Abstract: In order to realize the effective prediction of mine water inflow and improve the prediction accuracy, based on the monthly water inflow data of Hebi No.8 Mine from 2009 to 2019, this paper uses the time series analysis software Eviews9.0 to establish three models. They are the X12 seasonal adjust model, ARIMA(2, 0, 1) model and SARIMA(2, 0, 1)×(0, 1, 1)₁₂ model, the monthly water inflow in 2019 is used for verification. By comparing the prediction errors of the three models, the optimal model for predicting mine water inflow in Hebi No.8 Mine is discussed. The results show that the three models have good prediction effects on water inflow, among which ARIMA(2, 0, 1) has the highest prediction accuracy, followed by SARIMA(2, 0, 1) ×(0, 1, 1)₁₂ model, and X12 seasonal adjust model is slightly inferior. The possible error sources of the three models are studied and analyzed, which can provide a new idea for the prediction of mine water inflow.