Abstract: The paper had studied the effect of moisture content on the NIR predicted value of propylene glycol in tobacco leaf, and its effect on casing uniformity coefficient. Results showed that (1) the greater the difference in moisture content between testing sample and model sample, the greater the absolute deviation of the predicted value, when the testing sample’s water content is larger than it of the model samples, the predictive value is high; conversely, the predictive value is low. (2) There is a significant linear relationship between the water content and the absolute deviation of predicted value (R₂=0.938), therefore, the forecast value can be corrected according to the moisture content. (3) It showed that the evaluation result of the casing uniformity from a same batch cigarettes, because the water content of testing sample was about 1.70% than it of model sample, there is difference of 2.0% between the casing uniformity coefficient calculated according to the predicted value and adjusted value of propylene glycol. In a word, it is helpful to improve the evaluation results of casing uniformity by controlling the influence of moisture content on the NIR predicted value of propylene glycol.