Stochastic noise approach to traffic flow modeling
Author
Summary, in English
Traffic flow states are described as resulting from a stochastically driven system. Vehicles advance based on the energy profile of their surrounding traffic.
We create a stochastic process generated from an ergodicity satisfying Markov chain whose system dynamics sample from the Gibbs distribution. Specifically, we employ Arrhenius microscopic dynamics in order to also capture non-equilibrium behavior and monitor the states favored by the system through its time evolution.
Monte Carlo simulations of this traffic system provide information and statistics regarding free-flow, “synchronized” traffic, jam wave formation or dissipation, “stop and go” regimes and a variety of interesting such traffic behavior, summarized in, among others, the fundamental diagram. Generalizations to the current model and a number of ideas for further studies are proposed.
We create a stochastic process generated from an ergodicity satisfying Markov chain whose system dynamics sample from the Gibbs distribution. Specifically, we employ Arrhenius microscopic dynamics in order to also capture non-equilibrium behavior and monitor the states favored by the system through its time evolution.
Monte Carlo simulations of this traffic system provide information and statistics regarding free-flow, “synchronized” traffic, jam wave formation or dissipation, “stop and go” regimes and a variety of interesting such traffic behavior, summarized in, among others, the fundamental diagram. Generalizations to the current model and a number of ideas for further studies are proposed.
Publishing year
2004
Language
English
Pages
741-754
Publication/Series
Physica A: Statistical Mechanics and its Applications
Volume
342
Issue
3-4
Document type
Journal article
Publisher
Elsevier
Topic
- Mathematics
Keywords
- Traffic flow
- Stochastic Arrhenius microscopic dynamics
- Monte Carlo simulations
Status
Published
ISBN/ISSN/Other
- ISSN: 0378-4371