Design Optimization of Complex Structures Using Metamodels

Current engineering analyses rely on running expensive and complex computer codes. Statistical techniques are widely used in engineering design to construct approximate models of these costly analysis codes. These models referred as metamodels, are then used in place of the actual analysis codes to reduce the computational burden of engineering analyses. The intent of this study is to provide a comprehensive discussion of the fundamental issues that arise in design optimization using metamodels, highlighting concepts, methods, techniques, as well as practical implications. The paper addresses the selection of design of experiments, metamodel selection, sensitivity analysis and optimization. (C) 2010 Jordan Journal of Mechanical and Industrial Engineering. All rights reserved