Response surface methodology has been widely applied to optimize the process. However, it was rarely applied to dry digestion. We used central composite design to optimize the anaerobic dry co-digestion of food waste and manure. Mixture ratio and particle size of food waste and manure were selected as independent variables, and target surface response was the methane production yield (MPY). BMP tests were conducted, and MPY was fitted by a second-order polynomial quadratic model, which was found to be significant with higher coefficient (R-2=0.98). As results of F-value analysis, the mixture ratio was found to be more important than particle size. Finally, the optimum conditions of mixture ratio (food waste : manure=5.79 : 4.21) corresponding to 15.6 of C/N ratio and particle size 1.12 cm were determined. In addition, 313 mL CH4/g VSadded of MPY was anticipated under optimum conditions with 94.4% of desirability.