Abstract: Microwave popcorn was analyzed by using response surface methodology to find out the effect of different weight of charge, heating time and ingredient additions on popped kernels ratio (Y₁), expansion index (Y₂), and sensory score (Y₃) of microwave popping corn and the prediction models were established. The models of Y₁, Y₂ and Y₃ showed no significant lack of fit at 5%, 5% and 10% (p₁＞0.05, p₂＞0.05 and p₃＞0.1). But they were statistically significant at 5%, 1% and 1% (p₁＜0.05, p₂＜0.01 and p₃＜0.01) with R² values of 0.8139, 0.8737 and 0.9160 respectively. All the models were appropriate to express response variables. The preparing condition was optimized by using desirability function approach for simultaneous optimization of several variables levels that were 1.518 weight of charge, 0.518 heating time and 1.118 ingredient additions. The popped kernels ratio, expansion index and sensory score under the optimum condition were 97.5%, 11.0 and 7.2 respectively.