Unveiling City Jam-prints of Urban Traffic based on Jam Patterns
Zeng Guanwen,Serok Nimrod,Lieberthal Efrat Blumenfeld,Duan Jinxiao,Liu Shiyan,Sui Shaobo,Li Daqing,Havlin Shlomo
2024-01-27
Abstract:We analyze the patterns of traffic jams in urban networks of five large cities and an urban agglomeration region in China using real data based on a recently developed jam tree model. This model focuses on the way traffic jams spread through a network of streets, where the first street that becomes congested represents the bottleneck of the jam. We extended the model by integrating additional realistic jam components into the model and find that, while the locations of traffic jams can vary significantly from day to day and hour to hour, the daily distribution of the costs associated with these jams follows a consistent pattern, i.e., a power law with similar exponents. This distribution pattern appears to hold not only for a given region on different days, but also for the same hours on different days. This daily pattern of exponent values for traffic jams can be used as a fingerprint for urban traffic, i.e., jam-prints. Our findings are useful for quantifying the reliability of urban traffic system, and for improving traffic management and control.
Physics and Society