Analysis of models and principles of the system modeling in the construction of predictive models of cargo loading

The study of intra-annual dynamics of generalized indicators of railway production activity is an essential part of long-term forecasting, planning and analysis. The development of indicators of uneven operation of the car fleet is one of the important issues of solving the General problem of increasing the rhythmicity of operational work of railway transport. In the evaluation of seasonal variations of transport using conventional methods causes a significant error. As one of the ways to solve this problem, the authors propose an improved methodological tool for assessing the seasonal unevenness of cargo loading to the ports of the Far East. This method is based on the construction of a mathematical model of cargo loading, on the basis of which loading is predicted for the upcoming year. In the form of the completed work, the principles of mathematical modeling and methods of mathematical research, as well as the application of a systematic approach to solving the problem of predicting the volume of car traffic are considered. By means of scenario planning and expert forecasting, the results were adjusted and the conclusion was made about the need to develop throughput capacities in the busiest sections of the BAM and TRANS-Siberian railway. A comparison of the actual loading volumes with the forecast values showed that the presented forecast was justified. Deviations of forecast values from real ones are within acceptable limits. The proposed tools can significantly increase the accuracy of estimating seasonal unevenness of cargo loading and forecast the arrival of car traffic to seaports. All this will contribute to improve the quality of planning and analysis of the functioning and development of railways. The full range of measures includes the construction of process models of the production unit of the Russian Railways holding and the preparation of a forecast model of production activity, and can also contribute to the creation of an innovative system of operational indicators of polygons.