This study used the Taguchi method combined with Analytic Hierarchy Process (AHP) to present optimization design of the Photovoltaic-Thermal (PV/T) system for improving the two quality characteristics namely electrical and thermal efficiency. The control parameters for PV/T design including the collector plate material, collector azimuth, number of collector tubes, mass flow rate, collector angle, and storage tank volume to area (V/A) ratio are crucial factors affecting the performance of the PV/T. First, Taguchi method coupled with main effect analysis (MEA) and analysis of variance (ANOVA) was used to obtain the best parameters combination on each single quality characteristic of electricity and thermal efficiency. In order to achieve the maximum total efficiency of the PV/T system, multi-quality analysis AHP was used to obtain the optimal control parameters combination according to the contribution degrees of quality weight. Then, the computational fluid dynamic (CFD) software used to simulate the PV/T module thermal distribution before and after optimization. The optimization control parameters combination are: collector plate material is copper, collector azimuth is due south, number of collector tubes is 12, mass flow rate is 0.01 kg/s-m2, collector angle is 25°, and V/A ratio is 123. The effectiveness of the proposed method was evaluated by the confirmation experiments. The optimization properties under these parameters are electrical efficiency 14.29% and thermal efficiency 44.96% which is improved than traditional PV/T system (electrical efficiency 12.74% and thermal efficiency 34.06%). The findings of CFD showed that the temperature of the PV/T module decreased by about 10 °C after optimality analysis, and the coefficient of correlation between the simulated and actual PV/T water temperatures was 0.991 (very close to 1). The CFD simulated performance testing proposed in this study has a good verification effect.