A MACHINE LEARNING-BASED APPROACH TO ACCELERATING COMPUTATIONAL DESIGN SYNTHESIS

A modification-based framework for computational design synthesis augmented by machine learning is presented. The framework allows a wide range of engineering design problems to be addressed via a machine learning based search algorithm with minimal required adaptation of the search heuristics. Search is accomplished via two agents, a ‘data modeller’ and a ‘modification advisor’, that work together to guide a generate-and-test-oriented search with suggested actions based on past observation of the search. A proposed implementation of the search algorithm is discussed and the results of its application to two design examples are presented. Implications of the method for engineering design are discussed.

[1]  Kristina Shea,et al.  Learning Intelligent Modification Strategies in Design Synthesis , 2003 .

[2]  Kristina Shea,et al.  From discrete structures to mechanical systems: A framework for creating performance-based parametric synthesis tools , 2003 .

[3]  Steven M. Rubin,et al.  Computer Aids For VLSI Design , 1987 .

[4]  Peter J. Fleming,et al.  An Overview of Evolutionary Algorithms in Multiobjective Optimization , 1995, Evolutionary Computation.

[5]  Keith A. Seffen,et al.  A SIMULATED ANNEALING ALGORITHM FOR MULTIOBJECTIVE OPTIMIZATION , 2000 .

[6]  D. E. Grierson,et al.  Method for Conceptual Design Applied to Office Buildings , 2002 .

[7]  J. Cagan,et al.  A Shape Annealing Approach to Optimal Truss Design With Dynamic Grouping of Members , 1997 .

[8]  Michael J. Todd,et al.  Mathematical programming , 2004, Handbook of Discrete and Computational Geometry, 2nd Ed..

[9]  Maja J. Matarić,et al.  Action Selection methods using Reinforcement Learning , 1996 .

[10]  John L. Bresina,et al.  Integrating planning and reaction: A preliminary report , 1990 .

[11]  Xuemei Wang,et al.  Learning Planning Operators by Observation and Practice , 1994, AIPS.

[12]  John S. Gero,et al.  Evolutionary learning of novel grammars for design improvement , 1994, Artificial Intelligence for Engineering Design, Analysis and Manufacturing.

[13]  William Prager,et al.  Optimal layout of cantilever trusses , 1977 .

[14]  Richard S. Sutton,et al.  Sequential Decision Problems and Neural Networks , 1989, NIPS 1989.