Multi-directional continuous traffic model for large-scale urban networks

In this paper, the authors propose a new multi-directional two-dimensional continuous model for urban traffic. It is called the NEWS model, since it represents a system of four partial differential equations (PDEs) that describe propagation of vehicle density in cardinal direction layers: North, South, West and East. The NEWS model can be applied to predict traffic evolution on a general urban network of arbitrary size by knowing only its boundary flows, as well as its topology and infrastructure parameters such as roads speed limits, number of lanes and capacities. The flux direction is retrieved from turning ratios at intersections, which is then aggregated in four direction layers using geometrical projection matrices. The authors show its formal derivation step-by-step from the classical Cell Transmission Model at one intersection. The authors then show that the NEWS model is a hyperbolic PDE system that corresponds to a conservation law with bounded densities. The model prediction performance is validated using synthetic data from the microsimulator Aimsun. Finally, the model is also validated using real data collected from a network of sensors installed in Grenoble (a city in France).

Language

  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01842011
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Apr 7 2022 5:31PM