A stochastic driver distraction model for microscopic traffic simulations

In this paper we propose a novel approach towards integrating driver distractions with microscopic traffic simulations for the purpose of evaluating the impact of distracted driving on traffic flow in large-scale simulator studies. We use stochastic processes to model the frequency and duration of distractive tasks and, by doing so, generate distraction profiles for every individual vehicle in the simulation. We validate our model with the aid of an empirical data set and show that it is capable of reproducing the statistical properties derived from these data to a satisfying extent by performing dynamic simulations using the microscopic traffic simulator TraffSim.