Similar is Better: Speed Variability Reduces Traffic Flow

Abstract

Every driver knows that severe weather conditions cause traffic congestions. In many cases there is no direct reason for the congestion, and people tend to attribute it to the slow driving mode. Our computational study shows that the slow driving cannot account for this phenomenon. It comes out that the reason is not the decrease in mean velocity, but rather the change in the speed distribution on the road. Width of the distribution, namely - the number of "very fast" or "very slow" vehicles, affects both the need and the availability of lane change and bypassing. Besides providing insight and analyzing the underlying mechanism of a collective phenomenon, this example sheds light on a fundamental aspect of computational modeling. Although simple-as-possible models are desirable and provide better understanding of the real important component of the modeled system, they can very easily turn into over-simplifications and miss relevant qualitative phenomena.