The objective of this research is to obtain an optimal combination on improving the adsorption/photocatalysis capability of composite photocatalysts by means of design of experiments (DOE) and analysis of variance (ANOVA). Five control factors, including photocatalyst type, catalyst loading. support type, heat-treated temperature, and stirring speed, with four levels were considered to investigate the optimal parameter setting. The present work applied an L16 orthogonal array combined with ANOVA analysis to individually find out their improved parameters: adsorption and photocatalysis efficiencies. Two sets of parameter settings appeared a similar optimal combination, in which the type of support acts as the most crucial contribution to affect both of adsorption and photocatalysis capabilities. This can be attributed to the fact that vast pore structure of supports probably leads to uniform dispersion of catalysts, thus promoting the photocatalysis capability. The ANOVA result also confirmed that both the reliability of experiments and rationality of variances fall within a confidence level of 99%, indicating the robust design for these composite photocatalysts. The parameter settings via ANOVA and DOE methodologies provide an efficient approach not only in enhancing the adsorption/photocatalysis hybrid performance but also tuning the optimal controlling factors for preparation of the composite photocatalysts, based on economic and robust-design viewpoints. (C) 2007 Elsevier B.V. All rights reserved.