Sectional speed is an important operational indicator that is related to the size of train flows along a railway corridor. The number of trains traveling along a railway corridor significantly affects the speed of trains on freight-intensive routes, which affects the main performance indicators of the entire transportation system. Solving the problem of finding a rational number of trains on a railway corridor is possible by creating a regression model that will help to assess critical performance characteristics on railway sections and facilitate the automation of transportation planning processes in the context of the functioning of traffic control centers (TCCs). The study presents a predictive regression mathematical model based on the random forest method, which allows identifying the dependencies between the speed on sections, their congestion, and the number of trains traveling along the entire railway corridor. The application of the random forest method provides higher forecasting accuracy compared to a simple decision tree algorithm, the ability to efficiently process big data, and solves the problem of overfitting and automatically takes into account the interaction between various factors. Since the selection of the number of trains moving along the entire railway corridor is a task with a high level of uncertainty due to the variable operating conditions on the railway, this research proposes the use of a game-theoretic approach. As a model, we use a game with nature, which allows us to analyze the interactions between participants and external unpredictable factors known as “nature”.