An easy-to-use algorithm for simulating traffic flow on networks: theoretical study
For researchers in traffic flow modeling, this work offers an easy-to-implement algorithm that simplifies junction handling, though it is a methodological contribution rather than a breakthrough.
The paper presents a Godunov-based numerical scheme for traffic flow on networks that automatically selects a reasonable solution at junctions without external procedures, and shows its equivalence to classical traffic flow theory.
In this paper we study a model for traffic flow on networks based on a hyperbolic system of conservation laws with discontinuous flux. Each equation describes the density evolution of vehicles having a common path along the network. In this formulation the junctions apparently disappear since each path is considered as a single uninterrupted road. We consider a Godunov-based approximation scheme for the system which is very easy to implement. Besides basic properties like the conservation of cars and positive bounded solutions, the scheme exhibits nice properties, being able to select automatically a reasonable solution at junctions without requiring external procedures (e.g., maximization of the flux via a linear programming method). Moreover, the scheme can be interpreted as a discretization of the traffic models with buffer, although any buffer is introduced here. Finally, we show how the scheme can be recast in the framework of the classical theory of traffic flow on networks, where a conservation law has to be solved on each arc of the network. This is achieved by solving the Riemann problem for a modified equation, and showing that its solution corresponds to the one computed by the numerical scheme.