Abstract:
Response surface methodology is widely used for developing, improving, and optimizing processes in various fields. In this article, we present a method for constructing second order rotatable designs in order to explore and optimize response surfaces based on a set of fraction of factorial designs. The designs achieve both properties of rotatability and estimation efficiency. We shall concentrate on the moment matrices and the related information surfaces based on the parameter subsystem of interest on the Kronecker model and their corresponding rotatable designs. The set of rotatable designs based on the central composite designs shall be presented. These designs shall be shown to be A-, D-, E-and T-optimal.
