The Taguchi method uses orthogonal arrays
to screen a large number of controllable and noise factors and mark
the significant ones and their levels. For continuous factors, this
technique is not normally used for experimental optimization due
to its inability to determine the best combination of factor values
within the specified region of interest. This problem can be overcome
by using RSM that allows for finding an approximation
suitable to fit to the data from experiments at various points in the
experimental space defined by Central Composite Design (CCD) and
predict the optimum factor combination (Arteaga, Li-Chan, Vazquez-
Arteaga, & Nakai,1994).