Multi-class AR model: comparison with microsimulation model for traffic flow variables at network level of interest and the two-dimensional formulation

This paper discusses on the choice of a recently proposed multi-class, speed-area occupancy based, one-dimensional continuum model towards the measurement of network level traffic flow variables, namely, travel time and outflow. The parameters of the model are calibrated for a hypothetical mid-block with four vehicle classes, using the micro simulation data from VISSIM software. Analysis result indicates that the model produces travel time and outflow result closer to that by the VISSIM micro simulation. To explore this macroscopic model towards accurate modelling of lacking lane disciplined traffic, i.e. the traffic with frequent lateral movements, a two-dimensional formulation of the model is proposed. It is shown that the new formulation retains the properties of its original version as well as the desirable properties of any traffic flow model.

Automated summary

This paper discusses the choice of a recently proposed model for measuring network level traffic flow variables and calibrates the parameters using micro simulation data. The analysis results show that the model produces results closer to micro simulation results. A two-dimensional formulation of the model is proposed to accurately model traffic with lacking lane discipline.