MODELING THE ECONOMIC EFFECTS OF IMPLEMENTING ARTIFICIAL INTELLIGENCE TECHNOLOGIES IN RAIL TRANSPORT

Abstract

The current state of the transport sector necessitates the search for innovative solutions aimed at improving the efficiency and sustainability of freight transportation. A key area of focus is modeling the economic effects of implementing artificial intelligence technologies. The purpose of the research is to explore the directions of development and implementation of artificial intelligence tools in rail transport and to develop a scientific and methodological approach to modeling the economic effects of their application. To achieve the purpose of the article, a set of general scientific and specialized methods of economic analysis and modeling was applied (system approach, method of generalization and abstraction, structural-functional analysis, scenario approach, economic-mathematical modeling). A methodological toolkit is proposed, based on an NLP model with a confidence threshold mechanism that formalizes the allocation of requests between automated and manual processing. Three application scenarios are examined, with the potential economic impact assessed for each. The minimal scenario corresponds to the pilot phase of implementing the intelligent technology, characterized by conservative confidence thresholds and limited automation, which allows for cautious evaluation of benefits. The basic scenario reflects the stage of systemic scaling and the stabilization of algorithmic solutions following the retraining of the NLP model, ensuring more reliable integration into operational processes and demonstrating measurable improvements in efficiency. The maximal scenario describes the stage of mature, full-scale operation of the intelligent platform, where automation becomes dominant, human intervention is minimized. The practical significance of the developed methodological approach to modeling the economic effects of implementing artificial intelligence technologies in rail transport lies in its potential application to enhancing the freight transport management system – particularly through reducing delays, optimizing logistics flows, and improving the efficiency of infrastructure resource utilization.