Abstract: Accurate prediction of wagon flow routing is a prerequisite for making a high quality daily and shift freight traffic plan. By utilizing massive historical data of wagon flow from the transportation information integration platform, a simple dynamic prediction method of railway wagon routing is proposed, which is based on long-term and short-term wagon flow route selection preference parameters that are derived from historical data statistics of freight wagon path and can reflect the effects of many factors in the railway on wagon route selection. By combining wagon route selection preference parameters and real-time position tracking data of wagons, wagon flow routing can be predicted. Furthermore, an environment of big date application is set up and the program of wagon flow routing prediction is developed and deployed. In the end, two index of prediction hit rate and prediction accuracy are defined and calculated to evaluate the effects of the prediction and the calculations indicate that the prediction results have high accuracy and satisfactory practical value.