Project

Back to overview

Big data transport models: The example of road pricing

English title Big data transport models: The example of road pricing
Applicant Axhausen Kay W.
Number 167189
Funding scheme NRP 75 Big Data
Research institution Institut für Verkehrsplanung und Transportsysteme ETH Zürich
Institution of higher education ETH Zurich - ETHZ
Main discipline Other disciplines of Engineering Sciences
Start/End 01.04.2017 - 31.12.2020
Approved amount 661'478.00
Show all

All Disciplines (2)

Discipline
Other disciplines of Engineering Sciences
Information Technology

Keywords (6)

Road pricing; Mode detection; Transport model; Elasticity ; Optimal calibration; Big data

Lay Summary (German)

Lead
Smartphone-Nutzer erzeugen täglich eine grosse Menge an Bewegungsdaten. Städte könnten diese Daten anonymisiert nutzen, um ihre Verkehrssysteme zu optimieren. Dieses Projekt möchte mit der Entwicklung eines Mobility-Pricing-Ansatzes zu diesen Optimierungen beitragen.
Lay summary

 

Dank der Zusammenarbeit mit einem Mobilfunkbetreiber stehen dem Projekt vier aussergewöhnlich detaillierte Datensätze zur Verfügung:

  1. ein aggregierter Datensatz zum Fernreiseverhalten, der erstmals eine korrekte Schätzung der Touren über 50 km ermöglicht
  2. Anonymisierte 50'000 Personentage, um neue Algorithmen zur Erkennung der täglichen Verhaltensmuster zu entwickeln
  3. Anonymisierte 10'000 Personenwochen, um detaillierte Modelle des Verkehrsverhaltens zu schätzen
  4. Aggregierte durchschnittliche stündliche Nachfrage pro Monat von jeder Schweizer Gemeinde zu jeder anderen unter Einhaltung des Datenschutzes fliessen die Arbeiten in die Erstellung und Kalibrierung des Schweiz-Modells ein, das wir auf der Grundlage der individuen-basierten Software MATSim erstellen.

Auf vielen städtischen Strassen und Schienen kommt es in Spitzenzeiten zu Engpässen, doch in der übrigen Zeit ist die Infrastruktur nur schwach ausgelastet. Smartphone-Daten geben ein umfassendes Bild der Nutzung der Stadt. Dieses Projekt wird aufgrund von anonymisierten Daten von Mobilfunk-Abonnenten ein individuen-basiertes Modell der Schweiz verbessern – und einen Ansatz mit unterschiedlichen Transportpreisen entwickeln, das die Schweizer Verkehrspolitik unterstützen kann.

Wir werden beispielhaft auf der Grundlage von anomysierten Mobilfunk-Datensätzen neue Auswertemethoden entwickeln. Die Auswertung der elektronischen Spuren soll die bisherige individuen-basierte Simulation verbessern – und beispielhaft die Möglichkeiten, Kosten und Grenzen eines Schweizer Mobility-Pricing-Ansatzes erkunden und optimieren.

Direct link to Lay Summary Last update: 16.08.2017

Lay Summary (French)

Lead
Les utilisateurs de smartphones engendrent quotidiennement un très grand volume de données de mouvement. Autant de données anonymisées dont les villes pourraient se servir en vue d’optmiser leurs systèmes de transport. Ce projet s’inscrit dans cette logique développant une approche de tarification de la mobilité.
Lay summary

Grâce à une collaboration avec un opérateur réseau, le projet dispose de quatre jeux de données exceptionnellement détaillés:

  1. un jeu de données agrégées sur les voyages longue distance qui permet pour la première fois une estimation correcte des trajets supérieurs à 50 kilomètres;
  2. 50’000 jours-personnes anonymisées, pour développer de nouveaux algorithmes afin de cerner le modèle de comportement quotidien ;
  3. 10’000 semaines-personnes anonymisées, pour estimer des modèles détaillés de comportements en matière de transport ;
  4. demande horaire moyenne agrégée par mois de chaque commune suisse à chaque autre.

En accord avec la législation sur la protection des données, les travaux serviront à élaborer et calibrer la modélisation de la Suisse que nous réaliserons sur la base du logiciel individu-centré MATSim.

Dans les villes, l’infrastructure de transport est engorgée aux heures de pointe, alors qu’elle n’est que faiblement utilisée le reste du temps. Les données issues des smartphones donnent une image complète de l’utilisation de la ville. Sur la base de données anonymisées provenant d’abonnés d’un opérateur réseau, ce projet vise à améliorer une modélisation individu-centrée de la Suisse et à développer une nouvelle approche tarifaire de la mobilité qui pourrait se révéler utile à la politique suisse des transports.

A partir de jeux de données anonymisées de téléphones mobiles, nous allons développer de nouvelles méthodes d’évaluation. L’analyse de ces traces électroniques doit améliorer l’ancienne simulation individu-centrée ainsi qu’explorer et optimiser les possibilités, les coûts et les limites d’une tarification de la mobilité en Suisse.

Direct link to Lay Summary Last update: 16.08.2017

Lay Summary (English)

Lead
Each day, smartphone users generate an large volume of time-stamped location data. Cities could use this anonymised data to optimise their transport systems. The aim of this project is to develop a mobility pricing approach as a contribution to these optimisations.
Lay summary

The collaboration with a network operator gives the project access to four exceptionally detailed data sets:

  1. An aggregated data set on long-distance travel behaviour patterns allowing a precise estimate of trips over 50 kilometres for the first time;
  2. Anonymised 50,000 person-days permitting the development of new algorithms to identify daily behaviour patterns;
  3. Anonymised 10,000 person-weeks facilitating the estimation of detailed traffic behaviour models;
  4. Aggregated, average hourly demand by month from each Swiss commune to every other.

Work will comply with data-protection regulations and aims to improve and calibrate the Switzerland model, which we will construct using the individual-based software MATSim.

Many city road and rail trips face bottlenecks during peak periods, whereas for the remainder of the time infrastructure utilisation is low. Smartphone data provides a comprehensive picture of usage of a city. Drawing on anonymised data from network operator subscribers, this project aims to refine an individual-based model of Switzerland – and to develop an approach with dynamic transport prices that is capable of supporting Swiss transport policy.

By way of example, we will develop new modelling methods based on anonymised mobile phone data sets. The objective behind modelling these electronic traces is to enhance current individual-based simulations – and explore and optimise example scenarios for the possibilities, costs and limitations of a Swiss mobility pricing approach.

Transport models can be constructed faster and calibrated more reliably using large volumes of data. The models will serve to improve traffic system control and thus help to smooth infrastructure peak loads. The data-optimised MATSim model of Switzerland and the mobility pricing findings have the potential to make a key contribution to maximising transport system efficiency, which in turn will enhance the well-being of everyone.


Direct link to Lay Summary Last update: 16.08.2017

Responsible applicant and co-applicants

Employees

Publications

Publication
Undertaking mobility field experiments using GPS Tracking
Molloy J. (2021), Undertaking mobility field experiments using GPS Tracking, ETH Zürich, Switzerland.
Activity Estimation from Mobile Phone Data
Cik M., Lechner A., Hebenstreit C., Fellendorf M. (2020), Activity Estimation from Mobile Phone Data, in Paper presented at the Transportation Research Board Annual Meeting, January 12-16, Washington DC, U, Washington DC, United StatesTransportation Research Board, Washington, D.C.
Generating Synthetic mobile phone datasets using MATSim
Molloy Joseph, Cik Michael, Fellendorf Martin, Axhausen Kay W. (2020), Generating Synthetic mobile phone datasets using MATSim, in Paper presented at the 20th Swiss Transport Research Conference (STRC 2020) (virtual), May 13-14, Ascona, SwitzerlandSTRC, Switzerland.
A ticket-based public transport pricing model for Switzerland
Hörl Sebastian, Molloy Joseph (2019), A ticket-based public transport pricing model for Switzerland, in Paper presented at the 19th Swiss Transport Research Conference (STRC 2019), Ascona, Switzerland, Ma, Ascona, SwitzerlandSTRC, Switzerland.
Trip purpose imputation from mobile phone trajectories using an artificial Neural Network
Stadlbauer B. (2019), Trip purpose imputation from mobile phone trajectories using an artificial Neural Network, TU Graz, Graz, Austria.
Aktivitätenerkennung aus Mobilfunkdaten-Events
Lechner A. (2018), Aktivitätenerkennung aus Mobilfunkdaten-Events, TU Graz, Graz, Austria.
Calibration of Agent Based Transport Simulations with Multi-Fidelity Bayesian Optimization
Turchetta M., Makarova A., Beyeler S., Krause A. (2018), Calibration of Agent Based Transport Simulations with Multi-Fidelity Bayesian Optimization, in Paper presented at the 18th Swiss Transport Research Conference (STRC 2018), Ascona, Switzerland, Ma, Ascona, SwitzerlandSTRC, Switzerland.
Comparison of passive mobile traces and GPS data for the calculation of mobility indicators
Molloy J., Silm S., Ahas R., Axhausen K. W. (2018), Comparison of passive mobile traces and GPS data for the calculation of mobility indicators, in Paper presented at Mobile Tartu 2018, 27-29, University of Tartu, Estonia27-29.
Estimating the externalities of a sustainable mobility platform using GPS traces
Molloy J., Tchervenkov C., Axhausen K. W. (2018), Estimating the externalities of a sustainable mobility platform using GPS traces, in Paper presented at mobile.TUM 2018, Munich, Germany, 13-14 June 2018, Munich, GermanyTUM, Munich, Germany.

Collaboration

Group / person Country
Types of collaboration
EPFL Switzerland (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
Mobility Lab, University of Tartu Estonia (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Research Infrastructure
- Exchange of personnel
TU Berlin Germany (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
EPFL Switzerland (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
KTH Stockholm Sweden (Europe)
- in-depth/constructive exchanges on approaches, methods or results
California Institute of Technology United States of America (North America)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
- Exchange of personnel
University of Toronto Canada (North America)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
- Exchange of personnel
FCL (ETH Zürich) Singapore (Asia)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
- Exchange of personnel

Scientific events

Active participation

Title Type of contribution Title of article or contribution Date Place Persons involved
23rd EURO Working Group on Transportation Meeting Talk given at a conference An efficient approach to create agent-based transport simulation scenarios based on ubiquitous Big Data and a new, aspatial activity-scheduling model 16.08.2020 Paphos, Cyprus Nagel Kai;
20th Swiss Transport Research Conference (STRC 2020) (virtual) Talk given at a conference Generating Synthetic mobile phone datasets using MATSim 13.05.2020 Ascona, Switzerland Molloy Joseph;
Transportation Research Board Annual Meeting 2020 Poster Activity Estimation from Mobile Phone Data 12.01.2020 Washington D.C., United States of America Fellendorf Martin;
19th Swiss Transport Research Conference Talk given at a conference A ticket-based public transport pricing model for Switzerland 15.05.2019 Ascona, Switzerland Molloy Joseph;
Universitätstagung Verkehrswesen 2018 Talk given at a conference Berechnen der Externalitäten von GPS Tracks mittels Anwendung von Mikrosimulation 23.09.2018 Obergurgl, Austria Molloy Joseph;
IATBR 2018 Talk given at a conference Comparison of passive mobile traces and GPS data for the calculation of mobility indicators 22.07.2018 Santa Barbara, United States of America Axhausen Kay W.; Molloy Joseph;
Mobile Tartu 2018 Talk given at a conference Comparison of passive mobile traces and GPS data for the calculation of mobility indicators 27.06.2018 Tartu, Estonia Molloy Joseph;
mobil.TUM Talk given at a conference Estimating the externalities of a sustainable mobility platform using GPS traces 13.06.2018 Münich, Germany Molloy Joseph;
18th Swiss Transport Research Conference Talk given at a conference Microsimulation of Time Variant Road Pricing for Kanton Zug using MATSim 16.05.2018 Ascona, Switzerland Molloy Joseph;
18th Swiss Transport Research Conference Talk given at a conference Calibration of Agent Based Transport Simulations with Multi-Fidelity Bayesian Optimization 16.05.2018 Ascona, Switzerland Krause Andreas; Makarova Anastasiia;
Workshop on data processing and analytics of smartphone and GPS data Talk given at a conference Using high resolution passive mobile positioning data in transport models 11.09.2017 Tartu, Estonia Axhausen Kay W.; Molloy Joseph;
17th Swiss Transport Research Conference Talk given at a conference Improving destination choice modeling using location-based big data 17.05.2017 Ascona, Switzerland Molloy Joseph;


Self-organised

Title Date Place
NRP 75 "Big Data Transport Models": Fifth Periodic Meeting 18.02.2020 IVT, ETH Zürich, Switzerland
NRP 75 "Big Data Transport Models": Fourth Periodic Meeting 08.03.2019 Zürich, Switzerland
NRP 75 "Big Data Transport Models": Third Periodic Meeting 03.09.2018 Zürich, Switzerland
NRP 75 "Big Data Transport Models": Second Periodic Meeting 25.01.2018 ETH Zürich, Switzerland
NRP 75 "Big Data Transport Models": First Periodic Meeting 20.07.2017 ETH Zürich, Switzerland

Knowledge transfer events

Active participation

Title Type of contribution Date Place Persons involved
NRP 75 Big Data: Technology Transfer Event Talk 10.11.2020 Virtual, Switzerland Molloy Joseph; Krause Andreas;
Presentation to INFRAS Talk 13.10.2018 Infras AG, Bern, Switzerland Molloy Joseph;


Communication with the public

Communication Title Media Place Year
Media relations: print media, online media Connecting Large Scale Transport Models and Mobility Trace Data NSL Newsletter Italian-speaking Switzerland Western Switzerland Rhaeto-Romanic Switzerland German-speaking Switzerland 2018

Associated projects

Number Title Start Funding scheme
189001 Consumption and travel after the smartphone revolution 01.01.2020 Project funding (Div. I-III)
154310 MAXess: Measuring accessibility in policy evaluation 01.08.2015 ERAfrica
165900 Long Distance Travel Demand Simulation 01.04.2016 Project funding (Div. I-III)
153807 Sharing is Saving: how collaborative mobility can reduce the impact of energy consumption for transportation 01.10.2014 NRP 71 Managing Energy Consumption

Abstract

Transport models reproduce who is travelling when why and where to for the base year, but then also forecast them for a future year and for a transport system changed with regards to prices, capacities, alignments and regulations. Current big data studies are focussed on describing the status quo, but lack predictive power. The goal of this project is to merge the power of big data streams, here GSM traces, with the power of agent-based transport modelling to build models with forecasting power, but do this fast and with optimised parameters. We will use a hypothetical road pricing scheme for Switzerland to show the potential of the approach.
-