An empirical study on the traffic state evolution and stop-and-go traffic development on freeways

Abstract

Stop-and-go traffic, related to traffic breakdown and instability, is the core mechanism of traffic state transition to congestion. The purpose of this paper is to provide improved empirical understanding on the development of the stop-and-go traffic from asymmetric theory's viewpoint. We analyse traffic state transition before and after stop-and-go wave by observing platoon trajectories using NGSIM data. We observe that stop-and-go wave develops into growth (dissipation) in unstable (stable) traffic, which becomes stable (unstable) by passing the wave, respectively. Alternating patterns can explain recurring oscillatory traffic. Besides, we investigate the relationship between the evolution of stop-and-go wave and lane-changes. We can find that the growth region in the flow-density plane is located on the higher flow and density area above dissipation region. Also, as the number of lane-change increases (decreases), traffic will be developed to growth (dissipation), respectively. The findings of this research have potential for detailed traffic control over stop-and-go waves' development.