Abstract: With the continuous increase of coal seam mining depth, the coal mine production faced with complex water inrush mechanism and changeable main controlling factors, which makes the prediction of floor water inrush more difficult. In order to accurately predict the risk of floor water inrush, the vulnerability index method was used to obtain the risk weight of floor water inrush for each sub-factor, and the analytic hierarchy process（AHP） was used to divide the risk threshold of floor water inrush, so as to establish coal seam floor water inrush prediction model. We analyzed the geological and hydro-geological data of Henan Longmen Coal Industry Changcun Coal Mine from the three aspects of water inrush sources, water inrush channels and aquifers, the influencing factors of floor water inrush are divided into eight factors: aquifer water pressure, water output, grouting volume, fault similarity dimension, fault dip, fault strike, fault drop and effective aquifer thickness. Based on eight sub-factors, the risk of floor water inrush is divided into five areas: safe area 0．25， 0．32）, relatively safe area 0.32, 0.44）, the more dangerous area 0.44, 0.56）, the dangerous area 0.56, 0.70）, and the extremely dangerous area 0.70, 0.85. Finally, the vulnerability index method evaluation model based on AHP was established. Compared with the traditional water inrush coefficient method, the evaluation results of the vulnerability index method based on AHP have better accuracy, and the division is more refined andreasonable, which provides a scientific evaluation method and theoretical basis for the prediction of mine water hazards.