Integration of design of experiments and artificial neural networks for achieving affordable concurrent design
Wei Chen, Sriram Varadarajan, Wei Chen, Sriram Varadarajan · 38th Structures, Structural Dynamics, and Materials Conference · 1997
For designs involving computer intensive systems analyses, approximation techniques are commonly used to create a simplified approach to evaluating the system behavior. These techniques help in reducing the product development time and in finding the optimal solutions. Two important types of approximation techniques are the Design of Experiments (DOE) and the Artificial Neural Networks (ANN). While these techniques have their own unique features, they have certain important advantages as well as disadvantages over each other. In this paper, an integration strategy is presented in which both methods complement one another in achieving affordable current systems design. The proposed strategy is verified by comparing the DOE and ANN approaches to the approximations of typical nonlinear behaviors in design. The high speed civil transport (HSCT) aircraft design is used as an example in this study.