A new product family mining method based on PLM database

Product family (PF) is the most important part of product platform. A new method is proposed to mine PF based on multi-space product data in PLM database. Product structure tree (PST) and bill of material (BOM) are used as the data source. A PF can be obtained by mining physics space, logic space and attribute space of product data. In this work, firstly, a PLM database is described, consisting of data organization form, data structure, and data characteristics. Then the PF mining method introduces the sequence alignment techniques used in bio-informatics, which mainly includes data pre-processing, regularization, mining algorithm and cluster analysis. Finally, the feasibility and effectiveness of the proposed method are verified by a case study of high and middle pressure valve, demonstrating a feasible method to obtain PF from PLM database.