Quantized Hopfield Networks and Tabu Search for Manufacturing Cell Formation Problems
The use of neural networks in cell manufacturing system design is not new. This paper presents a new application of Hopfield neural networks in cell formation problems design: the quantized Hopfield neural networks. This quantized Hopfield neural networks was primarily used in a hybrid approach with the 玊abu search method?in order to solve the cell formation problems using big sizes industrial data. The problem is formulated as a zero/one linear and integer programming model in order to minimize the dissimilarities between machines and/or parts. Our hybrid approach allows us to obtain optimal or near-optimal solutions very frequently and much more quickly than traditional Hopfield networks. The effectiveness of the suggested approach is in the flexibility that it brings to us like the resolution time of problems of big sizes, and, in the execution speed when we applied it.
flexible manufacturing systems metaheuristics artificial intelligence
Barthélemy H.ATEME-NGUEMA Thiên-My DAO
Department of Mechanical and Manufacturing Engineering,école de Technologie Supérieure University of Quebec,Montreal (Canada)
国际会议
北京
英文
2007-05-30(万方平台首次上网日期,不代表论文的发表时间)