Using representations and data that are digital, we can create images about what happens where and when in cities, including mobility patterns that remained unaccounted until now. If properly analysed, big data for mobility can radically improve the socioeconomic and environmental analysis of public and sustainable transport. This session will discuss how big data is affecting mobility in terms of new travel behaviour and transport planning. At the user level, the relations between social networks, social media usage and travel behaviour in EU countries will be discussed. Scientific insight on the social media usage of millennial students in EU countries to understand their impact on social activities and mobility in urban areas will be presented. At the planer level, responses to changes in mobility patterns or unaccounted needs given by the analysis of public transport smart data will be presented. Advances on an integrated accessibility index will be discussed as a way for policy makers to improve current transport planning practices. Yet, big data in transport is not immune from some problems, especially those relating to statistical validity, bias and incorrectly imputed causality. This point will be discussed alongside liability, since Big data is gathered and manipulated by many different stakeholders. The proposed panel discussion therefore aims to provide to the audience a clear understanding on ways in which big data affects travel behaviour and transport planning, while accounting for data quality and pan European standardisation aspects.
3. 3
COST Actions
Network
4 years
Min 7 countries
Research coordination
and capacity building
activities
€ ~500 000 euros
over lifetime
Memorandum of
Understanding
4. 4
Scope of the session (updated):
Provide a clear understanding on ways in which big data
affects transport planning, travel behaviour and autonomous
transport, while accounting for data quality, privacy and pan
European standardisation aspects
5. The Big data analysis and planning
Floridea Di Ciommo
cambiaMO| changing Mobility
Action Chair Transport Equity Analysis COST Action TU1209
6
6. Mobility Patterns:
Big data analysis for eliciting needs and
planning
@CollectiuPunt6 cambiaMO | changing MObility
7. Big Data meaning for Madrid
• Random sample of 5,900 smart cards for 1 year
• Segmentation by age group
• Spatial disaggregation (users from each district )
Mode share:
• 42,4% PT
• 28,6% car
• 29% by foot and bike
Smart Card use per mode:
66% – 73%
cambiaMO | changing MObility
9. Where Smart cards‘ users are living and what is
their needs?
cambiaMO | changing MObility
10. Big Data analysis for planning
𝐼𝐴 𝑑
𝑟,𝑢,𝑠
= 1 −
σ 𝐝=𝟏
𝐫
𝑷𝑹 𝒅
𝒓
+ σ𝐢=𝟏
𝐮
𝑹𝑼 𝒃
𝒖
+ σ𝐢=𝟏
𝐬
𝑹𝑰 𝒅
𝒔
1
Inaccessibility index
DISTRICTS OF MADRID IARUS
Villaverde 0
Vallecas 0,384
How using Big data sets for eliciting needs and defining
mobility strategies? Could we create some automatism in the
mobility services system crossing Big data?
Thanks
Floridea Di Ciommo – cambiaMO| changing MObility -
floridea.diciommo@cambiamo.net
11. Big Data, Social Networks and Travel
Behaviour
Pnina Plaut
Associate Professor at the Faculty of Architecture and Town Planning Technion
Action Chair Social networks and Travel behaviour COST Action TU1305
12. New Social Structure
People have a wider spatial distribution of social networks than in the past.
A larger set of social contacts are active today than in the past and they overlap less in
spatial terms.
The appearance of virtual social networks such as Facebook and changes in working
patterns (home/hub-based, shorter working week days) have resulted in intertwining of
leisure activities with other daily routines.
The Result
More complex mobility patterns.
Strong links between lifestyles and personal travel
in the context of continuing social
and technological change.
13. Interacting with the transport system in real-time
- The emergence of smart-phones led to technological
developments in the form of diffusion of bottom-up
user-generated information
-Innovative mobility services
- moovit
Waze
- Uber
- Get –A- Taxi Where-is-Bus
All are combinations of cellphones with GPS
technology
14.
15. Survey participants
20 Countries
23 Universities
10 Languages
8250 Valid responses
Reporting on 3 leisure activities that were done jointly
with at least 2 other people in the last 2 weeks
16. ICT Social network
What is the size of your
largest SM network (%)
3
5
12
14
13
11
10
7
5
15
4
0
2
4
6
8
10
12
14
16
From the above network What is
the size of your “small circle”(%)
0
5
10
15
20
25
30
2 or
less
3 - 4 5 - 7 8 - 10 11 -
20
21 -
40
more
than
40
don't
use
19. Big Data
People are now generating new (large) data sets -
tagged to space and time.
We can extract from passive data sets, like smart travel
card data of individual trips in public transport systems
or from mobile phone service companies.
Using representations and data that are digital, we can
create images about what happens where and when in
cities.
21. The Data Challenges
We live in a world that is fast becoming digital in all its dimensions
Social networks and travel behaviour analysis can draw upon
active collection of data : surveys, questionnaires, or interviews.
-Ego-centric approach
-Whole network approach
Where do we draw the boundaries and who is included in the
analysis ?
22. Big Data in ACT
Nikolas Thomopoulos
Action Chair WISE-ACT COST Action CA16222
23. 24
Who has used Uber to complete a journey to date?
> 5 bn Uber trips in 633 cities of 78 countries >15 mil trips / day
Is this Big Data?
No, just a lot of data: AVs data = 2660 internet users (Zmud, 2018)
Question
25. 26
Uber (and others) have been piloting ACT globally Big Data
WISE-ACT Atlas includes 200 ACT trials
There is unanimous agreement that big data is revolutionizing
commerce in the 21st century. When it comes to business, big
data offers unprecedented insight, improved decision-making, and
untapped sources of profit (MIT TR, 2013)
Data is the new currency: Monetisation
Big Data revolution
27. 28
“Big data is a term describing the storage and analysis of
large and or complex data sets using a series of techniques
including, but not limited to: NoSQL, MapReduce and
machine learning” (Ward and Barker, 2013)
Definition of Big Data
28. 29
ACT Value Chain Data Flow
Need for data processing on the
move (Thomopoulos and Chang, 2014)
5G IRACON (CA15104)
Data storage: Cloud?
Essential for MaaS
Data Architecture: Open Data?
Developing country needs?Source: Simoudis, 2017
29. 30
Multi-stakeholder holistic approach is essential
Incorporate all actors to address:
o Government/ Regulatory
o Social
o Business
o Transport
o Evaluation
Follow WISE-ACT activities…
Future needs of ACT
30. Big Data in Mobility: Legal Concerns
Affecting Availability and Quality
Federico Costantini
Associate Professor Theory of Law and Legal InfromaticS
WG leader WISE-ACT COST Action CA16222
31. 32
Summary
(1) EU Legal framework of «Information Society» -> «Digital Single Market»
(2) First issue: (free) availability of data -> «New Copyright Directive»
(3) Second issue: data quality (and liability) -> «Non Personal Data»
(4) Implications in the Mobility sector
(5) Conclusions and final evaluations
32. 33
(1) Evolution of the European Union legal
framework concerning “Information Society”
DIR. 95/46/CE
Data protection
DIR. 2002/58/CE
e-privacy
DIR. 1999/93/CE
Electronic signatures
DIR. 2000/31/CE
Electronic commerce
Reg. (UE) 910/2014 “eIDAS”
“privacy package”
DIR. (UE) 2016/1148 «N.I.S.» REG. (UE) 2016/679 “GDPR”
DIR. (UE) 2016/680 «e-justice»
DIR. (UE) 2016/681 «PRN»
Proposal Reg. «E-Privacy»
COM(2017)10
Proposal Reg. «Non Personal
data» COM(2017)495
DIR. 2001/29/CE
Digital copyright
Proposal DIR. (UE) «copyright»
COM(2016)593
33. 34
(2) First issue: (free) availability of data
-> «New Copyright Directive»
[…] With the exception of XXX data
feeds and APIs, you will only
access the XXX web site with a
human-operated interactive web
browser and not with any program,
collection agent, or "robot" for the
purpose of automated retrieval or
display of content. […]
[…] You will only access the YYY
web site with an interactive web
browser (or other authorized agents,
which include general purpose media
players) and not with any program,
collection agent, or "robot" for the
purpose of automated retrieval of
content, unless you are granted
permission by YYY to do so. […]
excerpts from «Terms and conditions» of from mobility data websites
34. 35
(3) Second issue: data quality (and liability)
-> «Non Personal Data»
https://images.pexels.com/photos/17739/pexels-
photo.jpg?auto=compress&cs=tinysrgb&dpr=2&h=7
50&w=1260
https://www.maxpixel.net/Crop-Straw-Cereal-Pasture-
3099288
… sometimes a weather forecast is
a matter of money: who pays for the
damages if it is wrong?
35. 36
(4) Implications in the Mobility sector
https://en.wikipedia.org/wiki/Floating_car_data#/media/File:Trans
Core_RFID_reader_and_antenna.jpg
http://www.contemplatingdata.com/2017/10/10/big-data-
beginners-guide-for-non-technical-people/four-vs-2/
36. 37
(5) Conclusions and final evaluations
https://images.pexels.com/photos/75183/garden-back-vegetables-fruit-
75183.jpeg?cs=srgb&dl=back-fruit-garden-75183.jpg&fm=jpg
If Internet was a garden …
37. Privacy and Social Responsibility
Payal Arora
Associate Professor at Erasmus University Rotterdam
39. 40
Autonomy and Privacy
Personal vs. social norms
Data aggregation and decision-making
Mapping nodes of information disclosures
Transport behavior and de-anonymization
Standards and Privacy
Scalability and adaptability beyond national borders
‘Golden Standard’ to privacy rules
Privacy by design
Security and Privacy
Profiling, predictive privacy harms and public transport space
Fear of hacking
Storage and analysis of data- who is in charge?
Privacy and transport dilemmas
42. Big data in the transport sector:
needs for standardisation
Tatiana Kovacikova
ERA Chair Holder on ITS – university of Zilina
Scientific Committee member at COST
43. 44
Big Data Landscape 2017
Big Data + AI = The New Stack
Open Science, Open data - Data moving to the Cloud
http://mattturck.com/bigdata2017/
44. 45
Complex, multilevel topology corresponding to the various
aspects of transport research, planning, design and
operation
Different transport modes (road, rail, maritime, air, multi-
modal)
Different transport types (persons/freight,
urban/interurban/rural, domestic/international transport,
commuting/school/recreational, etc.)
Covering all phases of transport projects lifecycle (planning,
design, implementation, operation and management)
Variety of technologies: ITS, IoT, CAV, innovative technologies
(machine learning, artificial intelligence
All types of transport data (sensor generated data, traffic
management and traffic control data, user behaviour data,
tracking data, ticketing and fare collection systems …)
What is specific for big data in transport
45. 46
Transport data resources and mechanisms for
data collection
The Transport Data Revolution: https://ts.catapult.org.uk/
46. 47
Typology of mobility service types
influenced by transport data
The Transport Data Revolution: https://ts.catapult.org.uk/
47. 48
Develop and adhere interoperable (global) data standards
because of the wide range of systems from which these data are created
significant challenge to the development of data-driven intelligent mobility services
Understand content of data - description of the data available
each data collection is based on harmonised metadata profiles*
data elements (description of a dataset in a minimal but adequately way),
wordings and semantics,
predefined categorisations – transport specific
data field names,
data value type,
data field lengths
Understand structure of data – description of the technical format of the data sets
following data formats are currently foreseen for different transport domains**
DATEX II for road transport data
NeTEx and SIRI for public transport data
TN-ITS and Inspire for geographical data
Data formats currently valid and used for the creation of traveller information services
Further work required for the formats for other transport data
Where/why standardisation of transport data is
needed
*DCAT-AP (Application profile for data profiles in Europe
https://joinup.ec.europa.eu/release/dcat-ap-v11
** Delegated Regulations following the European ITS Directive (2010/40/EU)
48. 49
Big data and transport planning (Floridea)
Big data and travel behaviour (Pnina)
Big data and autonomous and connected transport (Nikolas)
Big data and data quality (Federico)
Big data and privacy (Payal)
Big data and ITS standards (Tatiana)
Panel discussion:
Benefit to the citizen? Which dimension is most important?