Digital twin of a multimodal transport hub for adaptive management of passenger and freight flows in real time

Abstract

The article investigates improving the efficiency of operational management of a multimodal transport hub under variable demand, competition between passenger and freight flows for shared resources, and random disturbances. The aim of the study is to develop a conceptual and methodological framework for a digital twin that supports real-time dispatching decisions through a combination of monitoring, short-term forecasting, scenario testing, and multi-criteria optimization. An integrated hub-state model is proposed, accounting for demand, infrastructure saturation, transfer time, cargo-handling duration, and data quality. Within the control loop, actions are selected by minimizing weighted losses with robust risk adjustment, which makes it possible to filter out unstable decisions. For verification, a computational experiment was conducted on a test hub in two modes: baseline rule-based control and scenario-adaptive control within the digital twin loop. The results demonstrate improvements in key indicators, including lower average delays, a smaller share of congested zones, shorter transfer and cargo-handling times, and a higher stability index. Under a stress scenario, the approach showed stronger robustness, with lower peak overloads and faster recovery to normal operation. The practical value lies in supporting dispatcher decisions and phased integration into existing urban transport monitoring platforms. The findings confirm that controlling telemetric data quality and coordinating information flows are critical for forecast accuracy, recommendation reliability, and dispatcher support during peak loads and uncertainty.