A data mining approach to forming generic bills of materials in support of variant design activities

Carol J. Romanowski, Rakesh Nagi

Research output: Contribution to journalArticle

Abstract

In variant design, the proliferation of bills of materials makes it difficult for designers to find previous designs that would aid in completing a new design task. This research presents a novel, data mining approach to forming generic bills of materials (GBOMs), entities that represent the different variants in a product family and facilitate the search for similar designs and configuration of new variants. The technical difficulties include: (i) developing families or categories for products, assemblies, and component parts; (ii) generalizing purchased parts and quantifying their similarity; (iii) performing tree union; and (iv) establishing design constraints. These challenges are met through data mining methods such as text and tree mining, a new tree union procedure, and embodying the GBOM and design constraints in constrained XML. The paper concludes with a case study, using data from a manufacturer of nurse call devices, and identifies a new research direction for data mining motivated by the domains of engineering design and information.

Original languageEnglish (US)
Pages (from-to)316-328
Number of pages13
JournalJournal of Computing and Information Science in Engineering
Volume4
Issue number4
DOIs
StatePublished - Dec 1 2004

    Fingerprint

Keywords

  • Data Mining
  • Generic Bills of Material (GBOM)
  • Tree Mining
  • Variant Design

ASJC Scopus subject areas

  • Software
  • Computer Science Applications
  • Computer Graphics and Computer-Aided Design
  • Industrial and Manufacturing Engineering

Cite this