Abstract
We study the dynamical process of congestion formation for large-scale urban networks by exploring a unique dataset of taxi movements in a megacity. We develop a dynamic model based on a reaction and a diffusion term that properly reproduces the cascade phenomena of traffic. The interaction of these two terms brings the values of the speeds on road network in self-organized patterns and it reveals an elegant physical law that reproduces the dynamics of congestion with very few parameters. The results presented show a promising match with an available real data set of link speeds estimated from more than 40 millions of GPS coordinates per day of about 20,000 taxis in Shenzhen, China.
Highlights
We study the dynamical process of congestion formation for large-scale urban networks by exploring a unique dataset of taxi movements in a megacity
While there is a vast literature on congestion dynamics, control and spreading in one-dimensional traffic systems with a single mode of traffic[12,13,14,15,16,17], most of the analysis at the network level is based on simplistic models or detailed simulation, which requires a large number of input parameters and cannot be solved in real-time
The shape of Macroscopic Fundamental Diagram (MFD) is a property of the network infrastructure and control and not very sensitive to the demand, but it depends on the spatial heterogeneity of congestion
Summary
We study the dynamical process of congestion formation for large-scale urban networks by exploring a unique dataset of taxi movements in a megacity. Combining realistic modeling of congestion and efficient traffic control for large scale urban systems remains a big challenge This is due to the high unpredictability of choices of travelers (in terms of route, time of departure and mode of travel), the uncertainty in their reactions to the actions of others (users or controllers), the spatiotemporal propagation of congestion, the lack of coordinated actions coupled with the limited infrastructure available[1,2] and the strong scatter of the data for link-level traffic sensors. The advantages of this type of models are the low computation cost and the possibility to use well-defined empirical relations between the main traffic variables: flow, density and speed, known as Fundamental Diagram (FD) Note that these relations are not for individual links as in the traditional link-level models, but for homogeneously congested regions of a city. It emerges, for example in ref. 33, the complexity to calibrate the huge amount of parameters (route and mode choice, traffic dynamics, demand generation and more) in the widely used micro-simulators for driving behavior models and the difficulty to find a universal and definitive values system
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
