Manufacturing Knowledge Subject Mining and Ranking for Mass Customization

Knowledge-based engineering has been recognized as an effective means to implement mass customization. This paper focuses on the mining and ranking of manufacturing documents to identify manufacturing knowledge with subject in order to support mass customization production effectively. With this view, a kind of manufacturing knowledge model based on manufacturing feature for mass customization was presented. A two-step procedure for manufacturing document was proposed namely subject mining and ranking. Manufacturing knowledge subject mining was performed through the processes of sample knowledge training, analyzing to unknown subject document, and subject similarity analysis and classification. At the same time, similarity measure for manufacturing feature attribute and feature manufacturing process term was formulated including similarity matrix and hybrid space vector model and so on. Then manufacturing knowledge subject mining flow was presented as well. Manufacturing knowledge ranking is to reorder manufacturing knowledge sequence by match priority within a subject based on their successful usage status so as to improve the reasonability of manufacturing knowledge utilization. Finally, we illustrate our approach with examples.