Abstract
In this study, two manufacturing systems, a kanban-controlled system and a multi-stage, multi-server production line in a diamond tool production system, are optimized utilizing neural network metamodels (tst_NNM) trained via tabu search (TS) which was developed previously by the authors. The most widely used training algorithm for neural networks has been back propagation which is based on a gradient technique that requires significant computational effort. To deal with the major shortcomings of back propagation (BP) such as the tendency to converge to a local optimal and a slow convergence rate, the TS metaheuristic method is used for the training of artificial neural networks to improve the performance of the metamodelling approach. The metamodels are analysed based on their ability to predict simulation results versus traditional neural network metamodels that have been trained by BP algorithm (bp_NNM). Computational results show that tst_NNM is superior to bp_NNM for both of the manufacturing systems.
Similar content being viewed by others
References
Alabas C, Altiparmak F and Dengiz B (2002). A comparison of the performance of artificial intelligence techniques for optimizing the number of kanbans . J Opl Res Soc 53: 907–914.
Alam F, McNaught KR and Ringrose TJ (2004). A comparison of experimental designs in the development of a neural network simulation metamodel . Simul Model Pract Theory 12: 559–578.
Altiparmak F, Dengiz B and Bulgak AA (2007). Buffer allocation and performance modeling in asynchronous assembly system operations: An artificial neural network metamodeling approach . Appl Soft Comput 7: 946–956.
Aytug H, Dogan CA and Bezmez G (1996). Determining the number of kanbans: A simulation metamodeling approach . Simulation 67: 23–32.
Blanning RW (1975). The construction and implementation of metamodels . Simulation 24: 177–184.
Chambers M and Mount-Campbell CA (2002). Process optimization via neural network metamodeling . Int J Prod Econ 79: 93–100.
Chan KK and Spedding TA (2001). On-line optimization of quality in a manufacturing system . Int J Prod Res 39: 1127–1145.
Chen MC and Yang T (2002). Design of manufacturing systems by a hybrid approach with neural network metamodelling and stochastic local search . Int J Prod Res 40: 71–92.
Dengiz B, Bektas T and Ultanir E (2006). Simulation optimization based DSS application: A diamond tool production line in industry . Simul Model Pract Theory 14: 296–312.
Dengiz B, Alabas-Uslu C and Dengiz O (2008). A tabu search algorithm for the training of neural networks. J Opl Res Soc, advance online publication, 16 January 2008, doi:10.1057/palgrave.jors.2602535.
Fonseca DJ, Navaresse DO and Moynihan GP (2003). Simulation metamodeling through artificial neural networks . Eng Appl Artif Intel 16: 177–183.
Fukuoka Y, Matsuki H, Minamitani H and Ishida A (1998). A modified back propagation method to avoid false local minima . Neural Networks 11: 1059–1072.
Glover F (1989). Tabu search—Part I . ORSA J Comput 1: 190–206.
Glover F (1990). Tabu search—Part II . ORSA J Comput 2: 4–32.
Haouani M, Ferney M, Zerhounin S and Elmoudni A (1995). Control of manufacturing systems using neural networks . In: Proceedings of the 1995 EUROSIM Conference. Elsevier: Austria.
Haykin S (1999). Neural Networks: A Comprehensive Foundation . Prentice-Hall: New Jersey.
Hurrion RD (1992). Using a neural network to enhance the decision making quality of a visual interactive simulation model . J Opl Res Soc 43: 333–341.
Hurrion RD (1997). An example of simulation optimization using a neural network metamodel: Finding the optimum number of kanbans in a manufacturing system . J Opl Res Soc 48: 1105–1112.
Hurrion RD and Birgil S (1999). A comparison of factorial and random experimental design methods for the development of regression and neural network simulation metamodels . J Opl Res Soc 50: 1018–1023.
Hush DR (1999). Training a sigmodial node is hard . Neural Comput 11: 1249–1260.
Jang KY, Yang K and Kang C (2003). Application of artificial neural network to identify non-random variation patterns on the run chart in automotive assembly process . Int J Prod Res 41: 1239–1254.
Kilmer RA, Smith AE and Shuman LJ (1994). Neural networks as a metamodeling technique for discrete event stochastic simulation . Intell Eng Syst Artif Neural Networks 4: 1141–1146.
Kirkpatrick S, Gelatt CD Jr and Vecchi MP (1983). Optimization by simulated annealing . Science 220: 671–680.
Kuo Y, Yang T, Peters BA and Chang I (2007). Simulation metamodel development using uniform design and neural networks for automated material handling systems in semiconductor wafer fabrication . Simul Model Pract Theory 15: 1002–1015.
Laguna M and Marti R (2002). Neural network prediction in a system for optimizing simulations . IIE Trans 34: 273–282.
Lee I and Shaw M (2000). A neural-net approach to real time flow-shop sequencing . Comput Ind Eng 38: 125–147.
Mollaghasemi M, LeCroy K and Georgiopoulos M (1998). Application of neural networks and simulation modeling in manufacturing system design . Interfaces 28: 100–114.
Monostori L and Viharos ZJ (2001). Hybrid AI-and simulation-supported optimization of process chain and production plants . Ann CIRP 50: 353–356.
Park Y, Kim S and Lee YH (2000). Scheduling jobs on parallel machines applying neural network and heuristic rules . Comput Ind Eng 38: 189–202.
Pierreval H and Huntsinger RC (1992). An investigation on neural network capabilities as simulation metamodels . In: Luker P. (ed). Proceedings of 1992 Summer Computer Simulation Conference. Society for Computer Simulation: Reno, Nevada, pp. 413–417.
Sabuncuoglu I and Touhami S (2002). Simulation metamodelling with neural networks: An experimental investigation . Int J Prod Res 40: 2483–2505.
Stone M (1974). Cross-validatory choice and assessment of statistical predictions . Journal of the Royal Statistical Society B36: 111–133.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Dengiz, B., Alabas-Uslu, C. & Dengiz, O. Optimization of manufacturing systems using a neural network metamodel with a new training approach. J Oper Res Soc 60, 1191–1197 (2009). https://doi.org/10.1057/palgrave.jors.2602620
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1057/palgrave.jors.2602620