A unified group technology implementation using the backpropagation learning rule of neural networks

Y. Kao, Y. B. Moon

Research output: Contribution to journalArticle

53 Scopus citations

Abstract

Two engineering problems in implementing Group Technology are part family formation and part classification. Regardless of the approach adopted for the formation and classification, a critical problem is how to maintain consistency. The consistency problem can be addressed most effectively if the formation and classification is a single procedure rather than two separate procedures. A feedforward neural network using the Backpropagation learning rule is adopted to automatically generate part families during the part classification process. The spontaneous generalization capability of the neural network is utilized in classifying the parts into the families and creating new families if necessary. A heuristic algorithm using the neural network is described with an illustrative example.

Original languageEnglish (US)
Pages (from-to)425-437
Number of pages13
JournalComputers and Industrial Engineering
Volume20
Issue number4
DOIs
StatePublished - 1991

ASJC Scopus subject areas

  • Computer Science(all)
  • Engineering(all)

Fingerprint Dive into the research topics of 'A unified group technology implementation using the backpropagation learning rule of neural networks'. Together they form a unique fingerprint.

  • Cite this