Integrated product engineering: a hybrid evolutionary framework

The systematic identification of new materials for specific: engineering ap plications with optimal values of thermophysical, mechanical and/or biologi cal properties is a key technical challenge with obvious commercial applica tions. The design of these new materials consists of two components: (i) th e forward problem, which involves the prediction of how changes in the basi c compositional units give rise to various engineering property and (ii) th e inverse problem, which involves discovery of viable formulations that are predicted to possess desired performance characteristics. This situation i s however complicated by the fact that in many industrial design situations , data are both sparse and noisy, the fundamental understanding of the syst em is limited and time and resource constraints are stringent. Thus, a syne rgistic approach employing first principle chemistry/physics modeling and s tatistical techniques like Neural networks seems promising for the forward problem. The inverse problem is addressed using ideas from evolutionary alg orithms. Two widely different industrial product-design problems are consid ered in this paper and the applicability of this new methodology is demonst rated. (C) 2000 Elsevier Science Ltd. All rights reserved.