Machine Learning in Engineering Design: Learning Generalized Design Prototypes from Examples

  • M. L. Maher
Conference paper
Part of the International Centre for Mechanical Sciences book series (CISM, volume 333)


The use of machine learning in engineering design should be based on a recognition of what makes design different from problem solving in general and should be guided by a representation paradigm that is useful in solving engineering design problems. Recent research in knowledge-based design has identified concept-based representation paradigms consistent with a model of the design process; this paper focusses on the representation paradigm called design prototypes. Conceptual clustering is a machine learning approach that provides techniques for structuring observations into generalized concepts. This paper describes how a conceptual clustering program is extended to learn engineering design knowledge by clustering function, structure, and behavior attribute-value pairs. These clusters are then used as the basis for learning associations between function, structure and behavior, resulting in generalized design prototypes.


Machine Learning Technique Design Knowledge Inductive Learning Conceptual Cluster Design Prototype 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Alem, L. and Maher, M.L. (1991). “Using Conceptual Clustering to Learn about Function, Structure and Behavior in Design”, Kmet’91, First international Conference on Knowledge Modeling and Expertise Transfert, Sophia Antipolis, French Riviera, France, April 22–24, 1991.Google Scholar
  2. Brown, D. and Chandrasekaran, B. (1985). “Expert Systems for a Class of Mechanical Design Activity”, in Knowledge Engineering in Computer-Aided Design, (editor) J. Gero, North-Holland, pp. 259–283.Google Scholar
  3. Carbonell, J.G. (1990). “Introduction: Paradigms for Machine Learning” in Machine Learning Paradigms and Methods, (editor) J. Carbonell, MIT/Elsevier, pp. 1–10.Google Scholar
  4. Chandrasekaran, B. (1990), “Design Problem Solving: A Task Analysis” in AI Magazine, Winter Issue.Google Scholar
  5. Feigenbaum, E.A. and Simon, H. (1984). “EPAM-like Models of Recognition and Learning”, Cognitive Science, 8, pp. 305–336.CrossRefGoogle Scholar
  6. Fisher, D.H (1987). “Knowledge Acquisition via Incremental Conceptual Clustering ” in Machine Learning, 2, pp. 139–172.Google Scholar
  7. Forgy, C.L. (1981). “OPS5 User’s Manual”, Technical Report CMU-CS-81–135, Carnegie Mellon University, Pittsburgh PA.Google Scholar
  8. Gero, J.S., Maher, M.L. and Zhang, W. (1988). “Chunking Structural Design Knowledge as Prototypes” in Artificial Intelligence in Engineering: Design, (editor) J. Gero, Elsevier/Computational Mechanics Publications, pp 3–21.Google Scholar
  9. Gero, J. (1990), “Prototypes: A Knowledge Representation Schema for Design” in AI Magazine, Winter.Google Scholar
  10. Gluck, M. and Corter, J. (1985). “Information, Uncertainty and the Utility of Categories”, in Proceedings Seventh Annual Conference of the Cognitive Sciences Society, Ivrine, CA. pp. 283–287.Google Scholar
  11. Lebowitz M. (1987). “Experiments with Incremental Concept Formation: UNIMEM ” in Machine Learning, 2, pp. 103–138.Google Scholar
  12. Maher, M.L. and Fenves. S.J., (1984). “HI-RISE: A Knowledge-Based Expert System for the Preliminary Structural Design of High Rise Buildings”, Technical Report, R-85–146, Department of Civil Engineering, Carnegie Mellon University.Google Scholar
  13. Maher, M.L. (1988). “Engineering Design Synthesis: A Domain Independent Representation” in Artificial Intelligence, for Engineering Design, Analysis and Manufacturing, 1 (3), pp. 207–213.Google Scholar
  14. Maher, M.L. (1990). “Process Models of Design Synthesis” in AI Magazine, Winter Issue.Google Scholar
  15. Marcus, S., Stout, J., and McDermott, J. (1988). “VT: An Expert Elevator Designer That Uses Knowledge-Based Backtracking”. in AI Magazine 9 (1), pp. 95–114.Google Scholar
  16. McDermott, J. (1980). “R1: A Rule-Based Configurer of Computer Systems”, Technical Report CMU-CS-80–119, Carnegie Mellon University, Pittsburgh PA.Google Scholar
  17. Michalski, R.S. and Kodratoff, Y. (1990). “Research in Machine Learning; Recent Progress, Classification of Methods, and Future Directions”, in Machine Learning An Artificial Intelligence Appproach Volume III, (editors) Y. Kodratoff and R. Michalski, Morgan Kauffmann, pp. 3–30.Google Scholar
  18. Mitchell, T.M., Steinberg, L.I., and Shulman, J.S. (1984). “A Knowledge-based Appproach to Design” in Proceedings of the IEEE Workshop of Principles of Knowledge-based Systems, IEEE pp 27–34.Google Scholar
  19. Nyberg, E.H. (1988). “The Framekit User’s Guide Version 2.0”, Technical Report CMUCMT-MEMO, Carnegie Mellon University, Pittsburgh PA.Google Scholar
  20. Reich, Y. (1990). “Design Knowledge Acquisition: Task Analysis and a Partial Implementation”, in The 5th Knowledge Acquisition for Knowledge-Besed Systems Workshop, Banff, Canada.Google Scholar
  21. Tham, K.W., Lee, H.S., and Gero, J.S. (1990). “Building Envelope Design Using Design Prototypes” in AI in Building Design: Progress and Promise, ASHRAE Symposium, St. Louis, Missouri.Google Scholar

Copyright information

© Springer-Verlag Wien 1998

Authors and Affiliations

  • M. L. Maher
    • 1
  1. 1.University of SydneySydneyAustralia

Personalised recommendations