Rethinking digital twins in transportation: From virtual model to Human-in-the-Loop control

Digital twins have emerged as a transformative concept in many fields, traffic management including. Yet their definition and practical applications remain ambiguous. This paper seeks to clarify what constitutes a true digital twin, distinguishing it from traditional traffic simulation models, and explores scenarios where digital twins offer unique advantages. Drawing on global best practices and state-of-the-art research, we identify key challenges and provide actionable insights into their resolution. A pivotal contribution of this study is the introduction of the Digital Twin Evropská (DTE), a real-world case study focused on a critical urban corridor in Prague. We detail the motivation for its development, the methodologies employed, including real-time path-flow estimation, continuous simulation calibration, and integrated data-assimilation workflows, and the resulting insights. Under the taxonomy proposed in this paper, the currently implemented DTE is classified as a Digital Shadow with predictive capabilities, while the Human-in-the-Loop decision-support architecture represents the proposed pathway towards a fully operational Digital Twin. The findings demonstrate the practical utility of continuously synchronized digital representations for urban traffic management and provide a basis for their further development towards operational Digital Twins. The paper concludes with recommendations for leveraging DTE in operational contexts and outlines potential next steps for advancing the digital twin paradigm.