Supply-demand diagrams and a new framework for analyzing the inhomogeneous Lighthill-Whitham-Richards model

TL;DR

This paper introduces a novel supply-demand diagram framework for analyzing the inhomogeneous LWR traffic model, simplifying the solution of Riemann problems and enabling analysis of complex road networks.

Contribution

The paper presents a new supply-demand diagram approach that allows for simpler, more intuitive analysis of inhomogeneous traffic models and the existence of interior states at boundaries.

Findings

Existence and uniqueness of stationary states are established.

A graphical scheme for solving Riemann problems is developed.

Interior states can exist next to linear boundaries, extending analysis capabilities.

Abstract

Traditionally, the Lighthill-Whitham-Richards (LWR) models for homogeneous and inhomogeneous roads have been analyzed in flux-density space with the fundamental diagram of the flux-density relation. In this paper, we present a new framework for analyzing the LWR model, especially the Riemann problem at a linear boundary in which the upstream and downstream links are homogeneous and initially carry uniform traffic. We first review the definitions of local supply and demand functions and then introduce the so-called supply-demand diagram, on which a traffic state can be represented by its supply and demand, rather than as density and flux as on a fundamental diagram. It is well-known that the solutions to the Riemann problem at each link are self-similar with a stationary state, and that the wave on the link is determined by the stationary state and the initial state. In our new framework, there can also exist an interior state next to the linear boundary on each link, which takes infinitesimal space, and admissible conditions for the upstream and downstream stationary and interior states can be derived in supply-demand space. With an entropy condition consistent with a local supply-demand method in interior states, we show that the stationary states exist and are unique within the solution framework. We also develop a graphical scheme for solving the Riemann problem, and the results are shown to be consistent with those in the literature. We further discuss asymptotic stationary states on an inhomogeneous ring road with arbitrary initial conditions and demonstrate the existence of interior states with a numerical example. The framework developed in this study is simpler than existing ones and can be extended for analyzing the traffic dynamics in general road networks.