Objective：The autoregressive integrated moving average（ARIMA）model was used to predict the number of registered tuberculosis cases in Changzhou and Yancheng in Jiangsu Province，in order to explore its feasibility for predicting tuberculosis epidemic situation in Jiangsu Province，and provide reference for future tuberculosis prevention and control work. Methods：The data on the number of registered tuberculosis cases in Changzhou and Yanchengof Jiangsu Province from January 2005 to December 2016 were collected，and the ARIMA models were established by the R3.5.2 software. The number of registered tuberculosis cases in January to December 2017 was predicted，with mean absolute percentage error（MAPE），root mean square error（RMSE）and mean absolute error（MAE）to evaluate the accuracy of ARIMA model predictions. Results：The optimal prediction models for Changzhou and Yancheng were ARIMA（0，1，1）（0，1，1）12 and ARIMA（1，1，1）（0，1，1）12，respectively. In the predicted number of registered tuberculosis cases in 2017，the MAPE values of the two cities were 8.718 6 and 16.787 8，the RMSE values were 14.061 7 and 39.487 2，the MAE values were 11.813 1 and 33.349 8，respectively. Conclusion：The ARIMA model has a relatively good fitting effect in predicting the monthly number of registered tuberculosis cases in Changzhou，so it is speculated that the model is suitable for short-term prediction and dynamic analysis of tuberculosis in the southern of Jiangsu.