Probabilistic modeling of acceleration in traffic networks as a function of speed and road type

Abstract

Statistical acceleration and deceleration distributions are developed as a function of speed and road type. The approach allows for the estimation of acceleration and deceleration variation among vehicles on a link with a given speed. Acceleration is shown to be a random variable that follows a probabilistic distribution that is practically independent of the road type. For the given data set, this distribution is a half-normal distribution for both acceleration and deceleration. Moreover, the standard deviation of the distributions decreases as the speed range increases. The developed model has a number of applications, especially where acceleration needs to be modeled as in the case of non-microscopic traffic models. In such context, instantaneous emission models benefit most from this analysis as these models account for engine operation, accelerations, or other power surrogate terms, which lead to the generation of tailpipe emissions. Results of this paper also have applications for designing and validating regulatory driving cycles.