Driver heterogeneity impact on aggregated traffic behavior

Abstract

A new traffic model is developed to evaluate the impact of driver heterogeneity, including driver behavior and braking distance. This is achieved by incorporating the effect of different drivers and changes in braking distance. Thus, real world traffic conditions are considered. It is shown that the proposed model is hyperbolic and stable, so it is well posed and admissible. The performance of this model is compared with the widely used Payne–Witham (PW) model over a 3000 m circular road in MATLAB using the FORCE scheme. This represents a worst-case traffic scenario. The results obtained indicate that the velocity and driver anticipation evolve realistically with the proposed model, so it can effectively characterize driver heterogeneity. Conversely, the PW model produces unreasonable results.