It is significance to predict coal production for balancing coal supply and demand in China. The primary goal of this research is the prediction of coal production in china. The method used in the study is known as the BP neural network. The BP neural network is designed with the MATLAB simulation software based on coal production historical data from 1980 to 2007. The studies we have performed showed that the prediction of coal production based on BP neural network is reasonable and valuable. Finally, we get the prediction of coal production from 2010 to 2015, and the prediction indicates that the coal production will increase in the next 5 years.