Abstract not found

An agent-based approach to simulation for transportation planning applications offers a lot of conceptual flexibility. Many millions of agents plus many hundreds of thousands of elements of transportation infrastructure need to be represented. For transportation planning applications, the demand needs to be sensitive to changes in supply, which implies that besides the realistic representation of the transportation system it is also necessaxy to represent people's decision-making process leading to the demand.

We study several deterministic one-dimensional traffic models. For integer positions and velocities we find the typical high and low density phases separated by a simple transition. If positions and velocities are continuous variables the model shows self-organized criticality driven by the slowest car.

Abstract: We study a model for freeway traffic which includes strong noise taking into account the fluctuations of individual driving behavior. The model shows emergent traffic jams with a self-similar appearance near the throughput maximum of the traffic. The lifetime distribution of these jams shows a short scaling regime, which gets considerably longer if one reduces the fluctuations when driving at maximum speed but leaves the fluctuations for slowing down or accelerating unchanged. The outflow from a traffic jam self-organizes into this state of maximum throughput. 1

Modeling and simulation of traffic has a long tradition. A vast number of different approaches have been used to simulate traffic, each of which has been calibrated and validated separately. This work aims at the way different models are interrelated. It is shown how one can, starting from very general modeling assumptions, construct a family of car following models that contains models closely related to well-known simulation models as special cases. Investigating this model family it can be shown, which models are qualitatively equivalent and which are not. This gives important hints as to which model approaches can in principle be unified.

A set of very simple rules for driving behavior used to simulate roadway traffic gives realistic results. Because of its simplicity, it is easy to implement the model on supercomputers (vectorizing and parallel), where we have achieved real time limits of more than 4 million kilometers (or more than 53 million vehicle sec/sec). The model can be used for applications where both high simulation speed and individual vehicle resolution are needed. We use the model for extended statistical analysis to gain insight into traffic phenomena near capacity, and we discuss that this model is a good candidate for network routing applications.

this paper is as follows: Firstly, the cellular automaton approach is introduced. Relevant quantities of the measurements are presented. Additionally, an overview of modifications of the original model is given which are needed to reproduce a variety of effects known from real traffic under several conditions. We will also present analytical treatments valid at least in certain limits or approximations. The underlying road network and database are described in the third section. Some measurements of the network and the reproduction of traffic states are discussed in section four.

The aim of this article is to show the applicability of a simple traffic simulator for the calculation of emissions. Traffic is simulated by using a queuing model (Q--model) originally introduced to solve the dynamic traffic assignment problem (DTA). Although vehicle dynamics is modeled on a quite coarse level it can be shown that, given certain conditions, emissions computed with the Q-model agree fairly good with those computed with more detailed models. For this purpose, we have done extensive comparisons of emissions computed with a car-following model which explicitly mimics single vehicle dynamics.

This review addresses recent developments in nonequilibrium statistical physics. Focusing on phase transitions from fluctuating phases into absorbing states, the universality class of directed percolation is investigated in detail. The survey gives a general introduction to various lattice models of directed percolation and studies their scaling properties, fieldtheoretic aspects, numerical techniques, as well as possible experimental realizations. In addition, several examples of absorbing-state transitions which do not belong to the directed percolation universality class will be discussed. As a closely related technique, we investigate the concept of damage spreading. It is shown that this technique is ambiguous to some extent, making it impossible to define chaotic and regular phases in stochastic nonequilibrium systems. Finally, we discuss various classes of depinning transitions in models for interface growth which are related to phase transitions into absorbing states. Keywords: nonequilibrium phase transitions, stochastic lattice models, directed percolation, contact process, absorbing state transitions, parity-conserving class, damage spreading, interface growth, depinning transitions, roughening transitions, nonequilibrium wetting

"Verkehrschaos" ist ein wohlbekannter Begriff der Umgangssprache, aber die wenigen Versuche, ihn auf eine wissenschaftliche Grundlage zu stellen, benutzen die Kontinuumsapproximation. Wir verwenden stattdessen einen zellularen Automaten, der für die recht hohe Granularität von Verkehr (individuelle Fahrzeuge) auf natürliche Weise geeignet erscheint. Unsere Simulationen laufen deutlich schneller als die anderer Mikrosimulationsprojekte. Wir zeigen, neben Performance-Ergebnissen, Resultate einer Simulation des gesamten Autobahnverkehrs von Nordrhein-Westfalen, Ansätze zur Klärung der Chaosfrage, sowie Experimente zu Advanced-Traffic-Management-Manahmen.