Neural Network를 이용한 기계-부품 그룹 형성에 관한 연구

Abstract

this paper is concerned with the solution of machine-part group formation based on similarity of operations, considering both inter-machine load balance and inter-cell movements. To form groups based on similarity of operations, ART2 (Adaptive Resonance theory 2) model which is a kind of Neural Network and is efficent in pattern clustering and adaptability to changes of patterns is used. and A module is connected with ART2 to reduce the load of bottlenect machine and inter-cell movements. Computational experience is provided to demonstrate the usefulness of the proposed model. And it is expected that flexibility of operations will be improved by alternative machine as a solution.