Abstract: In order to predict the yield of flue-cured tobacco in Shaanxi Province, the spatial distribution of the climatic characteristics for flue-cured tobacco planting were analyzed. The yield prediction model was established using multi-linear regression method based on the comprehensive climatic suitability index of representative counties. Several meteorological factors were considered, including the average temperature during maturity stage, precipitation in vigorous growing period, available time and sunshine duration during the growth period. The datasets of tobacco yield and meteorological data of the 13 representative counties where the tobacco planted continuously were used in this study. Results showed that the predicted and observed yield agreed well, with an average relative error of 3.74%. The accuracy of the yield prediction for 2014 was 92%, with a relative error of 7.96%. The prediction model was of high accuracy and practicability, and could be used as an effective tool to predict the tobacco yield production.