Abstract: In order to predict the arrival time of railway freight more accurately, this paper used Python to implement the prediction model of railway freight arrival time based on random forest algorithm. According to different vehicle attributes, the paper predicted the time of vehicles arriving at the terminal, took into account the various factors of vehicles, calculated the eigenvector, and constantly corrected the error in the process of driving, so that the prediction time to the destination was more accurate. The Zhanglan–Dingbian freight transportation was taken as an example to verify the model. The verification results show that accuracy of the model is high, which has the value of popularization and application.