Variant Design and Data Mining

•Description
•Grant Support
•Personnel
•Papers and Talks

Description

In the distributed and horizontally integrated manufacturing environment in Agile Manufacturing (AM), there is a great demand for new product development methods that are capable of generating new customized assembly designs based on mature component designs that might be dispersed at geographically distributed partner sites.

Assembly Modeling: Fundamental research in models that encapsulate pertinent assembly information to support a number of downstream activities (assembly sequence generation and evaluation, tolerance chain analysis, etc.) in addition to the variant design methodology. More specifically, the complementary assembly modeling concept is proposed with two kinds of assembly models, the Hierarchical Assembly Model (HAM) and the Relational Assembly Model (RAM). The HAM explicitly captures the hierarchical and functional relationships between constituent components while the RAM explicitly captures the mating relationships at the manufacturing feature level. While all concepts here are not new, the concept of open modeling and detailed mating relationships captured at the manufacturing feature level advance the current state of assembly modeling.
Assembly variant design methodology is based on these assembly models mentioned above. The matching components are searched and retrieved based on the Assembly Variants Model (an instance of HAM) and then the Constraint Groups (CG) are identified by manipulating the Assembly Mating Graph (an instance of RAM). Graph theoretic tools/algorithms are invoked for automatic CG extraction. Then the assembly variant design process encodes the CG information along with redesign cost/time data to formulate a Mixed Integer Linear Programming (MILP) problem which is solved using a standard solver (or heuristic). This methodology provides a systematic approach to facilitate the variant design of complex assembly products in the agile manufacturing environment. Finally, a prototype system is developed and examples are presented. While variant design of piece parts has been prevalent, to our belief, this work is the first systematic optimization-based methodology for mechanical assemblies.
Data-mining methods in variant design: Our research proposes using data mining to incorporate heterogeneous and distributed information into the beginning stages of design – thereby reducing iterations and lowering the cost of product design.

Grant Support

  • “Cold-box and Component Classification Scheme,” Praxair, Inc., Tonawanda, NY, $14,971 (June 1996 – August 1996); co-PI L. Lin. Two students supported in summer.
  • “Optimal Selection of Partners in Agile Manufacturing,” US Army Tank-Automotive Command Research Contract to University of Maryland; co-Principal Investigator with Harhalakis, G. and I. Minis (PI); Subcontract from University of Maryland, $68,199 of total $562,216 for 2 years (August 1993 – September 1995). Two graduate students supported.
  • Steptools, Inc. (student support)

Personnel

Doctoral Students

  • Aihu Wang; Graduated 11/01. Dissertation title: “Variant Design of Complex Assemblies in Agile Manufacturing.” Starting position: Faculty at School of Business Administration, South China University of Technology, Guangzhou, China.
  • Carol Romanowski; Graduated 5/04. Dissertation title: “A Data Mining-based Engineering Design Support System.” Recipient of the 1999 Department of Energy (DOE) Doctoral Fellowship (for three years). A-Exam: November 2000. Currently: Lecturer, Department of Industrial Engineering, University at Buffalo.

MS Students

  • Mangesh Bhandarkar; graduated 6/97. Thesis title: “Satisfying Data needs in Agile Manufacturing through Translation and Feature Extraction into STEP Product Models.” Currently at Netfish Technologies, Santa Clara, CA.
  • Vishwanathan Ramabhatta; graduated 7/97. Thesis title: “An Object Hierarchy Scheme for Complex Assemblies to aid Information Retrieval and Work Process Database Integration,” (co-advised with Li Lin). Currently Senior Consulting Engineer at Aspect Corp., Nashua, NH.
  • Shun L. Ng; graduated 7/97. Thesis title: “A Web-Centric System for Concurrent Engineering of Transfer Molded Components;” (co-advised with Li Lin).
  • Sanjay Ramanujan; graduated 1/01. Thesis title: “Decision Support System for Integrated Assembly Plan Generation and Task Scheduling.” Currently at ILOG, Mountainview, CA.
  • Sushant Bhadange; graduated 1/01. Thesis title: “Disassembly Sequence Generation based on Mating Graph and Access Directions.” Currently Product Development Engineer at Reveo, Inc., Hawthorne, NY.
  • Kedar Sambhoos; expected graduation 5/04. Tentative thesis title: “Precedence relationships in assembly mating graphs.” (co-advised with Bahattin Koc).

Papers and Talks

Papers:

Journal Papers:

  • Romanowski, C.J. and Nagi, R. “A Data Mining Approach to forming Generic Bills of Materials in support of Variant Design Activities,” submitted to ASME Journal of Computing and Information Science in Engineering, 2004, Vol. 4(4), pp. 316-328.Download(PDF)
  • Romanowski, C.J., Nagi, R. and Sudit, M., “Data Mining in an Engineering Design Environment: OR Applications from Graph Matching,” accepted to Computers & Operations Research, Dec 2004. Download(PDF)
  • Romanowski, C.J. and Nagi, R., “On Comparing Bills of Materials: A Similarity/Distance Measure for Unordered Trees,” IEEE Transactions on Systems, Man, and Cybernetics, Part A, 2005, Vol. 35(2), pp. 249-260. Download(PDF)
  • Wang, A., Koc, B. and Nagi, R. “Complex Assembly Variant Design in Agile Manufacturing. Part I: System Architecture and Assembly Modeling Methodology,” IIE Transactions on Design and Manufacturing, 2005, Vol. 37(1), pp. 1-16. Download(PDF)
  • Wang, A., Koc, B. and Nagi, R. “Complex Assembly Variant Design in Agile Manufacturing. Part II: Assembly Variant Design Methodology,” IIE Transactions on Design and Manufacturing, 2005, Vol. 37(1), pp. 17-34. Download(PDF)
  • Romanowski, C.J. and Nagi, R., “A Data Mining-based Engineering Design Support System: A Research Agenda,” Data Mining for Design and Manufacturing: Methods and Applications, Editor: D. Braha, Kluwer Academic Publishers, 2001, pp. 161-178. (Chapter)
  • Bhandarkar, M.P., Downie, B., Hardwick, M. and Nagi, R., “Migration from IGES to STEP: One-to-one Translation of IGES Drawing to STEP Drafting Data,” Computers in Industry, 2000, Vol. 41, pp. 261-277.
  • Bhandarkar, M.P. and Nagi, R., “STEP-based Feature Extraction from STEP geometry for Agile Manufacturing,” Computers in Industry, 2000, Vol. 41, pp. 3-24.
  • Iyer, S. and Nagi, R., “Automated Retrieval and Ranking of Similar Parts in Agile Manufacturing,” IIE Transactions on Design and Manufacturing, special issue on Agile Manufacturing, 1997, Vol. 29(10), pp. 859-876.
  • Bhandarkar, M.P. and Nagi, R., “STEP Product Information Models in Agile Manufacturing,” International Journal of Agile Manufacturing, 1997, Vol. 1(1), pp. 65-75.
  • Ramabhatta, V., Lin, L. and Nagi, R., “Object Hierarchies to aid Representation and Variant Design of Complex Assemblies in an Agile Environment,” International Journal of Agile Manufacturing, 1997, Vol. 1(1), pp. 77-90.

Submitted Journal Papers:

  • .

Conference Papers (Full Paper Reviewed):

  • Romanowski, C.J. and Nagi, R., “The Search for Similar Parts in Variant Design,” 12th Industrial Engineering Research Conference, Portland OR, May 2003.
  • Romanowski, C.J. and Nagi, R., “A data mining and graph theoretic approach to building generic bills of materials,” 11th Industrial Engineering Research Conference, Orlando FL, May 2002.
  • Romanowski, C.J. and Nagi, R., “A Data Mining Approach to Integrating Product Life Cycle Information in Engineering Design,”10th Industrial Engineering Research Conference, Dallas TX, May 2001.
  • Wang, A. and Nagi, R., “Assembly Modeling for Complex Assembly Variant Design,” 10th Industrial Engineering Research Conference, Dallas TX, May 2001.
  • Wang, A. and Nagi, R., “Framework for Complex Assembly Variant Design in Agile Manufacturing,” 10th Industrial Engineering Research Conference, Dallas TX, May 2001.
  • Ramabhatta, V., Lin, L. and Nagi, R., “An Object Hierarchy Scheme for Complex Assemblies to aid Information Retrieval and Work Process Database Integration,” 6th Industrial Engineering Research Conference, Miami, FL, May 1997, pp. 650-655.
  • Bhandarkar, M.P., Downie, B., Hardwick, M. and Nagi, R., “IGES to STEP Translation and Visualization,” 6th Industrial Engineering Research Conference, Miami, FL, May 1997, pp. 656-661.
  • Bhandarkar, M.P. and Nagi, R., “STEP Product Information Models in Agile Manufacturing,” International Conference on Agile Manufacturing ICAM’97, Lafayette, LA, February 1997.
  • Ramabhatta, V., Lin, L. and Nagi, R., “Object Hierarchies to aid Representation and Variant Design of Complex Assemblies in an Agile Environment,” International Conference on Agile Manufacturing ICAM’97, Lafayette, LA, February 1997.
  • Iyer, S. and Nagi, R., “Identification and Ranking of Similar Parts in Agile Manufacturing,” 4th Industrial Engineering Research Conference, Nashville, TN, May 1995, pp. 709-718.
  • Iyer, S. and Nagi, R., “Identification of Similar Parts in Agile Manufacturing,” ASME Winter Annual Conference: Concurrent Product Design DE-Vol. 74, Chicago, November 1994, pp. 87-96.

Conference Papers (Full Paper, Abstract Reviewed):

  • Romanowski, C.J. and Nagi, R., “Improving Preventive Maintenance Scheduling using Data Mining Techniques,” 8th Industrial Engineering Research Conference, Phoenix AZ, May 1999.
  • Candadai, A., Herrmann, J.W., Minis, I. and Nagi, R., “Variant Design Critiquing for Agile Manufacturing,” NSF Design and Manufacturing Grantees Conference, San Diego, January 1995.

Leave a Reply