An asymmetric cellular automata model for heterogeneous traffic flow on freeways with a climbing lane

Abstract

Traffic congestions frequently occur on uphill segments of four-lane freeways in China, which have become typical bottlenecks. Therefore, this paper focuses on the analysis, modeling, and simulation of heterogeneous traffic flow on uphill, in order to understand and eliminate such bottlenecks. The traffic characteristics were obtained from the realistic data, and a cellular automata model for longitudinal driving and lane changing was proposed and validated. The longitudinal driving rules were established based on the Nagel–Scheckenberg model. Lane changing was classified into active, inactive, and mandatory types which were used to clearly describe asymmetric lane-changing rules on two-lane segments and uphill with a climbing lane. The expressions of lane-changing motivation and safety were established. The measured results show that cars and 6-axle articulated trucks are the main types, and the speed difference between them is large. For normal slopes with a high truck ratio, even the total traffic is light, cars are unable to run freely. The simulated results prove that the realistic lane changing is asymmetric. The effects of uphill and climbing lanes on traffic flow are related to density. Setting up a climbing lane can alleviate or eliminate the uphill bottleneck effect. A critical density for distinguishing free flow from non-free flow does not exist on two-lane segments but exists on the uphill with a climbing lane. Vehicle segregation is significant under asymmetric lane-changing rules. The segregation degree is related to the traffic flow state.