Title: The future's back : first generic models of self-defeating road traffic forecasting
Authors : Ott, Thomas
Füchslin, Rudolf Marcel
Conference details: NDES 2017: "Novel technologies from knowing how networks work", Nonlinear Dynamics of Electronic Systems NDES, Zernez, Switzerland, 5–7 June 2017
Issue Date: 5-Jun-2017
Language : Englisch / English
Subjects : Modeling; Traffic
Subject (DDC) : 303: Soziale Prozesse
380: Kommunikation, Verkehr
Abstract: Traffic forecasts are critical for effective road traffic management. Driven by growing needs and the availability of more and new data and technology, traffic forecasting methods have become a vivid research field in the last couple of years (e.g. [1,2]). However, almost no attention has been paid so far to the effect of forecasting information on the traffic situation. From the perspective of traffic management, the desired effect often corresponds to a self-defeating prophecy; in the case of a predicted congestion, the goal of forecasting-based traffic management is to prevent or mitigate the predicted situation. In practice, the dynamics of this nonlinear feedback effect is hardly comprehended in traffic management. In our contribution, we elucidate the issue using the example of the South-to-North transit route in Switzerland and we derive reduced generic models of self-defeating traffic forecasting. They may provide a first step toward improved decision-making for traffic management, taking into account the repercussions of traffic recommendation on the system under consideration.
Departement: School of Engineering
Organisational Unit: Institut für Angewandte Mathematik und Physik (IAMP)
Institut für Angewandte Simulation (IAS)
Publication type: Konferenz: Sonstiges / Conference Other
Type of review: Keine Angabe / Not specified
URI: https://digitalcollection.zhaw.ch/handle/11475/2696
License (according to publishing contract) : Lizenz gemäss Verlagsvertrag / Licence according to publishing contract
Appears in Collections:Publikationen School of Engineering

Files in This Item:
There are no files associated with this item.

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.