IMPROVEMENT OF MODELING D-OPTIMAL DESIGN-DATA WITH NEURAL NETWORKS - RADIAL BASIS FUNCTION NEURAL NETWORKS ARE VERY ROBUST

D-Optimal design can be used to study the influence of numerous parame ters in a process, with a limited number of experimental paints, Howev er, the multiple regression techniques normally used to analyze the da ta coming from this design. can lend to a false interpretation. In thi s study, two different neural networks have been used to obtain a bett er modeling of the process. Advantages and disadvantages of both metho ds will be discussed.