Comparison of Highway Traffic Breakdown Features between U.S. and China Using Discrete Wavelet Transform: An Empirical Study

摘要

This paper provides empirical observations and measurements on highway traffic breakdown using discrete wavelet transform for revealing some important features related to breakdown activation and recovery process during transitions from free-flow traffic to congested states. The method effectively distinguishes useful breakdown information from nonlinear and nonstationary speed time series. Comparatively, the analysis of loop detector data from two highways (I80-W in Berkeley, CA, U.S. and G6-S in Beijing, China) shows that the wavelet-based energy approach enables effective identification of traffic breakdown occurrence time, recovery time, state transition time, and congested speed. Observations show that I80-W performs longer transition time and recovery time in traffic breakdown than G6-S, from which it is inferred that traffic breakdown occurs and recovers more abruptly on G6-S. The empirical study has important implications for understanding the breakdown process and statistical characteristics of traffic state transitions.