Enzymatic methanolysis of refined soybean oil with methanol was investigated using Rhizomucor miehei lipase, Lipozyme RM IM, in n-hexane for reaction times of 30 min. Response surface methodology (RSM) based on three-level, three-factor (variable) face-centered cube design was used for the optimization of methanolysis. The independent variables that affect the methanolysis reaction conducted in n-hexane are temperature (degrees C), enzyme/oil weight ratio, and oil/methanol molar ratio. A good quadratic model was obtained for the methyl ester production by multiple regression and backward elimination. A linear relationship was observed between the observed and predicted values (R-2 = 0.9635). The effects of temperature and enzyme amount, which affected methyl ester content of the product (response) positively, were significant (P < 0.01). The quadratic term of temperature and the interaction term of enzyme amount with temperature affected the response negatively (P < 0.01). The interaction term of conditions for the response at which methyl ester content of the product was 76.9% were determined to be 50 degrees C, 2.37 methanol/oil mole ratio, and 0.09 enzyme/oil weight ratio.