Content uploaded by Dominik Bucher
Author content
All content in this area was uploaded by Dominik Bucher on Jan 13, 2017
Content may be subject to copyright.
Swiss Competence Center for Energy Research
Efficient Technologies and Systems for Mobility
Partners
References
Transport Mode IdentificationIntroduction
The project GoEco! takes advantage of the wide availability of
smartphones, in order to overcome the traditional awareness-raising
approach used to foster sustainable mobility and exploit eco-feedback,
social norms and peer pressure elements in an ICT-based motivation
system. In particular, it uses a smartphone app to analyze how we can
encourage people to engage in more sustainable mobility lifestyles.
This poster discusses the various challenges we faced when deploying
GoEco! Tracker (an app using the Moves® fitness tracker to collect
mobility measurements), and provides a summary of results obtained
by one month of large scale testing within the GoEco! living lab
performed in Switzerland, allowing us to collect baseline mobility data
for the sample of voluntary participants of the GoEco! living lab.
Challenges and Results from Deploying the GoEco! Tracker App
The present urban transportation system, mostly
tailored for cars, has long shown its limitations [1]. Even
though many alternative and effective transport modes
are already available [2], they still tend to be neglected
due to adeep‐rooted car dependency [3].
How can we encourage people to engage in more
sustainable mobility lifestyles, reducing use of the car?
GoEco! tries to overcome the traditional awareness-
raising approach and, building on recent research in
social psychology and behavior studies, to take
advantage of the wide availability of smartphone
devices (cf. [4]).
This poster discusses results from one month of large
scale testing of the GoEco! Tracker app aimed at
automatic mobility tracking, with a few hundred users
within the GoEco! living lab performed in Zurich and
Ticino.
Dominik Bucher, David Jonietz, Martin Raubal
Institute of Cartography and Geoinformation, ETH Zurich
Stefano-Franscini-Platz 5, 8093 Zurich, Switzerland
dobucher@ethz.ch, jonietzd@ethz.ch, mraubal@ethz.ch
Francesca Cellina, Roman Rudel
Insitute for Applied Sustainability to the Built Environment, SUPSI
via Trevano, 6952 Canobbio, Switzerland
francesca.cellina@supsi.ch, roman.rudel@supsi.ch
Francesca Mangili, Claudio Bonesana, Andrea-Emilio Rizzoli
Dalle Molle Institute for Artificial Intelligence, USI - SUPSI
Galleria 2, 6928 Manno, Switzerland
francesca@idsia.ch, claudio@idsia.ch, andrea@idsia.ch
Living Lab Challenges and Choices
Building GoEco! Tracker on top of the Moves® app
guaranteed a high tracking accuracy, availability on
many devices, and low battery power consumption, at
the cost of higher complexity (multiple apps) and
privacy issues and communication delays (cf. [5]).
651 persons signed up for the living lab after a
widespread marketing campaign in January 2016, 461
actually downloaded the app, and around 200 provided
tracks during the whole baseline recording phase (4
weeks in March / April 2016).
As Moves® only provides mode identification for
walking,running,cycling,and a general transport mode,
we built our own classifier on top of Moves® data.
This naïve Bayes classifier refines the mode of transport
using route data, such as average speed,total distance,
heading change, and various metrics obtained by
comparison with street network and public transport
timetables. It uses new data to improve itself and
achieves an accuracy of around 80%.
Accuracy of Tracked Data
On average, we recorded 7.4 activities per user per day,
of which 77.1% were actively validated by users (31’782
out of 41’199).
User Reports and Change Potential
As a form of feedback, users receive reports on their
mobility behavior. Next to a detailed summary, we
provide an analysis of the current mobility patterns with
respect to possible improvements.
These possibilities for change are computed for
systematic routes (e.g., home to work and back), as well
as on a general level, i.e., for all routes a participant
took. The figure above shows the possibility for change
of a single user, who could travel more by public
transport, and thus save energy.
Expected Impact
The impact, which a system like GoEco! can have on
energy demand and GHG emissions, is mostly
determined by users who travel by car on a frequent
basis. In our currently ongoing analysis, we found that
many could realistically reduce the kilometers traveled
by car between 10%and 50%, resulting in energy
savings in the order of magnitude of multiple 10 kWh
per user per week. Of course, many people have good
reasons why they use the car, so it is yet to be seen if
GoEco! manages to incentivize people enough to
actually perform this behavior change.
An overall assessment is mostly determined by the
second study phase (Fall 2016), during which the full
GoEco! app is deployed. In case this phase successfully
manages to change mobility behavior, it would be
required to make the app available to a wide part of the
population, in order to have the desired overall energy
savings effect.
[1] Miller, H.J.: Beyond sharing: cultivating cooperative transportation
systems through geographic information science. Journal of
Transport Geography 31, 296–308 (2013)
[2] Kramers, A.: Designing next generation multimodal traveler
information systems to support sustainability-oriented decisions.
Environmental Modelling & Software 56, 83–93 (2014)
[3] Diekstra, R., et al.: Cars and behaviour: psychological barriers to car
restraint and sustainable urban transport. The greening of urban
transport: planning for walking and cycling in Western cities,
John Wiliy & Sons Ltd (1997)
[4] Weiser, P., et al.: Towards sustainable mobility behavior: research
challenges for location-aware information and communication
technology. GeoInformatica 20(2), 213–239 (2016)
[5] Bucher, D., et al.: Exploiting Fitness Apps for Sustainable Mobility-
Challenges Deploying the GoEco! App. ICT4S (2016)
Car
75.6%
Public
Transport
25.7%
Bicycle 9.2%
Other
19.4%
Current mobility patterns Potential mobility patterns
Car
42.7%
Walk
2.7%
Walk 1.1%
Bicycle 3.7%
Public
Transport 1.8%
Other 19.6%
Reasons given for leaving the study (by ~30 users)
Recorded Activities
Comparison with
Swiss mobility census
