Verkehrsmodellierung mit öffentlich zugänglichen Daten: Eine Evolution für Modelleingangsdaten

• Antragsteller: Professor Dr. Constantinos Antoniou; Professor Dr.-Ing. Rolf Moeckel (†)
• Fachliche Zuordnung: Verkehrs- und Transportsysteme, Intelligenter und automatisierter Verkehr
• Förderung: Förderung von 2019 bis 2024
• Projektkennung: Deutsche Forschungsgemeinschaft (DFG) - Projektnummer 415208373

• Erstellungsjahr: 2023

Zusammenfassung der Projektergebnisse

Data are indispensable for transport modeling and analyses, providing insights into travel patterns and behavior. Advancements in sensing and communication technologies have led to the development of non-conventional and emerging data sources. Due to their public availability, these data are also increasingly used in transport research and practice. The TraMPa project is a diverse exploration of the potential and limitations of open data in transport modeling on conceptual and methodological levels. Conceptually, we categorize various data sources, including mobile phone and social media data, and standardized formats like General Transit Feed Specifications (GTFS) according to their openness. Based on the scientific literature, we uncovered which data have been used for which transport modeling task. Through an online survey, we assessed the modelers’ motivations and skepticism towards using open data in transport modeling. From a methodological perspective, we developed novel approaches for processing and modeling data from open sources for specific applications in transport modeling. On the demand-side modeling, the project’s main methodological highlight is the creation of an open synthetic population (SP), providing a more flexible alternative to traditional weighted census microdata. The project evaluates the performance of the travel demand model MITO with open data, confirming the suitability of open synthetic populations. The developed approaches exploit state-of-the-art methods from data science to address the quality of novel data, indirectly predict traffic information and calibrate the traffic simulations. These methods stress the data scarcity and automation challenges during data processing and traffic simulation set-up. Open data sources contribute positively to transport modeling, although challenges remain. Data noise, unknown data processing methods by the data owner and uncertain future availability of the same data provide challenges for transport modeling. Nevertheless, the project confirms great potential of open data to at least complement traditional data sources. Overall, TraMPa offers an exploration of open data’s potential in transforming transport modeling, highlighting possibilities, preferences and limitations in the evolving landscape of data-driven transportation. The most imporant publication that came out of this project provides a systematic overview of available open data for transport modeling and analyses.

Projektbezogene Publikationen (Auswahl)

• Impact of simulation-based traffic noise on rent prices. Transportation Research Part D: Transport and Environment, 78, 102191.
  Kuehnel, Nico & Moeckel, Rolf
  
• Predicting Venue Popularity Using Crowd-Sourced and Passive Sensor Data. Smart Cities, 3(3), 818-841.
  Timokhin, Stanislav; Sadrani, Mohammad & Antoniou, Constantinos
  
• Prediction of Lane-Changing Maneuvers with Automatic Labeling and Deep Learning. Transportation Research Record: Journal of the Transportation Research Board, 2674(7), 336-347.
  Mahajan, Vishal; Katrakazas, Christos & Antoniou, Constantinos
  
• Data to the people: a review of public and proprietary data for transport models. Transport Reviews, 42(4), 415-440.
  Mahajan, Vishal; Kuehnel, Nico; Intzevidou, Aikaterini; Cantelmo, Guido; Moeckel, Rolf & Antoniou, Constantinos
  
• Explaining demand patterns during COVID-19 using opportunistic data: a case study of the city of Munich. European Transport Research Review, 13(1).
  Mahajan, Vishal; Cantelmo, Guido & Antoniou, Constantinos
  
• Temporal distribution of sociodemographic characteristics at transit stops. Transportation Planning and Technology, 44(2), 208-221.
  Faroqi, Hamed; Moeckel, Rolf & Mesbah, Mahmoud
  
• Crash Risk Estimation Due to Lane Changing: A Data-Driven Approach Using Naturalistic Data. IEEE Transactions on Intelligent Transportation Systems, 23(4), 3756-3765.
  Mahajan, Vishal; Katrakazas, Christos & Antoniou, Constantinos
  
• Predicting network flows from speeds using open data and transfer learning. IET Intelligent Transport Systems, 17(4), 804-824.
  Mahajan, Vishal; Cantelmo, Guido; Rothfeld, Raoul & Antoniou, Constantinos
  
• “Can location-based publicly available data to improve destination choice models?” In: mobil.TUM. Singapore.
  Llorca, C. & S. Arora, G. Chao, R. Moeckel
  
• Treating Noise and Anomalies in Vehicle Trajectories From an Experiment With a Swarm of Drones. IEEE Transactions on Intelligent Transportation Systems, 24(9), 9055-9067.
  Mahajan, Vishal; Barmpounakis, Emmanouil; Alam, Md. Rakibul; Geroliminis, Nikolas & Antoniou, Constantinos