Abstract
Measurement of traffic flow stability and resilience is a critical step toward evaluating the performance of transportation systems and implementing appropriate management strategies. Quantifying changes in the stability and resilience of transportation systems, however, is hampered by the complexity of real traffic dynamics and the diversity of infrastructures. Here, we demonstrate that changes in traffic flow stability and resilience are signaled by generic features, known as early warning signals in the theory of critical slowing down, observed before traffic instabilities occur. This finding is incorporated in an operational data-driven algorithm to evaluate the risk of traffic jams on highways. Theoretical findings and tests on simulated and empirical case studies support the premise of this approach and identify candidate statistical measures that are sensitive to changes in the stability and resilience of transportation systems. Our use of universal measures advances the monitoring capability, prediction and control of complex transportation systems.
Highlights
C ONTINUOUS growth of the number of motor vehicles has made traffic congestion a serious problem around the world
We assessed the performance of three time-varying early warning signals, namely variance, lag-1 autocorrelation, and spectral density ratio extracted from measured stochastic fluctuations of the observed flow dynamics on highways
In addition to warning impending instabilities, early warning metrics are shown to be sensitive to changes in the stability and resilience of the traffic dynamics making them a potential tool to evaluate the effectiveness of adopted management strategies
Summary
C ONTINUOUS growth of the number of motor vehicles has made traffic congestion a serious problem around the world. Traffic congestion is intertwined with other concerns that significantly affect quality of life, ranging from economic and environmental problems to behavioral and health consequences related with vehicle emission [1]–[3]. Recent concepts related to smart cities and flexible, multi-modal transportation further increase the complexity of the dynamics of transportation systems of the future. The study of the complex dynamics of traffic has emerged as a topic of considerable attention from engineers, planners, and policymakers [4]–[6], with the ultimate goal of improving the stability and resilience. Manuscript received August 16, 2020; revised March 2, 2021 and May 18, 2021; accepted June 30, 2021.
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
