A stochastic cellular automaton model for traffic flow with multiple metastable states

TL;DR

This paper introduces a stochastic cellular automaton model for traffic flow that captures multiple metastable states and reproduces complex flow-density relations observed in real traffic, enhancing understanding of traffic phase transitions.

Contribution

The paper extends existing CA models to include slow-to-start effects and driver perspective, revealing multiple metastable states and their stability in traffic flow.

Findings

Multiple metastable branches near transition density

Wide scattering in fundamental diagram

Explicit analysis of branch stability and velocity distributions

Abstract

A new stochastic cellular automaton (CA) model of traffic flow, which includes slow-to-start effects and a driver's perspective, is proposed by extending the Burgers CA and the Nagel-Schreckenberg CA model. The flow-density relation of this model shows multiple metastable branches near the transition density from free to congested traffic, which form a wide scattering area in the fundamental diagram. The stability of these branches and their velocity distributions are explicitly studied by numerical simulations.