Objective: To establish an autoregressive moving average model for the prediction of tuberculosis cases in students of Shanxi Province, and to provide scientific basic for the prevention and treatment of pulmonary tuerculosis among students. Methods: A optimized ARIMA model was set up based on reported monthly data of TB in students from January 2010 to September 2019 in Shanxi Province by SAS 9.3 software, and the incidence trend in the next two years was predicted. Results: The average reported rate of active TB in students of Shanxi Province was 23.52 per 100 000 from 2010 to 2019,showing an overall downward trend(χ2=999 980.46，P<0.01). The optimal model was SARIMA(0，1，1)(0，1，1)12,SBC=982.16. The fitted equation was (1-0.63B) (1-B12)Yt=(1-0.61B12)εt. The mean relative error was 19.35%,and the predicted incidence trend was consistent with the previous years,and the peak was from March to May. Conclusion Substantial progress has been made in student TB prevention of Shanxi Province. The ARIMA product season model is suitable for forecasting the TB incidence in students,so as to provide scientific guidance for its early prevention and control.