An extended non-lane-discipline-based continuum model through driver behaviors for analyzing multi-traffic flows

Abstract

Non-lane-discipline environment has an essential role in driver behavior. In this environment, drivers change their driving behavior through the traffic flow, causing perturbations and complexity. To evaluate the effect of driver behavior under non-lane-based discipline on the traffic flow, the present study investigated the physical and dynamic characteristics of drivers on an urban highway in Iran for each vehicle class based on the lateral separation distance (LSD) parameter and strength influence factor. Then, an extended non-lane-discipline-based continuum model was proposed for different traffic conditions. Further, the proposed model was simulated using the adjusted upwind scheme for lane analysis. The field data results indicated that non-lane-based discipline has an important role in changing car drivers’ driving behaviors from typical and sluggish to aggressive behavior compared with lane-based discipline. Moreover, under maximum LSD, due to an increase in the lateral distance, car drivers became aggressive. In addition, there was a high stability region in traffic flow due to the positive effect of driver behavior, under non-lane-discipline environment, on the critical region of density where there was an increase in the traffic flow. However, in the free flow similar to jam density, the effect of the behavior on traffic flow was insignificant. Furthermore, driver behavior under non-lane-based heterogeneous traffic conditions could improve traffic flow. For the case of the car’s traffic flow regarding aggressive behavior under non-lane-based discipline, the capacity of lane reached 1.18–1.45 times higher in cars than that in sluggish behavior under lane-based discipline, respectively. Accordingly, the capacity of the lane was observed to be 1.50–3.41 times higher than that of trucks as the truck proportions decreased from 80% to 0% due to different LSD parameters and strength influence factors. Thus, the extended continuum model can appropriately predict traffic flow considering driver behavior under non-lane-based discipline.