Author Name
Admin_niti
Choose Report Type
Publication Date
Report Upload
Download
(5.02 MB)
vertical
Infrastructure connectivity and E-mobility
PDF Text
DATA-DRIVEN
MOBILITY
IMPROVING PASSENGER TRANSPORTATION
THROUGH DATA Authors
NITI Aayog
Pranjal Dubey
Shikha Juyal
Ishita Kandra
Abhishek Saxena
Shweta Sharma
Anil Srivastava
Rocky Mountain Institute
Akshima Ghate
Emily Goldfield
Robert McIntosh
Greg Rucks
Samhita Shiledar
Clay Stranger
Independent contributor
Anand Shah
Contacts
NITI Aayog: adv.niti@gov.in
www.niti.gov.in
RMI: ocs@rmi.org
www.rmi.org
Acknowledgements
The authors would like to thank Dr. Rajiv Kumar,
Vice Chairman, NITI Aayog and Mr. Amitabh Kant,
CEO, NITI Aayog for their support that made this
report possible.
Suggested Citation
NITI Aayog and Rocky Mountain Institute.
Data-driven Mobility: Improving Passenger
Transportation Through Data. 2018.
“The views and opinions expressed in this
document are those of the authors and do not
necessarily reflect the positions of the institutions
or governments. While every effort has been made
to verify the data and information contained in this
report, any mistakes and omissions are attributed
solely to the authors and not to the organization
they represent.”
Authors and
acknowledgements Outline
1.0 INTRODUCTION
2.0 THE LANDSCAPE OF MOBILITY DATA
2.1 Defining mobility data and related terms
2.2 Primary stakeholder groups
3.0 MOBILITY DATA USE CASES
3.1 Traveller use cases
Mobility as a service
Ancillary trip information
3.2 City and government use cases
Safety and security
Transportation, route, and infrastructure planning
Real-time system management
Enforcement and regulation
3.3 Researcher use case
3.4 Using data to enable emerging technologies
Big data analytics
Blockchain
Artificial Intelligence (AI)
4.0 DATA ACQUISITION FOR MOBILITY USE CASES
4.1 Process of data acquisition for specific use cases
4.2 Challenges associated with data acquisition
Privacy and data security
Poor quality and incomplete data
Acquiring data from private data owners
5.0 DATA AGGREGATION
6.0 OPEN DATA
7.0 CONCLUSION
8.0 CITATIONS 6
1.0
/////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
Introduction
+++
Mobility is a key economic driver: moving people and goods is at the core of a
well-functioning and prosperous society. Increasing the efficiency and quality
of a country’s mobility system helps support a stronger economy and a higher
standard of living for its citizens.
Data is an essential piece for unlocking maximum value within a transportation
system. Data analytics, and data sharing between organizations, has the
potential to create more efficient passenger mobility, as well as allow for optimal
designing of transit routes and services, infrastructure, and regulations.
Data collection is also essential to enabling the use of a number of emerging
technologies in the mobility space, such as blockchain and artificial
intelligence. Using data to optimize commutes and transit infrastructure will
lead to lower levels of congestion, reduced tailpipe emissions, and less time
spent in transit, resulting in cities that are cleaner, safer, better designed, and
more economically prosperous.
Maximizing the benefits of data analytics for the mobility sector requires the
sharing of data between parties. Companies, public transit agencies, and
commuters are generating and collecting huge amounts of transport data.
However, this data is siloed between organizations and individuals, and
recorded with different standards and formats. Absence of transport data
ecosystem is an important concern in India as well. Encouraging and enabling
data collection and data sharing will allow linking of currently disparate
models of transport and unlock tremendous value within the transportation
systems in the country. 7
Data can enable travellers to benefit from seamless multimodal transportation,
and make it as convenient as private vehicle use. This paper describes how
data can be used to make passenger mobility more efficient, beginning by
outlining the relevant mobility data stakeholders and use cases and then
analyzing the flow of data in this system. By making possible such a data-
enabled future for transportation, mobility assets can be better utilized and
integrated, and India can bolster economic growth while building cleaner,
more liveable communities. 8
2.0
/////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
The landscape of
mobility data
+++
Mobility data encompasses a wide range of mobility-related information. The
landscape of mobility data involves multiple stakeholders ranging from data
owners and aggregators to data users. The different types of data owners and
beneficiaries, and the network of relationships between members of these two
key groups, is a core piece of understanding for crafting value propositions to
different stakeholders for varying use cases of data.
2.1 Defining mobility data and related terms
Data can be defined as “representation of information, facts, concepts, opinions,
or instructions in a manner suitable for communication, interpretation, or
processing by humans or by automated means
1
.” Mobility data can include a
wide range of data related to transport. This includes data of public transit
agencies, private mobility solution providers, and individual citizens. In addition
to information about transportation assets and trips, mobility data can also
include data that affects or goes in tandem with mobility, such as weather data
(can affect traffic, etc.), pollution and air quality data, and traffic violations data.
India’s proposed Personal Data Protection Act, described in more detail in
section 4.2, defines personal data. According to the proposed Act, personal
data refers to “data about or relating to a natural person who is directly or
indirectly identifiable, having regard to any characteristic, trait, attribute or any
other feature of the identity of such natural person, or any combination of such
features, or any combination of such features with any other information.” It goes
on to define the processing of personal data as “an operation or set of operations 9
performed on personal data, and may include operations such as collection,
recording, organisation, structuring, storage, adaptation, alteration, retrieval, use,
alignment or combination, indexing, disclosure by transmission, dissemination
or otherwise making available, restriction, erasure or destruction.”
2.2 Primary stakeholder groups
At the highest level, the three categories of stakeholders most critical to the
data landscape are data owners, beneficiaries, and government.
Data owners are companies, organizations, and individuals that produce and
own datasets. Beneficiaries are groups and individuals that can benefit from
using the data owned by the data owners. In addition to both benefiting from and
owning data, government can play a key role in enabling the interactions between
these two stakeholder groups and protecting their interests. It should be noted
that these categories are not rigid: data can flow from owners to beneficiaries,
but it can also flow within each of these categories, and some beneficiaries may
also qualify as data owners. There are many cases of business to business data
sharing; for example, Uber uses data from Google Maps to provide its own service.
Within the category of data ownership, there is one primary distinction: public
versus private data ownership. Public data is shared openly, irrespective of how
it is generated. Private data is kept within an organization and not shared with
the public. Data owners may have both private and public datasets. Data
generated by private service providers is important in India because of a
significant presence of privately-owned or operated transport services that
meet the passenger mobility needs in cities. 10
Matatus are the privately owned mini-buses in Nairobi, Kenya. Matatus
are very popular because they are cheap and convenient, but have
had the same problems as a typical informal transit system: difficulty
to access timetables, routes, and stops. The Digital Matatus Project
is designed to solve these problems using digitization. Digital Matatus
Project is a joint project between Columbia University, The Rockefeller
Foundation, and the technology sector in Nairobi, and is focusing on
developing a mobile routing application and new transit map for the city
using cellphone technology. First, students from the University of
Nairobi rode all of the bus route using an app that collected data in a
General Transit Feed Specification (GTFS) compatible format. Data
points included routes, stops, and visual notations (signs and shelters).
The data was then processed and released in the form of a paper map
and transit apps. Some of the transit apps include Ma3route, Flashcast,
sonar, digtitalmatatu, and matatumap. The City of Nairobi has recognised
the importance of the digitization and is using this data to create a new
trip planning tool for the city. Learning from the success story of Nairobi,
several other cities in Africa are planning to map their informal transit
sector as well. This case study may prove a relevant example for
addressing similar challenges related to the lack of data for India’s
informal transit system.
Case study: Digital Matatus Project
2
A large portion of data owned by private data owners is generated by the
movement of individual travellers. If the mobility service is provided through a
mobile application, travellers typically agree to allow the company to collect
and own the data that the traveller generates. A significant portion of private
mobility service providers in India do not have a digital presence. Despite this,
these companies are generating significant amounts of data that can be useful.
The main challenge in the latter case is finding out ways to collect the data
from these service provider’s operations.
Each traveller could also be considered a private data owner. Individually, this
has much less value than the aggregated data from many travellers, making
companies with large aggregated datasets much more relevant data owners
than individual travellers in today’s data landscape; however, new technologies
have the potential to change this relationship. Blockchain, for example,
disintermediates the aggregators of data and could allow travellers to have more
control over and more direct compensation for sharing the data they generate.
Data beneficiaries can be categorized into three main stakeholder groups:
travellers, cities and governments, and researchers. Each of these beneficiary
groups correlates with a set of use cases for mobility data, which are explored
in more detail in section 3. 11
» Travellers are any individuals needing to move from their current location
to an end destination. In this case, the primary use case for open mobility
data would be to optimize the route and mode of transport for increased
efficiency and ease of use.
» Cities and governments include planners, regulators, and operators in
charge of system-level design and operations for a city’s transportation
network, or policy design and operations at a state or national level. The
use cases for cities for mobility data include transportation planning
(infrastructure and services planning), improving safety, and city planning.
» Researchers include any organizations or individuals conducting research
in the area of mobility. Greater access to data allows researchers greater
insight into the mobility system to better analyze what is happening and draw
more data-based conclusions, which could in turn be of use to policy makers
and cities.
Travellers represent the individual benefits of data analytics and data sharing,
while cities represent system-level benefits. For each of these beneficiaries,
the type of data required to achieve the desired use case varies. For example,
a traveller needs data that will help him optimize his trip to his destination,
which likely only requires real-time data or potentially short-term projections
of his transport options when he chooses to depart. In contrast, a city planner
who is designing infrastructure for the future would benefit from historic
transit data so that she can examine past trends. 12
The use cases for mobility data can be examined from the perspective of each
beneficiary category. The type of data and analytics required varies between the
use case.
3.1 Traveller use cases
There are many potential use cases that apply to improving transportation
efficiency for individual travellers. Many of these can be grouped under the
general category of “mobility as a service.”
Mobility as a service
Mobility as a Service, or MaaS, refers to the technology-enabled, on-demand
availability of multimodal trip options, including multimodal trip planning and
seamless payment. Mobility as a service is an alternative to private vehicle
ownership; travellers should be able to order a ride to wherever they need to go,
at the time they need and in whatever size or type of vehicle meets their needs.
Under the MaaS paradigm, the entire transportation system functions as a
cooperative, interconnected system to meet travellers’ needs through a variety
of transport modes. To achieve this, infrastructure, technology platforms,
payment, transportation services, and data analysis must be capable of working
together
3
. Mobility data—as well as organizations’ willingness to share data—is
the key to unlocking MaaS. This paradigm has the potential to benefit both
travellers as well as transit providers who share data: travellers can enjoy
increased options and trip efficiency; while for transit providers, compiled data
3.0
/////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
Mobility data use
cases
+++ 13
can help open up markets and customer bases that are currently
disaggregated due to the lack of connection between service providers.
There are several data-supported elements that go into MaaS, several of which
are described in more detail below.
» Multimodal trip planning
A primary element of MaaS is the ability to see all available modes of
transport, and both choose the mode that is most optimal for the situation
and be able to easily link various modes of transport to get to the destination.
For example, a traveller could go onto a single platform and enter his
destination and be shown the best option for getting there out of all
available modes, which may include a portion of the trip using one mode
and another portion using a different mode. Currently, travellers lack easy
access to information about all their transportation options. Mobility data—
as well as transit providers’ willingness to share data—is essential to
enabling multimodal trip planning.
» Seamless Payment
Enabling travellers to seamlessly pay the various transportation providers
through a single portal will increase the accessibility of transport options
and promote multimodal trips. Implementing seamless payment requires
collection and integration of transit data, and also relies on transit companies
being willing to share data.
India Stack’s Unified Payment Interface
4
, which allows all bank account
holders in India to send and receive money instantly from their smartphones,
may uniquely position India to be able to quickly adopt a multimodal transit
application with integrated payment.
» Real-time mode connectivity and optimization
Mobility data can enable the real-time optimization of travel plans around
changing factors, such as weather and traffic, as well as travellers’ preferences
(e.g. least expensive, shortest time, etc.). For example, an algorithm can
take into account traffic data to reroute a traveller around an accident or
congested area, or update expected travel time based on weather conditions. 14
01
0203
Multimodal trip
planning
Seamless one-stop
payment
Mode connectivity,
real-time route
optimization
Figure 1: Three interdependent components of “mobility as a service.
5
”
Ancillary trip information
In addition to MaaS, there are a number of other services that access to data
can provide to increase the efficiency and ease of a traveller’s journey. This could
include accessing data on real-time conditions (i.e. traffic, weather, accidents, etc.),
information about interesting landmarks along the route, or any other sort of
ancillary information to enhance a trip.
Figure 2: Screenshot from BMTC’s trip planner app.
Case study: Bangalore’s bus tracking application
6 15
Bengaluru Metropolitan Transport Corporation (BMTC) has piloted a
mobile application using Intelligent Transportation System (ITS) to give
fixed and real-time information about BMTC buses to the passengers.
Users can easily access the trip planner tool to see the bus timetables
and route maps. BMTC buses are equipped with GPS and give a
real-time location of buses and estimated time of arrival. The mobile
application also gives additional information about the approaching
buses at the bus stop, such as bus number and the platform details on
which the bus is going to arrive. The app was developed in order to make
BMTC bus experience more comfortable for the passengers and promote
public transportation as the preferred mode of transport. Similar mobile
applications are also being developed in other cities in India, such as
Ahmedabad.
3.2 City and government use cases
There are several use cases that apply to optimizing passenger mobility from a
systems-level. Cities and governments around the world are realizing the value
of using mobility data to improve system safety and optimize transit planning
and city design around the efficient movement of people and goods. Some of
these use cases are discussed below.
Safety and security
Data can enable improved safety and security within the transportation system
in a number of manners. For example, increased access to data allows cities
to see where accident hotspots are, thus enabling them to respond more quickly
and also understand the issues in those areas. With increased understanding
of when and how accidents occur, cities can ensure a greater level of safety for
their citizens by responding faster when incidents occur and developing
solutions to systemic concerns.
One particular area of concern in India is women’s safety in the transportation
sector. Many women feel unsafe traveling alone, and frequently avoid using
public transport
7
. Improved tracking of vehicles and verification of drivers and
vehicles that are deemed safe are some examples of how data can allow women
to feel safer using transportation. Safety equipment such as panic buttons, GPS,
and CCTVs could be installed in vehicles to improve safety—and tracking of
crimes and harassment, should they occur—for women. 16
In 2016, Grab, Easy Taxi, and Le.Taxi formed a partnership with the
World Bank and other organizations to share the traffic data from their
drivers’ GPS streams through an open data license. The three ridesharing
companies combined cover more than 30 countries and serve millions
of people
8
. The partnership’s goal in sharing their data is to provide cities,
particularly in developing countries, with the tools to address traffic
congestion and plan for better mobility by developing better, evidence-
based solutions to road safety and traffic challenges. This unprecedented
multi-provider agreement to share data is an important counterexample
to the argument that sharing data undermines competitive advantage.
A core reason that this initiative was able to launch successfully is that
the private companies involved knew exactly what would happen with the
data, what type of interface people would see when the data was published,
and had assurance that the data would not be used for other purposes.
Case study: Open Transport Partnership
Transportation, route, and infrastructure planning
Transportation planners can leverage data analytics to better design and
maintain routes, public transit, and mobility infrastructure. There are many
specific use cases within this category: Analyzing traffic and commute patterns
allows planners to understand where to build infrastructure (including non-
motorized transport infrastructure) and add transit routes to ease stress in the
most highly-trafficked areas. Data analytics can aid planners in minimizing
congestion in cities by identifying the primary cause (i.e. poorly-timed signals,
insufficient parking, etc.). Data can also aid in monitoring the structural health
of transportation infrastructure, such as bridges and overpasses, and related
data, such as water and flood data, can inform how and where to build
transportation infrastructure.
Real-time system management
Data can aid in the real-time functioning of the mobility system. Operators can
remotely monitor the transportation system and manage system operations.
Increased access to data will give operators more real-time information that will
help them ensure the smooth functioning of the transportation system. One
way of collecting this data is by installing sensors to detect traffic and vehicle
movement. Additionally, as more and more private companies are collecting
data on travellers’ movement through GPS, cites are beginning to see an
opportunity to partner with these companies to acquire data. 17
Waze, a Google-owned traffic and navigation app, launched its
Connected Citizens Program in late 2014 as a two-way exchange of
information with cities, with Waze providing city partners with user
driving information in return for real-time and advanced notice of
construction and road closures. This partnership benefits both parties:
Waze sees the program as a way it can grow and improve its services,
while city and state governments can easily expand their view of their
roads and streets without having to invest in more road sensors and
traffic cameras. As part of the Connected Citizens Program, Waze
has partnered with over 100 cities around the world, who can now use
Waze’s data to help with city planning and transportation regulation.
In Rio de Janeiro, for example, the Waze API is completely embedded
in the city’s Control Center, to help with day-to-day monitoring of road
conditions
9
. In Boston, the city’s central Traffic Management Center uses
Waze’s real-time data to change the traffic signals in 550 of the city’s
intersections as needed to reduce congestion. Washington, D.C. has
used data supplied by Waze to aid in the city’s “war on potholes
10
.” In
all of these cases, the cities were able to increase access to data without
investing in additional monitoring equipment.
This example also demonstrates that private companies are more
willing to share data when the use case is very clear (i.e. city planning
and regulating) and the use of the data is transparent.
Case study: Waze Connected Citizens Program
Enforcement and regulation
Increased access to data affords regulators better visibility into the transportation
system, allowing them to improve the enforcement of regulations, and develop
new and modify existing regulations to ensure a smoothly functioning system.
For example, data can allow regulators to see where infractions incur most
frequently and consequently improve their enforcement ability in those areas;
it can also aid in the collection of road tolls and parking fees.
3.3 Researcher use case
The researcher use case includes anyone conducting mobility-related research,
such as academic institutions and think tanks. Increased access to data gives
these groups and individuals a greater ability to conduct in-depth quantitative
analysis on the transportation system, in order to draw conclusions about and
make recommendations for the mobility system. This allows researchers to
provide cities and other beneficiaries with more detailed insights, such as the
cause and effect relationships between investments and impact, in order to
inform what a city should invest in to improve its transportation system. 18
3.4 Using data to enable emerging technologies
A number of emerging technologies have potential applications in the mobility
sector. Increasing the collection and use of mobility data will help unlock the
potential of these technologies. This section examines some of the most notable
emerging technologies—big data analytics, blockchain, and artificial intelligence—
with respect to their potential applications in the mobility space.
Big data analytics
Data analytics refers to the practice of examining large amounts of data to
discover patterns, correlations, and other insights
11
. Data analytics and
technologies can allow businesses, governments, and other organizations to
analyze datasets and draw conclusions to help them make informed decisions.
Big data analytics includes a wide range of specific tools and techniques that
can be used to gather insights from data; some specific techniques include
data mining, predictive analytics, and text mining.
For mobility, big data analytics can help with planning and managing transportation
networks and designing and optimizing services to meet transportation needs.
Both the private and public sector can benefit from big data analytics for mobility.
Some areas where big data analytics can significantly impact the transportation
sector include
12
:
» Optimization of transit schedules by analyzing demand and predicting the
impact of maintenance, road work, congestion, and accidents,
» Increased safety and reduced environmental impact,
» Fleet optimization and predictive maintenance through real-time view of
fleet operating conditions, statistics around usage and weather patterns,
and maintenance cycles, and
» Freight movement and routing optimization.
Blockchain
Blockchain stands ready to revolutionize significant portions of the transportation
space and could be particularly applicable in the Indian context. Blockchain
allows for more decentralized transactions through a shared, standardized
ledger system, which could allow decentralized passenger and freight systems
in India to improve their efficiency without the oversight of a central governing
body. Some examples of this include:
» Improved tracking: Currently most elements of the transportation system—
such as freight, vehicles, and jobs—are tracked through a combination of
brokers, middlemen, and paperwork. Goods can be lost, vehicles unaccounted
for, or jobs duplicated or lost due to lack of accurate tracking. By using
blockchain, these items can be tracked on a shared, secured, decentralized,
and trusted standardized ledger system
13
. 19
» Open-source mobility as a service: Utilization of blockchain supports
multiple technologies on a shared, distributed, and transparent platform.
Local startups could provide customized services or fleets, and these could
be offered through smartphone applications. This would allow companies
to provide more unique, customized, and localized services while allowing
startups to take advantage of existing platforms and customer bases
14
.
» Secure data logging and tracking: For vehicle sharing, customers often
have a set of preferences and desires they wish in both their vehicle and
their fellow carsharing customers. Blockchain enables a customer to have
a secure area for these preferences to be logged, as well as previous
experiences in carsharing. Through blockchain encryption, this is a permanent,
immutable ledger. This could also be used for insurance purposes, as a
vehicle could keep a constant log of its experiences in order to document
fault in an accident (similar to the “black box” installed in most airplanes
for the purpose of facilitating the investigation of aviation accidents and
incidents
15
).
Artificial Intelligence (AI)
Artificial intelligence is expected to revolutionize many traditional industries
over the next decade. There are a number of areas in which artificial
intelligence is impacting the mobility sector, some of which are outlined below.
» Advanced Driver Assistance Systems (ADAS): Artificial intelligence can
help to automate, adapt, and enhance vehicle systems for safety and
better driving, to reduce the potential for human error. These technologies
may increase safety and reduce collisions and accidents by alerting the
driver to potential problems, implementing safeguards, and taking over
control of the vehicle to varying degrees. There are multiple levels of
automation in advanced driver assistance systems, the most mild being
a system that issues warnings to a human driver, and the most intensive
being fully autonomous vehicles. 20
Pursuit of machine learning in the mobility space
Companies worldwide are pursuing either new mobility industry areas
(such as autonomous driving and predictive maintenance) or old areas
with improvement through automation (fleet management, infrastructure
planning, supply and manufacturing management, quality assurance,
and in-vehicle experience). These companies are traditional mobility
companies, startups, and outside industries entering into the mobility
space. Given these trends, India could explore machine learning and
automation in order to stay competitive on the global scale
16
.
Figure 3: McKinsey 2017 assessment of companies pursuing mobility-related machine learning markets
» Supply chain management: In the logistics space, the use of artificial
intelligence will lead to more effective distribution of resources in the
freight sector as supplies are more effectively moved where they need
to go at the right time. Locations and movement of vehicles will be more
effectively determined.
» Infrastructure planning: As big data becomes more available, cities will
start to rely on learning machines and artificial intelligence to process
these sets into useful outputs. Artificial intelligence will be used to predict
where and how transportation should be built, determine effective mobility
corridors, and recommend infrastructure projects. 21
4.0
/////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
Data acquisition for mobility
use cases
+++
This section explores the process for acquiring data, as well as the challenges
associated with it, to support mobility-specific use cases. There are several
components that play a role in data acquisition, and the following sections explore
ways of addressing and minimizing some of the key challenges.
Data acquisition can be done in a primary manner, e.g. through an organization’s
own sensors or user application, or in a secondary manner, e.g. by acquiring
existing datasets from other parties. In the case of primary data collection,
organizations often collect data regardless of whether there is an immediate
use case for it, under the assumption that owning more data is an asset and
may provide useful insights in the future. With secondary data collection,
however, it can be helpful to first identify the desired use case before making
a request for data.
More and more transit organizations, both public and private, as well as
individuals, are collecting significant amounts of transit-related data. The range,
scope, and volume of data collection is expanding. However, a significant
number of mobility service providers in India still do not collect data, do not
have the ability to collect data, or do not operate through digital platforms to
facilitate the easy collection of data. Technology can be an enabler in cases
of such mobility providers to enable data collection. For example, many cities
are using sensors to collect data to improve signal synchronization.
This increase in data presents a massive opportunity to better integrate existing
transport systems, optimize transit options to users’ needs, and plan and regulate 22
cities to best support mobility patterns. The potential value that mobility data
can unlock has led some analysts to refer to data as “the new form of oil” for
transport systems. However, there are still many barriers to data acquisition that
need to be addressed in order to realize the full potential value of using data
analytics to improve mobility.
4.1 Process of data acquisition for specific use cases
When the desired use case for data is known, there are three primary steps
for acquiring the necessary data: identifying the sources for the required
data; surveying what data is available and identifying what gaps remain;
and collecting additional data or acquiring data from other data owners to
fill those gaps.
1. Identifying necessary data
The first step is to identify what sort of data is needed to fit the desired use
case. This is an essential step; thoroughly analyzing what types of datasets
are needed—and conversely, what data is not needed—will save time and
allow the organization to make specific asks for data, if additional datasets
are required from other data owners.
Example: Route Planning
A city government wants to use data to inform public transit route planning.
To do this, a team of transportation planners determines that they want
historic data on commute patterns and traffic congestion in order to
identify the most highly trafficked routes to determine where new mass
transit routes can be developed to ease congestion. They recognize
that they do not need real-time data or ancillary data such as weather
information, trip fares, etc.
2. Determining what data is available and what gaps remain
Once the necessary datasets have been identified, the organization can
survey which data are already available to them, either through data that they
already own or have access to, or through publicly available data. There are
already many sources of open data; surveying all of the data that is already
available will prevent an organization from collecting redundant data or
making unnecessary data requests. Once the organization has determined
which of the datasets are already available to them, they can then identify
where gaps still exist and what sort of data could be acquired to fill these
gaps.
Example: Route Planning
The transportation planners note that they have open access to historic
data on the ridership of current public transit options from the public transit
agency, and they also have data on traffic patterns from the several traffic
sensors that have been installed throughout the city. They then determine
that their existing traffic pattern data does not have as much detail as they 23
would like, and leaves off a few key areas of the city. They decide to acquire
additional data for commute and traffic patterns in the city.
3. Collecting remaining data
To fill in the gaps identified, an organization or individual has two options:
1. Acquire the data from another data owner: If someone already owns the
data needed, one option is to acquire the data from that data owner.
2. Collect the data: If no one already owns the needed data, or if the data
owner is unwilling to share the data, then the organization must devise a
way to collect the data themselves.
Example: Route Planning
The team of transportation planners evaluates the options of installing more
traffic sensors to collect their own data on commute and traffic patterns
or approach other data owners to acquire data that already exists. They
decide to take a combined approach of installing additional sensors, as well
as working with shared mobility providers to acquire existing datasets.
4.2 Challenges associated with data acquisition
There are a number of challenges associated with data acquisition,ranging from
the quality and availability of data to concerns around privacy and security once
data is collected or acquired. The following sections highlight some of these
key challenges. 24
Privacy and data security
As big data becomes more prevalent and necessary in developing effective
mobility systems, concerns have been raised about protecting individuals’ privacy.
Personally identifiable information (PII) is generally considered information that
can be used on its own or with other information to identify, contact, or locate
an individual person
17
. Individuals are generally concerned about leaks of PII
because they can result in negative consequences.
Data does not need to include personal identifiers to be useful in many cases.
Metadata—data that describes and gives information about other data and is
not attached to a particular individual—can be extremely useful in determining
how mobility systems are being used, by whom, and when, to help in
determining which systems should be built or supported more effectively.
The proposed Personal Data Protection Act of 2018 prohibits the processing
of sensitive personal data without explicit consent. Thus, any company or
government agency gaining access to data points about mobility users will
need to do one—or both—of two things:
» Remove any PII from the data before allowing it to be used publicly.
» If data is to be transferred, ensure the process is completely secure
so that PII remains only in the possession of the parties that have been
authorized to own or access it.
For the first solution, personal identifiers must be reliably removed from the
data so that individuals’ privacy is not compromised. Though many datasets
include PII, personal identifiers are rarely necessary for transportation planning.
For the second solution, data containing PII (or ideally all data) should be sent
through secure channels when it is transferred for use. If PII must be kept and
transferred, it should be done in such a way as to eliminate the possibility of
access by outside parties.
In July 2018, the Government of India released the drafts of two major
reports on data protection: “A Free and Fair Digital Economy”
by the Chairmanship of Justice B. N. Srikrishna, and the Personal
Data Protection Act, 2018 by the Ministry of Electronics and Information
Technology. The goal of these reports is to introduce standards for
data handling, processing, and privacy.
India’s draft data protection and privacy guidelines: Personal Data
Protection Act and “A Free and Fair Digital Economy” 25
B. N. Srikrishna’s report on data protection
18
defines personal data based
on its identifiability. The report elaborates: “Identifiability in circumstances
where the individual is directly identifiable from the presence of
direct identifiers such as names is perhaps uncontroversial and will
obviously be included within the scope of any definition of personal
data. The definition should also, in addition, apply to contexts where
an individual may be indirectly identifiable from data that contains
indirect identifiers
19
.” The report recommends that a data protection
law should be set up, that will be responsible for the enforcement and
effective implementation of the definition of personal data and sensitive
personal data, legal affairs, policy and standard setting, research, and
awareness.
B.N. Srikrishna’s report outlines seven key principles for effectively
designing a privacy policy. At a high level, the principles are that the
policy should be:
1) Technology agnostic
2) Holistic
3) Include language on informed consent
4) Recommend data minimization
5) Assign controller accountability
6) Structure enforcement
7) Include deterrent penalties
20
The Draft Personal Data Protection Act
21
focuses on the fair and
reasonable processing of data. The Draft Act specifies that there must
be a clear, specific, and lawful purpose behind data processing and
stipulates that only necessary data should be collected. The Act
proposes to replace traditional terms such as data controller and data
subject (i.e. person whose data is being collected) with data fiduciary
and data principal, respectively. The data fiduciary would be responsible
for ensuring the personal data quality as well as complete, accurate,
and not misleading data processing. Sensitive personal data may be
processed on the basis of explicit consent.
The reports are currently under review by the Ministry of Electronics
and Information Technology. The government is also seeking comments
on the drafts from relevant stakeholders and the public
22
. 26
Poor quality and incomplete data
Data collection in India is insufficient and in some cases the data that is
collected is of poor quality or incomplete. A dataset may be considered of poor
quality if it fails to provide adequate or accurate information. For example,
real-time location data is often inaccurate, collected infrequently, or restricted
to only certain services
23
; a transit agency may be inconsistent about stop and
route identifiers, or a bus equipped with GPS may have a system that is broken
or inaccurate. In order to craft better mobility systems, better data collection
will be necessary in India’s future.
Acquiring data from private data owners
Acquiring data from private data owners is often one of the biggest challenges
in collecting the data needed for a particular use case. These data owners tend
to be concerned primarily with jeopardizing their competitive advantage by
sharing their private data. The nature of their concern may vary with the use
case for the data; for example, data owners often feel less threatened by
providing only real-time data to help travellers optimize their transport options,
whereas they are more wary of providing historic data to planners and developers.
This concern must be addressed by making appropriate and transparent
requests for data, as well as making clear the value proposition of sharing
data to the data owner. 27
Data aggregation, as a general strategy, is any process in which information
is gathered and summarized in certain forms, usually for statistical analysis.
Commonly, it is used to obtain more information about particular groups by
determining patterns of behavior. It typically involves compiling and anonymizing
data from multiple sources, usually for the purposes of observing system-level
patterns or understanding broader trends.
In transportation data, aggregation strategies are particularly useful for two
reasons: research and privacy purposes. In the modern age of IOT (Internet-
of-Things), large volumes of transportation information is easier than ever to
collect. The advent of technologies such as GPS, ride-sharing, and navigation
and routing programs provides multiple avenues to track individual pathways
of movement.
Data aggregation in mobility allows researchers, planners, and private industries to
understand larger trends about user profiles of their city. Similar to how a store
might market specific products to a certain demographic, data aggregation in
mobility enables planners to see how certain user profiles might be using
transportation differently.
While granular information is immensely useful for urban planners to understand
city behavior, it also raises the question of individual privacy. Data aggregation
is a method that enables these city practitioners and developers to access
invaluable data that may otherwise not be available, due to the risk of privacy
or competitive risk. By aggregating based on certain districts, areas, or
5.0
/////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
Data aggregation
+++ 28
types of user profiles, data aggregation anonymizes individuals and removes
granularity to such a degree that the user information is now shareable. It is
often a compromise between private companies with mobility data or various
smartphone companies or routing systems—which contain privacy-sensitive,
individual user data—and public city practitioners who are seeking information
to better understand their constituents.
Though often a powerful tool, these aggregation strategies can run the risk of
failing to provide the information that users need. Private companies are more
willing to share their aggregated, rather than granular, data, but the aggregated
information may end up obscuring the information that is most necessary for
the use case, such as city planning. The goal in data aggregation should be
to protect individual privacy while providing useful context to understand and
improve the transportation demands of cities.
Uber’s Movement tool, launched in January 2017, provides anonymized
data to help urban planning in cities around the world
24
. The tool is free
and open to anyone interested in using it, and provides travel times
between any two points in a city at any time of day. The data provided
is an aggregation of many individuals’ trips, anonymized and compiled
to remove a level of granularity that would violate individuals’ privacy or
compromise Uber’s competitive advantage.
Uber Movement aggregates data in a way that is effective to visualize
traffic flows and major congestion points; Uber Movement’s data
aggregates travel time by city zone, which is useful to understand the
impact of major events, from road closures to natural disasters. However,
city and transit planners have pointed out that the tool leaves out data
that would be most useful to them, such as the most common pickup
and dropoff
25
locations in a city. This example represents the conflict
between private interests and the public benefits of data sharing. Despite
mixed reviews of the tool’s usefulness, however, Uber Movement presents
a good example of how data aggregation can be used to increase
public access to data that otherwise may not be shared at all due to
privacy and competition concerns.
Case Study: Uber Movement Tool 29
With a growing population and increasing demand for transport, India stands
to benefit tremendously from the collection and sharing of mobility data. Using
and promoting open data increases the availability of data to a wider audience.
According to the National Informatics Centre, “A dataset is said to be open if
anyone is free to use, reuse, and redistribute it—Open Data shall be machine
readable and it should also be easily accessible
26
.” Open data is inherently free
and available to the public.
Open data helps to enable the many potential use cases for data analytics in
the transport sector, as outlined in section 3. The Government of India has taken
steps toward increasing the amount of and access to open data in an effort to
unlock the many benefits of data analytics. While this supports multiple sectors,
the mobility sector in particular has the potential to benefit tremendously from
this initiative.
The Government of India, and a number of other countries and cities around the
world, host online open data portals with datasets for many sectors, including
buildings, education, public safety, and transportation. The goal of these portals
is to provide easy access to open data and information in order to “spark
innovation, promote public collaboration, increase government transparency,
and inform decision making
27
.” These portals have been successful in
collecting and hosting a wide range of data, and presenting it in an easy-to-
use and accessible manner. These central data portals can add a lot of value
by aggregating datasets in a single location and making them easily available
6.0
/////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
Open data
+++ 30
The Government of India has launched Open Government Data (OGD)
Platform (data.gov.in) to support the Open Data Initiative for nation-
wide data sharing. This initiative was launched as part of the National
Data Sharing and Accessibility Policy (NDSAP) as well as the Digital
India Initiative. OGD platform provides open access to datasets,
documents, services, tools, and applications collected by various
ministries, departments, and organizations of the Government of India
for public use. The main goal of OGD is to provide open access to the
data generated through public funds and to enhance transparency,
accountability, citizen engagement, collaboration, better governance,
decision-making, and innovation
28
. Currently, there are over 2 lakh data
points, which have cumulatively been viewed over 17 million times.
Case Study: Open Government Data (OGD) Platform in India
to any party who can derive value from them. The U.S. cities of Austin and
Chicago are two other notable global examples of well-organized, frequently
updated, and comprehensive open data portals; both websites have categories
for “transportation” with dozens of datasets, ranging from traffic camera and
sensor data to non-motorized transit infrastructure maps and plans to public
transit ridership numbers. 31
7.0
/////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
Conclusion
+++
Data has an important role to play in helping India achieve a mobility system
that is clean, efficient, and adequately supports the mobility needs of its
citizens. To ensure the maximum benefit of mobility data, steps should be
taken to collect and share data and ensure that the data collected by different
parties is made available as much as possible. Effective communication between
data owners and potential beneficiaries is at the core of reaching this outcome;
to effectively communicate and collaborate, all stakeholders involved must
understand the landscape of the system and the motivations and risks of the
various parties involved.
With the rapid pace of technological developments and urbanization, India
should be developing and enhancing its transportation system with an eye to
the future: as the nature of cities and transport changes, mobility plans need to
be designed to be flexible and adaptable to keep up with an evolving landscape.
Building a comprehensive practice of data collection and sharing in India will
form a strong foundation to support new technologies and innovations, such
as blockchain, in the mobility space.
Thoughtfully constructing a supportive framework for collecting and sharing
mobility data will enable India to dramatically improve the efficiency and strength
of its mobility system as well as urban planning and regulation, resulting in
communities that are cleaner, safer, and better support the needs of their citizens. 32
Government of India, “Personal Data Protection Bill 2018,” 2018. http://meity.
gov.in/writereaddata/files/Personal_Data_Protection_Bill%2C2018_0.pdf
Digital Matatus, 2018. http://www.digitalmatatus.com/map.html
Carlin, Kelly, Bodhi Rader, and Greg Rucks, “Interoperable Transit Data:
Enabling a Shift to Mobility as a Service,” Rocky Mountain Institute,
October 2015. https://rmi.org/wp-content/uploads/2017/03/Mobility-
InteroperableTransitData-Report.pdf
“About UPI API,” IndiaStack. http://indiastack.org/upi/
Carlin, Kelly, Bodhi Rader, and Greg Rucks, “Interoperable Transit Data:
Enabling a Shift to Mobility as a Service,” Rocky Mountain Institute,
October 2015. https://rmi.org/wp-content/uploads/2017/03/Mobility-
InteroperableTransitData-Report.pdf
“BMTC: Easy Travel Information Planner,” Google Play, 2018. https://play.
google.com/store/apps/details?id=com.bmtc.mybmtc&hl=en
Sonal Shah, Kalpana Viswanath, Sonali Vyas and Shreya Gadepalli, “Women
and Transport in Indian Cities,” ITDP and Safetipin, December 2017. https://
smartnet.niua.org/sites/default/files/resources/171215_women-and-transport-
in-indian-cities_final.pdf
1
2
3
4
5
6
7
8.0
/////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
Citations
+++ 33
“The World Bank Launches New Open Transport Partnership to Improve
Transportation through Open Data,” The World Bank, December 19, 2016.
http://www.worldbank.org/en/news/press-release/2016/12/19/the-world-
bank-launches-new-open-transport-partnership-to-improve-transportation-
through-open-data
Stern, Noah, “Waze Partners With Local Governments to Enhance Mobility,”
Government Technology, February 12, 2016. http://www.govtech.com/dc/
articles/waze-partners-with-local-governments-to-enhance-mobility.html
Ibid.
“Big Data Analytics: What it is and why it matters,” SAS. https://www.sas.
com/en_us/insights/analytics/big-data-analytics.html
“How Big Data & Advanced Analytics Will Change Transportation,” Datafloq,
December 2016. https://datafloq.com/read/big-data-advanced-analytics-
transportation/1747
Winnesota, “How Blockchain is Revolutionizing the World of Transportation
and Logistics,” 2017. https://www.winnesota.com/blockchain
Cohen, Boyd, “How to power shared mobility startups with blockchain
technology,“ Shareable, March 2018. https://www.shareable.net/blog/
empowering-shared-mobility-startups-with-internet-of-mobility-blockchain-
technology
Sandner, Phillip, “Analysis of Blockchain Technology in the Mobility Sector,”
Medium, April 2018. https://medium.com/@philippsandner/analysis-of-
blockchain-technology-in-the-mobility-sector-1078e429615f
Cornet, Andreas et al, “The road to artificial intelligence in mobility—smart
moves required,” McKinsey & Company, September 2017. https://www.
mckinsey.com/industries/automotive-and-assembly/our-insights/the-road-to-
artificial-intelligence-in-mobility-smart-moves-required
“Personally Identifiable Information (PII),” Cornell Law School. https://www.
law.cornell.edu/cfr/text/2/200.79
Srikrisna, B.N. “A Free and Fair Digital Economy Protecting Privacy,
Empowering Indians,” 2018. https://www.medianama.com/wp-content/
uploads/Data_Protection_Committee_Report-comp.pdf
Mandhani, Apoorva, “Govt. Releases Justice Srikrishna Committee Report
And Draft Personal Data Protection Bill, 2018,” July 2018. https://www.
livelaw.in/justice-srikrishna-committee-releases-report-and-personal-data-
protection-bill-2018/
8
9
10
11
12
13
14
15
16
17
18
19 34
Singh, Pratap Vikram,“Seven Principles for Data Protection: BN
Srikrishna Committee,” Governance Now, 27 November 2017. http://
www.governancenow.com/news/regular-story/seven-principles-for-data-
protection-bn-srikrishna-committee-
Government of India, “Personal Data Protection Bill 2018,” 2018. http://meity.
gov.in/writereaddata/files/Personal_Data_Protection_Bill%2C2018_0.pdf
Agarwal, Surabhi, “Government seeks public comments on Data
protection bill; September 10 deadline,” The Economic Times, August
2018. https://economictimes.indiatimes.com/tech/internet/government-
seeks-public-comments-on-data-protection-bill-september-10-deadline/
articleshow/65404228.cms
Crane, Jackson and Greg Rucks, “A Consortium Approach to Transit Data
Interoperability,” Rocky Mountain Institute, 2016. https://rmi.org/wp-content/
uploads/2017/03/consortium_approach_to_ITD_report2016.pdf
Gilbertson, Jordan and Salzberg, Andrew, “Introducing Uber Movement,”
Uber Newsroom, January 9, 2017. https://www.uber.com/newsroom/
introducing-uber-movement-2/
Peterson, Andrea and Scola, Nancy, “Data is Uber’s business. But
protecting it may be its biggest weakness,” The Washington Post,
November 18, 2014. https://www.washingtonpost.com/news/the-switch/
wp/2014/11/18/data-is-ubers-business-but-protecting-it-may-be-its-largest-
weakness/?noredirect=on&utm_term=.5b9ef8330e4f
“Open Government Data,” Government of India Ministry of Electronics and
Information Technology. http://sdtckochi.nic.in/OGD.aspx
“The official City of Austin open data portal.” https://data.austintexas.gov
“Open Data,” Ministry of Electronics and Information Technology, June 2016.
http://meity.gov.in/open-data
20
21
22
23
24
25
26
27
28 35
MOBILITY
IMPROVING PASSENGER TRANSPORTATION
THROUGH DATA Authors
NITI Aayog
Pranjal Dubey
Shikha Juyal
Ishita Kandra
Abhishek Saxena
Shweta Sharma
Anil Srivastava
Rocky Mountain Institute
Akshima Ghate
Emily Goldfield
Robert McIntosh
Greg Rucks
Samhita Shiledar
Clay Stranger
Independent contributor
Anand Shah
Contacts
NITI Aayog: adv.niti@gov.in
www.niti.gov.in
RMI: ocs@rmi.org
www.rmi.org
Acknowledgements
The authors would like to thank Dr. Rajiv Kumar,
Vice Chairman, NITI Aayog and Mr. Amitabh Kant,
CEO, NITI Aayog for their support that made this
report possible.
Suggested Citation
NITI Aayog and Rocky Mountain Institute.
Data-driven Mobility: Improving Passenger
Transportation Through Data. 2018.
“The views and opinions expressed in this
document are those of the authors and do not
necessarily reflect the positions of the institutions
or governments. While every effort has been made
to verify the data and information contained in this
report, any mistakes and omissions are attributed
solely to the authors and not to the organization
they represent.”
Authors and
acknowledgements Outline
1.0 INTRODUCTION
2.0 THE LANDSCAPE OF MOBILITY DATA
2.1 Defining mobility data and related terms
2.2 Primary stakeholder groups
3.0 MOBILITY DATA USE CASES
3.1 Traveller use cases
Mobility as a service
Ancillary trip information
3.2 City and government use cases
Safety and security
Transportation, route, and infrastructure planning
Real-time system management
Enforcement and regulation
3.3 Researcher use case
3.4 Using data to enable emerging technologies
Big data analytics
Blockchain
Artificial Intelligence (AI)
4.0 DATA ACQUISITION FOR MOBILITY USE CASES
4.1 Process of data acquisition for specific use cases
4.2 Challenges associated with data acquisition
Privacy and data security
Poor quality and incomplete data
Acquiring data from private data owners
5.0 DATA AGGREGATION
6.0 OPEN DATA
7.0 CONCLUSION
8.0 CITATIONS 6
1.0
/////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
Introduction
+++
Mobility is a key economic driver: moving people and goods is at the core of a
well-functioning and prosperous society. Increasing the efficiency and quality
of a country’s mobility system helps support a stronger economy and a higher
standard of living for its citizens.
Data is an essential piece for unlocking maximum value within a transportation
system. Data analytics, and data sharing between organizations, has the
potential to create more efficient passenger mobility, as well as allow for optimal
designing of transit routes and services, infrastructure, and regulations.
Data collection is also essential to enabling the use of a number of emerging
technologies in the mobility space, such as blockchain and artificial
intelligence. Using data to optimize commutes and transit infrastructure will
lead to lower levels of congestion, reduced tailpipe emissions, and less time
spent in transit, resulting in cities that are cleaner, safer, better designed, and
more economically prosperous.
Maximizing the benefits of data analytics for the mobility sector requires the
sharing of data between parties. Companies, public transit agencies, and
commuters are generating and collecting huge amounts of transport data.
However, this data is siloed between organizations and individuals, and
recorded with different standards and formats. Absence of transport data
ecosystem is an important concern in India as well. Encouraging and enabling
data collection and data sharing will allow linking of currently disparate
models of transport and unlock tremendous value within the transportation
systems in the country. 7
Data can enable travellers to benefit from seamless multimodal transportation,
and make it as convenient as private vehicle use. This paper describes how
data can be used to make passenger mobility more efficient, beginning by
outlining the relevant mobility data stakeholders and use cases and then
analyzing the flow of data in this system. By making possible such a data-
enabled future for transportation, mobility assets can be better utilized and
integrated, and India can bolster economic growth while building cleaner,
more liveable communities. 8
2.0
/////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
The landscape of
mobility data
+++
Mobility data encompasses a wide range of mobility-related information. The
landscape of mobility data involves multiple stakeholders ranging from data
owners and aggregators to data users. The different types of data owners and
beneficiaries, and the network of relationships between members of these two
key groups, is a core piece of understanding for crafting value propositions to
different stakeholders for varying use cases of data.
2.1 Defining mobility data and related terms
Data can be defined as “representation of information, facts, concepts, opinions,
or instructions in a manner suitable for communication, interpretation, or
processing by humans or by automated means
1
.” Mobility data can include a
wide range of data related to transport. This includes data of public transit
agencies, private mobility solution providers, and individual citizens. In addition
to information about transportation assets and trips, mobility data can also
include data that affects or goes in tandem with mobility, such as weather data
(can affect traffic, etc.), pollution and air quality data, and traffic violations data.
India’s proposed Personal Data Protection Act, described in more detail in
section 4.2, defines personal data. According to the proposed Act, personal
data refers to “data about or relating to a natural person who is directly or
indirectly identifiable, having regard to any characteristic, trait, attribute or any
other feature of the identity of such natural person, or any combination of such
features, or any combination of such features with any other information.” It goes
on to define the processing of personal data as “an operation or set of operations 9
performed on personal data, and may include operations such as collection,
recording, organisation, structuring, storage, adaptation, alteration, retrieval, use,
alignment or combination, indexing, disclosure by transmission, dissemination
or otherwise making available, restriction, erasure or destruction.”
2.2 Primary stakeholder groups
At the highest level, the three categories of stakeholders most critical to the
data landscape are data owners, beneficiaries, and government.
Data owners are companies, organizations, and individuals that produce and
own datasets. Beneficiaries are groups and individuals that can benefit from
using the data owned by the data owners. In addition to both benefiting from and
owning data, government can play a key role in enabling the interactions between
these two stakeholder groups and protecting their interests. It should be noted
that these categories are not rigid: data can flow from owners to beneficiaries,
but it can also flow within each of these categories, and some beneficiaries may
also qualify as data owners. There are many cases of business to business data
sharing; for example, Uber uses data from Google Maps to provide its own service.
Within the category of data ownership, there is one primary distinction: public
versus private data ownership. Public data is shared openly, irrespective of how
it is generated. Private data is kept within an organization and not shared with
the public. Data owners may have both private and public datasets. Data
generated by private service providers is important in India because of a
significant presence of privately-owned or operated transport services that
meet the passenger mobility needs in cities. 10
Matatus are the privately owned mini-buses in Nairobi, Kenya. Matatus
are very popular because they are cheap and convenient, but have
had the same problems as a typical informal transit system: difficulty
to access timetables, routes, and stops. The Digital Matatus Project
is designed to solve these problems using digitization. Digital Matatus
Project is a joint project between Columbia University, The Rockefeller
Foundation, and the technology sector in Nairobi, and is focusing on
developing a mobile routing application and new transit map for the city
using cellphone technology. First, students from the University of
Nairobi rode all of the bus route using an app that collected data in a
General Transit Feed Specification (GTFS) compatible format. Data
points included routes, stops, and visual notations (signs and shelters).
The data was then processed and released in the form of a paper map
and transit apps. Some of the transit apps include Ma3route, Flashcast,
sonar, digtitalmatatu, and matatumap. The City of Nairobi has recognised
the importance of the digitization and is using this data to create a new
trip planning tool for the city. Learning from the success story of Nairobi,
several other cities in Africa are planning to map their informal transit
sector as well. This case study may prove a relevant example for
addressing similar challenges related to the lack of data for India’s
informal transit system.
Case study: Digital Matatus Project
2
A large portion of data owned by private data owners is generated by the
movement of individual travellers. If the mobility service is provided through a
mobile application, travellers typically agree to allow the company to collect
and own the data that the traveller generates. A significant portion of private
mobility service providers in India do not have a digital presence. Despite this,
these companies are generating significant amounts of data that can be useful.
The main challenge in the latter case is finding out ways to collect the data
from these service provider’s operations.
Each traveller could also be considered a private data owner. Individually, this
has much less value than the aggregated data from many travellers, making
companies with large aggregated datasets much more relevant data owners
than individual travellers in today’s data landscape; however, new technologies
have the potential to change this relationship. Blockchain, for example,
disintermediates the aggregators of data and could allow travellers to have more
control over and more direct compensation for sharing the data they generate.
Data beneficiaries can be categorized into three main stakeholder groups:
travellers, cities and governments, and researchers. Each of these beneficiary
groups correlates with a set of use cases for mobility data, which are explored
in more detail in section 3. 11
» Travellers are any individuals needing to move from their current location
to an end destination. In this case, the primary use case for open mobility
data would be to optimize the route and mode of transport for increased
efficiency and ease of use.
» Cities and governments include planners, regulators, and operators in
charge of system-level design and operations for a city’s transportation
network, or policy design and operations at a state or national level. The
use cases for cities for mobility data include transportation planning
(infrastructure and services planning), improving safety, and city planning.
» Researchers include any organizations or individuals conducting research
in the area of mobility. Greater access to data allows researchers greater
insight into the mobility system to better analyze what is happening and draw
more data-based conclusions, which could in turn be of use to policy makers
and cities.
Travellers represent the individual benefits of data analytics and data sharing,
while cities represent system-level benefits. For each of these beneficiaries,
the type of data required to achieve the desired use case varies. For example,
a traveller needs data that will help him optimize his trip to his destination,
which likely only requires real-time data or potentially short-term projections
of his transport options when he chooses to depart. In contrast, a city planner
who is designing infrastructure for the future would benefit from historic
transit data so that she can examine past trends. 12
The use cases for mobility data can be examined from the perspective of each
beneficiary category. The type of data and analytics required varies between the
use case.
3.1 Traveller use cases
There are many potential use cases that apply to improving transportation
efficiency for individual travellers. Many of these can be grouped under the
general category of “mobility as a service.”
Mobility as a service
Mobility as a Service, or MaaS, refers to the technology-enabled, on-demand
availability of multimodal trip options, including multimodal trip planning and
seamless payment. Mobility as a service is an alternative to private vehicle
ownership; travellers should be able to order a ride to wherever they need to go,
at the time they need and in whatever size or type of vehicle meets their needs.
Under the MaaS paradigm, the entire transportation system functions as a
cooperative, interconnected system to meet travellers’ needs through a variety
of transport modes. To achieve this, infrastructure, technology platforms,
payment, transportation services, and data analysis must be capable of working
together
3
. Mobility data—as well as organizations’ willingness to share data—is
the key to unlocking MaaS. This paradigm has the potential to benefit both
travellers as well as transit providers who share data: travellers can enjoy
increased options and trip efficiency; while for transit providers, compiled data
3.0
/////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
Mobility data use
cases
+++ 13
can help open up markets and customer bases that are currently
disaggregated due to the lack of connection between service providers.
There are several data-supported elements that go into MaaS, several of which
are described in more detail below.
» Multimodal trip planning
A primary element of MaaS is the ability to see all available modes of
transport, and both choose the mode that is most optimal for the situation
and be able to easily link various modes of transport to get to the destination.
For example, a traveller could go onto a single platform and enter his
destination and be shown the best option for getting there out of all
available modes, which may include a portion of the trip using one mode
and another portion using a different mode. Currently, travellers lack easy
access to information about all their transportation options. Mobility data—
as well as transit providers’ willingness to share data—is essential to
enabling multimodal trip planning.
» Seamless Payment
Enabling travellers to seamlessly pay the various transportation providers
through a single portal will increase the accessibility of transport options
and promote multimodal trips. Implementing seamless payment requires
collection and integration of transit data, and also relies on transit companies
being willing to share data.
India Stack’s Unified Payment Interface
4
, which allows all bank account
holders in India to send and receive money instantly from their smartphones,
may uniquely position India to be able to quickly adopt a multimodal transit
application with integrated payment.
» Real-time mode connectivity and optimization
Mobility data can enable the real-time optimization of travel plans around
changing factors, such as weather and traffic, as well as travellers’ preferences
(e.g. least expensive, shortest time, etc.). For example, an algorithm can
take into account traffic data to reroute a traveller around an accident or
congested area, or update expected travel time based on weather conditions. 14
01
0203
Multimodal trip
planning
Seamless one-stop
payment
Mode connectivity,
real-time route
optimization
Figure 1: Three interdependent components of “mobility as a service.
5
”
Ancillary trip information
In addition to MaaS, there are a number of other services that access to data
can provide to increase the efficiency and ease of a traveller’s journey. This could
include accessing data on real-time conditions (i.e. traffic, weather, accidents, etc.),
information about interesting landmarks along the route, or any other sort of
ancillary information to enhance a trip.
Figure 2: Screenshot from BMTC’s trip planner app.
Case study: Bangalore’s bus tracking application
6 15
Bengaluru Metropolitan Transport Corporation (BMTC) has piloted a
mobile application using Intelligent Transportation System (ITS) to give
fixed and real-time information about BMTC buses to the passengers.
Users can easily access the trip planner tool to see the bus timetables
and route maps. BMTC buses are equipped with GPS and give a
real-time location of buses and estimated time of arrival. The mobile
application also gives additional information about the approaching
buses at the bus stop, such as bus number and the platform details on
which the bus is going to arrive. The app was developed in order to make
BMTC bus experience more comfortable for the passengers and promote
public transportation as the preferred mode of transport. Similar mobile
applications are also being developed in other cities in India, such as
Ahmedabad.
3.2 City and government use cases
There are several use cases that apply to optimizing passenger mobility from a
systems-level. Cities and governments around the world are realizing the value
of using mobility data to improve system safety and optimize transit planning
and city design around the efficient movement of people and goods. Some of
these use cases are discussed below.
Safety and security
Data can enable improved safety and security within the transportation system
in a number of manners. For example, increased access to data allows cities
to see where accident hotspots are, thus enabling them to respond more quickly
and also understand the issues in those areas. With increased understanding
of when and how accidents occur, cities can ensure a greater level of safety for
their citizens by responding faster when incidents occur and developing
solutions to systemic concerns.
One particular area of concern in India is women’s safety in the transportation
sector. Many women feel unsafe traveling alone, and frequently avoid using
public transport
7
. Improved tracking of vehicles and verification of drivers and
vehicles that are deemed safe are some examples of how data can allow women
to feel safer using transportation. Safety equipment such as panic buttons, GPS,
and CCTVs could be installed in vehicles to improve safety—and tracking of
crimes and harassment, should they occur—for women. 16
In 2016, Grab, Easy Taxi, and Le.Taxi formed a partnership with the
World Bank and other organizations to share the traffic data from their
drivers’ GPS streams through an open data license. The three ridesharing
companies combined cover more than 30 countries and serve millions
of people
8
. The partnership’s goal in sharing their data is to provide cities,
particularly in developing countries, with the tools to address traffic
congestion and plan for better mobility by developing better, evidence-
based solutions to road safety and traffic challenges. This unprecedented
multi-provider agreement to share data is an important counterexample
to the argument that sharing data undermines competitive advantage.
A core reason that this initiative was able to launch successfully is that
the private companies involved knew exactly what would happen with the
data, what type of interface people would see when the data was published,
and had assurance that the data would not be used for other purposes.
Case study: Open Transport Partnership
Transportation, route, and infrastructure planning
Transportation planners can leverage data analytics to better design and
maintain routes, public transit, and mobility infrastructure. There are many
specific use cases within this category: Analyzing traffic and commute patterns
allows planners to understand where to build infrastructure (including non-
motorized transport infrastructure) and add transit routes to ease stress in the
most highly-trafficked areas. Data analytics can aid planners in minimizing
congestion in cities by identifying the primary cause (i.e. poorly-timed signals,
insufficient parking, etc.). Data can also aid in monitoring the structural health
of transportation infrastructure, such as bridges and overpasses, and related
data, such as water and flood data, can inform how and where to build
transportation infrastructure.
Real-time system management
Data can aid in the real-time functioning of the mobility system. Operators can
remotely monitor the transportation system and manage system operations.
Increased access to data will give operators more real-time information that will
help them ensure the smooth functioning of the transportation system. One
way of collecting this data is by installing sensors to detect traffic and vehicle
movement. Additionally, as more and more private companies are collecting
data on travellers’ movement through GPS, cites are beginning to see an
opportunity to partner with these companies to acquire data. 17
Waze, a Google-owned traffic and navigation app, launched its
Connected Citizens Program in late 2014 as a two-way exchange of
information with cities, with Waze providing city partners with user
driving information in return for real-time and advanced notice of
construction and road closures. This partnership benefits both parties:
Waze sees the program as a way it can grow and improve its services,
while city and state governments can easily expand their view of their
roads and streets without having to invest in more road sensors and
traffic cameras. As part of the Connected Citizens Program, Waze
has partnered with over 100 cities around the world, who can now use
Waze’s data to help with city planning and transportation regulation.
In Rio de Janeiro, for example, the Waze API is completely embedded
in the city’s Control Center, to help with day-to-day monitoring of road
conditions
9
. In Boston, the city’s central Traffic Management Center uses
Waze’s real-time data to change the traffic signals in 550 of the city’s
intersections as needed to reduce congestion. Washington, D.C. has
used data supplied by Waze to aid in the city’s “war on potholes
10
.” In
all of these cases, the cities were able to increase access to data without
investing in additional monitoring equipment.
This example also demonstrates that private companies are more
willing to share data when the use case is very clear (i.e. city planning
and regulating) and the use of the data is transparent.
Case study: Waze Connected Citizens Program
Enforcement and regulation
Increased access to data affords regulators better visibility into the transportation
system, allowing them to improve the enforcement of regulations, and develop
new and modify existing regulations to ensure a smoothly functioning system.
For example, data can allow regulators to see where infractions incur most
frequently and consequently improve their enforcement ability in those areas;
it can also aid in the collection of road tolls and parking fees.
3.3 Researcher use case
The researcher use case includes anyone conducting mobility-related research,
such as academic institutions and think tanks. Increased access to data gives
these groups and individuals a greater ability to conduct in-depth quantitative
analysis on the transportation system, in order to draw conclusions about and
make recommendations for the mobility system. This allows researchers to
provide cities and other beneficiaries with more detailed insights, such as the
cause and effect relationships between investments and impact, in order to
inform what a city should invest in to improve its transportation system. 18
3.4 Using data to enable emerging technologies
A number of emerging technologies have potential applications in the mobility
sector. Increasing the collection and use of mobility data will help unlock the
potential of these technologies. This section examines some of the most notable
emerging technologies—big data analytics, blockchain, and artificial intelligence—
with respect to their potential applications in the mobility space.
Big data analytics
Data analytics refers to the practice of examining large amounts of data to
discover patterns, correlations, and other insights
11
. Data analytics and
technologies can allow businesses, governments, and other organizations to
analyze datasets and draw conclusions to help them make informed decisions.
Big data analytics includes a wide range of specific tools and techniques that
can be used to gather insights from data; some specific techniques include
data mining, predictive analytics, and text mining.
For mobility, big data analytics can help with planning and managing transportation
networks and designing and optimizing services to meet transportation needs.
Both the private and public sector can benefit from big data analytics for mobility.
Some areas where big data analytics can significantly impact the transportation
sector include
12
:
» Optimization of transit schedules by analyzing demand and predicting the
impact of maintenance, road work, congestion, and accidents,
» Increased safety and reduced environmental impact,
» Fleet optimization and predictive maintenance through real-time view of
fleet operating conditions, statistics around usage and weather patterns,
and maintenance cycles, and
» Freight movement and routing optimization.
Blockchain
Blockchain stands ready to revolutionize significant portions of the transportation
space and could be particularly applicable in the Indian context. Blockchain
allows for more decentralized transactions through a shared, standardized
ledger system, which could allow decentralized passenger and freight systems
in India to improve their efficiency without the oversight of a central governing
body. Some examples of this include:
» Improved tracking: Currently most elements of the transportation system—
such as freight, vehicles, and jobs—are tracked through a combination of
brokers, middlemen, and paperwork. Goods can be lost, vehicles unaccounted
for, or jobs duplicated or lost due to lack of accurate tracking. By using
blockchain, these items can be tracked on a shared, secured, decentralized,
and trusted standardized ledger system
13
. 19
» Open-source mobility as a service: Utilization of blockchain supports
multiple technologies on a shared, distributed, and transparent platform.
Local startups could provide customized services or fleets, and these could
be offered through smartphone applications. This would allow companies
to provide more unique, customized, and localized services while allowing
startups to take advantage of existing platforms and customer bases
14
.
» Secure data logging and tracking: For vehicle sharing, customers often
have a set of preferences and desires they wish in both their vehicle and
their fellow carsharing customers. Blockchain enables a customer to have
a secure area for these preferences to be logged, as well as previous
experiences in carsharing. Through blockchain encryption, this is a permanent,
immutable ledger. This could also be used for insurance purposes, as a
vehicle could keep a constant log of its experiences in order to document
fault in an accident (similar to the “black box” installed in most airplanes
for the purpose of facilitating the investigation of aviation accidents and
incidents
15
).
Artificial Intelligence (AI)
Artificial intelligence is expected to revolutionize many traditional industries
over the next decade. There are a number of areas in which artificial
intelligence is impacting the mobility sector, some of which are outlined below.
» Advanced Driver Assistance Systems (ADAS): Artificial intelligence can
help to automate, adapt, and enhance vehicle systems for safety and
better driving, to reduce the potential for human error. These technologies
may increase safety and reduce collisions and accidents by alerting the
driver to potential problems, implementing safeguards, and taking over
control of the vehicle to varying degrees. There are multiple levels of
automation in advanced driver assistance systems, the most mild being
a system that issues warnings to a human driver, and the most intensive
being fully autonomous vehicles. 20
Pursuit of machine learning in the mobility space
Companies worldwide are pursuing either new mobility industry areas
(such as autonomous driving and predictive maintenance) or old areas
with improvement through automation (fleet management, infrastructure
planning, supply and manufacturing management, quality assurance,
and in-vehicle experience). These companies are traditional mobility
companies, startups, and outside industries entering into the mobility
space. Given these trends, India could explore machine learning and
automation in order to stay competitive on the global scale
16
.
Figure 3: McKinsey 2017 assessment of companies pursuing mobility-related machine learning markets
» Supply chain management: In the logistics space, the use of artificial
intelligence will lead to more effective distribution of resources in the
freight sector as supplies are more effectively moved where they need
to go at the right time. Locations and movement of vehicles will be more
effectively determined.
» Infrastructure planning: As big data becomes more available, cities will
start to rely on learning machines and artificial intelligence to process
these sets into useful outputs. Artificial intelligence will be used to predict
where and how transportation should be built, determine effective mobility
corridors, and recommend infrastructure projects. 21
4.0
/////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
Data acquisition for mobility
use cases
+++
This section explores the process for acquiring data, as well as the challenges
associated with it, to support mobility-specific use cases. There are several
components that play a role in data acquisition, and the following sections explore
ways of addressing and minimizing some of the key challenges.
Data acquisition can be done in a primary manner, e.g. through an organization’s
own sensors or user application, or in a secondary manner, e.g. by acquiring
existing datasets from other parties. In the case of primary data collection,
organizations often collect data regardless of whether there is an immediate
use case for it, under the assumption that owning more data is an asset and
may provide useful insights in the future. With secondary data collection,
however, it can be helpful to first identify the desired use case before making
a request for data.
More and more transit organizations, both public and private, as well as
individuals, are collecting significant amounts of transit-related data. The range,
scope, and volume of data collection is expanding. However, a significant
number of mobility service providers in India still do not collect data, do not
have the ability to collect data, or do not operate through digital platforms to
facilitate the easy collection of data. Technology can be an enabler in cases
of such mobility providers to enable data collection. For example, many cities
are using sensors to collect data to improve signal synchronization.
This increase in data presents a massive opportunity to better integrate existing
transport systems, optimize transit options to users’ needs, and plan and regulate 22
cities to best support mobility patterns. The potential value that mobility data
can unlock has led some analysts to refer to data as “the new form of oil” for
transport systems. However, there are still many barriers to data acquisition that
need to be addressed in order to realize the full potential value of using data
analytics to improve mobility.
4.1 Process of data acquisition for specific use cases
When the desired use case for data is known, there are three primary steps
for acquiring the necessary data: identifying the sources for the required
data; surveying what data is available and identifying what gaps remain;
and collecting additional data or acquiring data from other data owners to
fill those gaps.
1. Identifying necessary data
The first step is to identify what sort of data is needed to fit the desired use
case. This is an essential step; thoroughly analyzing what types of datasets
are needed—and conversely, what data is not needed—will save time and
allow the organization to make specific asks for data, if additional datasets
are required from other data owners.
Example: Route Planning
A city government wants to use data to inform public transit route planning.
To do this, a team of transportation planners determines that they want
historic data on commute patterns and traffic congestion in order to
identify the most highly trafficked routes to determine where new mass
transit routes can be developed to ease congestion. They recognize
that they do not need real-time data or ancillary data such as weather
information, trip fares, etc.
2. Determining what data is available and what gaps remain
Once the necessary datasets have been identified, the organization can
survey which data are already available to them, either through data that they
already own or have access to, or through publicly available data. There are
already many sources of open data; surveying all of the data that is already
available will prevent an organization from collecting redundant data or
making unnecessary data requests. Once the organization has determined
which of the datasets are already available to them, they can then identify
where gaps still exist and what sort of data could be acquired to fill these
gaps.
Example: Route Planning
The transportation planners note that they have open access to historic
data on the ridership of current public transit options from the public transit
agency, and they also have data on traffic patterns from the several traffic
sensors that have been installed throughout the city. They then determine
that their existing traffic pattern data does not have as much detail as they 23
would like, and leaves off a few key areas of the city. They decide to acquire
additional data for commute and traffic patterns in the city.
3. Collecting remaining data
To fill in the gaps identified, an organization or individual has two options:
1. Acquire the data from another data owner: If someone already owns the
data needed, one option is to acquire the data from that data owner.
2. Collect the data: If no one already owns the needed data, or if the data
owner is unwilling to share the data, then the organization must devise a
way to collect the data themselves.
Example: Route Planning
The team of transportation planners evaluates the options of installing more
traffic sensors to collect their own data on commute and traffic patterns
or approach other data owners to acquire data that already exists. They
decide to take a combined approach of installing additional sensors, as well
as working with shared mobility providers to acquire existing datasets.
4.2 Challenges associated with data acquisition
There are a number of challenges associated with data acquisition,ranging from
the quality and availability of data to concerns around privacy and security once
data is collected or acquired. The following sections highlight some of these
key challenges. 24
Privacy and data security
As big data becomes more prevalent and necessary in developing effective
mobility systems, concerns have been raised about protecting individuals’ privacy.
Personally identifiable information (PII) is generally considered information that
can be used on its own or with other information to identify, contact, or locate
an individual person
17
. Individuals are generally concerned about leaks of PII
because they can result in negative consequences.
Data does not need to include personal identifiers to be useful in many cases.
Metadata—data that describes and gives information about other data and is
not attached to a particular individual—can be extremely useful in determining
how mobility systems are being used, by whom, and when, to help in
determining which systems should be built or supported more effectively.
The proposed Personal Data Protection Act of 2018 prohibits the processing
of sensitive personal data without explicit consent. Thus, any company or
government agency gaining access to data points about mobility users will
need to do one—or both—of two things:
» Remove any PII from the data before allowing it to be used publicly.
» If data is to be transferred, ensure the process is completely secure
so that PII remains only in the possession of the parties that have been
authorized to own or access it.
For the first solution, personal identifiers must be reliably removed from the
data so that individuals’ privacy is not compromised. Though many datasets
include PII, personal identifiers are rarely necessary for transportation planning.
For the second solution, data containing PII (or ideally all data) should be sent
through secure channels when it is transferred for use. If PII must be kept and
transferred, it should be done in such a way as to eliminate the possibility of
access by outside parties.
In July 2018, the Government of India released the drafts of two major
reports on data protection: “A Free and Fair Digital Economy”
by the Chairmanship of Justice B. N. Srikrishna, and the Personal
Data Protection Act, 2018 by the Ministry of Electronics and Information
Technology. The goal of these reports is to introduce standards for
data handling, processing, and privacy.
India’s draft data protection and privacy guidelines: Personal Data
Protection Act and “A Free and Fair Digital Economy” 25
B. N. Srikrishna’s report on data protection
18
defines personal data based
on its identifiability. The report elaborates: “Identifiability in circumstances
where the individual is directly identifiable from the presence of
direct identifiers such as names is perhaps uncontroversial and will
obviously be included within the scope of any definition of personal
data. The definition should also, in addition, apply to contexts where
an individual may be indirectly identifiable from data that contains
indirect identifiers
19
.” The report recommends that a data protection
law should be set up, that will be responsible for the enforcement and
effective implementation of the definition of personal data and sensitive
personal data, legal affairs, policy and standard setting, research, and
awareness.
B.N. Srikrishna’s report outlines seven key principles for effectively
designing a privacy policy. At a high level, the principles are that the
policy should be:
1) Technology agnostic
2) Holistic
3) Include language on informed consent
4) Recommend data minimization
5) Assign controller accountability
6) Structure enforcement
7) Include deterrent penalties
20
The Draft Personal Data Protection Act
21
focuses on the fair and
reasonable processing of data. The Draft Act specifies that there must
be a clear, specific, and lawful purpose behind data processing and
stipulates that only necessary data should be collected. The Act
proposes to replace traditional terms such as data controller and data
subject (i.e. person whose data is being collected) with data fiduciary
and data principal, respectively. The data fiduciary would be responsible
for ensuring the personal data quality as well as complete, accurate,
and not misleading data processing. Sensitive personal data may be
processed on the basis of explicit consent.
The reports are currently under review by the Ministry of Electronics
and Information Technology. The government is also seeking comments
on the drafts from relevant stakeholders and the public
22
. 26
Poor quality and incomplete data
Data collection in India is insufficient and in some cases the data that is
collected is of poor quality or incomplete. A dataset may be considered of poor
quality if it fails to provide adequate or accurate information. For example,
real-time location data is often inaccurate, collected infrequently, or restricted
to only certain services
23
; a transit agency may be inconsistent about stop and
route identifiers, or a bus equipped with GPS may have a system that is broken
or inaccurate. In order to craft better mobility systems, better data collection
will be necessary in India’s future.
Acquiring data from private data owners
Acquiring data from private data owners is often one of the biggest challenges
in collecting the data needed for a particular use case. These data owners tend
to be concerned primarily with jeopardizing their competitive advantage by
sharing their private data. The nature of their concern may vary with the use
case for the data; for example, data owners often feel less threatened by
providing only real-time data to help travellers optimize their transport options,
whereas they are more wary of providing historic data to planners and developers.
This concern must be addressed by making appropriate and transparent
requests for data, as well as making clear the value proposition of sharing
data to the data owner. 27
Data aggregation, as a general strategy, is any process in which information
is gathered and summarized in certain forms, usually for statistical analysis.
Commonly, it is used to obtain more information about particular groups by
determining patterns of behavior. It typically involves compiling and anonymizing
data from multiple sources, usually for the purposes of observing system-level
patterns or understanding broader trends.
In transportation data, aggregation strategies are particularly useful for two
reasons: research and privacy purposes. In the modern age of IOT (Internet-
of-Things), large volumes of transportation information is easier than ever to
collect. The advent of technologies such as GPS, ride-sharing, and navigation
and routing programs provides multiple avenues to track individual pathways
of movement.
Data aggregation in mobility allows researchers, planners, and private industries to
understand larger trends about user profiles of their city. Similar to how a store
might market specific products to a certain demographic, data aggregation in
mobility enables planners to see how certain user profiles might be using
transportation differently.
While granular information is immensely useful for urban planners to understand
city behavior, it also raises the question of individual privacy. Data aggregation
is a method that enables these city practitioners and developers to access
invaluable data that may otherwise not be available, due to the risk of privacy
or competitive risk. By aggregating based on certain districts, areas, or
5.0
/////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
Data aggregation
+++ 28
types of user profiles, data aggregation anonymizes individuals and removes
granularity to such a degree that the user information is now shareable. It is
often a compromise between private companies with mobility data or various
smartphone companies or routing systems—which contain privacy-sensitive,
individual user data—and public city practitioners who are seeking information
to better understand their constituents.
Though often a powerful tool, these aggregation strategies can run the risk of
failing to provide the information that users need. Private companies are more
willing to share their aggregated, rather than granular, data, but the aggregated
information may end up obscuring the information that is most necessary for
the use case, such as city planning. The goal in data aggregation should be
to protect individual privacy while providing useful context to understand and
improve the transportation demands of cities.
Uber’s Movement tool, launched in January 2017, provides anonymized
data to help urban planning in cities around the world
24
. The tool is free
and open to anyone interested in using it, and provides travel times
between any two points in a city at any time of day. The data provided
is an aggregation of many individuals’ trips, anonymized and compiled
to remove a level of granularity that would violate individuals’ privacy or
compromise Uber’s competitive advantage.
Uber Movement aggregates data in a way that is effective to visualize
traffic flows and major congestion points; Uber Movement’s data
aggregates travel time by city zone, which is useful to understand the
impact of major events, from road closures to natural disasters. However,
city and transit planners have pointed out that the tool leaves out data
that would be most useful to them, such as the most common pickup
and dropoff
25
locations in a city. This example represents the conflict
between private interests and the public benefits of data sharing. Despite
mixed reviews of the tool’s usefulness, however, Uber Movement presents
a good example of how data aggregation can be used to increase
public access to data that otherwise may not be shared at all due to
privacy and competition concerns.
Case Study: Uber Movement Tool 29
With a growing population and increasing demand for transport, India stands
to benefit tremendously from the collection and sharing of mobility data. Using
and promoting open data increases the availability of data to a wider audience.
According to the National Informatics Centre, “A dataset is said to be open if
anyone is free to use, reuse, and redistribute it—Open Data shall be machine
readable and it should also be easily accessible
26
.” Open data is inherently free
and available to the public.
Open data helps to enable the many potential use cases for data analytics in
the transport sector, as outlined in section 3. The Government of India has taken
steps toward increasing the amount of and access to open data in an effort to
unlock the many benefits of data analytics. While this supports multiple sectors,
the mobility sector in particular has the potential to benefit tremendously from
this initiative.
The Government of India, and a number of other countries and cities around the
world, host online open data portals with datasets for many sectors, including
buildings, education, public safety, and transportation. The goal of these portals
is to provide easy access to open data and information in order to “spark
innovation, promote public collaboration, increase government transparency,
and inform decision making
27
.” These portals have been successful in
collecting and hosting a wide range of data, and presenting it in an easy-to-
use and accessible manner. These central data portals can add a lot of value
by aggregating datasets in a single location and making them easily available
6.0
/////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
Open data
+++ 30
The Government of India has launched Open Government Data (OGD)
Platform (data.gov.in) to support the Open Data Initiative for nation-
wide data sharing. This initiative was launched as part of the National
Data Sharing and Accessibility Policy (NDSAP) as well as the Digital
India Initiative. OGD platform provides open access to datasets,
documents, services, tools, and applications collected by various
ministries, departments, and organizations of the Government of India
for public use. The main goal of OGD is to provide open access to the
data generated through public funds and to enhance transparency,
accountability, citizen engagement, collaboration, better governance,
decision-making, and innovation
28
. Currently, there are over 2 lakh data
points, which have cumulatively been viewed over 17 million times.
Case Study: Open Government Data (OGD) Platform in India
to any party who can derive value from them. The U.S. cities of Austin and
Chicago are two other notable global examples of well-organized, frequently
updated, and comprehensive open data portals; both websites have categories
for “transportation” with dozens of datasets, ranging from traffic camera and
sensor data to non-motorized transit infrastructure maps and plans to public
transit ridership numbers. 31
7.0
/////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
Conclusion
+++
Data has an important role to play in helping India achieve a mobility system
that is clean, efficient, and adequately supports the mobility needs of its
citizens. To ensure the maximum benefit of mobility data, steps should be
taken to collect and share data and ensure that the data collected by different
parties is made available as much as possible. Effective communication between
data owners and potential beneficiaries is at the core of reaching this outcome;
to effectively communicate and collaborate, all stakeholders involved must
understand the landscape of the system and the motivations and risks of the
various parties involved.
With the rapid pace of technological developments and urbanization, India
should be developing and enhancing its transportation system with an eye to
the future: as the nature of cities and transport changes, mobility plans need to
be designed to be flexible and adaptable to keep up with an evolving landscape.
Building a comprehensive practice of data collection and sharing in India will
form a strong foundation to support new technologies and innovations, such
as blockchain, in the mobility space.
Thoughtfully constructing a supportive framework for collecting and sharing
mobility data will enable India to dramatically improve the efficiency and strength
of its mobility system as well as urban planning and regulation, resulting in
communities that are cleaner, safer, and better support the needs of their citizens. 32
Government of India, “Personal Data Protection Bill 2018,” 2018. http://meity.
gov.in/writereaddata/files/Personal_Data_Protection_Bill%2C2018_0.pdf
Digital Matatus, 2018. http://www.digitalmatatus.com/map.html
Carlin, Kelly, Bodhi Rader, and Greg Rucks, “Interoperable Transit Data:
Enabling a Shift to Mobility as a Service,” Rocky Mountain Institute,
October 2015. https://rmi.org/wp-content/uploads/2017/03/Mobility-
InteroperableTransitData-Report.pdf
“About UPI API,” IndiaStack. http://indiastack.org/upi/
Carlin, Kelly, Bodhi Rader, and Greg Rucks, “Interoperable Transit Data:
Enabling a Shift to Mobility as a Service,” Rocky Mountain Institute,
October 2015. https://rmi.org/wp-content/uploads/2017/03/Mobility-
InteroperableTransitData-Report.pdf
“BMTC: Easy Travel Information Planner,” Google Play, 2018. https://play.
google.com/store/apps/details?id=com.bmtc.mybmtc&hl=en
Sonal Shah, Kalpana Viswanath, Sonali Vyas and Shreya Gadepalli, “Women
and Transport in Indian Cities,” ITDP and Safetipin, December 2017. https://
smartnet.niua.org/sites/default/files/resources/171215_women-and-transport-
in-indian-cities_final.pdf
1
2
3
4
5
6
7
8.0
/////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
Citations
+++ 33
“The World Bank Launches New Open Transport Partnership to Improve
Transportation through Open Data,” The World Bank, December 19, 2016.
http://www.worldbank.org/en/news/press-release/2016/12/19/the-world-
bank-launches-new-open-transport-partnership-to-improve-transportation-
through-open-data
Stern, Noah, “Waze Partners With Local Governments to Enhance Mobility,”
Government Technology, February 12, 2016. http://www.govtech.com/dc/
articles/waze-partners-with-local-governments-to-enhance-mobility.html
Ibid.
“Big Data Analytics: What it is and why it matters,” SAS. https://www.sas.
com/en_us/insights/analytics/big-data-analytics.html
“How Big Data & Advanced Analytics Will Change Transportation,” Datafloq,
December 2016. https://datafloq.com/read/big-data-advanced-analytics-
transportation/1747
Winnesota, “How Blockchain is Revolutionizing the World of Transportation
and Logistics,” 2017. https://www.winnesota.com/blockchain
Cohen, Boyd, “How to power shared mobility startups with blockchain
technology,“ Shareable, March 2018. https://www.shareable.net/blog/
empowering-shared-mobility-startups-with-internet-of-mobility-blockchain-
technology
Sandner, Phillip, “Analysis of Blockchain Technology in the Mobility Sector,”
Medium, April 2018. https://medium.com/@philippsandner/analysis-of-
blockchain-technology-in-the-mobility-sector-1078e429615f
Cornet, Andreas et al, “The road to artificial intelligence in mobility—smart
moves required,” McKinsey & Company, September 2017. https://www.
mckinsey.com/industries/automotive-and-assembly/our-insights/the-road-to-
artificial-intelligence-in-mobility-smart-moves-required
“Personally Identifiable Information (PII),” Cornell Law School. https://www.
law.cornell.edu/cfr/text/2/200.79
Srikrisna, B.N. “A Free and Fair Digital Economy Protecting Privacy,
Empowering Indians,” 2018. https://www.medianama.com/wp-content/
uploads/Data_Protection_Committee_Report-comp.pdf
Mandhani, Apoorva, “Govt. Releases Justice Srikrishna Committee Report
And Draft Personal Data Protection Bill, 2018,” July 2018. https://www.
livelaw.in/justice-srikrishna-committee-releases-report-and-personal-data-
protection-bill-2018/
8
9
10
11
12
13
14
15
16
17
18
19 34
Singh, Pratap Vikram,“Seven Principles for Data Protection: BN
Srikrishna Committee,” Governance Now, 27 November 2017. http://
www.governancenow.com/news/regular-story/seven-principles-for-data-
protection-bn-srikrishna-committee-
Government of India, “Personal Data Protection Bill 2018,” 2018. http://meity.
gov.in/writereaddata/files/Personal_Data_Protection_Bill%2C2018_0.pdf
Agarwal, Surabhi, “Government seeks public comments on Data
protection bill; September 10 deadline,” The Economic Times, August
2018. https://economictimes.indiatimes.com/tech/internet/government-
seeks-public-comments-on-data-protection-bill-september-10-deadline/
articleshow/65404228.cms
Crane, Jackson and Greg Rucks, “A Consortium Approach to Transit Data
Interoperability,” Rocky Mountain Institute, 2016. https://rmi.org/wp-content/
uploads/2017/03/consortium_approach_to_ITD_report2016.pdf
Gilbertson, Jordan and Salzberg, Andrew, “Introducing Uber Movement,”
Uber Newsroom, January 9, 2017. https://www.uber.com/newsroom/
introducing-uber-movement-2/
Peterson, Andrea and Scola, Nancy, “Data is Uber’s business. But
protecting it may be its biggest weakness,” The Washington Post,
November 18, 2014. https://www.washingtonpost.com/news/the-switch/
wp/2014/11/18/data-is-ubers-business-but-protecting-it-may-be-its-largest-
weakness/?noredirect=on&utm_term=.5b9ef8330e4f
“Open Government Data,” Government of India Ministry of Electronics and
Information Technology. http://sdtckochi.nic.in/OGD.aspx
“The official City of Austin open data portal.” https://data.austintexas.gov
“Open Data,” Ministry of Electronics and Information Technology, June 2016.
http://meity.gov.in/open-data
20
21
22
23
24
25
26
27
28 35