Large capacity of railway cargo transportation forecasting

forecasting the number of wagons with various goods, following various routes. The topology of the railway network is given — for all possible pairs of railway lines, information about all blocks of wagons, which have moved from one line to another, including the number of wagons in a block, the type of cargo, and the date of the route, is provided. The algorithm, based on convolution of the empirical density distribution of the values of time series with the loss function is used for prediction. Previously, forecasting was carried out for each railway junction separately. It is proposed to be improved by the quality of forecasting predicting by pairs of lines instead of predicting departure of all wagons from the given junction. The algorithm is illustrated by the daily data on transportation of 38 types of cargo collected during a year and a half.