Abstract: Objective To establish time series Autoregressive Integrated Moving Average (ARIMA) model to predict brucellosis incidence in Shanxi, and provide evidence for the prevention and control of brucellosis. Methods The time series ARIMA model was established by using brucellosis surveillance data in Shanxi from 2006 to 2010, and short term prediction of brucellosis incidence was made. Results The ARIMA model established fitted well. The medium term predictive incidence was similar to actual one and the result interpretation was reliable. Conclusion It is feasible to use the established ARIMA model in the prediction of the brucellosis incidence in Shanxi.