From Location Tracking to Personalized Eco-Feedback: a Framework for
Geographic Information Collection, Processing and Visualization to
Promote Sustainable Mobility Behaviors
Dominik Buchera, Francesca Mangilib, Francesca Cellinac, Claudio Bonesanab, David Jonietzd, Martin
aInstitute of Cartography and Geoinformation, ETH Zurich, Stefano-Franscini-Platz 5, 8093 Z¨urich, Switzerland
Dalle Molle Institute for Artiﬁcial Intelligence (IDSIA), USI-SUPSI, Galleria 2, Via Cantonale 2c, 6928 Manno, Switzerland
cInsitute for Applied Sustainability to the Built Environment, SUPSI, Via Trevano, 6952 Canobbio, Switzerland
dHERE Technologies, Seestrasse 356, 8038 Z¨urich, Switzerland
Nowadays, most people carry around a powerful smartphone which is well suited to constantly monitor the
location and sometimes even the activity of its user. This makes tracking prevalent and leads to a large
number of projects concerned with trajectory data. One area of particular interest is transport and mobility,
where data is important for urban planning and smart city-related activities, but can also be used to provide
individual users with feedback and suggestions for personal behavior change. As part of a large-scale study
based in Switzerland, we use activity tracking data to provide people with eco-feedback on their own mobility
patterns and stimulate them to adopt more energy-eﬃcient mobility choices. In this paper we explore the
opportunities oﬀered by smartphone based activity tracking, propose a general framework to exploit location
data to foster more sustainable mobility behavior, describe the technical solutions chosen and discuss a
mobility patterns from users’ trajectories, computes credible alternative transport options, and presents the
results in a concise and clear way. The resulting eco-feedback helps people to understand their mobility
choices, discover the most non-ecological parts of their travel behavior, and explore feasible alternatives.
Keywords: Mobility Tracking, Mode Detection, Trajectory Clustering, Eco-feedback, Sustainability
Preprint submitted to Elsevier February 21, 2019
Many smartphone apps perform automatic mobility and activity tracking, with increasing precision both
in path recognition as well as in the identiﬁcation of transport modes [
]. While many applications focus
on health and ﬁtness aspects [
], or passively track in the background, for example to give location-based
recommendations (e.g., [
]), there has been a recent surge in applications concerned with individual mobility
]. Such apps are often driven by ecological incentives such as reducing CO
from travel or generally increasing the sustainability of personal transport. The key components aim at
making people aware of their mobility behavior, i.e., at showing them the impact of their behavior on the
environment, on their ﬁnancial budget or on the time spent traveling.
Still, in many applications users just check the individual routes they travel and do not get a complete,
yet simple and immediate understanding of their mobility patterns. It is left to the app users to assess which
journeys are suitable for optimization and what the eﬀects of a behavioral change on the environment are.
Getting the right eco-feedback, deﬁned as “feedback on individual or group behaviors with a goal of reducing
environmental impact” [
], and making users aware of their mobility patterns and the consequences they
entail, is acknowledged as a necessary—though not suﬃcient—condition towards more sustainable mobility
patterns (cf. [
]). For example, current eco-feedback technologies lack the distinction between systematic
and unsystematic travel behavior, the provision of a personalized and meaningful assessment of possible
changes and the presentation of the various eﬀects behavioral changes entail.
In the Swiss-based GoEco! project
], we addressed these issues, taking advantage of the wide
availability of smartphone devices to encourage people to reduce their use of cars. Using a speciﬁcally
developed smartphone app [
], we performed a large scale ﬁeld test, monitoring the activities of several
hundred volunteer citizens from Southern Switzerland and the City of Zurich. During three distinct mobility
tracking periods distributed over a year, each one lasting at least six weeks, the movements and transport
mode choices of participants were recorded. To assess the eﬀects of eco-feedback on mobility behavior,
no feedback was given to the study participants in the ﬁrst and last period (i.e., they were only tracked
to collect a baseline dataset). Starting after the ﬁrst period, they were also stimulated with eco-feedback
and gamiﬁed activities, with the aim of persuading them to improve their mobility behavior. While the
gamiﬁcation elements (such as goal-setting, challenges, badges or leaderboards) were integrated directly into
the smartphone app, eco-feedback was additionally given in the form of a report booklet, sent to the study
participants via e-mail. In this paper, we mainly focus on this non-interactive eco-feedback report which was
sent to the participants at the end of the ﬁrst period.
This report contained basic elements such as summaries of distances traveled, durations spent on diﬀerent
modes of transport or energy consumption and CO
emitted as eﬀect of the recorded travel behavior.
These summaries were available at various time scales, such as the overall study period or individual weeks.
Additionally, the report contained a section about detected systematic journeys (appearing multiple times
within a few weeks, such as traveling to work or school), highlighting which of them could be optimized with
little loss in travel convenience by adopting a proposed alternative. An estimate of the possible impacts on
the environment (energy consumption and CO
emissions) rounded oﬀ the report. In post-experimental
surveys we assessed the plausibility of the detected systematic routes and the proposed alternatives. Here, we
present the algorithms used to detect systematic travels and to compute feasible alternatives. In addition, we
provide an estimate what a sample of individuals like the one of the GoEco! project could potentially save in
terms of CO
emissions by changing mobility behavior to more sustainable modes of transport, according to
the proposed alternatives.
To the best of our knowledge, this is one of the ﬁrst experiments of its type. Therefore, it required the
development of a framework for collecting, analyzing and reporting individual mobility patterns and potential
for changes towards more sustainable mobility choices. The proposed framework includes: (i) location
tracking; (ii) identiﬁcation of transport modes; (iii) assessment of individual mobility patterns (modal share
and mobility footprint); (iv) identiﬁcation of systematic itineraries; (v) identiﬁcation of more sustainable
1See goeco-project.ch for details and recent updates.
mobility choices; (vi) presentation of the outcomes as personalized feedback. Section 2 introduces related
work from the domain of mobility behavior assessment and eco-feedback provision. Section 3 presents the
proposed framework, structurally following the six parts mentioned. In Section 4 we provide additional
details, illustrating the core methods of our framework. The results of the application of the above framework
and methods to data gathered as part of this study are presented in Section 5. Finally, we discuss the
strengths and weaknesses of our framework and point out remaining open questions in Sections 6 and 7.
2. State of the Art in the Analysis of Activity Data to Understand Mobility Behavior
In the past, several studies have aimed at not only analyzing people’s mobility behavior, but also at making
it more environmentally sustainable by using mobile applications and a feedback loop. Examples include
], Tripzoom [
], MatkaHupi [
], SuperHub [
], PEACOX [
] and Traces [
these studies often involve small sample sizes (up to several dozens of users), short study periods (up to a few
weeks), lack a control group or solely perform qualitative assessments [
]. For example, UbiGreen [
was one of the ﬁrst smartphone-based eco-feedback applications and featured a visual representation of the
ecological impact of travel behavior in the form of a tree that lost or gained leaves. If one choses sustainable
transport modes on a certain day, the tree would grow greener. The thirteen participants mostly responded
in positive ways to the application, but a systematic quantiﬁcation of the UbiGreen eﬀects was not performed
in the paper. Building on these previous studies, our research tries to apply their ﬁndings to a large sample
over the duration of one year, and studies the adoption of eco-feedback related to mobility and the resulting
changes in behavior.
However, before eco-feedback can be provided or sustainable travel alternatives can be proposed, movement
resp. mobility needs to be recorded. This is a topic which has attracted interest from various disciplines
and has also proven to be useful for numerous applications other than the promotion of sustainable mobility
behavior, including transportation planning (e.g., [
]) or personalized routing (e.g., [
movement is tracked using Global Navigation Satellite System (GNSS) receivers, e.g., the GPS sensor in the
mobile phone of a user. Numerous studies have proposed approaches to augment this tracking data, e.g., by
inferring the performed activities or the used transport modes by considering additional sensor data (see for
]). Such research has allowed the development of several apps for tracking users’ itineraries
and activities [
]. To obtain the results in Section 5, we applied the methods presented within this paper
to GPS data tracked by such an app (namely Moves
), which automatically recorded user trajectories
and identiﬁed visited places and routes. To improve the transport mode recognition, an additional new
classiﬁer was developed that learned from previously tracked and validated data and additionally considered
spatio-temporal context (cf. ).
Once such movement data is available, it must be analyzed and mined for observable and interpretable
individual mobility patterns. The most basic of these patterns include modal splits and aggregate values (such
as total CO
emissions or travel duration). Most of the previously mentioned studies used such aggregate
values, e.g., in UbiGreen [
], where the daily number of “eco-friendly” journeys is compared to the number
of car journeys. In our framework, we provide methods to analyze mobility on a much more detailed (and
individual) level by inspecting individual journeys, with the aim of providing users with a more thorough
feedback and thus oﬀering more assistance for behavior change. For this task, several trajectory pattern
mining techniques can be employed [
]. Previous studies on mobility data mining have particularly focused
on the identiﬁcation of visited places from raw GPS-data by assessing prolonged permanence in a small area of
the moving person [
], and the derivation of traveled routes between these locations by map matching
]. By assessing the similarities between trajectories of several individuals, these methods
generally discover knowledge about collective mobility behaviors, like common places of interest, frequently
visited locations, the way of moving, etc. This is useful information in urban planning and transportation
management for instance [
]. Other works, closer to our aim of generating meaningful eco-feedback,
2A ﬁtness tracking app discontinued in July 2018.
use pattern mining techniques to detect individual mobility habits by looking for regularities in personal
trajectory data [54, 36, 27].
After ﬁnding patterns in the trajectory data, personal feedback and proposed sustainable travel alternatives
can be provided. Typically, this requires a personalized, multi-modal routing application. When computing
optimized routes, personalization is often achieved by incorporating individual preferences and constraints as
parameters of route segment traversal costs into the routing algorithm (e.g., [
]), or using a preceding
]. A range of techniques has been developed especially for computing routes on large-scale,
dynamic multi-modal transportation networks (e.g., [
]). An appropriate aggregation of the output of
such planning systems can then be given to the user as feedback on possible more sustainable alternatives.
Giving meaningful eco-feedback is a well-researched topic (cf. [
]). Commonly, one tries to satisfy
psychological (autonomy, competence and relatedness) and social needs (achievement, aﬃliation and intimacy,
and leadership and followship) and should respect the current context of a user (e.g., how motivated a person
is, if a user has the ability to perform a certain behavior, if the timing of the feedback is right, and if a
user is in the correct stage of a behavioral change process) [
]. In terms of eco-feedback particularly for
mobility behavior, there are many more open questions, as mobility is highly individual and to a large degree
depending on personal, temporal and environmental context [
]. As it is generally diﬃcult to passively
collect data on a user’s context (e.g., if the user needs to carry heavy luggage for a certain journey), in
looking for available more sustainable alternatives we here primarily respect temporal and environmental
context factors, such as the available transport modes at a given date and time.
3. A Framework for the Generation of Eco-Feedback
Generating eco-feedback is a complex procedure involving diﬀerent steps. Figure 1 shows the individual
steps of our framework to induce behavioral change in the context of mobility choices. The system includes a
mobility tracker installed on the personal smartphone of each user. The tracker’s task is to automatically and
passively register all routes traveled by each participant. The system then assigns a transport mode (TM) to
each detected journey, based on a transport mode classiﬁcation algorithm and/or on direct interrogation of
the user (e.g., the recipients of eco-feedback are often given the possibility to validate and adjust a recognized
transport mode). The collected routes are then stored in a database and analyzed following three main goals:
•to proﬁle the user mobility patterns, in terms of modal share and related footprint (global analyses);
to evaluate the potential improvement each user can achieve regarding overall sustainability of her
to identify similar trajectories (i.e., systematically traveled routes), for whom targeted and realistic
suggestions for more sustainable mobility choices (alternatives) can be proposed.
Finally, the outcomes of this threefold analysis (eco-feedback ) are presented to the user in a report.
3.1. Mobility Tracking
In our mobility tracking framework, a trackpoint is the most basic object. It consists of a longitude, a
latitude, and a timestamp. Trackpoints are aggregated into routes which are deﬁned as transfers from a point
of interest (POI) to another, where a POI is a place visited on purpose, to perform or attend some activity
(such as working or shopping). Each route represents a process, which has a well-deﬁned start and end point,
both in space as well as in time (these start and end points are central for several parts of our method, while
the actual trackpoints of the route play a minor role). The distance traveled is the sum of the distances
between all trackpoints, and the duration of a route is the diﬀerence between the timestamps of the end and
start positions. The longitude and latitude coordinates of the start and end points of a route will be referred
to as visited positions, whereas the POIs connected by a route are physical places that cannot be described
by a single pair of GPS coordinates; indeed, also because of the inaccuracies of the tracking system, diﬀerent
(yet close) visited positions may refer to the same POI.
Figure 1: The proposed personalized eco-feedback framework generates a variety of feedback components from the classiﬁed
movement trajectories: global analyses, individual transport mode alternatives, and concrete suggestions for systematically
Each route can be traveled with more than one transport mode. Segments, on the other hand, are parts
of a route covered with a single mode of transport. A stay point is any point where a user spends some
minutes, e.g., when changing from one transport mode to another. For example, a route traveled by [walk
walk] would consist of two walk segments, and one train segment and will include two stay points
(i.e., the train stations).
The goal of a tracking app is therefore to collect a thorough yet concise set of trackpoints, and to identify
the visited places and stay points which allow to partition the whole set of trackpoints into routes and
segments. Common issues in the development of tracking apps for large-scale mobility data collection studies
]) include battery consumption, availability on diverse operating systems, and varying classiﬁcation
accuracies due to diﬀerent sensors (cf. [
] for an overview). There are several apps that promise oﬀ-the-shelf
solutions to these problems (with varying success). For example, in our study, the commercial app Moves
was employed, which allowed custom apps to connect and download tracking data through an application
programming interface (API). Moves pre-processed the data from mobile sensors and provided trajectories
as sequences of trackpoints already partitioned into segments, routes and stay points, which were of suﬃcient
quality for eco-feedback and mobility analyses [
]. While the exact workings of Moves were not disclosed
publicly, from our experience it recorded a single trackpoint approx. every 5 minutes (depending on the
transport mode and GPS availability), used the accelerometer for transport mode identiﬁcation and treated
places where users spend more than approx. 10 minutes as staypoints. Alternative options to employing
Moves would have been to either use a more pricey paid app, integrate an open-source solution, or develop
a new tracking app from scratch. As the focus of this study was on the exploration and analysis of an
eco-feedback intervention, there were not suﬃcient resources available for any of the alternatives (especially
the development of a reliable tracking app from scratch is a non-trivial task).
3.2. Transport Mode Detection
In order to assess the ecological impact of peoples’ mobility, a transport mode has to be allocated to
each segment. Similarly to most commercial trackers currently available, Moves only distinguishes between a
small set of transport modes, speciﬁcally walk,bike and motorized transport. As a more detailed classiﬁcation
is necessary for ecological footprint analyses, we developed a classiﬁer with the ability to distinguish between
thirteen transport modes (car,electric car,motorbike,scooter,train,bus,tram,plane,ship,bicycle,electric
bicycle,kick scooter and foot). Details of the algorithm for transport mode classiﬁcation are given in .
To improve the reliability of the assigned transport modes, we present all the routes identiﬁed by Moves
to the users, showing the transport modes assigned by the classiﬁer. The users are then asked to validate
the routes, i.e., conﬁrm or change the mode of transport and delete a route if it is “particularly wrong”,
either regarding the places visited or the travel time. This possibility to delete routes was mainly put in
place as sometimes the recorded routes are completely unrealistic (e.g., on the border of a country, where
due to changing mobile phone network carriers, the cell tower positioning method yields bad data) and
would greatly distort following data processing. On the one hand, this manual validation supplies a set of
labeled data that can be fed back into the algorithm to improve the accuracy of the future transport mode
classiﬁcation, while on the other hand, it ensures a high accuracy of the collected data. As the user validation
continued along the whole study period, the dataset of routes and transport modes used as a base for all our
computations can be considered as reasonably correct and not aﬀected by the accuracy of the classiﬁer.
3.3. Mobility Patterns: Modal Share and Mobility Footprint
The most basic individual mobility features, which summarize the multitude of data automatically
gathered by the app, are the total distance traveled and the related travel duration, averaged over a speciﬁc
time period. In this study, we opted for a weekly time period, as it is easily understood by users, it allows
including both systematic and non systematic routes (i.e., routes that are resp. are not traveled several times
within a short time frame, such as a week), and accounts for the variability of daily mobility needs arising
from individual weekly schedules.
A second set of mobility features includes the individual shares of the diﬀerent modes of transport used.
Our eco-feedback shows this modal share based on the average weekly kilometers traveled by each mode of
transport, as this is directly related to energy consumption and CO
emissions. Of course, it would also be
possible to show the modal share based on traveling times or the number of routes, however, these aspects
are more relevant when optimizing schedules than for improving ecological footprints.
To simplify the feedback, some transport modes are aggregated, such as car (both internal combustion
engine and electric car), public transport (train, tram and bus), bicycle (both conventional and electric),
and foot, while other encompasses all the remaining modes of transport. Finally, the eco-feedback includes
information on users’ ecological footprints, expressed in terms of energy consumption (kWh) and CO
emissions. Both their average values per week and per kilometer traveled are shown to the user. To provide
such estimates, we relied on the Mobitool
consumption and emission factors [
], which depend on the
mode of transport, refer to a single kilometer traveled in Switzerland and take into account the consumption
and emissions of the full life-cycle (see Table 1). For many transport modes these factors heavily depend
on the involved power generation systems (e.g., power from renewable sources leads to much fewer CO
emissions than power generated by fossil-fuel power plants). As such, these factors and the resulting energy
consumption and CO
emission values are speciﬁc to Switzerland. In order to apply the framework to
another region, the power generation systems within that region have to be analyzed and taken into account
3.4. Systematic Mobility
A central element of our eco-feedback are the suggestions of mobility alternatives speciﬁc to the itineraries
traveled by each individual user. Since it is not possible (and neither useful) to suggest alternatives for all
ever traveled routes, we focus on the systematic ones, i.e., those traveled multiple times within a certain
(short) time frame. Systematic itineraries are of interest, as a behavioral change in these situations is easier,
since they can be planned in advance and become part of the daily routines. Also, they have a larger potential
to reduce energy consumption and CO2emissions, as they are repeated over time.
To provide applicable eco-feedback and suggest credible alternative modal choices, our approach moves
from individual routes to loops of routes starting and ending at the user’s home place. Without this step (i.e.,
if suggestions of alternatives are based on routes only), the system might end up suggesting unreasonable
or conﬂicting solutions within the same loop. For example, in a car-based [home
3mobitool.ch is a Swiss platform for mobility management tools and curated environmental data.
Transport Mode Energy Required CO2Produced
Car 0.91 197.23
Bus 0.64 145.41
Tram 0.55 37.47
Train 0.14 7.32
Bicycle 0.04 7.64
Walk 0.00 0.00
Table 1: Energy consumption and CO2emission factors used to estimate a user’s ecological footprint.
home] loop, suggesting to replace the car-based route [workplace
workplace] with a bicycle-based one is not plausible, since it cannot be assumed that a bicycle is available
at the user’s workplace. Or, in a car-based loop [home
public transport to reach the workplace might not be acceptable, if soon after work the user directly goes to
a large supermarket in the suburbs, which is not served by public transport. To avoid such inconsistencies,
we need to explicitly take into account the constraints that choosing a means of transport on a speciﬁc route
imposes on the following routes. In conclusion, by considering clusters of systematic loops, we can account
for systematic mobility patterns of users and at the same time identify reasonable alternatives.
The speciﬁc algorithm adopted for the identiﬁcation of systematic and non-systematic loops is described
in Section 4.1. Instead of blindly searching for clusters of similar itineraries based on standard trajectory
clustering algorithms, we focus on the needs and the limitations of this study and develop a more interpretable
algorithm by answering to some questions such as “how should a loop be deﬁned?”, “how diﬀerent can loops
be, which belong to the same cluster?”, “how many times and how frequently has a loop to be traveled, in
order to be considered systematic?”. Answering these questions, we arrived at our deﬁnition of systematic
loops and at the selection of heuristics adopted by the algorithm, by keeping in mind the ﬁnal goals of the
intervention, i.e., to suggest feasible and meaningful alternatives that can eﬀectively induce a behavioral
change, and to assign realistic values to the individual potentials for change. The developed algorithm also
accounts for the limitations and inaccuracies of the tracking system and tries to correct them.
3.5. Alternatives and Potential for Change
The assessment of someone’s potential for change is primarily based on the comparison to optimal travel
behavior, where optimal in this case is computed with respect to CO
emissions. While many factors could
be included in this computation (such as possibilities to telecommute, or the rescheduling of activities),
here we only consider the individual routes and the modes of transport chosen. This choice is primarily
made based on the available data, which is not detailed enough to infer accurate schedules and contexts of
users. We consider the contribution of all the systematic loops and all other routes (i.e., routes for which no
systematic loop could be constructed) to the overall CO
production and, for each of them, we identify an
optimal alternative solution, i.e., a diﬀerent itinerary and modal choice that allows reaching all the POIs
visited in the original loop or route, while reducing the overall CO
production. Note that this means that
we consider the alternatives for all traveled journeys to compute the optimal mobility behavior. However, in
the ﬁnal feedback, suggestions for concrete alternatives are only given for routes that are frequently traveled.
The potential for change is then computed as the diﬀerence between the actual emissions and those that
would have been achieved by replacing all loops in each cluster of systematic loops and all other routes by the
corresponding optimal alternative. When no alternative is found, the original loops or routes are retained.
Therefore, given any route consisting of a number of segments annotated with a mode of transport, we
compute the CO
produced as the sum of all the segment lengths multiplied by an emission factor which
depends on the mode of transport. Again, we refer to the Mobitool values already introduced in Table 1,
speciﬁc to Switzerland. Duration, distance, and start and end points of routes and segments are deﬁned as
in Section 3.1.
A range of diﬀerent optimization criteria could have been used instead of CO
emissions, such as the
total covered distance, the duration spent traveling, or the required energy. We focus on CO
as we are concerned about giving eco-feedback to people (which ultimately should result in a reduction of
greenhouse gases emissions), and this has the beneﬁt of reducing the required energy as well, as there is a
strong correlation between energy consumption and CO2emissions.
Concerning the identiﬁcation of alternative solutions, in order to successfully replace a route, an alternative
has to respect the following requirements:
It cannot use a mode of transport which is not available to the user (e.g., if someone does not own
a bicycle, alternatives containing bike segments are invalid). While this can be extracted from the
validated routes themselves, the study participants were also initially asked which modes of transport
they have access to.
Its duration should not be excessively long (e.g., if traveling by public transport takes a user three
times as long as traveling by car, she will most likely ignore the suggestion). This is less problematic
for short routes, as there the metric of interest is usually the additional time an alternative takes (e.g.,
replacing a 10 minutes car route with a 20 minutes bicycle route is acceptable, even though it increases
the traveling time by 100%).
production should be substantially lower than the original one (e.g., more than 5% lower, to
account for tracking and positioning inaccuracies). This is usually the case when switching to a more
eco-friendly mode of transport.
Requirement 1 is particularly important when considering loops, especially those including car and bicycle
transport modes. When starting a loop with one of these modes of transport, we need to make sure it is
brought back to the starting location (i.e., the user’s home place). On the other hand, if the ﬁrst route
does not start with any of these two modes, they cannot be used later on in the loop either. While the
ﬁrst requirement imposes a strict criterion for removing unacceptable alternatives, the other two form an
optimization problem. The algorithm to identify alternative solutions for routes and loops that meets the
requirements above is described in Section 4.2.
The analyses described above lead to an eco-feedback which is presented to each user in the form of a
short report. The report is divided into two parts: the ﬁrst part analyzes the potential for change when
only systematic loops are replaced by the proposed alternatives, while the second part includes the complete
set of loops and routes recorded for each user. As changes induced in systematic itineraries may have a
larger impact on the overall energy savings and may be more easily undertaken and maintained by the user,
the ﬁrst part of the eco-feedback details both the identiﬁed loops, and the suggested alternatives. For each
systematic loop recognized by our algorithm, users receive a feedback similar to the one depicted in Figure 2.
On the left hand side, a representative loop, selected among the original loops of the systematic cluster, is
shown on the map. In the example of Figure 2, the representative loop consists of three POIs and starts with
a short walk, followed by tram segments. On the right hand side, the suggested alternative is presented. It is
suggested to take the bicycle, which is not only more eco-friendly, but even faster in this case. It produces
0.18 kg CO
each time the loop is traveled and can be completed in 48 minutes, while the original modal
choice produces 0.28 kg CO2and takes 1 hour.
In addition to these individual loop assessments, the second part of the report provides aggregate results
about the overall travel duration and CO
emissions that could have been achieved if the user had always
adopted the optimal alternative suggested by the system. An example of the summary displaying the total
potential for change of a user is shown in Figure 3. In this example, a user, originally choosing the car
for 93.9% of her routes, can see that she could potentially reduce its use to 32.6%, resulting in average
savings of 56.7 kg per week (from 78.5 to 21.8 kgCO
/week). In a similar way, on average the energy
consumption could be reduced by 238.6 kWh per week (from 369.6 to 131 kWh/week).
While the feedback about systematic loops shows concrete suggestions for improvement, the overall
potential for change (Fig. 3) is a less tangible indication. For this reason, in the second part of the report we
Figure 2: View of the alternative for a systematic loop as displayed in the eco-feedback report. On the left is the original loop,
on the right the alternative.
Figure 3: The modal choice of a single user, as identiﬁed based on six weeks of monitoring using a tracking app (A), and the
modal choice suggested in the provided eco-feedback (B). Next to each ﬁgure, the average weekly traveled distances are shown.
also provide general rules for making non-systematic routes more sustainable. All these routes are divided by
mode of transport and then into three categories (less than 1 km long, between 1 and 3 km long, and longer
than 3 km), and for each of them an alternative route is computed. The routes are then presented in tabular
form, showing the number of times a certain combination was tracked (e.g., the number of car routes shorter
than 1 km), and how many of those could be replaced with a more sustainable alternative, thus giving the
user an idea of how eﬀectively her usual choices of transport modes could be improved. This table contains
all non-systematic routes traveled by a user, irrespective of their appearance within a loop.
In this section we provide more details about the methods to identify loops and systematic loops and to
suggest valid, realistic and more sustainable alternatives for each user mobility choices.
4.1. Identiﬁcation of Systematic Loops
In a ﬁrst step, individual trajectories must be aggregated into loops, which then can be clustered into
systematic loops, based on the frequency in which they appear.
To identify loops, we ﬁrst introduce the concept of connected routes. Two consecutive routes
are connected if the ﬁrst one ends at the same location where the second one starts (here we are assuming
that routes are chronologically ordered). In practice, the GPS coordinates of the end position of a route
seldom coincide with those of the start position of the subsequent route, and their distance can even be
rather large, due to the (occasionally substantial) inaccuracies produced by the activity tracking tool used in
this project. Therefore, we consider as the same place any two positions whose distance is less than 500 m.
For the loop identiﬁcation, all routes of a user are considered.
We begin from the ﬁrst route that starts at the user’s home place (which is known from a study pre-survey)
and follow the sequence of connected routes until we ﬁnd a route that ends at the user’s home. It can
happen that the ending point of a route (hereafter referred to as route 1) and the starting point of the
next one (hereafter referred to as route 2) have a distance larger than 500 m, for instance due to GPS
malfunctioning. In this situation, one could artiﬁcially connect the two consecutive unconnected routes by
adding an unobserved route going from the end point of route 1 to the start point of route 2. However, we
preferred not to add unobserved routes, partially because the cause of such a situation is not necessarily
the loss of a route, but it may also be an error in the GPS positioning system that localizes the user in a
diﬀerent position than she actually is, and opted for deleting unconnected routes. To this purpose, we used
the following algorithm.
If route 2 starts at the end point of a previous route of the loop under construction, we remove
the intermediate routes and continue the construction of the loop.
If the starting point of route 2 has never been reached before in the loop under construction,
route 2 is discarded and the construction of the loop continues with the next route.
To better understand this algorithm, consider the following example (see Figure 4).
Let us assume that user X has visited the sequence of points [home, work, restaurant, work, home].
If only the [home
restaurant] and [work
home] routes are recorded by the
activity tracker, we will have a route ending at restaurant (the second one) followed by one starting at
work (the third one), and it would not be possible to complete the loop. However, we notice that work
has already been visited before, and thus, by removing the [work
restaurant] route, we complete the
home] loop. The identiﬁed loop is just a subsection of the correct one, yet it has
been actually traveled and should be taken into account.
If, instead, the activity tracker missed the [work
restaurant] route, the second, i.e., [restaurant
work] route will start from a position which was not seen before in the loop under construction.
Then, the [restaurant
work] route is removed and the construction of the loop continues with the
home] route, which is connected to [home
work]. Once again, the subset [home
work →home] of the original loop is obtained.
Figure 4: Two possible ways a loop of routes is mistakenly recorded by the activity tracker.
Once all loops have been reconstructed, to identify clusters of systematic loops
we look for groups of loops that are similar. To compare loops, we represent them as sequences of POIs in
chronological order. POIs are identiﬁed by analyzing the geographical distribution of the visited places. As
already mentioned, several diﬀerent (yet close) visited places may correspond to the same POI, not only
because of the spatial extent of the POI but also because of the inaccuracies of the GPS and tracking systems.
Therefore, it is necessary to deﬁne a threshold distance
below which diﬀerent GPS coordinates are assumed
to represent the same POI. Once again, the threshold
= 500 m is adopted. Such a choice implies a low
resolution in distinguishing two close POIs. Clearly, this is an application speciﬁc setting that should depend
both on the scope of the analysis and the precision and accuracy of the collected data.
The aggregation of similar items (in our case, geographical positions of visited places) is typically
done using a clustering algorithm. As the number of clusters is unknown in this case, we use hierarchical
], which allows building the set of clusters by setting a maximum acceptable distance between
two items belonging to the same cluster. The POIs of a user are obtained by applying hierarchical clustering
to the set of her
visited places, i.e., the start and end points of all the routes included in her loops.
Hierarchical clustering builds a hierarchy of clusters by starting from
clusters containing a single place
each and aggregating at each next level of the hierarchy the two clusters of the previous level which are
closest, based on a speciﬁed dissimilarity measure. We apply the average linkage distance strategy (cf. [
that deﬁnes the dissimilarity between two clusters as the average distance between pairs of elements (one
in each cluster). The aggregation process is stopped when the distance between the two closest clusters
exceeds the predeﬁned threshold of
= 500 m. Finally, each POI is identiﬁed by the centers of mass of the
corresponding cluster of visited places.
Each loop can then be translated into an ordered sequence of POIs, so that loops represented by identical
sequences of POIs can be assigned to the same cluster of similar loops. All clusters including at least three
loops are considered to identify the user’s systematic loops. The criteria deﬁning a systematic loop are
subjective and depend on the meaning assigned to the word systematic. Depending on the goal of the
analysis, one may consider as systematic only loops traveled daily, or include also loops traveled once every
two weeks. In this work, having used a monitoring period of only six weeks, and considering that some loops
have been lost due to the limited accuracy of the tracking system, the number of three occurrences implies
that a systematic loop has been traveled on average at least once every two weeks, which was considered
enough to accept it as part of a user’s routines. We used a sample of ten users to test that the detected
systematic loops were actually considered systematic by the users themselves.
A second loop processing step accounts for situations such as the one presented in the example below.
Let us assume that user X travels twice the loop [home
home], once the loop [home
home] and twice the loop [home
(see Figure 5). With the threshold set up above, none of these loops would be identiﬁed as systematic.
However, we believe that the loop [home
home] is relevant, although sometimes it has
some additional stop-over, and would like to be able to identify it as a systematic loop.
Figure 5: Similar loops treated as the same systematic loop.
If a loop does not belong to a cluster of systematic loops (that is, its sequence of POIs has been observed
less than three times), but it includes all POIs of another loop in the same order, the two loops are counted as
if belonging to the same cluster of loops, described by the sequence of POIs of the shortest loop. Indeed, the
former is considered an extension of the latter. In the example above, the sequences [home
home] and [home
home] will be counted as if belonging to
home] cluster, which will thus be identiﬁed as a systematic cluster, as it now
includes ﬁve loops. This way some point of interest may be lost, yet more sequences of points of interest that
are actually part of the user mobility patterns and routines can be identiﬁed. Our choice is also motivated by
the fact that the activity tracker is not always capable of distinguishing between real POIs and trackpoints,
such as stations, bus stops, gas stations or other stay points where users just temporarily stop along the way
to their actual destination. This creates a number of spurious loops that are, indeed, formally equivalent to
systematic loops, but would not be taken into account in the assessment of the user mobility patterns. Also,
in this case, processing would not be necessary (being, instead, detrimental) whenever the POIs were more
reliably identiﬁed, for instance, by a more accurate GPS tracking device or by asking the user to validate
4.2. Identiﬁcation of Alternatives
In this section, we present the algorithm developed to identify optimal, low-carbon alternatives for all
Identiﬁcation of Alternatives for Routes.
In a ﬁrst step, a route planner computes a set of eligible
routes, by considering the diﬀerent modes of transport (MOT) mentioned in Table 1. As diﬀerent MOT can
often be combined with a positive eﬀect, we additionally use the OpenTripPlanner (OTP) route planner
and an implementation based on the model introduced in [
] to ﬁnd multi-modal trips. The OTP uses an
implementation of Round-Based Public Transit Routing [
] for schedule-based transport modes, but is able
to combine this method with street-level routing to compute multi-modal journeys. As the implementation
takes into account the exact departure times of public transport, we picked a representative time (the time
of the route that represents the loop “best”, in the sense that it shares most POIs with other routes in
the same loop and has the smallest departure time diﬀerence to them) for each systematic route when
querying the OTP. The exact departure and arrival time of the resulting routes were discarded, though,
and only the overall duration was kept for analysis. The approach presented in [
] is primarily used for
less frequently used combinations, which cannot be handled by the OTP (such as combinations of bicycle
and public transport). It essentially uses a heuristical two-step procedure, where ﬁrst potential transfer
nodes are computed (between diﬀerent modes of transport), which are evaluated for feasibility in a second
round (e.g., determining if there is actually a train connecting two potential transfer nodes). In terms of the
overall framework, the concrete routing application could be replaced with any routing service. We primarily
relied on OTP as it is one of the few freely available public transport routing applications (which removed
any constraints on the number of alternatives we could retrieve from it). For all the alternatives retrieved
by the route planning algorithms as well as for the original route, we then compute key features, such as
the duration and the total CO
production. For each original route
, this results in a set of alternatives
denotes the alternative route with its trackpoints and modes, which has an associated
total duration daand CO2production ca) that could be used to replace the original route ro.
In a second step, impossible and highly unlikely suggestions are removed from
, if they fulﬁll any of the
•Any of their modes of transport appears in the user’s list of unavailable modes of transport.
The original and alternative routes are too similar. To assess this, we deﬁne a similarity measure
between the two routes as the fraction of distances between the trackpoints of the original route and
the closest trackpoints of the alternative that are below 200 meters. Two routes are considered too
similar whenever the modes of the original route and the alternative coincide and the similarity measure
•The alternative does not reduce the CO2production by more than 5% (compared to the original).
•The total slope (in elevation) of a bicycle-based alternative is more than 1.5%.
The OpenTripPlanner (opentripplanner.org) is an open source route planner that is able to ﬁnd multi-modal public transport
•The duration of the alternative is too long, i.e., it exceeds the threshold dth:
dth = (do+tmax)−tmax
1 + do·tmax
where the parameter
represents the maximal increase in duration and
denotes the duration
in hours of the original route. In our implementation, we set the maximal prolongation of the route
2 hours (=1:12 h). Of course, this limit is not reached for shorter durations (e.g.,
a route with an original length of 5 min can maximally be replaced by one of 11.5 minutes). Figure 6
shows the value of dth as a function of do.
Figure 6: Threshold dth for the alternative duration.
It would be possible to ﬁlter out routes with long distances as well, even diﬀerentiating them by MOT
(e.g., preventing alternatives where one would have to walk for 10 km), but in practice such routes are already
ﬁltered out by the duration constraint, except if the original route is itself in a similar range. For this reason,
we do not use any distance ﬁlter.
Finally, the remaining alternatives must be ranked according to some criteria. While more elaborated
choices could be considered (especially if one wished to account for user preferences, such as wanting to lead
a healthy lifestyle, or disliking some form of transport), here we simply rank the remaining routes according
to the CO2produced, in order to give feedback about the sustainability potential of an alternative.
Identiﬁcation of Alternatives for Systematic Loops.
For each loop, we can identify alternatives
in a similar way as we did for routes. There are some circumstances that have to be specially considered,
though: when generating an alternative loop, the sequence of transport modes has to be respected. Moreover,
the total additional duration of an alternative loop should be further bounded. Using above formula, for
example, a loop of four routes of 30 minutes (a total of 2 hours) could be increased up to 3:48 hours, which
is clearly unacceptable for users. Therefore, requirement 2 of Section 3.5, is satisﬁed by bounding the total
time of an alternative loop using Formula 1 with a maximal time increase of
= 1:24 hours. Then, in the
above loop of four routes, which originally lasted 2 hours, the total loop duration can be increased by at
most 1:02 hours, instead of the original 1:48 hours.
The requirements are implemented by sequentially processing the routes of the original loop, and building
a graph of options, storing, at each node, the visited location and all available modes of transport. Figure 7
shows an exemplifying graph. The blue POI depicts the user’s home, where the transport modes walk,car,
bike and public transport (PT) are available. While in A) the algorithm suggests to take public transport to
the ﬁrst POI, in B) the use of the car is suggested. In the ﬁrst case, all consecutive route segments need to
be of type public transport or walk, as neither the car nor the bicycle are available anymore. In the second
case, taking the car would, in principle, force the second segment to be of either walk,car or public transport
type; however, since the car needs to return to the user’s home in the end, all modal choices except the car
are pruned after processing the last segment.
Figure 7: Example of computation for loop alternatives.
To get a deeper insight in the methods described above and to validate them, we have interrogated the
participants of the here presented study. Having obtained a conﬁrmation of their validity, we have then
analyzed the aggregated results of the performed computations, by estimating the overall potential for change
of the participants: such aggregated analyses can provide a (rough) upper bound of the potential eﬀects of a
5.1. Survey-based User Assessment
To assess the eﬀectiveness and accuracy of the data analyses presented above, we performed two surveys,
targeting the users themselves. The ﬁrst survey aimed at assessing the users’ perceptions about their overall
satisfaction throughout the project, and dedicated a few questions to investigating the quality and usefulness
of the individual eco-feedback reports. It was delivered to all participants, by means of an online questionnaire;
between 102 and 104 responses were collected for the subset of questions regarding the eco-feedback report,
equivalent to 39.5% of the sample of 261 study participants. This complete sample was recruited by means
of a public communication campaign, attracting around 700 voluntary people living in two Swiss regions
(the city of Zurich, a dense urban context characterized by high quality public transport provision, and
the Canton Ticino, a much less dense context, characterized by urban sprawl, where car is the dominant
transport mode). Filtering due to data incompleteness and for representativeness purposes led to the ﬁnal
261 participants. Referring to a 7-point Likert scale (where 1 = “totally disagree” and 7 = “totally agree”),
we investigated the level of agreement of users with the following sentences:
The reports correctly identiﬁed my reference mobility patterns, in terms of percentage of use of the
means of transport:
M = 5.81, SD = 0.98 (n= 104);
•The reports correctly identiﬁed my reference mobility patterns, in terms of systematic journeys:
M = 5.73, SD = 1.26 (n= 103);
•The reports suggested realistic and feasible alternatives for my systematic journeys:
M = 4.31, SD = 1.47 (n= 102);
Overall, these responses show a deﬁnitely positive assessment by users, especially about the system’s
capacity of automatically identifying the means of transport and the systematic loops (for the sake of
simplicity, called “systematic journeys” in the survey questions). To get a deeper insight, we developed a
second survey, again administered by means of an online questionnaire, aimed at collecting the perception of
accuracy by each user on each single systematic loop we had identiﬁed and on the related alternatives found.
We obtained responses by 85 users, namely 32.6% of the overall sample of study participants (261). Users
were shown a map for each systematic loop we had identiﬁed, presenting the sequence of points of interest
deﬁning each of them, and were asked (Q1) whether they had “traveled between the points shown in the
ﬁgure, for at least three times during the ﬁrst monitoring period (March-April 2016)”. Possible responses
where Yes/No/Partially. Then, they were shown an updated map, also reporting route connections between
such POIs, and were asked a second question (Q2): “Would you classify this route as systematic (namely, a
route that you frequently travel)?” Possible responses ranged from 1 to 5, where 1 = “deﬁnitely no” and 5 =
The collected answers refer to a total of 651 systematic loops, by 85 diﬀerent users. Question Q1 obtained
responses for 550 systematic loops: 86% of them were “Yes” and 6.8% were “No”. This conﬁrms a high
accuracy in the identiﬁcation of repeatedly traveled loops. Moreover, from responses to question Q2 (again
obtained for 550 systematic loops), we got the conﬁrmation that the frequent loops we had identiﬁed were
generally perceived as systematic: the mean value of Q2 responses was 3.98 (standard deviation = 1.40),
with 58% of the identiﬁed loops classiﬁed as deﬁnitely correct and 69% of them classiﬁed as either correct or
deﬁnitely correct (Likert scores equal to 4 and 5, see Figure 8). From the comments to questions Q1 and Q2,
we noticed that several users did not perceive those loops frequently traveled for shopping or leisure reasons
as systematic; others have judged loops as non-systematic because they have changed their routines, but
admitted that those loops where systematic during the monitoring period. This explains why the percentage
of positive answers to Q2 is smaller than for Q1.
For each loop for which a CO
-eﬃcient alternative was automatically found, the questionnaire also
showed the user the alternative (consisting of the suggested route and the means of transport between the
same POIs). We identiﬁed 235 such alternatives for the 651 systematic loops (36.1%). Based on these 235
alternatives, we asked a ﬁnal question (Q3): “Do you think the suggested route and means of transport are
plausible?”. Again, possible responses ranged from 1 to 5, where 1 = “deﬁnitely no” and 5 = “deﬁnitely yes”.
Overall, we received responses for 42 out of the 235 alternatives for systematic loops (17.9%). In this case,
the perceived accuracy by the users was lower, indicated by a mean value of 3.38 (SD = 1.62) in answers to
Q3. However, if we consider as plausible alternatives those assessed with a score of 4 or 5, they result in 50%
of the alternatives automatically found (see 8). A score of 3 (neutral assessment) is attributed to 17% of the
alternatives, thus leaving a 33% of alternatives that are assessed as not plausible. Due to the way the Q3
question was framed, these answers to Q3 could at least partially be aﬀected by a “social desirability bias”
]. As a consequence, we acknowledge that the actual percentage of plausible alternatives might be
slightly lower than indicated by current ﬁgures, though we are not able to quantify how much.
Figure 8: Assesment provided by survey respondents for (a) each systematic loop and (b) each alternative to a systematic loop
we had identiﬁed, according to a 5-point Likert scale.
5.2. Analysis of the Overall Potential for Change
Finally, we analyzed the results obtained when applying the previously introduced methods to all the 261
users of our study. Although aﬀected by the approximations necessarily involved in real data collection, these
results give the scale of potential energy savings that can be achieved by fostering changes in the mobility
behavior of comparable communities or groups of individuals. Taking into account the above limitations
related to the automatic identiﬁcation of alternatives, it is important to note that the results presented below
should be interpreted as upper boundaries for the potential savings, which can be achieved only in the ideal
situation where no other constraints limit user choices beside those applied here to the travel duration and
mode availability. Further, plane trips have been removed from the collected data, as they strongly impact
Tot. number of users: 261
Tot. number of routes: 33’855
Tot. number of SL: 1’375
Average SL repetitions: 5.2
Tot. number of SL alternatives: 638
Original SL Alternatives SL Original NS Alternatives NS
Weekly distances 76.9 km 45.5 km 310.2 km 273.3 km
Weekly energy consumption 45.2 kWh 17.8 kWh 200.4 kWh 159.9 kWh
Weekly CO2emissions 8.6 kg 3.0 kg 39.1 kg 30.5 kg
Use of car 43.6 % 18.9 % 53.1 % 39.3 %
Use of public transport 32.3 % 40.2 % 33.6 % 43.3 %
Use of bicycle 12.5 % 25.0 % 4.2 % 7.2 %
Use of walking 8.1 % 13.1 % 5.9 % 7.4 %
Other 3.5 % 2.7 % 3.2 % 2.8 %
Table 2: Analysis of the mobility patterns and potentials for change over all users. SL stands for “systematic loops” while NS
stands for “non systematic” and refers to routes. Reported are the average values per user, if not indicated otherwise (e.g., the
average user travels 387.1 km (76.9 + 310.2) per week, but could potentially reduce this to 318.8 km (45.5 + 273.3) per week).
the overall CO
emissions obtained for the short monitoring period (six weeks), and at the same time can
hardly be substituted by any realistic alternative. Also, in the computation of alternative values, all routes
were considered, namely those included in systematic loops (for which a loop alternative was computed), and
those which could not be included in any loop (for which a route alternative was computed). This means
that for every route an alternative was computed and taken in the case it was found to be feasible (according
to the criteria introduced in Section 4.2).
Table 2 shows some global results obtained when applying the methods described above to the routes
collected for all 261 users monitored during the project. The ﬁrst part of the table gives the total number
(tot) of users monitored, routes recorded and loops reconstructed, the average number (avg) of systematic
loops (SL) identiﬁed for each user and the average number of times each SL is traveled (SL repetitions).
One can see that on average, mobility patterns of a user can be characterized by about 5.3 systematic
loops, repeated about 5.2 times in a six week period. The second part of the table presents the distances
traveled, the energy consumed and the CO
emitted each week, reported as the mean (over all users) of the
average weekly values. All values are separately reported for systematic loops (SL) and for all the other
non-systematic routes (NS). The “alternative” columns compare these values to what could be achieved by
using the alternative solutions identiﬁed by our algorithms. Finally, in its third part, Table 2 analyzes the
overall modal choice, i.e., the fraction of the total distance (summed over all users) traveled with each mode
of transport. Figure 9 provides more detail about the distribution of such indicators among users, showing
box-plots for the modal choices and the average distances traveled, energy consumed and CO
Each row shows both the orig inally tracked data and the computed alternatives, considering the systematic
loops as well as all remaining non-systematic routes. The individual dots in the ﬁgure are outlier users, while
the boxes represent 50% of all users. Looking at the ﬁrst rows, for example, one can see that the parts of
systematic loops that are traveled by car could be replaced by more sustainable means of transport in many
cases, while the non-systematic car routes have a much smaller potential for replacement (often due to the
fact that they cover long distances to places far away from public transport stops, preventing both the use of
a bicycle as well as public transport). It can be seen how a large part of the potential energy savings stem
from a transition from car to public transport, bicycles, and walking. Energy savings, however, can also
come from a decrease in the weekly traveled distance. These savings come from situations similar to the
example in Figure 2, where the detours of a public transport journey are replaced by a more direct and thus
shorter bicycle path.
Figure 9: Distribution of modes of transport, distances traveled, energy consumption, and CO
emissions for the actual (orig)
and alternative (alt) routes, both for systematic loops (sys) and non systematic routes (nonsys).
As presented in the previous section, the user survey highlights that the systematic loops found were
generally of satisfactory quality. It is more diﬃcult to propose meaningful alternatives to users, partially
shown by the fact that we “only” found 235 alternatives for 651 systematic loops, partially by the lower
score of Q3. There are multiple reasons why it is impossible to ﬁnd suitable alternatives for all systematic
loops, but we would like to highlight two important ones here. First, as the loop length increases, it
becomes more diﬃcult to ﬁnd a suitable alternative, because each segment of the loop requires a suitable
alternative, and in sequence they must satisfy the mode availability constraints. This implies that ﬁnding
alternatives for individual routes is much easier than for complete loops. Second, in several surveys and
interviews we conducted, we found that our study sample is biased towards people who regard ecological
sustainability as a requirement for the future of humanity (pro-environmental attitude). Such people seem to
be naturally inclined to participate in experiments that try to make behavior more ecologically sustainable.
This attitude manifests in their systematically traveled loops, which are shorter and often covered more with
sustainable modes of transport than their non-systematic counterparts. In our two study regions, we found
that non-systematic routes are 3.4 (Zurich; urban) and 4.1 (Ticino; rural) times longer than routes appearing
in systematic loops. In Ticino, 74% of the routes of systematic loops are traveled by car, compared to 71% of
the non-systematic ones. In Zurich, however, 36% of the routes of systematic loops are covered by car, in
contrast to 51% of the non-systematic ones. This means that in general we have higher chances to ﬁnd a
promising alternative for a non-systematic route.
To explain the lower score for the automatic identiﬁcation of alternatives (Q3), one should consider
that our methods do not include many factors which may inﬂuence mobility choices and that the analyses
performed are necessarily aﬀected by the inaccuracies of the tracking system. For instance, if a visited place
is not recognized properly by our method, the suggested alternative does not visit this place of actual interest
for the user. This in turn will cause the user to assess it as “not useful”. Moreover, users might have speciﬁc
constraints, related to the need for accompanying family members, for having to carry weights, or for being
capable of suddenly changing schedules and routes due to family or job reasons. Such speciﬁc elements are
highly user dependent and cannot be taken into account by the system. Therefore, even if theoretically
speaking a more eﬃcient alternative from the point of view of CO
emissions is available to connect the
visited points of interest, speciﬁc needs of single users might lead to assess it as unfavorable and not realistic.
As this is highly dependent on the context of a user, it can also happen that someone classiﬁes an alternative
as feasible, but still is not able to use the alternative in a majority of cases. While in the here presented
study we only looked at the overall CO
and energy savings, it would be interesting to analyze the individual
route choices of people over a certain time span. In a “living lab” without particular constraints on the
participants, observing such changes is diﬃcult, though, as people change their places of living and work,
their personal circumstances, etc.
The eco-feedback provided only marginally accounts for individual diﬀerences between users. As mentioned
above, personal needs and preferences cannot always be recognized from automatic tracked data. As such,
some reasonable requirements constraining the modal choice have been assumed to apply indiscriminately to
all users. A more involved interaction with the users (e.g., asking people to state the purpose of each trip), as
well as improved tracking accuracy, would improve the accuracy of the suggestions and allow for some degree
of personalization of the feedback. In retrospective, a more accurate tracking app, allowing the inclusion
of travel survey techniques (e.g., asking for additional data about each trip) could have been beneﬁcial,
in particular for the derivation of more meaningful alternatives. Relying on Moves not only discouraged
asking users for additional travel data, but also meant relying on the given accuracy for the identiﬁcation of
points of interest, and thus recognizing of loops. Ultimately, these circumstances also inﬂuenced the upper
bounds presented in Table 2 and Figure 9, as the tracking inaccuracies can lead to points of interest not
being recognized, and thus not being respected in the computation of alternatives.
However, increasing the frequency of interactions between the mobility tracking app and its users would
need a careful assessment. In fact, in order to make smartphone-based tracking a viable solution for real world,
large-scale mobility monitoring, very little or no interaction with the user should be requested. This also
made it impossible to employ a “traditional” travel survey (where people manually enter traveled journeys
into a system). Additionally, in projects and studies aimed at assessing the eﬀectiveness of behavior change
interventions in controlled trials, namely by comparing mobility patterns before and after an intervention, and
with respect to a control group, any interactions between the app and its users would potentially inﬂuence
their behavior and, thus, the outcome of the study. As shown by high drop-out rate registered through
the three phases of our project and by ﬁnal interviews we performed with project participants, there is a
tangible risk that users feel under pressure by the interaction eﬀort requested by the app and, to avoid it,
directly opt for quitting app use. Therefore, the availability of a smartphone-based tracking tool requiring
none or very limited interaction with the users would be largely beneﬁcial also for other research purposes,
besides pure mobility tracking. This induces that the tracking system essentially includes a reliable transport
mode classiﬁer, as this would remove the need of asking for the actual transport mode used (due to the
unavailability of a highly accurate classiﬁer, in our study, users were asked to manually validate each recorded
The eco-feedback presented here was primarily delivered to users in terms of a static report as well
as with the use of several gamiﬁcation elements. Especially the potential for change can be used as an
individual starting point for goal setting, concrete suggestions, or rankings of users, as these elements are
highly user- and context-speciﬁc (e.g., creating a leaderboard of “most eco-friendly” users would be heavily
biased towards people living in cities or having short journeys to work). Alternative or supplementary choices
to the above presented eco-feedback include monetary and educational (any impersonal information about the
ecological sustainability of certain travel behaviors) incentives. Within this study we focused on automatically
delivered personalized feedback for various reasons: monetary incentives are purely extrinsic motivators,
and are thus known to only show temporary eﬀects [
]. Additionally, while educational incentives are
especially valuable for uninformed people, they show smaller eﬀects for people who are already aware of a
certain problem, and are considering a change, though are still caught in a typical “attitude-behaviour” or
“value-action” gap [
]. As several surveys and interviews accompanying the study showed, our participants
were already inclined to optimize their travel behavior with respect to ecological sustainability, but did not
know how to improve it or lacked discipline in changing their behavior.
7. Conclusion and Outlooks
We have presented a method to analyze automatically tracked location trajectories and provide eco-
feedback to their users. The method builds upon the identiﬁcation of systematic mobility patterns in user
trajectories and the suggestion of alternative modal choices for such frequent travels. In particular we have
focused on the identiﬁcation of systematic loops starting and ending at a user’s home. We argue that these
loops build a solid foundation to compute eco-feedback, as they reﬂect the actual mobility needs of each user,
allow providing reasonable and feasible solutions, and oﬀer a good level of abstraction to communicate to
users. For each loop, the proposed approach computes an alternative optimal solution that minimizes the
emissions while respecting a set of constraints (e.g., on the total duration of the travel) that can make it
acceptable to the average user. This approach can eﬀectively be applied in other countries than Switzerland,
where we originally developed it, provided that country-speciﬁc
emissions factors for each transport
mode are considered, for example by retrieving them from the ecoinvent Life Cycle Assessment database [
The generated eco-feedback is mostly of educational nature, i.e., it does not create an incentive for a
person to change her mobility behavior, but simply shows possible ecological gains from such a change. Of
course, this assessment can be used as a base to create incentives (e.g., by building gamiﬁcation elements on
top of it), thus making the suggestions provided to users more eﬀective [
]. Alternatively, behavior change
could be fostered by pointing out beneﬁts of modal changes. This can include time for other activities (such
as reading or working on a train) as well as economic aspects, ranging from simply showing the user how
much money she would save by performing a certain behavior to actively encouraging certain behavior by
travel vouchers, beneﬁts, etc. In order to give people more meaningful travel alternatives and thus better
eco-feedback, however, more accurate tracking technology as well as more detailed data about the individual
journeys (e.g., the purpose, or additional constraints like luggage or people the user is traveling with) is
required. A challenge for future research activities is therefore to look for a closer collaboration with app
users, in order to better understand their mobility needs and receive a more detailed feedback about the
quality of loop recognition and the applicability of alternatives, while at the same time trying to limit the
amount of compulsory interactions with the app as much as possible.
Another line of research can concern the method itself. Transport options such as carpooling or car-
or bikesharing make viable alternatives for many mobility demands, which were not included in this work.
As the exact journeys people travel are known, an automated assessment of possible carpools and shared
vehicles would be possible. This could be used to create eco-feedback in the form of suggestions to work
together with other app users. Another extension could consider rescheduling user’s activities, in order to
create an optimal daily schedule in terms of mobility and sustainability. Again, this is highly dependent
on various individual factors (e.g., trip purpose, or interaction with other people’s schedules) which would
require a much stronger interaction of the user with the app.
Concerning the eco-feedback itself, its immediacy plays a big role in making it an eﬀective tool for
fostering more sustainable mobility behaviors [
]. While in our case the feedback was delivered passively
after a certain period of time, the prompt recognition of a starting journey and the real-time suggestion of
an alternative could greatly increase its eﬀectiveness. The main challenge in this real-time approach is to
anticipate the intentions (e.g., the destination) from the tracked trajectories of a user. This research area has
recently received a lot of attention though, and will increase in importance as location based services get
more and more prevalent.
This research was supported by the Swiss National Science Foundation (SNF)
within NRP 71 “Managing energy consumption” and is part of the Swiss Competence Center for Energy
Research SCCER Mobility of the Swiss Innovation Agency Innosuisse.
Jillian Anable, Ben Lane, and Tanika Kelay. An evidence base review of public attitudes to climate change and transport
behaviour. The Department, 2006.
Theo A Arentze. Adaptive personalized travel information systems: A bayesian method to learn users’ personal preferences
in multimodal transport networks. IEEE Transactions on intelligent transportation systems, 14(4):1957–1966, 2013.
Daniel Ashbrook and Thad Starner. Using GPS to learn signiﬁcant locations and predict movement across multiple users.
Personal and Ubiquitous computing, 7(5):275–286, 2003.
Erel Avineri. On the use and potential of behavioural economics from the perspective of transport and climate change.
Journal of Transport Geography, 24:512–521, 2012.
Hannah Bast, Daniel Delling, Andrew Goldberg, Matthias M¨uller-Hannemann, Thomas Pajor, Peter Sanders, Dorothea
Wagner, and Renato F Werneck. Route planning in transportation networks. In Algorithm Engineering, pages 19–80.
Jing Bie, Marcel Bijlsma, Gregor Broll, Hu Cao, Anders Hjalmarsson, Frances Hodgson, Paul Holleis, Y Houten, J Koolwaaij,
D Kusumastuti, et al. Move better with Tripzoom. International journal on advances in life sciences, 4(3&4):125–135,
 Dominik Bucher, Francesca Cellina, Francesca Mangili, Martin Raubal, Roman Rudel, Andrea Emilio Rizzoli, and Omar
Elabed. Exploiting ﬁtness apps for sustainable mobility–Challenges deploying the GoEco! app. ICT for Sustainability
Dominik Bucher, David Jonietz, and Martin Raubal. A heuristic for multi-modal route planning. In Georg Gartner and
Haosheng Huang, editors, Progress in Location-Based Services 2016, pages 211–229. Springer International Publishing,
Francesca Cellina, Dominik Bucher, Martin Raubal, Roman Rudel, Vanessa de Luca, and Massimo Botta. GoEco!–A set
of smartphone apps supporting the transition towards sustainable mobility patterns. Change-IT Workshop at ICT for
Sustainability (ICT4S), 2016.
Francesca Cellina, Dominik Bucher, Roman Rudel, Martin Raubal, and Andrea-Emilio Rizzoli. Promoting sustainable
mobility styles using eco-feedback and gamiﬁcation elements: introducing the GoEco! living lab experiment. In Behave
2016 - 4th European Conference on Behaviour and Energy Eﬃciency, At Coimbra, Portugal, 2016.
 Daniel Delling, Thomas Pajor, and Renato F Werneck. Round-based public transit routing. page 11.
Daniel Delling, Thomas Pajor, and Renato F Werneck. Round-based public transit routing. Transportation Science,
Julian Dibbelt, Thomas Pajor, and Dorothea Wagner. User-constrained multimodal route planning. Journal of Experimental
Algorithmics (JEA), 19:3–2, 2015.
Brian J Fogg. Persuasive technology: using computers to change what we think and do. Ubiquity, 2002(December):5, 2002.
Elisha R Frederiks, Karen Stenner, and Elizabeth V Hobman. Household energy use: Applying behavioural economics to
understand consumer decision-making and behaviour. Renewable and Sustainable Energy Reviews, 41:1385–1394, 2015.
Jon Froehlich. Gamifying green: gamiﬁcation and environmental sustainability. In Steﬀen P Walz and Sebastian Deterding,
editors, The Gameful World: Approaches, Issues, Applications, pages 563–596. Mit Press, London, 2015.
Jon Froehlich, Tawanna Dillahunt, Predrag Klasnja, Jennifer Mankoﬀ, Sunny Consolvo, Beverly Harrison, and James A
Landay. UbiGreen: investigating a mobile tool for tracking and supporting green transportation habits. In Proceedings of
the SIGCHI Conference on Human Factors in Computing Systems, pages 1043–1052. ACM, 2009.
Jon Froehlich, Leah Findlater, and James Landay. The design of eco-feedback technology. In Proceedings of the SIGCHI
Conference on Human Factors in Computing Systems, pages 1999–2008. ACM, 2010.
S Gabrielli, R Maimone, P Forbes, and S Wells. Exploring change strategies for sustainable urban mobility. In Designing
Social Media for Change at the ACM SIG-CHI Conference on Human Factors in Computing Systems (CHI 2013), 2013.
Fosca Giannotti, Mirco Nanni, Fabio Pinelli, and Dino Pedreschi. Trajectory pattern mining. In Proceedings of the 13th
ACM SIGKDD international conference on Knowledge discovery and data mining, pages 330–339. ACM, 2007.
Juho Hamari, Jonna Koivisto, and Tuomas Pakkanen. Do persuasive technologies persuade?–a review of empirical studies.
In International conference on persuasive technology, pages 118–136. Springer, 2014.
Chris Harding, S. Srikukenthiran, Z. Zhang, K. Nurul Habib, and EJ Miller. On the user experience and performance of
smartphone apps as personalized travel survey instruments: Results from an experiment in toronto. In 11th International
Conference on Transport Survey Methods (ISCTSC), Est´erel, Qu´ebec, 2017.
Stephane Hess, Mohammed Quddus, Nadine Rieser-Sch¨ussler, and Andrew Daly. Developing advanced route choice models
for heavy goods vehicles using GPS data. Transportation Research Part E: Logistics and Transportation Review, 77:29–44,
 Stephen C Johnson. Hierarchical clustering schemes. Psychometrika, 32(3):241–254, 1967.
Antti Jylh¨a, Petteri Nurmi, Miika Sir´en, Samuli Hemminki, and Giulio Jacucci. Matkahupi: a persuasive mobile application
for sustainable mobility. In Proceedings of the 2013 ACM conference on Pervasive and ubiquitous computing adjunct
publication, pages 227–230. ACM, 2013.
Anja Kollmuss and Julian Agyeman. Mind the gap: why do people act environmentally and what are the barriers to
pro-environmental behavior? Environmental education research, 8(3):239–260, 2002.
John Krumm and Eric Horvitz. LOCADIO: Inferring motion and location from Wi-Fi signal strengths. In mobiquitous,
pages 4–13, 2004.
Julia Letchner, John Krumm, and Eric Horvitz. Trip router with individualized preferences (trip): Incorporating
personalization into route planning. In Proceedings of the National Conference on Artiﬁcial Intelligence, volume 21, page
1795. Menlo Park, CA; Cambridge, MA; London; AAAI Press; MIT Press; 1999, 2006.
Quannan Li, Yu Zheng, Xing Xie, Yukun Chen, Wenyu Liu, and Wei-Ying Ma. Mining user similarity based on location
history. In Proceedings of the 16th ACM SIGSPATIAL international conference on Advances in geographic information
systems, page 34. ACM, 2008.
Long Liu, Jin Xu, Stephen Shaoyi Liao, and Huaping Chen. A real-time personalized route recommendation system for
self-drive tourists based on vehicle to vehicle communication. Expert Systems with Applications, 41(7):3409–3417, 2014.
Yin Lou, Chengyang Zhang, Yu Zheng, Xing Xie, Wei Wang, and Yan Huang. Map-matching for low-sampling-rate GPS
trajectories. In Proceedings of the 17th ACM SIGSPATIAL international conference on advances in geographic information
systems, pages 352–361. ACM, 2009.
 Harvey J Miller and Jiawei Han. Geographic data mining and knowledge discovery. CRC Press, 2009.
Lara Montini and Kay W Axhausen. Preliminary results: Route choice analysis from multi-day GPS data. In 15th Swiss
Transport Research Conference, Ascona, 2015.
Anton J Nederhof. Methods of coping with social desirability bias: A review. European journal of social psychology,
Andrey Tietbohl Palma, Vania Bogorny, Bart Kuijpers, and Luis Otavio Alvares. A clustering-based approach for discovering
interesting places in trajectories. In Proceedings of the 2008 ACM symposium on Applied computing, pages 863–868. ACM,
Donald J Patterson, Lin Liao, Dieter Fox, and Henry Kautz. Inferring high-level behavior from low-level sensors. In
International Conference on Ubiquitous Computing, pages 73–89. Springer, 2003.
Santi Phithakkitnukoon, Teerayut Horanont, Giusy Di Lorenzo, Ryosuke Shibasaki, and Carlo Ratti. Activity-aware
map: Identifying human daily activity pattern using mobile phone data. In International Workshop on Human Behavior
Understanding, pages 14–25. Springer, 2010.
Mohammed Quddus and Simon Washington. Shortest path and vehicle trajectory aided map-matching for low frequency
GPS data. Transportation Research Part C: Emerging Technologies, 55:328–339, 2015.
Sasank Reddy, Jeﬀ Burke, Deborah Estrin, Mark Hansen, and Mani Srivastava. Determining transportation mode on
mobile phones. In Wearable Computers, 2008. ISWC 2008. 12th IEEE International Symposium on, pages 25–28. IEEE,
Lior Rokach and Oded Maimon. Clustering methods. In Data mining and knowledge discovery handbook, pages 321–352.
Johann Schrammel, Sebastian Prost, and Manfred Tscheligi. Personalized persuasion: Addressing target behavior,
persuadability and persuasive agents within PEACOX. CHI ’14 Workshop: Personalizing Behavior Change Technologies,
Nadine Schuessler and Kay Axhausen. Processing raw data from global positioning systems without additional information.
Transportation Research Record: Journal of the Transportation Research Board, (2105):28–36, 2009.
Monika Sester, Udo Feuerhake, Colin Kuntzsch, and Lijuan Zhang. Revealing underlying structure and behaviour from
movement data. KI-K¨unstliche Intelligenz, 26(3):223–231, 2012.
Leon Stenneth, Ouri Wolfson, Philip S Yu, and Bo Xu. Transportation mode detection using mobile phones and GIS
information. In Proceedings of the 19th ACM SIGSPATIAL International Conference on Advances in Geographic
Information Systems, pages 54–63. ACM, 2011.
Varsolo Sunio and Jan-Dirk Schm¨ocker. Can we promote sustainable travel behavior through mobile apps? evaluation and
review of evidence. International Journal of Sustainable Transportation, 11(8):553–566, 2017.
M Tuchschmid and M Halder. mobitool–grundlagenbericht: Hintergrund, methodik & emissionsfaktoren. Tuchschmid und
M. Halder im Auftrag von SBB, Swisscom, BKW und ¨
Yang Wang, Abdulbaki Uzun, Ulrich Bareth, and Axel K¨upper. Tracommender–Exploiting continuous background
tracking information on smartphones for location-based recommendations. In International Conference on Mobile Wireless
Middleware, Operating Systems, and Applications, pages 250–263. Springer, 2012.
Paul Weiser, Dominik Bucher, Francesca Cellina, and Vanessa De Luca. A taxonomy of motivational aﬀordances for
meaningful gamiﬁed and persuasive technologies. In Proceedings of the ICT4S 2015, 7-9 September 2015, Copenhagen,
 Paul Weiser, Simon Scheider, Dominik Bucher, Peter Kiefer, and Martin Raubal. Towards sustainable mobility behavior:
Research challenges for location-aware information and communication technology. GeoInformatica, 20(2):213–239, 2016.
Simon Wells, Paula Forbes, Judith Masthoﬀ, Silvia Gabrielli, and Antti Jyllha. SUPERHUB: integrating digital behaviour
management into a novel sustainable urban mobility system. In Proceedings of the 27th International BCS Human
Computer Interaction Conference, page 62. British Computer Society, 2013.
Thomas Wernbacher, Alexander Pfeiﬀer, Mario Platzer, Martin Berger, and Daniela Krautsack. Traces: A pervasive app
for changing behavioural patterns. In European Conference on Games Based Learning, page 589. Academic Conferences
International Limited, 2015.
Gregor Wernet, Christian Bauer, Bernhard Steubing, J¨urgen Reinhard, Emilia Moreno-Ruiz, and Bo Weidema. The
ecoinvent database version 3 (part i): overview and methodology. The International Journal of Life Cycle Assessment,
Joshua H West, P Cougar Hall, Carl L Hanson, Michael D Barnes, Christophe Giraud-Carrier, and James Barrett. There’s
an app for that: content analysis of paid health and ﬁtness apps. Journal of medical Internet research, 14(3):e72, 2012.
Yang Ye, Yu Zheng, Yukun Chen, Jianhua Feng, and Xing Xie. Mining individual life pattern based on location history. In
Mobile Data Management: Systems, Services and Middleware, 2009. MDM’09. Tenth International Conference on, pages
1–10. IEEE, 2009.
Yu Zheng, Yukun Chen, Quannan Li, Xing Xie, and Wei-Ying Ma. Understanding transportation modes based on GPS
data for web applications. ACM Transactions on the Web (TWEB), 4(1):1, 2010.