. H.,H. C. Fu,(1993).A hybrid neural model for solving optimisation problems.IEEE Transactions on Computer,42,219-227.
11. Wang, L.,D. Zheng(2001).An effective hybrid optimization strategy for job-shop scheduling problems.Computers & Operations Research,28,585-596.
12. Watanabe, Y.,K. Yoshimo,T. Kakeshita(1997).Solving combinatorial optimisation problems using oscillatory neural networks.IEICE Transaction on Information and Systems,E80-D,72-77.