Statistical experimental designs using a response surface methodology (RSM) have been most widely applied to optimization experiments (Myers et al., 2009), which provides information about direct effects, pairwise combined interacting effects and curvilinear variable effects of variable factors.