Macroscopic Investigation of Microscopic Merging Models at Unsignalized Intersections

A classical way to represent vehicle interactions at unsignalized intersections is to combine a gap-acceptance model with a car-following algorithm. However, outputs of such a combination may appear inconsistent at a macroscopic level when compared to well-established analytical merging models. On the one hand, simulated results can be sensitive to the simulation time-step which specifies the number of times the decision model is evaluated. Yet, this issue is straightforward to overcome and does not generally affect classical microscopic simulation packages. On the other hand, the interactions between the car-following algorithm and the gap-acceptance model trigger more intricate shortcomings. Firstly, the outcome of the decision process is modified by the numerical viscosity appearing in the car-following algorithm. Secondly, the priority sharing process observed in congestion cannot be correctly reproduced because of the constraints imposed by the car-following on the gap-acceptance model. To get over these issues, a new microscopic framework in which the insertion process is continuous and independent of the car-following algorithm is proposed in this article. It is able to model in a unified way uncongested and congested unsignalized intersections and produces accurate capacity outputs.