Purpose. To improve the distribution methods of freight flows on the railway network under conditions of capacity constraints. Methodology. Methods of railway operation theory and operations research were used. The problem of cargo flow distribution on the railway network was solved as a multi-product (interchangeable cargoes) transport linear programming problem with capacity constraints in the network form. The average cost of freight transportation is established by the methods of probability theory. Findings. In the course of the research the methods of cargo traffic calculation on the railway network are improved. The proposed approach to the formalisation of the task makes it possible to take into account the presence of different cargoes to be transported, capacity constraints of some railway network sections, as well as non-linear nature of dependence between the volume and cost of transportation. To meet the requirement of consignors equal access to public carrier services, the method for calculating the cost of transportation for an individual consignor is improved. Originality. Scientific novelty of the paper consists in improvement of methods of cargo traffic distribution on the railway network and tariffing of cargo transportation services in conditions of restricted supply capacity of separate sections. Practical value. The extractive industries have a steady flow of goods. The main logistic tasks in this case are to ensure transportation of the given volume of cargo and reduce the cost of transport services. Using the proposed mathematical methods of optimization allows you to get higher-quality solutions compared to the method of technical-economic comparison of options, which provides an overall cost reduction for cargo transportation. The improved method of tariffication of transportation can reasonably allocate the savings from reducing the cost of transportation among cargo consignors and thus reduce their logistics costs.