Reasonable design space approach to response surface approximation

Response surface approximations have become very popular as a tool for mult idisciplinary optimization, However, the accuracy of these models is a majo r concern. One method to increase the accuracy of response surface approxim ations is to construct them only for the portion of the design space that r epresents reasonable designs, Our paper describes the methodology underlyin g this reasonable design space approach. This methodology is illustrated us ing two examples involving multidisciplinary optimization of a high-speed c ivil transport model, In these examples, low-fidelity analyses are used to estimate the boundary of the reasonable design space. Within the reasonable design space, response surface models are constructed, based on results fr om higher-fidelity analyses. The accuracy of the response surface models cr eated with and without using the reasonable design space approach is compar ed, and the advantages of employing response surface models created via the reasonable design space approach are demonstrated.