Uncertain association rule mining algorithm for the cell formation problem in cellular manufacturing systems

Abstract

Although data mining has enjoyed popularity in recent years with advances in both academia and industry, the application of data mining to cellular manufacturing, one of the most powerful management innovations in job-shop or batch-type production, is still under-utilized. Based on association rule mining, Chen initially developed a cell formation approach. One problem of such a cell formation algorithm is that various real-life production factors were ignored. In this paper we propose a new cell formation algorithm by way of uncertain association rule mining. The proposed algorithm incorporates several key production factors, such as operation sequence, production volume, batch size, alternative process routings, cell size, the number of cells, and the path coefficient of material flow. The efficacy and efficiency of the proposed algorithm were tested using several numerical problems.

Keywords:

• Data mining
• Uncertain association rule mining
• Cellular manufacturing
• Cell formation

Acknowledgements

We gratefully acknowledge Professor Mu-Chen Chen for his help. We thank the anonymous referee for his many constructive criticisms and helpful comments. We also express our gratitude to the Editor, John Middle for his thoughtful and careful editorial work and valuable advice concerning this paper. The first author is partially supported by a Grant-in-Aid for Special Research from the 21st Century Center of Excellence (COE) Program ‘Exploring New Science by Bridging the Particle–Matter Hierarchy’ from the Ministry of Education, Culture, Sports, Science and Technology.