Transportation engineering in the era of emerging technologies and big data
Dr. Mehdi Keyvan Ekbatani
University of Canterbury
Time & Place
Thu, 15 Jul 2021 13:00:00 NZST in Link 309
All are welcome
Traditionally, traffic data are obtained from fixed inductive loop detectors which are costly to be installed and maintained. To monitor a large-scale network one cannot rely only on one type of data since each source of data carries partial information about the network traffic state. Thus, integrating, cleansing and filtering the big data collected from multiple sources such as loop detectors, surveillance cameras, Connected and Automated Vehicles trajectory, public transport smart cards and GPS devices to estimate principal parameters of traffic flow is crucial and still an open problem. Due to the increasingly availability of traffic data, adaptive control, which is intrinsically more effective and robust than the fixed-time control, is receiving more concern in both research and practical implementation. Current traffic optimization strategies are mostly unable to mitigate congestion in large-scale urban networks. In addition, the existing control strategies are either dealing with motorway or urban traffic flow dynamics. Mathematical modelling of the traffic dynamics of each element in a large urban network with a high number of links and intersections is a complex task. As the data-driven approaches usually deal with single data source while lacking mobility models, i.e., missing causality and physics, such methods are not robust and fragile to data error. Novel approaches have been developed to use multi-source data for traffic estimation as a basis of traffic management and monitoring. This led to better understanding of temporal-spatial evolution of traffic dynamics at local and network level. Innovative optimisation-based algorithms have been implemented to mitigate congestion, reduce emissions and improve urban mobility.
Dr. Mehdi Keyvan-Ekbatani completed his graduate studies in the field of transportation engineering at the Sharif University of Technology, Iran, in 2010. In 2010, he received a prestigious fellowship from the European Network of Excellence for Advanced Road Cooperative Traffic Management in the Information Society (NEARCTIS*) for his PhD project which was hosted at the Dynamic Systems and Simulation Laboratory (DSSL), TU Crete, Greece and Urban Transport Systems Laboratory (LUTS), EPFL, Switzerland. DSSL is among the leading institutions in Intelligent Transportation Systems (ITS); the lab was awarded with the IEEE ITS Institutional Lead Award in 2011. Right after finishing his PhD studies, he won the Canadian National Award for Top- Ranked Post-doctoral Fellows with his proposal on Motorway Traffic Control and Monitoring but he decided to accept an offer for a post-doc position at the Faculty of Civil Engineering and Geosciences, TU Delft. He was a Post-Doctoral Researcher with the Department of Transport and Planning, Delft University of Technology, The Netherlands, from 2014 to 2017. TU Delft was awarded with the IEEE ITS Institutional Lead Award in 2017. He is currently a Senior Lecturer at the University of Canterbury, New Zealand. His main research interests include traffic flow theory, traffic control, traffic estimation, network optimization, intelligent transportation systems, driving behaviour modelling, and data analytics in transportation. He is an Active Reviewer of more than 15 journals, and a member of several editorial boards and international scientific committees. He was the recipient of the Best Paper Award at the European Transportation Research Arena 2012 and the 2014 IEEE ITSS Best Ph.D. Dissertation Award.
*NEARCTIS consists of top-quality research groups in the area of Traffic Management in Europe: German Aerospace (DLR); EPFL (Switzerland); Imperial College London (UK); TU Crete (Greece); TU Delft (The Netherlands); University College London (UK); University of Southampton (UK); IFSTTAR (France)