Analysis of Experience and Prospects for the Application of Artificial Intelligence to Railway Transport Operations Management

At the present stage, the digitalization of the transport sector has created prerequisites for the transformation of railway transport, making it necessary to conduct a detailed study of the potential of artificial intelligence (AI) for improving transportation efficiency. The aim of this paper is to systematize the methods, practical experience, and prospects for the use of AI in railway transport operations management. A review of regulatory documents, scientific publications, and implemented projects in this field over the past decade has been conducted, during which AI has become one of the key drivers of transport development. The analysis has identified technological trends in the integration of AI into railway operations, infrastructure design, and logistics, including machine learning, computer vision, and digital twins. The areas of AI application have been identified, ranging from demand forecasting and strategic-level planning to neural-network-based traction calculation modeling. The advantages of AI implementation (more efficient utilization of line capacity, reduced transportation costs, and improved reliability and safety of infrastructure) as well as barriers to its implementation, including data quality and heterogeneity, regulatory requirements, and the specific characteristics of the transportation process, have been systematized. The feasibility of using AI to reduce uncertainty in selecting an effective train traffic interval regulation system for single-track railway sections is substantiated. Further research into the implementation of AI in railway transport operations management should focus on developing adaptive control systems, integrating heterogeneous data, developing an industry-specific standard, and incorporating training environments into the educational process for railway engineering students.