No abstract

This paper discusses response surface methods for approximation model building and multidisciplinary design optimization. The response surface methods discussed are central composite designs, Bayesian methods and Doptimal designs. An over-determined D-optimal design is applied to a configuration design and optimization study of a wing-body, launch vehicle. Results suggest that over determined Doptimal designs may provide an efficient approach for approximation model building and for multidisciplinary design optimization.

No abstract

Determine the possible range for each parameter.The difficulty with this one-variable-at-a-time approach is that patience and money are likely to run out before the list of parameters is completed. Furthermore, parameter interaction sensitivities cannot be identified using a one-variable-at-a-time approach (Gunter 1990). Consider the situation with ten parameters, each of which is tested for a high, a medium, and a low