한국정밀공학회:학술대회논문집 (Proceedings of the Korean Society of Precision Engineering Conference)
- 한국정밀공학회 1997년도 춘계학술대회 논문집
- /
- Pages.432-437
- /
- 1997
- /
- 2005-8446(pISSN)
경쟁 학습 신경회로망을 이용한 기계-부품군 형성에 관한 연구
Machine-Part Cell Formation by Competitive Learning Neural Network
초록
In this paper, Fuzzy ART which is one of the competitive learing neural networks is applied to machine-part cell formation problem. A large matrix and varios types of machine-part incidence matrices, especially including bottle-neck machines, bottle-neck parts, parts shared by several cells, and machines shared by several cells are used to test the performannce of Fuzzy ART neural network as a cell formation algorithm. The result shows Fuzzy ART neral network can be efficiently applied to machine-part cell formation problem which are large, and/or have much imperfection as exceptions, bottle-neck machines, and bottle-neck parts.
키워드