A digital twin in transportation: Real-time synergy of traffic data streams and simulation for virtualizing motorway dynamics. (January 2023)

• Record Type: Journal Article
• Authors: Kušić, Krešimir; Schumann, René; Ivanjko, Edouard
• Is Part Of: Advanced engineering informatics. Volume 55(2023)
• Publication Date: 2023-01
• Subjects: Digital twin -- Microscopic traffic simulation -- Real-time Big Data analytics -- Calibration -- Traffic sensors -- Smart roads; Computer-aided engineering -- Periodicals; Engineering -- Data processing -- Periodicals; 620.00285
• DOI: 10.1016/j.aei.2022.101858
• Languages: English
• ISSNs: 1474-0346

Abstract

Abstract: The introduction of digital twins is expected to fundamentally change the technology in transportation systems, as they appear to be a compelling concept for monitoring the entire life cycle of the transport system. The advent of widespread information technology, particularly the availability of real-time traffic data, provides the foundation for supplementing predominated (offline) microscopic simulation approaches with actual data to create a detailed real-time digital representation of the physical traffic. However, the use of actual traffic data in real-time motorway analysis has not yet been explored. The reason is that there are no supporting models and the applicability of real-time data in the context of microscopic simulations has yet to be recognized. Thus, this article focuses on microscopic motorway simulation with real-time data integration during system run-time. As a result, we propose a novel paradigm in motorway traffic modeling and demonstrate it using the continuously synchronized digital twin model of the Geneva motorway (DT-GM). We analyze the application of the microscopic simulator SUMO in modeling and simulating on-the-fly synchronized digital replicas of real traffic by leveraging fine-grained actual traffic data streams from motorway traffic counters as input to DT-GM. Thus, the detailed methodological process of developing DT-GM is presented, highlighting the calibration features of SUMO that enable (dynamic) continuous calibration of running simulation scenarios. By doing so, the actual traffic data are directly fused into the running DT-GM every minute so that DT-GM is continuously calibrated as the physical equivalent changes. Accordingly, DT-GM raises a technology dimension in motorway traffic simulation to the next level by enabling simulation-based control optimization during system run-time that was previously unattainable. It, thus, forms the foundation for further evolution of real-time predictive analytics as support for safety–critical decisions in traffic management. Simulation results provide a solid basis for the future real-time analysis of an extended Swiss motorway network.