Effect of driver distraction on traffic flow stability via desired safety margin model

Car-following is the most common traffic phenomenon in freeway. Driver behavior characteristics play important role in determining the qualitative dynamics of vehicles in car-following process. A number of driver behavior characteristics associated with uniform traffic flow may be affected by driver distraction. Therefore, this study aims to explore the effect of driver distraction on traffic flow stability from the microscopic viewpoints. Based on the risk homeostasis theory and stimulus-response concept, a desired safety margin car-following model was employed to investigate the variation of traffic flow. Participants make corresponding driving behaviors in the simulated environment based on the motion state of the preceding car with differing traffic scenarios. The motion information of simulated vehicles is collected by driving simulation experiments to analyze the driving behavior parameters under both no task and secondary task situations. Five driving behavior parameters, including the level of perceived risk, the acceleration and deceleration habits, and the reaction characteristics of the participant, are obtained. Analytical results indicate that driver distraction leads to different driving behaviors, thereby affecting the performance of traffic flow and reducing traffic efficiency. Furthermore, we analyze the string stability of traffic flow by numerical simulations. Compared with no task, numerical simulations indicate that driver distraction causes the traffic flow to become unstable. Those findings imply that the use of in-vehicle cell phone or car radio and associated public policy may need to further inform.