USING MODELING KNOWLEDGE TO GUIDE DESIGN SPACE SEARCH

Automated search of a space of candidate designs is an attractive way to improve the traditional engineering design process. To make this ap proach work, however, an automated design system must include both kno wledge of the modeling limitations of the method used to evaluate cand idate designs and an effective way to use this knowledge to influence the search process, We argue that a productive approach is to include this knowledge by implementing a set of model constraint functions whi ch measure how much each modeling assumption is violated. The search i s then guided by using the values of these model constraint functions as constraint inputs to a standard constrained nonlinear optimization numerical method. A key result of our work is a successful demonstrati on of the application of AI techniques to an important engineering pro blem. In an empirical study of parametric conceptual aircraft design, we observed a cost improvement of two orders of magnitude. The princip al contribution of our work is a new design optimization methodology w hich makes explicit the interaction between models of artifacts, and v alidity models of artifact models. (C) 1998 Elsevier Science B.V. All rights reserved.