Оценка результатов среднесрочного прогнозирования эксплуатационных характеристик железной дороги

Abstract

The results of the so-called descriptive forecast calculations of the future values of the key performance indicators of the transport industry are of considerable interest. The article presents a comparison of the results of the multivariate forecast of the performance indicators of the Krasnoyarsk Railway for 2015-2018, published in one of the previous works by the authors, with their actual values. These results were derived from a regressive mathematical model, which is an open dynamic recursive discrete system consisting of thirteen equations. The output variables in this model are the transport freight turnover, the locomotive productivity, the sector speed, the average weight of the freight train, the local car demurrage, the demurrage at the technical station, and loading volumes. In total, pessimistic, neutral and optimistic versions of the forecast have been developed. The comparison of the actual (real) and forecasted (estimated) values of the endogenous (internal, output, dependent, explicable) variables showed that 23 of the 28 values fell into the forecast ranges, which makes it possible to conclude that the model is highly adequate and can be used effectively for a wide range of mid-term forecasting tasks, connected with mid-term forecasting. Some ways of enhancing the predictive capability of the model are proposed.