0
Research Papers

Learning Design Rules With Explicit Termination Conditions to Enable Efficient Automated Design

[+] Author and Article Information
Kevin Rawson

Department of Mechanical Engineering, University of California, Riverside, CA 92521Kevin.Rawson@ngc.com

Thomas F. Stahovich

Department of Mechanical Engineering, University of California, Riverside, CA 92521stahov@engr.ucr.edu

LearnIT-II can be used with any modeling and analysis software that provides API’s for reading and writing the values of the design parameters and for reading the values of the design constraints. To simplify software development, we utilized our own analysis software in our experiments.

If there are n constraints, there will be 2n possible values for the combined constraint state. In theory, this could lead to overfitting of the decision tree. In practice, it is likely that only a small number of the possible values would be observed, and thus overfitting will be avoided. If necessary, a decision tree pruning technique could be used if this sort of overfitting becomes an issue.

When learning decision trees with continuous-valued attributes, there are no unique choices for the best inequality tests for those attributes. The rule loosening approach is intended to help with this ambiguity.

J. Mech. Des 131(3), 031011 (Mar 09, 2009) (11 pages) doi:10.1115/1.3066681 History: Received February 16, 2008; Revised October 28, 2008; Published March 09, 2009

We present a two-step technique for learning reusable design procedures from observations of a designer in action. This technique is intended for the domain of parametric design problems in which the designer iteratively adjusts the parameters of a design so as to satisfy the design requirements. In the first step of the two-step learning process, decision tree learning is used to infer rules that predict which design parameter the designer is likely to change for any particular state of an evolving design. In the second step, decision tree learning is again used, but this time to learn explicit termination conditions for the rules learned in the first step. The termination conditions are used to predict how large of a parameter change should be made when a rule is applied. The learned rules and termination conditions can be used to automatically solve new design problems with a minimum of human intervention. Experiments with this technique suggest that it can reproduce the decision making process observed from the designer, and it is considerably more efficient than the previous technique, which was incapable of learning explicit rule termination conditions. In particular, the rule termination conditions allow the new program to automatically solve design problems with far fewer iterations than previously required.

FIGURES IN THIS ARTICLE
<>
Copyright © 2009 by American Society of Mechanical Engineers
Topics: Design
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Figure 1

Wind anemometer. Initial values of the parameters in millimeters are shown in parentheses.

Grahic Jump Location
Figure 2

Rules and termination conditions learned for the design of an anemometer

Grahic Jump Location
Figure 3

LearnIT-II system

Grahic Jump Location
Figure 4

Top: Hypothetical decision tree trained to identify when parameter Pk should be increased. Pi and Pj are other parameters. CCS is a bit string representing the states of the four design constraints. Bottom: Corresponding production rules.

Grahic Jump Location
Figure 5

(Top) Typical training data for learning termination conditions. (Bottom) Typical termination conditions.

Grahic Jump Location
Figure 6

Automated design process using rule termination conditions

Grahic Jump Location
Figure 7

“User procedure” for the anemometer design problem. The design steps are applied in the order listed.

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In