Personal Data in Transport: Exploring A Framework For The Future
Uploaded by
Open Data InstitutePersonal Data in Transport: Exploring A Framework For The Future
Uploaded by
Open Data Institute
About
This report has been researched and produced by the Open Data Institute, and
published in June 2018. Its lead author was Libby Young, with contributions from
Gillian Whitworth, Jared Robert Keller, Jamie Fawcett, Miranda Marcus, Jessica
Weereratne, Caley Dewhurst, Anna Scott, Peter Wells and Jeni Tennison.
The purpose of the report is to help transport companies make better decisions
about how they use personal data in their business models for the future. Our
focus is journey data, a form of personal data the sector has always needed, now
accessible in new ways and at new scales.
The report coincides with the implementation of the EU’s General Data Protection
Regulation (GDPR). It is one of the most comprehensive pieces of regulation yet to
deal with personal data, and is changing the way businesses across sectors
collect and use data.
The GDPR is just one part of a growing trend towards people exercising more
control over data about them, so we take a long term view on this as well as other
trends – the increasing collection and use of journey data in transport, and the
benefits this can create for businesses who foster and earn trust by engaging with
us as they use data about us ethically and equitably.
We hope businesses find the research and learning shared here helpful in
exploring how to address and lead change around these trends, and as they work
to find ways to make travel more accessible, affordable and sustainable for all.
If you would like to send us feedback, please get in
touch by email at [email protected].
This report is published under the Creative Commons Attribution-ShareAlike 4.0
International licence. See: https://creativecommons.org/licenses/by-sa/4.0.
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 1
Contents
About 1
Contents 2
A note on language: 3
We don’t ‘own’ data 3
Executive summary 4
Data is infrastructure, like transport 5
Why is journey data personal data? 6
Defining terms 7
Legal and human definitions of personal data 7
Transport needs data 8
Defining terms 9
Transport, mobility, MaaS: explaining the differences 9
Companies are thinking about sharing data 10
Open standards in transport 11
Consumers and policy makers are also thinking about sharing data 12
Case studies for the future 13
Transport in Finland 13
Fostering trust in transport 14
How tighter data protection regulation impacts transport businesses 15
Beyond regulation – ethics, equity, engagement 16
1. Using data ethically 16
2. Supporting equitable access to and benefits from data 17
Rural mobility: an overlooked opportunity? 19
3. Engaging to make data work for all 20
Case studies for the future 21
TOC Ability 21
How transport uses data 22
Trains 22
Buses 23
Case studies for the future 23
Smart buses in the UK 23
Cars 24
Ridesharing 26
Bikesharing 27
Journey planners 27
Case studies for the future 28
Digital Matatus 28
Smart ticketing 29
A framework for the future 31
Recommendations 32
Appendix: Methodology 38
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 2
A note on language:
We don’t ‘own’ data
At the ODI, we believe that data does not belong to any one person, not
least because data about us is often also about other people. Our DNA, for
example, reveals information about our parents and family. Utility bills
reveal who we live with. Health records contain information about medical
professionals. This is why we do not refer to ‘my or your data’ but instead
use ‘data about me, you, us or them’.
Many people can have rights to – and responsibilities for – data. For
instance, many people and organisations may have intellectual property
rights due to their contributions to the data. In transport, people often take
journeys with family or friends, so ticket or location data can be connected
to or used to identify more than one person. In connected vehicles such as
cars, drivers and their passengers may be connected with other traffic
participants.
By avoiding possessive pronouns we hope to encourage understanding of
the wider network of relationships involved in data collection and use. Put
simply, neither individuals or organisations should think of themselves as
owning data. Instead they should think of themselves as holding or
stewarding data with the goals of both protecting the rights of people
impacted by the data and ensuring that people who have the right to
access and use it can get the data when they need it to make a decision.
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 3
Executive summary
Using journey data to meet a universal need
Transport is a universal need: almost everyone travels regularly, be it walking,
taking a bus, using a car or going by train. These journeys often generate data that
can reveal things about us – where we work, where our loved ones live, what we
like to do. This journey data is personal data.
The role of journey data in transport is changing. It is being collected in real time
and used on new scales, creating insights into individual and aggregated mobility.
Businesses and public services can use this data to make operational and policy
decisions to improve transport services, user experience or infrastructure.
Innovative new mobility services are also emerging
New ways of operating transport, such as bikesharing and ridesharing, and new
ways of accessing transport, like journey planners and smart ticketing, are
changing the sector. These on-demand, personalised services can help people
save time and money, be more (or less) active, own fewer or no means of
transport, lower their carbon footprint, or travel in other ways that meet their
individual or societal needs. This also opens up ways to make transport systems
more user-centred for more people and for the full length of their desired journeys.
In the future, travellers could complete trips door-to-door using a single planner
and ticket and access varied public and private transport as their journey requires.
Through our interviews with industry, workshops and research we found that
journey data can be used and shared to benefit people, organisations and
our transport systems more broadly in three ways:
● to increase accessibility and personalisation for people
● to improve operational efficiency and innovation for organisations
● to tackle systemic transport issues for the benefit of everyone
We also uncovered areas that must be explored to help realise this potential
equitably, using data about us ethically, and in ways that support our systems.
For instance, just as open data from public transport and other sources supports
new mobility services, these new services now hold journey datasets that can
support better decision-making about transport infrastructure and operations.
Many companies are thinking about ways to make aggregated and de-identified
journey data more open for this purpose, and some of them are testing ways to do
so. Further exploration is encouraged.
Our research also showed that sharing journey and other transport data more
widely requires us to address technical barriers including inconsistent open
standards, data silos and poor interoperability of data. Breaking down these
barriers will require collaboration and investment across the sector.
Most importantly, we found that journey data is transforming relationships
between companies and customers, drastically increasing the importance of trust,
and raising critical questions of ethics, equity and engagement which cannot go
unanswered. The impact of these trends will only increase as people grow more
aware of these issues and as they gain more rights over data about them.
These shifts create both opportunities and challenges for businesses. Based on
these and our other findings, we recommend transport organisations:
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 4
i) Develop a strategic and ethical approach to data by looking beyond
compliance with regulation to develop governance policies and processes,
investing in training and resources, and engaging with sector initiatives.
ii) Lay the foundations for and engage in data-enabled innovation by
supporting open standards and APIs, practising open innovation, and
exploring new ways to partner with public services.
iii) Foster trusted relationships with customers by communicating with them
clearly, finding ways to build more informed consent into the user
experience and explaining the benefits created by data sharing.
iv) Explore sharing more journey data safely and securely in three ways:
1. As open data – with careful and appropriate aggregation and
anonymisation, and accompanied by privacy impact assessments
2. In secure environments – in particular with accredited researchers,
who should openly publish their results
3. With people the data is about – through open APIs and in ways that
enable innovation.
There is much at stake for us all in the future of data and sharing data in transport.
For economies, the Transport Systems Catapult says data-enabled innovations in
transport may be worth £900bn by 2025, and conversely, not sharing transport
data may cost a country like the UK £15bn1.
For individuals, sharing journey and other data to increase personalisation of and
access to transport can improve our quality of life. Conversely, if personal data is
used in ways that erode trust it can create data wastelands (where data is not
collected or used), limiting the potential for transport systems to benefit everyone.
A future in transport where data is shared and used in ways that combine
openness and innovation with ethics and trust has powerful potential to benefit
our societies and economies.
Data is infrastructure, like transport
Infrastructure powers our societies. It provides the fundamental services and
systems that enable our economies to function, allow us to communicate,
and improve and support our lives.
When we think of infrastructure we first think of roads, bridges, water
supplies and electrical grids. But infrastructure also takes less tangible forms,
such as ideas, basic research and the internet.
Data is becoming as important as our physical infrastructure – it underpins
every sector of our economies and impacts every aspect of society. Roads
help us navigate to a place; data helps us navigate to a decision.
Data infrastructure consists of data assets, the organisations that operate
and maintain them, and the guides describing how to use and manage the
data. Trustworthy data infrastructure is sustainably funded and is directed to
maximise data use and value, meeting society’s needs.
Data infrastructure that is as open as possible creates the most value. Like
other types of infrastructure it is invisible (when it works). Making it as open
1
Transport Systems Catapult, h
ttps://ts.catapult.org.uk
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 5
as possible, while respecting privacy, helps to support a huge variety of uses
and ecosystems.
Governments, businesses and communities plan essential physical
infrastructure strategically – including our highways, electricity lines, water
courses and broadband connections – and we should treat data
infrastructure in the same way.
Why is journey data personal data?
Our journeys from location to location through time can be used to infer where we
work and shop, what we earn and what we do in our free time. These are all things
about us. Data collected when we make journeys is revealing more about us than
ever, such as when and where we travel with friends, family or colleagues2. For
these reasons, journey data is personal data.
Historically, public and private transport providers collected journey data as
‘counts’, i.e. the number of people on a train, bus tickets sold or cars using a road.
These counts were usually not tied to individuals, except when people took part in
customer surveys. The data helped transport organisations make decisions about
things like timetables or the location of bus stops.
In the last few years, GPS-enabled devices, smartphones and NFC-enabled smart
tickets and payment cards have made journey data easier and cheaper to collect.
Parts of the sector have moved on from ‘counts’ to data collected using these
technologies, at much larger scales and in real time, across modes of transport.
These richer journey datasets can help decision-makers improve transport
planning, personalise services and tackle systemic issues. They have increased
innovation by established operators and allowed new mobility services to emerge.
Our increasing use of these technologies to plan journeys and access transport is
allowing banks and smartphone makers to collect journey data. Banks can do this
when providing payment services for transport. And smartphone makers can
collect data each time we enable our location services during a journey, or when
we access mobility services on our phones3. Telecom providers have historically
collected customer location data, generally by triangulating mobile phone signals,
but this data is not considered reliable enough to use as journey data supporting
decision-making in transport.
All this journey data can be pseudonymised with identifiers like random numbers,
allowing journeys by one individual to be linked without using direct identifiers like
names. However, even when data is pseudonymised, there is still a risk of
re-identification when it is linked to other datasets. One study found that four data
points on time and location can identify 95% of people in a dataset4. Others have
mapped a cyclist’s journeys with TfL data5, or identified trips made by taxi drivers
in New York6.
Consumers and regulators are becoming more engaged with the issues around
this and personal data more broadly. How organisations develop their personal
2
This report focuses on transport customers, but we note that employees of transport
companies, like bus drivers and station guards, are also affected by most of these issues.
3
While this report is on transport, we note location data raises issues for many industries
4
Yves-Alexandre de Montjoye, César A. Hidalgo, Michel Verleysen & Vincent D. Blonde
(2013), ‘Unique in the Crowd’, https://www.nature.com/articles/srep01376
5
James Siddle (2014), ‘I Know Where You Were Last Summer: London's public bike data is
telling everyone where you've been’,
https://vartree.blogspot.com/2014/04/i-know-where-you-were-last-summer.html
6
Vijay Pandurangan (2014), ‘On Taxis and Rainbows: Lessons from NYC’s improperly
anonymized taxi logs’, https://tech.vijayp.ca/of-taxis-and-rainbows-f6bc289679a1
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 6
data strategies could increasingly become a differentiator for transport
organisations. We will look at some of the opportunities and risks this can create
for businesses later in this report.
Defining terms
Legal and human definitions of personal
data
In debates about personal data there is often an undeclared tension
between its legal definition and its human definition.
Data protection regulations like the GDPR define personal data as specific
bits of information about ‘an identifiable person’, such as name or location7.
People often feel differently, thinking of personal data more broadly, as any
information which reveals something about them, such as where they work,
what they like and who their loved ones are.
In this report, we use the human definition because, while we will always
need strong regulation, it is ultimately people who choose whether or not to
trust a business with personal data about them.
7
https://ico.org.uk/for-organisations/guide-to-the-general-data-protection-regulation-gdpr/
key-definitions
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 7
Transport needs data
Access to journey data is transforming the sector
The role of data in transport is changing, as the sector shifts from data-poor to
data-rich. In particular, journey data, a form of personal data the sector has always
needed but has historically found hard to get, is now accessible in new ways and
at new scales. Companies are collecting more of this data, more frequently, and in
real time – as technology makes this more feasible and viable – and using it to
innovate. Customers expect more personalisation and communication, and are
sharing more real time data in order to get it.
Our research and interviews with industry show transport companies are
collecting data in a variety of ways. While cash-only taxis and buses collect little or
no journey data, public transport operators such as trains and new private
operators such as bikesharing and ridesharing companies can connect customer
account details with real time data collected via smartphones, smart cards and
other devices to gather journey data for single modes of transport. End-to-end
mobility services like journey planners and smart ticketing services can collect
journey data (and intended journey data, in the case of journey planners) across
many journeys and modes of transport, creating multimodal or overarching views
of transport in real time. Public transport providers who operate multiple modes of
transport, such as Transport for London (TfL) can also create these powerful
overviews.
This real time journey data can be used and shared to benefit businesses, people
and public services, potentially in ways that meet the needs of all three groups.
This highlights that transport is an ecosystem of organisations, individuals and
infrastructure depending on each other to achieve their goals. Our interviews,
workshops and research called attention to three key areas of potential benefit:
Increasing accessibility and personalisation for passengers
– people can benefit from more personalised or affordable services as
organisations use journey data to better understand and serve their needs.
For example, services like Waze can crowdsource and share car journey data to
help drivers choose their route based on live traffic patterns. TfL can use journey
data in a personalised way, to warn passengers of relevant disruptions affecting
their route and at times that allow them to adjust their journeys. Services like
Moovit or Digital Matatus can help people use public transport in cities with less
formal or discoverable networks (see Digital Matatus case study).
Improving operational efficiency and innovation for organisations
– private and public organisations can benefit by using journey data to
make better decisions about how to use resources to develop or improve
services.
For example, TfL can use journey data collected through its Oyster cards to
improve and plan its transport interchanges in London8. Rideshare services can
use journey planners – enabled by open data and journey data – to find and
connect with customers. Bikeshare services can use journey data to help locate
their bikes where customers want them, in turn leading to innovation in
complementary services like Stage Intelligence which can help mobility services
like bikesharing make these operational decisions9.
8
https://www.citymetric.com/transport/data-helps-us-provide-better-transport-tfl-oyster-ca
rds-big-data-and-contactless-payments-1396
9
https://stageintelligence.co.uk
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 8
Tackling systemic transport issues for the benefit of everyone
– public services and governments can use data to better meet the travel
needs of underserved communities, and address issues like the ‘last mile’,
congestion and pollution10. It can also help make previously expensive
services more affordable for businesses to provide.
For example, Stratford-upon-Avon startup Enable ID is collaborating with UK train
operators to use personal data to explore intelligent accessibility for people with
disabilities (see TOC Ability case study). Private companies are working with local
authorities to develop smart bus services in rural areas currently underserved by
public transport (see Smart Buses in the UK case study). Finnish startup MaaS
Global is using multimodal transport and journey data to help tackle the last mile
problem in Helsinki and the West Midlands (see Transport in Finland case study).
Defining terms
Transport, mobility, MaaS: explaining the
differences
In this report, the term transport is used to describe the whole ecosystem
of moving people or goods from place to place11. Transport is made up of
physical infrastructure, as well as the organisations and individuals who own
or operate the different modes of transport, such as trains, buses, bikes
and cars.
On the other hand, mobility refers to an individual’s journey from location to
location through the transport system, either using one mode of transport,
such as walking, or multiple modes of transport, such as a car + train + bus
trip to a destination (also referred to as multimodal travel).
The term mobility is often used in the context of intelligent mobility, which
is using ‘emerging technologies to enable the smarter, greener and more
efficient movement of people and goods’12. More user-focused or newer
transport companies often call themselves mobility services.
Some tie these concepts together as ‘mobility as a service’, or MaaS.
MaaS is an evolving concept which aims to combine the benefits of
multimodal journey planners and smart ticketing to help people more easily
and efficiently plan and pay for multimodal journeys. MaaS could change
the way people consume public and private transport, enabling more
multimodal journeys, reducing the need for private car ownership and
improving end-to-end mobility.
End-to-end mobility relates to the ‘last mile’ problem, one of the systemic
transport challenge we reference in this report. The last mile describes the
movement of people to a final destination such as a home or place of work
(in reverse order, this is also a first mile problem). This challenge highlights
how hard it is to deliver physical transport infrastructure that meets the
needs of all travellers, each with individual destinations.
10
For examples from transport and other sectors, see: Ed Parkes and Mandy Costello
(2018), ‘Using open data for public services’,
https://theodi.org/article/patterns-for-using-open-data-in-the-delivery-of-public-services
11
This report focuses on the movement of people through transport systems, not goods.
12
https://ts.catapult.org.uk/intelligent-mobility/introduction
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 9
Companies are thinking about sharing data
Our interviews with transport companies showed a keen awareness of these
trends in the collection and use of data, as well as an appreciation of the
opportunities these trends create to develop services, earn the trust of customers
to support data sharing, as well as support transport ecosystems overall.
“ “We want to be able to share aggregated data
to help innovation.”
– Chris Perry, MaaS Global
Some stakeholders want to put the data they collect to wider use, and early data
sharing pilots are generating compelling results. For example, Waze’s data
partnership with TfL has reduced delays on London’s most congested route, the
Blackwall Tunnel, by using journey data to suggest drivers at risk of running out of
fuel reroute to a nearby petrol station13.
“
“Urban mobility is a universal asset.”
– Ed Parsons, Google
There is also recognition from the sector that it needs to find safe and secure ways
to make journey data it collects more open in ways that protect individual privacy
and support transport ecosystems. Many businesses think better collaboration
between organisations and clearer communication with people can help do this.
“ “We need to work out how to share in a way
that supports the wider system.”
– Chris Perry, MaaS Global
A growing number of businesses believe services that meet the needs of both
individuals and communities can foster and earn the trust of users, especially
when they tell users about these wider benefits. They observe that users share
journey data more when they know it has a communal benefit, because they value
the benefits that the journey data of others has given them.
13
Theo Chapple (2018), ‘Waze partnership: Reducing congestion at Blackwall Tunnel’,
https://blog.tfl.gov.uk/2018/01/11/waze-partnership-reducing-congestion-at-blackwall-tunn
el/
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 10
“
“Once you enable the GPS to share
anonymised data, you are getting better
information from us, and you are also
contributing this information to other people in
your community”
– Roy Bick, Moovit
Some mobility services explicitly communicate this message to customers,
sometimes as part of the user experience.
“
“People are more open when it is for the
greater good and if it is feeding back into their
own travel experience.”
– Max Stewart, Caution Your Blast
Open standards in transport
Open standards are an important part of our local, national and global data
infrastructure14. They help us to use, share and publish data in consistent and
easy-to-access ways, and can help to change markets. Data-enabled
innovation in transport depends on open standards. In order to work, most
mobility services need to link journey data with real time transport data. Even
companies like Waze and Moovit, who crowdsource data, need it.
Transport has long been a leader in open data, supporting the creation of
new mobility services around the world. For example, the General Transit
Feed Specification (GTFS) is an open standard for public transport schedules
originally started by Google and Portland, Oregon’s transit agency, TriMet15.
The Service Interface for Real Time Information (SIRI) is an open protocol to
exchange real time information about public transport services and vehicles16
(GTFS also has a real time extension). More recently, the General Bikeshare
Feed Specification (GBFS) developed by the North American Bike Share
Association shares real time data for docked bikeshare services17.
14
https://standards.theodi.org
15
Bibiana McHugh (2013), ‘Pioneering Open Data Standards: The GTFS Story’,
http://beyondtransparency.org/chapters/part-2/pioneering-open-data-standards-the-gtfs
-story
16
https://streamdata.io/blog/understanding-service-interface-real-time-information-siri-sta
ndard
17
Adam Russell (2016), ‘New bikeshare data standard opens doors for apps and analysis’,
https://mobilitylab.org/2016/05/04/new-bikeshare-data-standard-gbfs
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 11
SharedStreets is an open data standard for describing streets, and has been
developed specifically to support transport systems18.
Despite all this, open standards in transport are not yet universal. The UK is a
case in point. For timetables, TransXChange is not used by National Rail,
GTFS is not used by TfL19. For stations, the UK has no open address register
to support NaPTAN20. Fare data is very fragmented (for instance, the Rail
Delivery Group estimates there are currently 55 million different train fares21).
For ticketing, some find the ITSO open standard hard to implement22.
Friction in interoperability can increase costs for data users and limit
innovation, for instance by making it harder for new entrants to develop
complementary services to support existing transport. In particular, poor
interoperability limits innovation around MaaS.
Consumers and policy makers are also thinking
about sharing data
Companies’ attitudes to sharing in transport align with the findings of a recent ODI
survey of UK consumer attitudes to sharing personal data23. The survey of over
2000 consumers found that if a consumer trusts and knows an organisation, they
are more likely to share personal data about them with that organisation. 94% of
respondents said trust was important in deciding whether to share personal data.
64% said they would share personal data with an organisation they know,
compared to just 36% for an organisation they do not.
The survey also showed consumers are prepared to make trade-offs to share data
about them if it benefits them and society. While 15% said they would share data
to get insights about themselves, even more (26%) would share personal data to
help decide which public services to fund. 37% would share some data to help
academic research, while the most popular ‘data trade-off’ in the survey was
similarly for the greater good: nearly 50% said they would share personal data if it
helped develop new medicines and treatments. We use ‘greater good’ in this
report to refer to these types of trade offs, when an individual puts a community’s
needs ahead of their own, or sees the two as connected, even inseparable.
37% of people would share data about them
to advance academic research
Recent policy conversations about transport in the UK have focused on how to
share data more. The 2017 Bus Services Act requires private bus operators to
make fare, timetable and route data open, to give people access to more
information. The Transport Systems Catapult, working with the ODI and Deloitte,
identified ways for government to incentivise data sharing to support new mobility
services24. And a government inquiry is exploring ways to support MaaS.
18
https://www.sharedstreets.io
19
http://naptan.dft.gov.uk/transxchange
20
https://data.gov.uk/dataset/ff93ffc1-6656-47d8-9155-85ea0b8f2251/national-public-tran
sport-access-nodes-naptan
21
https://www.raildeliverygroup.com/media-centre/press-releases/2018/469773920-2018-
05-08.html
22
https://www.itso.org.uk
23
ODI (2018), ‘ODI survey reveals British consumer attitudes to sharing personal data’,
https://theodi.org/article/odi-survey-reveals-british-consumer-attitudes-to-sharing-personal
-data/
24
See ‘The Case for Government Involvement to Incentivise Data Sharing in the UK
Intelligent Mobility Sector’ briefing paper on https://ts.catapult.org.uk
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 12
Case studies for the future
Transport in Finland
Connecting personal data with open data infrastructure for
mobility-as-a-service
Finland’s groundbreaking Act on Transport Services, effective July 2018,
requires every transport group, public or private, to share an open API25. This
legislation aims to support a shift to transport as ‘mobility-as-a-service’,
meeting the travel needs of all Finnish citizens through single ticket,
end-to-end, multimodal journeys.
The Act builds on the success of Helsinki Region Infoshare (HRI), launched in
201126. HRI shares hundreds of open datasets to support academic research,
business innovation and public engagement. It also uses meet-ups,
hackathons and social media to drive these efforts. HRI has created many
benefits. For instance, the Helsinki Regional Transport Authority journey
planner open API has enabled many mobility services like:
WHIM – MaaS Global’s application to help people in Helsinki plan
and complete journeys easily and affordably without owning a car.
The service is also in beta in Birmingham in the UK in a collaboration
with Transport for West Midlands, National Express and Gett27.
TrafficSense – Aalto University’s research project collecting data on
users’ regular journeys and linking it with real time and multimodal
data on speeds, congestion and other data in order to give users
more personalised suggestions when planning travel routes28.
Blindsquare – a GPS application helping visually impaired people
travel independently. The voice-enabled app collects a user’s
location and other personal data in order to personalise and share
data such as street intersections, cafés and favourite locations29.
Alongside these initiatives, the government also supports MyData, an alliance
exploring ways to give people more control over data about them while
helping businesses use this data to develop services built on trust30.
This emerging national framework for transport data infrastructure –
integrating public and private services, supporting equity and ethics – could
suggest a blueprint for other countries, especially those exploring MaaS.
25
Ministry of Transport and Communications (2017), ‘Good and flexible transport services
through a new act’,
https://www.lvm.fi/en/-/good-and-flexible-transport-services-through-a-new-act-933165
26
Helsinki Region Infoshare (2018), ‘What is HRI?’,
https://hri.fi/en_gb/hri-service/what-is-hri
27
https://maas.global
28
Helsinki Region Infoshare (2017), ‘TrafficSense’,
https://hri.fi/data/en_GB/showcase/trafficsense
29
http://www.blindsquare.com
30
https://mydata.org/finland
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 13
Fostering trust in transport
Trust is a high stakes issue when using personal
data
Trust is a high stakes issue for transport companies who rely on their ability to
collect and use data about people, in particular journey data. While companies in
many sectors can collect location data during a person’s journey – from banks
and smartphone makers to telecom providers and social media – it is transport
companies whose relationships with customers are most defined by journey data,
and whose business models are most closely aligned with it. These stakes will
likely grow higher as journey data is collected and used more, as more people
think about how and why they share data, and as we gain more rights over data
about us.
In any sector using personal data, trust is strongest when people think, and feel,
that an organisation uses data about them in ways that meet both their needs for
a service and their expectations in relation to privacy and receiving a fair share of
the benefits for data about them. People become more willing to share data about
them and less likely to move to competitors. But when this trust goes, for instance
because data is used in a misleading way, we risk a backlash taking us into a
‘data wasteland’. In such cases, people may restrict their privacy settings, adding
bias into what might have otherwise been useful datasets. They may also
withdraw consent by changing service provider, by ceasing to use all providers of
a service, or stopping sharing personal data altogether, even for the greater good.
For transport – a universal asset which can benefit greatly from people’s
willingness to share journey data – such a loss of trust could be detrimental to
both transport providers and passengers.
Current levels of trust in how location data is being used across a range of sectors
should give us all pause for thought. One global consumer study by HERE
showed only one in five people feel they have full control over data about their
location, while 44% have shared it unintentionally and 76% feel stressed or
vulnerable doing so31. On the other hand, using journey data about a person to
provide a service for them at the right time and in the right place can create trust.
For example, interviewees reported that transport services who tell people about
disruptions affecting their journey can help them adjust routes in real time and
build trust in the service as a result.
“
“(If I) manage that disruption… that makes you
view me as a trusted partner.”
– Stuart Walker, Arriva
Our interviews with industry showed an interesting range of views about the role
of trust. Those who think of it as an abundant value to foster and grow take a
‘show and tell’ approach, aiming to communicate and demonstrate to users the
individual and communal benefits of sharing journey data. Others see trust more
in terms of scarcity, as a fixed asset to preserve, and commit to not sharing any
personal data (potentially without asking if and how users might be willing to share
it). A third group, increasingly a minority, and now forced to change by tighter data
31
‘Privacy and Location Data’ survey, h
ttps://www.here.com/en/company/
newsroom/press-releases/2018-05-03
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 14
regulation, do not proactively communicate about data at all, but may still share or
sell it, according to their terms and conditions.
How tighter data protection regulation impacts
transport businesses
As organisations have adopted new technologies to collect and use data in new
ways and at new scales, regulators have increasingly focused on how personal
data is collected, stored and used, and how people can have more control over it.
The most recent culmination of this trend is the GDPR, a European Union
regulation. It applies to any company processing data about EU citizens, even
companies based outside the EU. The GDPR creates some new rights around
data, but mostly builds on existing rights in existing data protection regulations.32
Right Description Did it exist What are its key changes
already? and elements?
The right to Organisations must give people Yes The information must be
be informed ‘fair processing information’ on concise, transparent and
the use of data about them, written in clear and plain
primarily through privacy notices language
The right of Gives people access to Yes Information must be available
access data an organisation holds free of charge for the first
about them copy, unless the request is
‘manifestly unfounded’, and
provided without undue delay,
generally within one month of
its receipt
The right to Lets a person obtain a copy of No This is a new right under the
data data about them to share with GDPR. It applies in specific
portability other services. situations where data was
processed automatically, and
the justification for processing
it was that the individual had
consented to that use, such as
loyalty card data.
Under the GDPR, companies must also apply the principles of data protection by
design and by default, minimising the data they collect, ensuring good governance
of personal data processing and demonstrating compliance. In situations deemed
high-risk, they must also carry out Data Protection Impact Assessments (DPIAs).
‘It is all about strengthening trust and updating the law to the 21st century.’
(Garreth Cameron, The Information Commissioner’s Office (ICO))
Fines for non-compliance under the GDPR could be substantial, with businesses
potentially liable for up to 4% of annual global turnover, or €20 million (whichever
is higher). But fines are not the focus of the regulation: ‘Thinking that GDPR is
about crippling financial punishment misses the point... it’s about putting the
consumer and citizen first.’ (Elizabeth Denham, UK Information Commissioner).
Most transport companies welcome the GDPR, as guidelines and benchmarks that are
necessary to manage long-term trends in the collection and use of personal data.
‘There have to be actual laws that providers are accountable to in order for
transport to remain sustainable’, says Eanna Lalor of Urbo.
Many also see it as an opportunity.
32
https://ico.org.uk/for-organisations/data-protection-reform/overview-of-the-gdpr
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 15
“
“GDPR helps you think about the customer
experience in more detail.”
– Stuart Walker, Arriva
The GDPR’s introduction of the right to data portability may also increase
innovation and competition, as people can use their right to ‘port’ data about them
to support the development of new services striving to meet their needs33. For
instance, a transport startup working to fill a gap in a specific location could
prompt people in that location to share historic journey data about them by
changing permission settings in a journey planner service they use.
Some businesses may try to minimise the risk of liabilities created by the GDPR by
collecting and using personal data less. In a sector increasingly reliant on real time
data from networks and users, this could reduce customer insight, create skewed
datasets and erode the quality of a company’s services over time. Trying to
minimise a legal risk could create a greater, more strategic business risk.
Beyond regulation – ethics, equity, engagement
The more an organisation needs personal data, the more it needs to change how it
manages risk. The complex issues raised by the commercial use of personal data
need to be assessed and managed proactively and consistently across a
company’s business model, user experience and communication. Risks which
might have been considered reputational, ancillary, or both in the past become
strategic and operational, changing how a company makes decisions. To manage
this, organisations need decision-making frameworks that support innovation and
are guided by three core principles of ethics, equity and engagement.
1. Organisations and people must use data ethically
To earn trust it is vital that organisations and people do not collect, use or
share data about people in unfair, discriminatory or misleading ways.
2. Organisations must support equitable access to, and benefits from,
data
This promotes fair competition, informed markets, and empowers people
as citizens, creators and consumers.
3. Organisations must engage with people to make data work for all
To help meet the needs of all communities, organisations must
communicate clearly and work collaboratively with them.
1. Using data ethically
Trust and ethics are inseparable – to earn trust, organisations must act ethically. A
decision can be lawful but still be unethical. Ethics is knowing the difference
between what you have a right to do and what is right to do (a line sometimes
attributed to US Supreme Court Justice Potter Stewart). The ethical issues raised
by the use of personal data are complex, and as a result some businesses are
exploring the rapidly emerging discipline of data ethics. These businesses see the
GDPR as a stepping stone to this.
33
Jack Hardinges and Gillian Whitworth (2018), ‘Will GDPR and data portability support
innovation?’, https://theodi.org/article/will-gdpr-and-data-portability-support-innovation
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 16
“
“We really see GDPR as being the baseline and
very much the minimum that has to be done…
We want to take it a step further not just with
ticking the boxes but by using our data in an
ethical manner.”
– Eanna Lalor, Urbo
In transport, the potential ethical issues raised by the use of personal data are
many and highly complex. In particular, more discussion is urgently needed about
what users expect from new mobility services, and the appropriate ways to use
and share data about them.
For instance, one study of rideshare services analysed cancellation rates and wait
times to suggest some drivers were discriminating against passengers based on
the perceived race or gender of the user profiles provided by rideshare services34.
The study also noted some female passengers were taken on longer trips. Should
a driver using their own car to offer trips via a rideshare service be given
information about race and gender before being given the option to cancel a trip?
And if drivers’ cancellations of a certain group make customer demand from that
group appear lower than it is, would this bias future innovation using rideshare
services’ journey datasets? For instance, would it encourage investment in
services for groups less discriminated against, because their customer demand
appears proportionally greater than it is? (This is as much a question of equity as
ethics, a reminder of how closely the two principles can relate).
There are also urgent ethical questions around the potential for data to be used in
misleading ways. Commentators like the Data Society have asked how the use of
crowdsourced traffic data can impact drivers in potentially misleading ways, for
instance if services recommend under-tested routes to gain more data about road
conditions35. This issue was taken to tragic extremes when a person’s death in
Niteroi, Brazil was attributed to misdirection caused by street names36. How can
services accessed while people are focused on activities such as driving ensure
they are not misleading? Even if a driver is not being misled, is it right to be given
the option to use an under-tested route in exchange for a free service?
2. Supporting equitable access to and benefits from data
Transport is a global need and a universal asset, so it is critical that data is used in
ways that meet the transport needs of everyone.
Some groups of people and modes of transport are now intensively monitored and
measured, and data about them can be used to strengthen the case for investment
in infrastructure, subsidies or innovation. On the other hand, because of the ways
journey data is collected, there are groups we have little or no data about who are
increasingly at risk of being excluded from such decision-making processes. These
34
Yanbo Ge, Christopher R. Knittel, Don MacKenzie, Stephen Zoepf (2016), ‘Racial and
Gender Discrimination in Transportation Network Companies’,
http://www.nber.org/papers/w22776
35
Alex Rosenblat and Tim Hwang (2016), ‘The Wisdom of the Captured’,
https://datasociety.net/output/the-wisdom-of-the-captured
36
British Academy (2018), ‘Data management and use: case studies of technologies and
governance’ report,
https://www.britac.ac.uk/sites/default/files/Data%20Governance%20-%20Case%20studie
s.pdf
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 17
‘data wastelands’ can in turn mean businesses miss opportunities to develop new
services or enter new markets.
Groups of people most at risk of being overlooked include those who:
● cannot access transport (for example wheelchair users, or people in rural
areas without transport)
● pay in cash for transport (including those who use cash-only taxis or buses)
● do not use monitored transport (for example walkers and runners who don’t
use fitness trackers and are outside connected city centres)
● are digitally and/or financially excluded (for example the very young or old)
● choose not to share data about themselves
This raises important questions about how data can harm, as well as help. For
instance, how will a lack of data on walkers and runners relative to other more
resource-intensive modes of transport impact the development of our transport and
cities over time? Will transport be skewed to deliver more resource-intensive modes
of transport because there is more data about users of those services? Will it be
harder to make a case for low carbon services and support active mobility?
Potholes offer another opportunity to think about the problem of data wastelands.
New mobility services for cyclists such as See.sense lights and Blubel navigators
collect data on potholes from users and share it with local authorities, either for free
or paid for as a service37. Will this make local authorities prioritise the needs of
those cyclists who can afford this equipment and share pothole data over other
people who use roads? Or will it help local authorities redirect the time and
resources saved through this new service into collecting other data to help make
better decisions about pothole repairs overall?
The example of people who choose not to share data about them raises policy
issues around personal choice which need to be explored. As new mobility services
integrate more with public transport, should there be options for people who do not
want data about them to be stored or used? (Some smart ticketing services offer
unregistered or anonymous versions which offer fewer benefits than cards which
are personalised to one user). It could even be a question of free will. Can you still
use transport without choosing to share data at all? If the answer is no, it is not a
genuine choice, so simply asking for consent could be considered misleading38.
Societal debates and democratic decisions help determine the choices that citizens
and consumers can make or, to put it another way, the sets of individual user needs
that will be met. If societies determine that they want consumers to have access to
a range of transport services that can be used without sharing personal data but
market forces do not lead to the emergence of these services then it may take
government intervention to make it happen.
37
https://seesense.cc and https://blubel.co
38
https://ico.org.uk/for-organisations/guide-to-the-general-data-protection-regulation-gdpr
/lawful-basis-for-processing/consent
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 18
Rural mobility: an overlooked opportunity?
In rural communities, a person’s independence is closely linked to their access
to transport, but to date new mobility services have done little to meet these
critical travel needs39. This lack of rural mobility services relative to cities is
effectively creating journey data wastelands in many areas. Left unchanged,
this risks limiting innovation in rural transport even further.
To date, new mobility services have focused most on cities, which offer more
of the infrastructure these services need to work, and are big, growing
markets. This makes sense for a startup, but once a service is established it is
important it extends outside urban areas where possible. This supports equity
in transport and helps businesses access markets to meet the needs of the
growing number of commuters who cross both rural and urban areas.
Journey data could play a critical role in improving rural transport. Historically,
one of the obstacles to developing rural transport was the cost of running it –
operations in less densely populated areas need to cover more miles for fewer
people. Journey data can help organisations understand demand more,
making services more feasible and viable.
Some organisations are exploring opportunities like this. For instance, in the
UK, Arriva operates smart buses in Kent. Ride’s smart bus service in Essex
and Suffolk is a FutureGov partnership with local authorities (see Smart Buses
in the UK case study). Community led groups like Western Dales Buses also
play a valuable role in meeting the needs of some local communities, but most
do not collect journey data.
Public sector initiatives may also help reduce rural data wastelands. For
example, a government investment to introduce smart ticketing across
England and Wales will give train operators the ability to collect more journey
data40. This could be used to support better investment and planning in rail
infrastructure and services. With appropriate anonymisation and consent, it
could be linked with other data to help understand the overarching travel
needs of people in rural areas beyond trains alone.
While some journey planners only cover cities, Google Maps can collect
journey data in rural areas, and offers suggested routes across large distances,
including some public transport. Could such rural journey data be aggregated
to develop overarching multimodal views of rural travel needs?
MaaS services could also play a valuable role in meeting rural travel needs. For
example, Navigogo is working with 16-25 year olds to design and test a new
service in Dundee and North East Fife, Scotland41. This collaborative initiative
aims to give young people personalised travel data across multiple modes of
transport, such as walking, buses, trains, taxis, car clubs and bike schemes,
and offer concessions and discounts in a single payment process.
39
See Transport Systems Catapult research on rural transport at
https://ts.catapult.org.uk/readyforinnovation
40
https://www.gov.uk/government/news/government-plans-80-million-smart-ticketing-rail-
revolution
41
https://www.the-espgroup.com/project/navigogo
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 19
3. Engaging to make data work for all
Organisations who encourage people to share data about them must earn trust by
making good on their word. They should do this by ensuring their commercial and
communications strategies for how they collect and use personal data match up.
Companies who let gaps grow between the two, or who are opaque about the
benefits of personal data to the business itself, can create strategic risks around
trust that business models using personal data cannot afford.
People currently feel poorly informed about how personal data is being used. A
DotEveryone survey of people in the UK found that 70% recognise that a service
they share personal data with can use it to target advertising, but only 56% know
that the data can be sold to other companies. And while 94% of people said it is
important to be able to choose how much data they share with companies and
know how that data is used, half of them do not know how to do this42.
For location data specifically, the findings are even more troubling. The global
survey by HERE mentioned earlier in this report found that 80% of consumers are
not clear about why location data about them is needed most of the time, what it
is used for, where it is stored, and which third parties have access to it.
To address these issues and risks, businesses are encouraged to think more
holistically about engagement, exploring ways to explain their business models,
how data about people is collected, used and stored, and the benefits and
trade-offs involved in clear and relevant ways. People can understand the
fundamentals of how a system works when given the relevant information
(systems thinker Donella Meadows used the example of a bathtub to show this43).
This sort of communication can help users make informed choices, either as
individuals, or with the support of third parties like data representatives44.
Figure 2. A seat on the Central Line, Transport for London
42
See http://attitudes.doteveryone.org.uk and http://understanding.doteveryone.org.uk
43
http://donellameadows.org/wp-content/userfiles/bathtubs101.pdf
44
Tom Steinberg (2018), ‘Personal Data Representatives: An Idea’,
https://medium.com/@TomSteinberg/personal-data-representatives-an-idea-d2274d1af69b
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 20
Case studies for the future
TOC Ability
Using personal data to improve equality in transport in the UK
The spending power of households with people with disabilities in the UK is
£250 billion, but businesses often miss opportunities to deliver services for
this group45. 9% of adults have difficulty with mobility, but represent only 5%
of train users, put off by poor accessibility at stations and during journeys46.
To address this market failure, Stratford-upon-Avon intelligent mobility
startup Enable ID is piloting TOC Ability47. The startup is using a collaborative
and iterative process to design and test the service. It involves a panel of
people with disabilities, Transport for London, Arriva, Loughborough
University (providing customer experience expertise), the University of
Surrey’s cyber security experts, and other groups.
Tests will run at sites including Wembley Central and Harrow & Wealdstone
(London Underground), Willesden Junction and Carpenders Park (London
Overground), and other stations outside London.
In particular, the pilot will test the feasibility of a data hub which delivers
better rail services to passengers with accessibility needs by sharing real time
data about them with Train Operating Companies (TOCs), journey planners,
bus and taxi firms, and even non-transport services like food retailers and
chemists.
To provide these services viably, the hub will help TOCs track and balance
the profitability of providing these services alongside the related costs of
running them. To support trust in the service, the hub will include a consent
management system to give users control over data about them, including
how it is used and shared.
How could an approach such as the one being used by TOC Ability be
applied to test ways to increase accessibility and improve user experience for
other groups with specific travel needs?
45
BBC News (2017), ‘New voice for disabled consumers’,
http://www.bbc.co.uk/news/business-39028509
46
The Guardian (2016), ‘Disabled people are still being treated as second class on public
transport’,
https://www.theguardian.com/public-leaders-network/2016/sep/27/disabled-people-secon
d-class-citizens-public-transport
47
https://enableid.com/tocability
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 21
How transport uses data
This section looks at some major land transport for people, outlining how they
collect and use data, particularly journey data. We do not aim to give a
comprehensive global view covering all types of transport – instead we hope to
give an overview of some key stakeholders in interconnected transport systems.
We have been struck by two observations while doing this. Firstly, categories in
this innovative sector are not clear cut. For this report, we look at modes of
transport (often thought of as operations) and new ways of accessing transport
(often called services), but the two constantly overlap. For example, buses are a
mode of transport that may also use digital services to give people new ways to
access their operations. A bikeshare service is a new way of accessing transport
that runs physical operations to enable that new form of access. A journey planner
is a new service for sharing data about multiple modes of transport that can also
use that data to develop physical operations in different modes of transport.
Our second observation is that new mobility services tend to collect and use
journey data far more than established public transport operations. If this trend
continues it could create imbalances in our transport systems, perhaps even
oligopolies or monopolies, in the form of journey data for specific locations. There
are private companies and public services already exploring ways, often
collaboratively, to address these imbalances and make data more open, and these
efforts must continue to ensure transport systems serve the needs of everyone.
Trains
Historically, intercity rail operators collected little personal data, focusing instead
on the technologies and data required to validate payments and prevent ticket
fraud. However, as new mobility services and other sectors show the benefits of
personal data, train operators are exploring ways to collect and use it.
In a growing numbers of places, such as Europe, North America and India, train
operators are collecting personal data by asking people to create accounts before
buying tickets online; third party retailers such as Trainline and GoEuro can do the
same. An increasing number of train operators are also moving to mobile tickets
and smart ticketing. Many train operators can also collect personal data through
wifi logins offered onboard (wifi providers can collect it too). All this data can be
used to improve core operations, customer service and complementary services.
Many train operators still don’t know when specific people board specific trains,
because cash payments and paper tickets transferable between people or trains
remain common. This is changing as countries like Finland, Holland, Germany,
France and the UK use more mobile tickets and smart tickets. Ticket barriers or
onboard staff can scan these tickets to collect real time journey data linked to
individuals. Even in these situations, it can still be hard to complete journey data,
as tickets are not always scanned at the end of journeys, and passengers may get
off at a station not reflected in their original ticket purchase.
The UK government’s investment in rail smart ticketing across England and Wales
mentioned earlier in this report will give train operators the ability to collect more
accurate journey data and may encourage them to think more about how they use
it and other personal data. Operators could find ways to increase engagement
with customers and address service gaps, for instance by combining appropriately
anonymised journey data with other data to understand overcrowding beyond
simple counts. Done ethically, this could help reduce overcrowding and pressure
on infrastructure by helping train operators understand what trade-offs individuals
might make to adjust when they travel to help the system and themselves.
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 22
Buses
Buses, like trains, have historically collected little data and are working to change
this. Bus services and their oversight are generally more fragmented than trains,
particularly outside major cities. This fragmentation can make it harder for individual
organisations to collect data at the scales needed to deliver benefits, and reduce
investment in data as a result. Access to more open data infrastructure, perhaps
including safely and securely shared journey data, could help address this issue.
The UK’s highly fragmented and shrinking bus network is a case in point48. There is
currently no consistent way for private sector operators of buses to share
timetables and fare data with public sector operators of bus lanes and bus stops.
This makes it hard for people to get the data they need, potentially making them
less likely to trust and use buses. Bus operators’ smart cards, which can collect
customer data, are still relatively new and equally fragmented, and it is unclear how
much they are used. Even when they are used, accurate journey data can still be
difficult to collect, as passengers may only scan or tap a smart ticket when they
board a bus, not when they get off. As a result of all this, there is little innovation
using data, limiting the public and private sectors’ ability to understand and serve
people’s needs.
More open transport data and engagement with local communities is not a magic
solution but it could help overcome these obstacles and there are signs of positive
change in these directions – the Bus Services Act requires UK bus operators to
provide open data by 2020, and some companies are now testing ‘smart buses’.
Case studies for the future
Smart buses in the UK
Using personal data to develop bus services for underserved locations
Formalised carpool-type services like Lyft Shuttle, Uber Pool and BlaBlaCar
show people will combine sharing data about them with businesses and
sharing small private spaces with drivers and passengers they don’t know if it
helps them make necessary journeys like work commutes or getting home
late at night. Now established operators, new entrants and local councils are
developing similar services for bigger groups in cities like Manchester,
London and Bristol, and in rural areas like Kent, Suffolk and Essex.
Often called smart buses or shared transport, these minibuses and minivans
offer routes with pre-determined or on-demand pick-up points and
destinations. In particular, these flexible, adaptable vehicles could help meet
rural travel needs, improving equity in transport.
The UK’s first smart bus pilot was launched by public transport operator
Arriva in Sittingbourne, Kent. ArrivaClick collects users’ journey plans via a
smartphone application and then maps a ‘corner-to-corner’ minibus route
guaranteeing seats for people, including wheelchair users. The year-long pilot
reported promising results: 12% of local residents downloaded the
48
BBC News (2018), ‘Britain's bus coverage hits 28-year low’,
http://www.bbc.co.uk/news/uk-england-42749973
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 23
application, and 52% of users switched from private car transport as a result.
49
In Essex and Suffolk, Ride’s on-demand bus service was started by
FutureGov in partnership with Essex and Suffolk County Councils and the
support of Transport Systems Catapult. In urban areas, Paris public transport
provider RATP’s international arm, RATP Dev, developed Slide Bristol to
deliver shared ride-to-work services in Bristol. San Francisco mobility startup
Chariot, bought by Ford in 2016, is running a TfL-approved pilot with fixed
pick-up points in parts of south London50.
Citymapper is an example of a journey planner using personal data to expand
into transport operations. It began as a London journey planner using TfL
open data, and Google and Apple maps and platforms. It combined this data
with users’ journey data to build an algorithm called ‘Simcity’. Simcity
analysed demand for routes, mapped it over existing public transport routes
to identify gaps, and forecast how new routes might change how people
travelled. A smart bus pilot tested these insights and collected customer
feedback. A late-night Smart Ride service launched as a result, and
Citymapper shares open data on it51. Now TfL, Google and Apple display
Citymapper’s service in their applications, as well as vice versa.
Some people argue that bus networks must remain fundamentally public to
ensure infrastructure is built for all, including people less able to pay or
persuade. Others say that the private sector should play a role developing
business models to fill public service gaps.
Cars
Since the 1990s, most cars have automatically collected location, speed, emissions
and other data via event data recorders and diagnostics systems installed directly in
cars. This data could usually only be retrieved via physical access to the car. It was
used by car makers to measure and improve safety and performance in car models,
and by mechanics to find and fix issues in specific cars. Car owners can now also
connect devices such as smartphones or telematics ‘black boxes’ to their cars.
These devices can collect and share journey data and other data with third parties
who use it to provide services such as navigation, insurance and entertainment.
As a result, cars collect large amounts of data – 25Gb of data an hour, likely to grow
to 4 terabytes with self-driving cars52. This includes a lot of personal data linked to
many people and interested parties.53 (Connected cars help demonstrate why no
one owns data54). In future, autonomous cars could even collect personal data that
is emotional and physical, like heart rate, facial expression and voice analysis.
As cars become more autonomous, people will be able to rent them as they need
them for specific journeys between places, improving end-to-end mobility. This
49
See https://www.arrivabus.co.uk/arrivaclick and https://www.intelligenttransport.com/
transport-news/65899/arrivas-on-demand-public-transport-service-a-success
50
See https://www.bookyourride.co.uk, http://www.slidebristol.com and
https://www.chariot.com/cities/london
51
Citymapper (2018), ‘Good Bus - What happens when an app company runs a bus’,
https://medium.com/citymapper/good-bus-part-1-3-77d65e6f8ce3
52
MIT Alumni (2016), ‘Cars, Data, and Internet of Things’,
https://alum.mit.edu/slice/cars-data-and-internet-things and Networked World (2016), ‘Just
one autonomous car will use 4,000 GB of data/day’,
https://www.networkworld.com/article/3147892/internet/one-autonomous-car-will-use-400
0-gb-of-dataday.html
53
For more detail, see the Future of Privacy Forum’s guide ‘Personal Data in Your Car’,
https://fpf.org/wp-content/uploads/2017/01/consumerguide.pdf
54
UC Hastings (2018), ‘No One Owns Data’, https://ssrn.com/abstract=3123957
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 24
rental model will mean cars are more likely to be owned by governments or
businesses than people. It will become more important that people know how data
about them is collected and used, and that they trust the organisations that do this.
Making data from cars as open as possible could help solve many shared
problems. Perhaps most compellingly, it could help reduce road deaths. 94% of car
crashes are due to human error55. Data from cars in accidents could be anonymised
and shared to help researchers and public authorities analyse and, where possible,
fix factors causing them. This data could also be linked with other open data on
road accidents, such as UK road safety data or New York motor vehicle collision
data56. It could also help car makers develop safer software for automated cars
more quickly.
Accidents could also be reduced by sharing data including journey data between
cars or with transport infrastructure. The EU’s C-Roads project and the Nevada
Center for Advanced Mobility’s partnership with Nexar are testing ways to create
these vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) networks57. They
could also use this data to reduce traffic and improve real time road management,
as well as help make roads more suitable for self-driving cars.
When an accident occurs, telematics insurers can already use a customer’s location
data to get help there quickly. Telematics insurers collect data from boxes installed
in customers’ cars or via their smartphones and use the data to personalise
insurance premiums. This can make insurance more affordable for drivers in higher
risk groups, such as young drivers, if they are better-than-average. Businesses can
also use this data to identify insurance fraud. Many telematics algorithms are not
transparent in how they track and measure driving patterns, and should explain to
customers how data about them is being used and the benefits it can create. More
clarity is also needed on how the data they collect can be shared58.
There are many other ways cars sharing data could help solve problems. Flourish is
exploring ways both connected and autonomous vehicles might help address the
mobility needs of older people in the UK59. Connected cars could share anonymised
location data to help maintain open mapping data which benefits all drivers, or help
make open address registers more accurate by sharing missing data with them.
Given the amount of personal data cars can collect, and the number of
organisations who can use it, it is critical privacy is considered and protected at all
times. Major car makers like Volvo, Ford, Nissan and BMW have signed up to the
Privacy Principles for Vehicle Technologies and Services60. One name not
mentioned is Tesla, who some commentators have criticised for how it shares data
about people61. As cars use more data themselves, car hacking will also become a
growing risk and increase organisations’ focus on the security of all data62.
55
https://crashstats.nhtsa.dot.gov
56
See https://data.gov.uk and https://opendata.cityofnewyork.us
57
See https://www.c-roads.eu/platform and https://www.getnexar.com/nevada
58
The Telegraph (2014), ‘Insurers admit 'black box' data may be handed to police’,
https://www.telegraph.co.uk/finance/personalfinance/insurance/motorinsurance/11217690/
Insurers-admit-black-box-data-may-be-handed-to-police.html
59
Shergold, Wilson & Parkhurst, G. (2016), ‘The mobility of older people, and the future role
of Connected Autonomous Vehicles’, http://www.flourishmobility.com/publications
60
https://autoalliance.org/connected-cars/automotive-privacy
61
The Guardian (2017), ‘The customer is always wrong: Tesla lets out self-driving car data –
when it suits’,
https://www.theguardian.com/technology/2017/apr/03/the-customer-is-always-wrong-tesla
-lets-out-self-driving-car-data-when-it-suits
62
The Register (2018), ‘Newsflash: Car cyber-security still sucks’,
https://www.theregister.co.uk/2018/01/26/car_hacking_wireless
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 25
Ridesharing
While most traditional taxis remain data-poor, ridesharing services are built on data.
Companies like Uber and Lyft collect journey data, user ratings, contact and bank
details for both passengers and drivers, to connect them on-demand and manage
payments between them. These mobility services can also link this data with other
data to understand customers more and develop adjacent services, such as food
deliveries63. They may also sell this data to third parties.
These popular services have met travel needs and created jobs in cities around the
world. They have helped people meet their last mile travel needs, and made things
easier for people who can’t access a car or public transport, or whose travel needs
cannot be met by public transport. All this benefits people who can afford to pay for
such services, but it does not assist people who cannot. It is critical that the impact
of these new services does not reduce equal access to transport over time.
These fast-growing mobility services have also highlighted the need to make
decisions about data which foster and earn trust at all times. Some commentators
are concerned about how rideshare services store the large amounts of personal
data they collect, and Uber is working to rebuild trust after concealing data hacks
last year64. Other commentators are concerned about the demands these services
put on transport infrastructure65, and debate their net effect on economies overall66.
As a result, there is growing pressure for rideshare businesses to engage with us
more, and make their journey data more open – San Francisco even compelled it67.
Now, Uber Movement shares aggregated, anonymised data for Paris, Sydney and
other cities to help policy makers understand and meet urban mobility needs68.
Several ridesharing services have also signed Zipcar co-founder Robin Chase’s
Shared Mobility Principles for Livable Cities, supporting collaboration and open data
to develop transport in the interests of all69. Concepts such as data observatories
may be one way to start turning these commitments into action and provide insights
that help local areas make more informed decisions70.
If rideshare services continue to grow, the imperative to share insights such as
usage data will grow with it. The speed of adoption of these new services,
combined with ongoing poor data collection in some public transport, could create
journey data monopolies, for instance in cities where fragmented bus services not
collecting data compete with a single ridesharing service. This asymmetry between
public and private operators could limit the ability of public services to invest and
innovate over time, and be exacerbated as public transport data continues to
become more open. For example, the Bus Services Act in the UK mandates smart
buses with a bus operating license to provide open data by 2020, whereas
rideshare services which can effectively compete with them but operate under a
private hire license are not required to do the same.
63
https://www.ubereats.com
64
Bloomberg (2017), ‘Uber Paid Hackers to Delete Stolen Data on 57 Million People’,
https://www.bloomberg.com/news/articles/2017-11-21/uber-concealed-cyberattack-that-e
xposed-57-million-people-s-data
65
CItyLab (2017), ‘The Ride-Hailing Effect: More Cars, More Trips, More Miles’,
https://www.citylab.com/transportation/2017/10/the-ride-hailing-effect-more-cars-more-tri
ps-more-miles/542592/
66
Thor Berger, Chinchih Chen & Carl Benedikt Frey (2017), ‘Drivers of Disruption?
Estimating the Uber Effect’,
https://www.oxfordmartin.ox.ac.uk/downloads/academic/Uber_Drivers_of_Disruption.pdf
67
BBC News (2017), ‘Uber and Lyft ordered to share data with San Francisco’,
http://www.bbc.co.uk/news/technology-40172017
68
https://movement.uber.com
69
https://www.sharedmobilityprinciples.org
70
Open Data Institute (2018), ‘Understanding the impacts of peer-to-peer accommodation,
the role of data and data observatories (report)’,
https://theodi.org/article/understanding-the-impacts-of-peer-to-peer-accommodation-the-r
ole-of-data-and-data-observatories-report
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 26
Bikesharing
Bikesharing has great potential to support the travel needs of urban populations
and has been embraced by people in cities around the world71. In the UK alone, 20
bikeshare operators have launched in 10 years. These services let people who can
cycle travel more flexibly and actively while reducing congestion and pollution.
Docked and dockless services like Urbo, obike and ofo72 use applications and bank
cards to collect account and bank details as well as some journey data. Operators
collect GPS data directly from bikes to locate and move them to match customer
demand. Some also collect journey data from users, via a smartphone or connected
device linking them to the bike, sharing that data with the user, and sometimes
selling it to third parties.
Some bikeshare services are exploring ways of sharing data that help urban
planners make better planning and investment decisions. For instance, Urbo aims
to work with local authorities whenever they enter a new city. Services increasingly
share usage data with local authorities, and it is now often a condition for dockless
services seeking local authority consent.
“
“We always open a line of dialogue with the
government and local council… it’s very much
a collaborative relationship”
– Eanna Lalor, Urbo
Some bikeshare services are working to increase accessibility for excluded groups.
For example, Germany’s Nextbike service can be accessed via an RFID card
bought in shops, to give access to people with no smartphone or bank card.
Hourbike offers electric bikes in some locations, to give access to those less able to
cycle.
Open standards and industry bodies are helping these services share more data
and in ethical and equitable ways. The General Bikeshare Feed Specification,
started in the US, is a new operator-led standard for sharing real time data on bike
availability from different services. It increases accessibility to bikeshare services by
helping journey planners include them in their applications, and improves urban
planners’ decision-making by giving them aggregated mobility patterns.
Industry groups are also emerging to help these new services engage with key
stakeholders in the public and private sectors, including large employers and
community groups. For example, Bikeplus, the representative body for bikeshare
services in the UK, analyses and shares data and works to ensure social and
environmental benefits are maximised and evenly spread73.
Journey planners
Journey planners are growing in popularity around the world, transforming how we
access and analyse transport ecosystems. Applications like Google Maps, Moovit
and Citymapper, typically accessed via smartphones, combine transport data from
many modes of transport with maps to help people choose and monitor trips from
and to precise locations. This makes them ‘end-to-end’ and ‘multimodal’ services.
71
http://bikes.oobrien.com
72
See http://www.myurbo.com, https://www.o.bike and https://www.ofo.com
73
https://www.carplusbikeplus.org.uk/projects/bikeplus
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 27
Users can give these applications permission to collect journey data, anonymously
or linked with other data about them. As well as helping to plan journeys, it can help
users track other things, such as distances travelled, calories burned or money
saved. It can also help improve the accuracy of the application’s data for all users,
and help the application itself develop other services, such as data analytics for
public and private mobility providers, or new transport operations.
Moovit crowdsources data from users to plot public transport stations and routes in
cities with little or no open transport data. This helps people in those cities access
and engage with public transport more easily, and those cities can license this
information to improve their data infrastructure74. Citymapper encourages cities to
partner with it to develop open data, and also uses the journey data it collects to
identify and fill gaps in transport services (see Smart Buses in the UK case study).
The benefit of open data infrastructure to these companies has influenced how they
communicate. They strive to explain the different ways and value of sharing data to
users, building communities and making the ‘greater good’ part of their service and
ethos. This can increase people’s engagement as they are motivated to help
improve a system for the benefit of others as well as themselves. Companies with
this type of engagement and business models that back it up foster and earn trust.
Journey planners have real time network views of how people move through
transport ecosystems. These overarching views could be of great value to public
services that often provided the open transport data which helped these private
services start. For instance, Moovit’s public transit index shares anonymised and
aggregated data on how people experience public transport in cities around the
world, from wait times, to number of changes in a single journey and the average
length of a commute75.
Further exploration of such sharing is encouraged, particularly where it could help
public transport fill service gaps that private companies have identified but can’t or
won’t fill themselves. If MaaS develops to position journey planners as centralised
gateways into our transport ecosystems, it will also be important their displays
continue to enable innovation by supporting awareness of and enabling access to
new mobility services as they emerge, using open APIs.
Case studies for the future
Digital Matatus
Collecting journey data for open data infrastructure in informal transport
systems
Nairobi’s Digital Matatus was the first group to develop open data
infrastructure for the informal transport systems common in cities across
Africa, Asia and Latin America76. The University of Nairobi’s C4D Lab, MIT’s
Civic Data Design Lab and Boston design agency Groupshot collaborated to
map the city’s network of matatus, privately-owned minibuses used by most
Nairobi residents77.
To do this, university students collected journey data using mobile phones
and GPS technology. They identified routes and stops, as well as fares, all of
which can vary journey to journey. To standardise and open the data, they
74
https://www.solutions.moovit.com
75
https://www.moovitapp.com/insights
76
http://www.digitalmatatus.com
77
‘The Digital Matatu Project’, https://doi.org/10.1016/j.jtrangeo.2015.10.005
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 28
asked Google to help them modify the GTFS open standard, built for formal
transport systems with fixed routes and stops, to fit informal transport
systems with a lot of variable data.
The group used the data to publish route maps in both digital and paper form.
Because the data was standardised using GTFS, Google and OpenStreetMap
could also include the routes in their maps. Making the routes visible and
accessible helped passengers save time by finding routes to meet their travel
needs easily. It also changed travel experiences, for instance helping women
feel safer at night78.
Making the data open and easy to use allowed others to create further
benefits. Matatu drivers could see underserved areas and try new routes.
Third-party developers used the data to create new applications such as
ma3route, which crowdsources location and other data to provide updates on
traffic and driving conditions79. New mobility services like this also expanded
the city’s transport ecosystem to include its technology community. The data
can also support transport planning to help combat the estimated $1 billion
Nairobi losses a year to wasted gridlock hours80.
Similar initiatives are also running in Manila, Accra, Dhaka and Mexico City81.
Linking resources and skills across cities, Digital Transport 4 Africa is a
collaborative initiative using open data to improve urban transport throughout
Africa82. Could learnings from these initiatives also be applied to cities with
more formal public transport where frequent journeys via new mobility
services may be evolving into informal transport systems?
Smart ticketing
Ticketing is a core function of any transport system and has changed significantly in
recent years. Many public transport operators have moved from paper to magnetic
tickets, then barcodes, and now mobile ticketing, smart cards and contactless
payments. These new technologies have simplified travel for people and allowed
transport operators to collect journey data and use it to make better planning and
investment decisions, as well as lower the cost of their ticketing solutions.
Smart cards using RFID and NFC technology let passengers in cities such as Hong
Kong, Sydney, Stockholm and London access services across multiple modes of
transport with a single card83. In London, people can use contactless bank cards.
Other cities, such as Oslo and Helsinki, focus on mobile ticketing and smartphones.
While many rural communities do not yet have access to smart ticketing, some
countries have moved to a national ticketing platform that can be used in urban and
rural area and for all modes of public transport. This solution is seen, for example, in
78
Wired (2015), ‘How Nairobi Got its Ad-hoc Bus System on Google Maps’,
https://www.wired.com/2015/08/nairobi-got-ad-hoc-bus-system-google-maps
79
https://www.ma3route.com
80
Reuters (2015), ‘Nairobi's billion dollar traffic blues’,
https://www.reuters.com/video/2015/03/23/nairobis-billion-dollar-traffic-blues?videoId=363
596944
81
See https://sakay.ph, http://data.afd.fr/accramobility, Albert Ching (2012), ‘A
User-Flocksourced Bus Experiment in Dhaka: New Data Collection Technique with
Smartphones’, http://web.mit.edu/czegras/
www/Flocksource_JUT.pdf and Quartz (2016), ‘Mexico City is attempting to map its more
than 1,000 unwieldy bus routes with a crowdsourcing app’,
https://qz.com/598895/mexico-city-is-attempting-to-map-
its-more-than-1000-unwieldy-bus-routes-with-a-crowdsourcing-app/
82
http://digitaltransport4africa.org
83
https://en.wikipedia.org/wiki/List_of_smart_cards
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 29
The Netherlands via OV-Chipkaart and in Denmark via Rejsekort. These systems are
spreading the benefits of smart ticketing more equitably and could provide rich
journey datasets to help policy makers make better planning and investment
decisions at a national level. Some systems, such as OV-Chipkaart, let users
choose between an anonymous version of the card, which does not store personal
data, and a personal version which does, and offers benefits as a result84.
Smart cards dedicated to specific groups of people can also help public services
ensure their travel needs are met. In the UK, West Yorkshire used journey data
collected on Senior Pass smart cards to see how people over 60 use buses there85.
Smart ticketing also means smartphone makers and banks are now part of our
transport infrastructure. Smartphones have helped new mobility services to emerge
and enabled the collection and sharing of real time data. Smartphone makers can
already collect large amounts of location data from users, storing or using it in
different ways. If bank cards become a common form of ticketing, they can also
build large journey datasets and link it with other data. It is paramount that
organisations with such large amounts of personal data use it ethically and
equitably at all times, and in ways users can understand.
84
https://www.ov-chipkaart.nl/purchase-an-ov-chipkaart/anonymous-ov-chipkaart.htm
85
Data Mill North (2017), ‘English National Concessionary Travel Scheme boarding data for
West Yorkshire’,
https://datamillnorth.org/dataset/english-national-concessionary-travel-scheme-boarding-d
ata-for-west-yorkshire
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 30
A framework for the future
Data-enabled transport must combine innovation with ethics in
open ways that build trust
The future value of transport to our societies and economies is much more than
financial. Quantifying the potential of data-enabled transport can yield some striking
results: as noted earlier, intelligent mobility may be worth £900bn by 2025; not
sharing transport data may cost an economy like the UK £15bn. But it is how
transport can change our lives and how we feel, its qualitative value, and
businesses’ reliance on access to data about us to make this possible, that makes
the need for an approach to data in transport that combines openness and
innovation with ethics and trust so urgent and great.
As shown earlier, personal data like journey data can help transport organisations
be more efficient by understanding and meeting travel needs in more targeted
ways. This can change our quality of life – better transport can help us live longer,
sleep more, even divorce less86. It can also affect how we feel – when transport
gives us a greater sense of control it can increase our feelings of independence or
happiness87.
Businesses can build sustainable models delivering these services by being more
open, as well as fostering and earning the trust of people who share data about
them. As services like journey planners and bikesharing show, businesses can in
turn make the data infrastructure they build more open to support the public
transport infrastructure and interoperability they need to continue to operate and
innovate. This can help make transport better for everyone and gain people’s trust
while also raising awareness of a business’s services with potential customers.
This sharing in all directions and the innovation and trust it can foster is core to how
transport works. Public and private transport services operate interdependently as a
person’s single journey will often involve several modes of transport. New mobility
services innovate using data made openly accessible by public services which have
a special mandate to serve the public good. New businesses like ridesharing and
bikesharing find innovative ways to share resources, and can help public services
and accredited researchers better understand people’s travel needs. In areas with
less public transport, private transport services can even help create foundations for
data infrastructure which leads to new physical infrastructure (see the Digital
Matatus case study). In transport, a more open approach can create value for all.
Contrast this with the HERE survey noted earlier which found that while nearly 70%
of us will share location data if we feel in control, only 20% currently feel that way,
and 80% of us feel nervous or vulnerable. A loss of trust is bad for any business,
but for consumer-facing groups using personal data, the risk can be far more than
reputational and the cost material. For example, research suggests 5% of Facebook
users deleted their account in just the first few weeks of the Cambridge Analytica
crisis88. As new services become more commoditised, and as data about people
becomes more portable, these risks and costs for businesses using personal data
86
Lifehacker (2012), Long Commutes Are Sucking the Life Out of You: Shortening Yours by
20 Minutes Could Save Your Health’,
https://lifehacker.com/5908879/long-commutes-may-be-killing-us-shorten-yours-by-20-
minutes-to-save-your-health
87
See https://ts.catapult.org.uk/readyforinnovation and The Happiness Index (2017), ‘How
your morning commute affects your happiness and wellbeing’,
https://the-happiness-index.com/does-your-morning-commute-affect-your-happiness-and-
wellbeing/
88
Campaign (2018), ‘One in 20 Brits delete Facebook accounts after the Cambridge
Analytica scandal’,
https://www.campaignlive.co.uk/article/one-20-brits-delete-facebook-accounts-cambridge
-analytica-scandal/1460836
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 31
are likely to go up, not down.
Some may assume that the compelling nature of their service makes a loss of trust
affordable in the short term, others that the cost of a loss of trust is low anyway
because customers get hooked to convenience. Even if both of these assumptions
are true when tested, shareholders with long-term investment targets and ESG
(environmental, social and governance) mandates are unlikely to tolerate them. And
if customers and shareholders do nothing, regulators will intervene on their behalf.
Similarly, as personal data becomes part of business models, commercial
imperatives and social responsibilities are harder to separate. An ‘us versus them’
approach can erode trust when exposed89. And for a universal need such as
transport, how personal data is used and shared affects employees and investors
just as it does customers. A large-scale loss of trust can create losses for all,
through increased risk-aversion, s ocial instability and even violence90.
For these reasons, we urge transport businesses who use personal data to become
more open and build trust now. Doing so will both reduce business risks and
increase growth opportunities.
Recommendations
1. Use existing open transport data to get a competitive edge
This additional insight at low cost can support internal decision-making, improve
your operations and help your organisation engage with and lead sector change.
Actions could include:
● using third-party sources of data in your internal processes
● combining your data with other data from public services, other
businesses or community groups
● engaging with existing external groups enabling open data for transport.
Businesses hoping to gain a competitive edge in this way can explore open
transport data provided by governments, public transport operators and open
data groups91, or on platforms like GitHub. Examples of existing sector initiatives
at the national level include the UK’s Transport Systems Catapult and Belgium’s
Intelligent Transport Systems. Broader alliances include Digital Transport 4 Africa
and Open Transport Net92.
2. Turn your transport data into infrastructure that can benefit all
This can let other organisations integrate your transport services with theirs, help
people access your services more easily, and inform decision-making for public
transport infrastructure. It can also help your teams collaborate more internally.
89
TechCrunch (2018), ‘Facebook retracted Zuckerberg’s messages from recipients’,
https://techcrunch.com/2018/04/05/zuckerberg-deleted-messages
90
Pew Research Centre (2017), ‘The Fate of Online Trust in the Next Decade’,
http://www.pewinternet.org/2017/08/10/the-fate-of-online-trust-in-the-next-decade
91
https://github.com/theodi/shared/wiki/Finding-Open-Data
92
See https://ts.catapult.org.uk, http://its.be, http://www.digitaltransport4africa.org and
http://www.opentnet.eu
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 32
For instance:
● Adopt existing open standards and create open APIs that help your
services integrate into journey planners and public planning processes
● Find other groups with a common or adjacent need that can be addressed
using data and partner with them to build new data infrastructure
● Identify data you manage anyway that could be shared or made open
safely and at low cost to benefit customers or public services
See how TransportAPI and Navitia.io use open APIs in transport93, how Thomson
Reuters has built open infrastructure for and with its customers94, and an idea for
how location data could be shared with local authorities to improve air quality95.
CRITICAL QUESTIONS
What does ‘as open as possible’ mean for journey data?
Making data as open as possible can create benefits for all, and as open
as possible means different things when dealing with different data. For
personal data, a balance must be struck between the benefits of providing
access to data – and individuals’ rights to privacy and being protected
from harm. It means this data can only be shared in very safe ways and, if
the person it is about has not consented, not at all. Only in very rare cases
– such as politicians’ expenses – should identifiable personal data be
open for everyone to access and use.
Developing ways to share journey data safely and securely and in
ways that foster trust with people will be hard, but we encourage
transport organisations to explore doing it in three ways:
1. As open data – with careful and appropriate aggregation and
anonymisation, and accompanied by privacy impact assessments
2. In secure environments – in particular with accredited researchers,
who should openly publish their results
3. With people the data is about – through open APIs, in ways that
enable innovation
There are several examples of ways this can be done in transport and
other sectors. For instance, Transport for London makes network
statistics
such as Oyster card journey data open to support innovation96, while Uber
Movement makes anonymised and aggregated journey data open to help
urban planning97. Open banking in the UK is using open APIs to give
customers more control over data about them98, and Projects by IF have
explored other prototypes for open APIs using personal data99.
93
See https://www.transportapi.com and https://www.navitia.io
94
Open Data Institute (2016), ‘Open enterprise: How three big businesses create value with
open innovation’,
https://theodi.org/article/open-enterprise-how-three-big-businesses-create-value-with-oPe
n-innovation
95
Projects By If (2018), Improving a city’s air quality using bulk location data from mobile
phones,
https://openapis.projectsbyif.com/scenario-3-improving-a-citys-air-quality-using-bulk-locat
ion-data-from-mobile-phones
96
https://tfl.gov.uk/info-for/open-data-users/our-open-data
97
https://movement.uber.com
98
https://www.openbanking.org.uk
99
https://newdigitalrights.projectsbyif.com
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 33
3. Use impetus from the GDPR to drive data literacy and ethics
This can differentiate your business from others and build customer loyalty. It can
also help your teams make more informed decisions, and help your organisation
stay ahead of further potential regulation on the use of personal data.
Consider:
● going beyond technical and regulatory requirements to develop
governance guidelines and processes for the collection and use of
personal data which give people more control over data about them
● providing training and resources to support different levels of data literacy
across your technology, strategy, operations, marketing and other teams
● generating debate and discussion, internally and externally, about the use
of personal data and exploring different ways to use and store it
The Data Ethics Canvas developed by the ODI can help identify, debate and
address issues around how data is used by your organisation100. Co-op is using
the GDPR to drive a broader debate on the use of personal data101. Caution Your
Blast and Birmingham City Council have piloted a journey planner that stores
personal data differently102.
CRITICAL QUESTIONS
What’s next in the regulation of personal data?
The GDPR is one of the most comprehensive pieces of regulation yet to
deal with personal data and will likely lead to more regulation in future,
both from the EU and individual countries. Just as technology can evolve
quickly and policy discussions around the use of personal data can move
fast, so organisations have a better chance of informing and being
prepared for further regulatory change if they engage in public debate.
Potential focuses of future regulation may be group privacy and actions103
(while the GDPR allows group actions, national laws like the UK’s Data
Protection Bill do not), inspection measures and conditions around when
(if at all) people can sell data about them. Authorities like the UK’s ICO are
also likely to make new recommendations about how personal data can be
used in light of issues around Facebook and Cambridge Analytica. In
future, businesses may also be able to use data protection sandboxes to
develop new services in a way regulators can assess as they emerge104.
4. Engage openly with people on how data is collected and used
This can let your organisation show customers how you consider and care for their
privacy needs, for instance explaining how personal data is safeguarded as well as
when and why it is shared. This can build trust and address privacy concerns so
100
https://theodi.org/article/data-ethics-canvas
101
Co-op Digital (2017), ‘How the ODI is helping Co-op Digital put data at the centre of the
organisation’,
https://digitalblog.coop.co.uk/2017/12/12/how-the-odi-is-helping-co-op-digital-put-data-at
-the-centre-of-the-organisation/
102
https://www.cautionyourblast.com/smart-routing
103
Linnet Taylor (2017), ‘Group Privacy: The Next Generation of Privacy Problems’,
https://linnettaylor.wordpress.com/2017/01/10/group-privacy-a-new-book-on-the-next-gen
eration-of-privacy-problems/
104
The Register (2018), ‘UK regulator moots data protection sandbox for organisations to
play in’,
https://www.theregister.co.uk/2018/03/05/ico_promises_data_protection_sandbox_for_org
anisations_to_play_in
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 34
people continue to share data about them, and strengthen customer relationships
more broadly.
Ways to engage more openly include:
● designing a user experience for your service so people know when and
how sharing data about them creates benefits so they can make more
informed decisions about whether or not to use the service
● enabling more granular control over types of personal data and sharing
● engaging with customers through surveys, focus groups and other
feedback forums to understand what data-sharing different groups are
comfortable with, for example urban commuters in cities with multiple
mobility services, or remote rural communities with few transport services
● embedding a user experience for any automated decision-making using
personal data, in order to explain why a decision was taken105. This both
complies with the GDPR and builds trust with users (self-driving vehicles
are an obvious focus here)
● clearly telling users which, if any, third parties their personal data is shared
with, what is shared, and how it is used. For example, a mobility service
may need to share personal data with a bank to take payment
● if both collecting personal data and selling it to third parties, making every
effort to explain the trade-offs and risks to both the individuals the data is
about and the third parties buying a product based on it. Over time this
may even suggest a new kind of caveat emptor that is closer to caveat
venditor, ‘seller beware’, for business models that offer a free service in
exchange for personal data, and use that personal data to sell another
service to third parties. This type of shift in responsibility becomes more
important the greater the asymmetry of power between buyer and seller in
both types of transaction.
● publishing more detailed assessments on the use of personal data in an
accessible way, whether they are mandatory privacy impact106 or
legitimate interests107 assessments, or voluntary ones such as the data
ethics canvas mentioned earlier.
See how journey planners like Moovit and C itymapper engage with users on how
they use data. Consider the ODI’s openness principles for organisations handling
personal data108. Explore Projects By IF’s open source catalogue of design
patterns for consent and sharing data109.
5. Use open innovation to be more efficient and equitable
Opening up your data for research and development can help address your
organisation’s specific challenges more efficiently, as well as systemic challenges
in a wider context. It can also widen your networks and access to new ideas. The
engagement and accessibility around open innovation, and its potential to support
transport systems more broadly, can also support your implementation of other
recommendations made here, by increasing your ability to capture the benefits of
open data and building more trust with people.
Organisations can engage in open innovation by:
105
https://ico.org.uk/for-organisations/guide-to-the-general-data-protection-regulation-gdp
r/individual-rights/rights-related-to-automated-decision-making-including-profiling
106
https://ico.org.uk/for-organisations/guide-to-the-general-data-protection-regulation-gdp
r/accountability-and-governance/data-protection-impact-assessments
107
https://ico.org.uk/for-organisations/guide-to-the-general-data-protection-regulation-gdp
r/lawful-basis-for-processing/legitimate-interests
108
Open Data Institure (2016), ‘Openness principles for organisations handling personal
data’,
https://theodi.org/article/openness-principles-for-organisations-handling-personal-data
109
https://catalogue.projectsbyif.com
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 35
● working with open data startups
● giving academics access to their data
● exploring new ways to collaborate with public services around data
● getting involved with sector innovation programmes using open data
● using data portability to help customers share data to support innovation
For examples in transport, see how data accelerator Data Pitch is working with
Deutsche Bahn110 and explore TfL’s innovation portal111. Explore how GTFS began
with collaboration between public and private organisations in the US (see Open
Standards in Transport), and expanded by partnering with academics in Kenya
(see the Digital Matatus case study). In other sectors, see how Arup works with
open data startups112, how OpenActive.io’s sector innovation programme is
developing new ways to help people get active113, and how the Open Up
Challenge is using data portability in UK banking to support innovation114.
CRITICAL QUESTIONS
How can re-identification risks be mitigated?
Our research showed that a number of transport organisations want to share
anonymised and aggregated journey datasets to support our transport systems
and enable innovation, but they are very concerned about re-identification risks115.
The technical complexity of these risks was highlighted again recently when
anonymised and aggregated open data shared by Strava was subsequently
shown to raise concerns about national security116.
A lot more work needs to be done around de-identification processes and
re-identification risks, and there are initiatives organisations can join now
to help do this. For example, the UK Anonymisation Network has set out an
anonymisation decision-making framework117 which can be adopted alongside
techniques from the ICO’s Anonymisation Code of Practice118. In the US, the
Future of Privacy Forum has developed an assessment model for preserving
privacy in open data and suggested strategies to mitigate re-identification risks119.
More must be done to reduce these risks to preserve trust in services using
journey data and to enable open data that can make transport better for all.
6. A collaborative approach to transport data could benefit the sector
Sharing insights and resources across the sector can maximise the benefits of
transport data for our economies and societies, reduce costs and risks for individual
companies, and build public trust.
Transport businesses can pilot working together around shared technical problems
such as re-identification risks, around more complex and ultimately philosophical
110
https://datapitch.eu
111
https://tfl.gov.uk/info-for/business-and-commercial/innovation-portal
112
Open Data Institute (2016), ‘Open enterprise: How three big businesses create value with
open innovation’,
https://theodi.org/article/open-enterprise-how-three-big-businesses-create-value-with-
Open-innovation
113
https://www.openactive.io
114
https://www.openupchallenge.io
115
‘De-Identification of Personal Information’, http://dx.doi.org/10.6028/NIST.IR.8053
116
The Guardian (2018), ‘Fitness tracking app Strava gives away location of secret US army
bases’,
https://www.theguardian.com/world/2018/jan/28/fitness-tracking-app-gives-away-location-
of-secret-us-army-bases
117
http://ukanon.net
118
https://ico.org.uk/media/for-organisations/documents/1061/anonymisation-code.pdf
119
Future of Privacy Forum (2018), ‘FPF Publishes Model Open Data Benefit-Risk Analysis’,
https://fpf.org/2018/01/30/fpf-publishes-model-open-data-benefit-risk-analysis
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 36
issues like informed consent, or campaigns raising public awareness of and trust in
the collection and use of data.
These organisations can also collaborate locally, nationally or even globally, for
instance joining initiatives mentioned earlier like the UK’s Transport Systems
Catapult. Organisations can also collaborate within a single mode of transport or
type of mobility service, like Bikeplus in the UK, or as a multimodal group such as
the MaaS Alliance120. These different groups can then share knowledge or
coordinate around broader shared problems relevant to them all as they emerge.
Such collaborative groups can also more effectively engage other stakeholders, like:
● public bodies responsible for transport policy, infrastructure and regulation
● civic society, consumer rights and other groups representing passengers
● newer stakeholders in transport like smartphone makers and banks
● open data and personal data experts
Sector-wide groups can also drive change programmes by creating a forum to:
● develop, advocate and measure transport-wide data strategies
● give guidance on best practices in transport data governance
● research the role of and impact of journey data and other data in transport
● standardise open data protocols for transport and provide tools for their use
● facilitate data training via face-to-face workshops, eLearning and webinars
● run competitions and accelerators for data-enabled innovation in transport
120
https://maas-alliance.eu/the-alliance
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 37
Appendix: Methodology
For this research, we commissioned the TravelSpirit Foundation to produce an
overview of the relevant literature related to the use of journey data within the
transport ecosystem. In addition to submitting regular reports of their findings,
TravelSpirit planned and conducted a workshop for ODI staff in order to help us
identify the modes of transportation worth focusing on and the types of themes
worth exploring.
We then conducted a landscape review which involved extensive research of the
current use of journey data by organisations involved in the transport ecosystem.
We identified a list of startups and large firms using journey data to improve
services for customers, which we analysed and categorised according to the way
that journey data was deployed within each organisation. A number of businesses
were then selected for targeted, in-depth interviews.
We conducted 11 interviews with participants from a range of backgrounds,
striving for a balance between people from startups, SMEs, public sector
organisations, research institutions and academia. Our interviewees were mainly
from companies based in the UK (7), though we also interviewed one person each
from Ireland, Singapore, Israel and Finland.
Open Data Institute 2018 Personal data in transport: exploring a framework for the future 38
