In order to predict the difficulty degree on filling pipeline blocking of filling system in deep mine, the comprehensive evaluation model of filling pipeline blocking was established on the basis of fuzzy mathematics and the entropy theory. Ten evaluation indexes, such as volume concentration of filling slurry and shape of filling aggregate, were selected as the discriminant ones of the fuzzy synthetic judge model. The corresponding membership matrix was established according to in-situ data of mine and three evaluation standard of filling pipeline blocking risk. The entropy theory was used to calculate the weight of each discriminated index, and the predictive value of filling pipeline blocking was finally achieved and sorted using the rules of fuzzy synthetic judge model. Regarding Longshou mine of Jinchuan Company, Dahongshan copper mine, Hedong gold mine and Xincheng gold mine as evaluation objects of filling pipeline blocking risk, the result shows that the risk grade of Longshou mine of Jinchuan company is a little more than level II, the risk grade of Xincheng gold mine is a little more than level III, the risk grades of Hedong gold mine and Dahongshan copper mine are both less than the level III. The comprehensive evaluation model based on fuzzy mathematics and the entropy theory can reflect the distance between the sample and the evaluation standard, and the evaluation result is objective and reasonable, which has good practicability.