ЗАСТОСУВАННЯ ТЕХНОЛОГІЙ ШТУЧНОГО ІНТЕЛЕКТУ ДЛЯ ОПТИМІЗАЦІЇ УПРАВЛІННЯ ЗАЛІЗНИЧНИМИ ВАНТАЖНИМИ ПЕРЕВЕЗЕННЯМИ

The purpose of the article is to study the factors influencing the automation process of providing a freight car for loading and to substantiate the directions of using artificial intelligence technologies for its optimization. It is shown that the main problem in the field of freight rail transportation is the limited availability of the working fleet of cars of JSC "Ukrzaliznytsia". External and internal factors of the development of automated processes of feeding a freight car for loading are considered. It is determined that the main external factor of rail freight transportation is the need and volume of military rail transportation. Internal factors that directly affect the success of automation in the transport industry are highlighted. The main internal factors influencing the automation processes of transport companies are minimizing human influence on automated processes, automation of certain types of wagons (gondola wagons, grain and mineral wagons), priority for cargo-intensive transportation directions, consideration of the economic component of individual transportations, the need for "manual" control in the event of accidents, theft of components, etc., insufficient attention to customer reactions to changes and innovations, lack of a centralized "history" of cooperation with the client, and insufficient level of cyber protection of automated systems. Organizational steps of JSC "Ukrzaliznytsia" regarding the introduction of artificial intelligence technologies to optimize the process of providing a freight car have been identified, namely: increasing the level of cyber protection of its information resources and databases, synchronization of databases, development of preliminary verification stages and implementation of artificial intelligence in automated processes. A general algorithm for the application of artificial intelligence as an innovative solution for the service of providing wagon loading has been developed. The main risks of implementing artificial intelligence in railway transport have been identified, namely: technical failures and malfunctions, cybersecurity and hacker attacks, excessive automation, ethical and social challenges, and unpredictability of system behavior.