Nash Flows over Time with Spillback and Kinematic Waves

TL;DR

This paper extends the deterministic queuing model for traffic networks to include spillback and kinematic waves, providing a characterization of Nash flows over time and an algorithm for computing dynamic equilibria.

Contribution

It introduces a generalized model incorporating spillback and kinematic waves, and proves the existence of dynamic equilibria with a constructive approach.

Findings

Characterization of Nash flows via spillback thin flows

Existence proof of dynamic equilibria

Algorithm for computing equilibria

Abstract

Modeling traffic in road networks is a widely studied but challenging problem, especially under the assumption that drivers act selfishly. A common approach is the deterministic queuing model, for which the structure of dynamic equilibria has been studied extensively in the last couple of years. The basic idea is to model traffic by a continuous flow that travels over time through a network, in which the arcs are endowed with transit times and capacities. Whenever the flow rate exceeds the capacity the flow particles build up a queue. So far it was not possible to represent spillback or kinematic waves in this model. By introducing a storage capacity arcs can become full, and thus, might block preceding arcs, i.e., spillback occurs. Furthermore, we model kinematic waves by upstream moving flows over time representing the gaps between vehicles. We carry over the main results of the original model to our generalization, i.e., we characterize Nash flows over time by sequences of particular static flows, so-called spillback thin flows. Furthermore, we give a constructive proof for the existence of dynamic equilibria, which suggests an algorithm for their computation. This solves an open problem stated by [13].