668 IEEE INTERNET OF THINGS JOURNAL, VOL. 5, NO.

2, APRIL 2018

Empowering Citizens Toward the Co-Creation

of Sustainable Cities
Verónica Gutiérrez, Dimitrios Amaxilatis, Georgios Mylonas, and Luis Muñoz, Senior Member, IEEE

Abstract—Urban ecosystems are becoming one of the most Information and communication technologies (ICTs) will
potentially attractive scenarios for innovating new services and play an increasingly important role in the transformation of
technologies. In parallel, city managers, urban utilities, and cities in the forthcoming decades. Some rationality in the use
other stakeholders are fostering the intensive use of advanced
technologies aiming at improving present city performance and of technology needs to be applied by decision makers when
sustainability. The deployment of such technology entails the gen- defining their future strategies. More sustainable management
eration of massive amounts of information which in many cases of the city services, such as transportation, energy, and waste
might become useful for other services and applications. Hence, management, health and social care for dependent people is
aiming at taking advantage of such massive amounts of infor- required, avoiding the creation of an unsustainable sea of ver-
mation and deployed technology as well as breaking down the
potential digital barrier, some easy-to-use tools have to be made tical solutions for every urban service. For this purpose, the
available to the urban stakeholders. These tools integrated in use of the Future Internet (FI) together with IoT technologies
a platform, operated directly, or indirectly by the city, provide is considered an enabler to foster the transformation toward
a singular opportunity for exploiting the concept of connected the smart city paradigm.
city whilst promoting innovation in all city dimensions and mak- A prerequisite for building smarter cities is the participa-
ing the co-creation concept a reality, with an eventual impact on
government policies. tion of society. The involvement and education of the different
stakeholders in the new initiatives need to be considered from
Index Terms—Co-creation, ecosystem, experimentation, the very beginning to consolidate the different communities
Internet of Things (IoT), OrganiCity (OC), services, smart cities,
urban platform. within the city innovation ecosystem. To this end, cities need
to develop their own smart city vision, defining the strategies
to engage and mobilize final users of their services to partici-
pate in the new initiatives. In parallel, they should also be part
of the ongoing discussions for the development of their local
I. I NTRODUCTION
environment.
NTERNET of Things (IoT) technology is steadily
I becoming an omnipresent reality, leveraging connected
ecosystems, where ubiquitous connectivity and big data appli-
No urban environment is the same, since each city comes
with its own demographic, social, cultural, and geographi-
cal contexts and challenges. The diversity of potential areas
cations are enabling innovation all around the world. It is of application for IoT and FI technologies and the complex
marketed like a panacea, being employed in many application interdependencies between the urban services within the cities
fields including, among others, smart home solutions, self- makes it difficult to find a single, truly scalable, pragmatic, and
driving cars as the future of the automotive industry and smart applicable solution, which enables the management of the city
cities. services together with the stimulation of a co-creation ecosys-
Nowadays, most people in Europe live and work in cities tem. In this paper, the Experimentation as a Service (EaaS)
and soon, it is expected that the population living in urban framework conceived within the OrganiCity (OC) project [1]
environments will further increase. Thus, cities will need to is presented as an option to tackle this challenge.
deal with the growing demands for services of their citizens This paper is organized as follows. Section II describes
and visitors while their urban infrastructures naturally age and the urban ecosystems, where the IoT infrastructures appear
become outdated. as enablers for the sustainable management of smart cities
Manuscript received January 31, 2017; revised July 11, 2017 and through the urban platform. In Section III, the OC EaaS con-
August 8, 2017; accepted August 15, 2017. Date of publication cept is presented, describing the co-creation process with the
August 24, 2017; date of current version April 10, 2018. This work was sup- different stakeholders and the different components that have
ported by the European Union under Grant Agreement 645198 of the Horizon
2020 Research and Innovation Program. (Corresponding author: Luis Muñoz.) been developed to create more inclusive innovation ecosys-
V. Gutiérrez was with the Department of Communications Engineering, tems. The results of two different experiments are presented in
University of Cantabria, 39005 Santander, Spain. She is now with Telefonica, Section IV in order to validate the approach and illustrate how
39003 Santander, Spain (e-mail: [email protected]).
D. Amaxilatis and G. Mylonas are with the Computer Technology Institute the co-creation tools help users with different skills to carry out
and Press “Diophantus,” 26504 Patras, Greece (e-mail: [email protected]; their experiments obtaining interesting results, and also how
[email protected]). cities with or without existing IoT infrastructure are aided by
L. Muñoz is with the Communications Engineering Department, University
of Cantabria, 39005 Santander, Spain (e-mail: [email protected]). OC. Finally, Section V provides the main conclusions, along
Digital Object Identifier 10.1109/JIOT.2017.2743783 with several future steps.
2327-4662 c 2017 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
GUTIÉRREZ et al.: EMPOWERING CITIZENS TOWARD CO-CREATION OF SUSTAINABLE CITIES 669

II. U RBAN E COSYSTEM Moreover, the intensive use of technology within all the
In this section, we explain how the use of IoT and FI tech- facets of the city should empower smarter citizens. To this
nologies are clear candidates to enable the consolidation of end, cities are obliged to encompass the deployment of new
urban ecosystems within cities. IoT infrastructures with the development of specific plans
for people to become aware of the progress toward the
smart city. They should engage and motivate citizens to par-
A. IoT Infrastructure As Enabler for Setting Up the Smart ticipate, as they are the natural influencers of the future
City Paradigm co-creation process within the urban landscape. Apart from
While ICT is in the heart of many modern cities, existing empowering them to make smarter decisions and personal
infrastructures need to be revamped to enable new ways to choices in their daily lives, this could be a way in which
monitor and control urban services. This should be accompa- citizens increase their ownership sense and pride in belong-
nied by the installation of new IoT infrastructures, as basic ing to the cities, having a better understanding of their
components for the integrated management of the city. neighborhoods.
One essential requirement for the deployment of sustain-
able infrastructures is to handle the traditional inefficient use
of communication infrastructures. Due to the broad range C. Urban Platform
of heterogeneous solutions available in cities, a level of Apart from the deployment of new IoT infrastructures
abstraction of the data processed is required in order to guar- and citizen participation initiatives, sustainable management
antee the interoperability of the information generated by IoT of smart cities demands a platform that harmonizes their
devices of different vendors and applications, independently urban infrastructures and services. This is what we call the
and regardless of the underlying technologies. Urban Platform, made up of vertical pillars linked to the city
Setting up an IoT infrastructure often entails the deploy- services.
ment of a huge number of devices to support various urban Fig. 1 presents a high-level architecture of the Urban
services. Although this will enable the development of novel Platform, addressing the holistic vision of smart city man-
solutions with a positive impact on the quality of city ser- agement. As can be seen, legacy systems, existing platforms,
vices, one of the major challenges for the cities in the future or other IoT-enabled services need to coexist within the
will be how to manage the increasing number of these hetero- same platform. The classic approach, where services are pro-
geneous and geographically dispersed devices. Maintenance of vided in a vertical manner, is discarded to avoid the creation
the IoT infrastructure should be included as part of the strategy of isolated silos, because innovation is constrained in that
toward future smarter cities. Additionally, IoT infrastructures way. The adaptation of urban data streams to a common
within cities can be augmented with citizen participation [2]. data model that harmonizes and guarantees the interoper-
Using applications on their smartphones or wearables, citi- ability among the city services is fundamental for building
zens will not only be consumers of digital services, but will smarter cities. To achieve this, on top of the aforementioned
also generate data as producers within possible smart city services and solutions, the Urban Platform includes an adap-
scenarios. tation/interoperability layer that deals with the integration of
Finally, the definition and use of common and harmonized data streams and incoming events from lower layers, so they
data models, using semantic descriptions, enables the concep- can be stored within a common data repository. The transver-
tion of new innovative solutions to interact with city services sal intelligence layer is placed in the upper tier, allowing
and to manage the urban environment, which can be replicable cities to implement business logic for provision of real intel-
in other urban ecosystems. ligence for smart cities. The implementation of this logic
depends on the context of each city, as it needs to con-
sider the traditional management of urban services and the
B. Services and Citizen Engagement ways predictive and decision models can be implemented to
Cities can be considered complex systems, where a wide create new rules for integrated city management. On top of
range of urban services are provided to the different stake- that, the Urban Platform proposes the creation of an open
holders, organizations, and businesses. The design and imple- city environment, with common and well-established APIs,
mentation of smart vision embraces technological, economical, enabling the implementation of city dashboards, the develop-
and sociological developments within the cities, where citizens ment of useful services and applications for citizens, as well
play a major role. as development of enablers for consolidating the innovation
Progress toward citizens’ better quality of life is a goal ecosystem.
shared by every city. Moreover, cities need to improve the The development of the smart city vision, together with the
operational efficiency of urban services. Hence, a better implementation of the Urban Platform is a process that will
understanding of cities is required, carrying out an analysis take several years. It requires the constant involvement of the
of urban services in order to extract tangible requirements different stakeholders and coordinated actions to overcome the
for their sustainable management; both local authorities and different obstacles that might appear in the process.
other stakeholders need to collaborate and their involvement In Europe, cities often delegate the provision of local
will be crucial to success in the implementation of holistic services to private companies, aiming to reduce their oper-
management of cities. ational costs and improve the quality of the services offered.
670 IEEE INTERNET OF THINGS JOURNAL, VOL. 5, NO. 2, APRIL 2018

III. E XPERIMENTATION A S S ERVICE :

M OBILIZING THE E COSYSTEM
A. Rationale
Traditionally, cities have been the meeting point between
societal challenges and technological innovation. The main
ambition of OC is to make the creation and design of
technologies and services for cities more inclusive for any
user of the smart city. It tackles the aspects of how smart
cities can grow organically with the involvement of different
stakeholders (citizens, communities, policy makers, activists,
scientists, researchers, developers, city service, and technology
providers), and not be driven solely by engineering visions.
OC has developed an EaaS facility, as a research and inno-
vation environment for the co-creation of FI-enabled urban
knowledge and services. The facility leverages emerging tools
and technologies to provide mechanisms that allow users
to extract knowledge from different cities, based on the
Fig. 1. Urban platform for the sustainable management of smart cities. data streams that are generated in the diverse urban ecosys-
tems. To make this more inclusive, the facility delivers a set
of tools and enablers to empower any user to be part of the
co-creation process. Moreover, the facility provides various
Companies have the exclusive right to run the urban services in
means for the engagement of the citizens participating in their
specific areas of the city, as they have been awarded by com-
validation.
petitive tender. However, if the adoption of IoT technologies
Taking into consideration the resource constraints of cities
is not included in the contracts with the urban providers, cities
in times of austerity, it is difficult to create new city infras-
should invest part of their budget to that end. As these types of
tructures and/or develop new “experiments.” Setting up the OC
contracts last for many years, inappropriate decisions can hin-
facility on top of the urban platform should permit city makers
der the development toward the smart city paradigm. Every
and service providers to address not only the exploration of
city is handling this challenge, developing different actions.
new pathways for their services, but also to understand any
As an example, in Santander, Spain, every time a competi-
societal implications in a rapid and more iterative manner.
tive tender to subcontract urban services is instigated, the city
In general, as can be seen in Fig. 2, the federation API
elaborates public procurements in which, it is mandatory to
enables cities to integrate different data streams, or any other
update the existing infrastructure, in order to incorporate and
city asset, into the OC facility. To this end, they should imple-
deploy IoT technologies for the management of city services.
ment what we call the OC city site, a component that makes
Moreover, citizen engagement activities are also included in
use of the city platform’s APIs to gather data from the Urban
the public procurement, aiming to contribute to the common
Platform and feed them as new data assets into the OC facil-
smart city strategy. These activities allow service providers to
ity. To this end, the facility defines a common data model,
create trustable environments, in which both city officials and
based on the FIWARE NGSI specification [5] that harmonizes
end-users can interact to improve the existing services.
and abstracts the information coming from the federated cities,
For supervision of the urban services, it is common that
aiming to guarantee the interoperability of smart city solutions.
municipalities set up operational and environmental regula-
tions, as well as principles that should be considered as
guidelines by the service providers. The Urban Platform can B. ORGANICITY: EaaS Platform
facilitate the work of municipality officials, as within their 1) Scope: OC offers a flexible EaaS framework, allowing
dashboards, the indicators for the different services can be researchers and developers of urban infrastructures, services
monitored. Moreover, the use of IoT technologies within the and knowledge, full exploitation of emerging FI technologies.
city services and citizen participation through apps can be Research and experimentation on top of the OC facility should
combined to implement both objective and subjective indi- lead to the creation of sustainable, effective, and replicable
cators, which can be measured in order to improve the quality smart city solutions and urban technology developments that
of the services and therefore, citizens’ satisfaction. are successfully adopted by citizens, local communities and
Although existing regulations within public administrations society as a whole.
are slowing down the process in some cities, with all these The design phase of the facility considered questions like
considerations in mind, we believe that we have identified how to provide data generated within the cities in a more pub-
a solid model for integrated city management that relies on lic domain, whether existing infrastructures are reliable enough
the synergies between the public and private sector. In the for data scale collection with citizens’ devices, how creation of
next sections, this paper focuses on the implementation of actionable knowledge from urban sensors can be scaled, how
the co-creation ecosystem and how the OC EaaS facility is we can incentivize and technically facilitate effective citizen-
established in different cities. city-academia–industry co-creation. Different stakeholders,
GUTIÉRREZ et al.: EMPOWERING CITIZENS TOWARD CO-CREATION OF SUSTAINABLE CITIES 671

provide a basis for exploring new business models derived

from the emerging solutions.
3) Architecture: The OC platform uses the microservices
architecture pattern [6]. Each service deployed in the context
of OC is a standalone application that uses a central authen-
tication service [7] (based on OAuth2.0) and provides either
a user interface for end-users or a programmatic API (usu-
ally RESTful). This allows us to develop different applications
based on their requirements and using tools, programming
languages, and frameworks that best suit each case (i.e.,
node.js for frontend applications, Java for backend services).
Additionally, the microservices pattern allows us to better fine
tune different parts of our infrastructure based on usage at
runtime levels and thus provide better user experience and
responsiveness to any number of clients, a major requirement
for big cross-city IoT applications.
4) Components: Instead of building the EaaS facility from
scratch, the project has reused components and experiences
Fig. 2. Co-creation ecosystem of OC complementing the urban platform for
sustainable management of cities. from three well established European urban ecosystems:
1) Aarhus, Denmark; 2) London, U.K.; and 3) Santander,
Spain, which are developing different smart city initiatives and
such as developers, data analysts, IoT solution manufacturers, already have infrastructures in place. Fig. 3 depicts the high-
urban service providers, activists, sociologists, economists, and level architecture of the OC facility, in which three different
citizens were involved, in order to extract requirements for the tiers can be identified: 1) the OC site tier; 2) the OC platform
EaaS facility. As a result, OC has implemented a set of tools tier; and 3) the OC experimentation tier.
that empowers citizens to be part of the co-creation process On top of the OC architecture, the OC experimentation
at different stages of the urban service lifecycle and provide tier offers a set of tools and enablers to be used within the
different means for their participatory engagement. experiments. Practical guidance about the use of the facil-
2) Stakeholders: The OC facility supports different types ity is available in [7]. First, through the accounts component,
of users, permitting them to manage the whole experiment life citizens, or any other stakeholder can register as OC users.
cycle and offering a set of tools and enablers that make the OC users can request the “providers” or “experimenters” role,
co-creation and validation of new solutions more inclusive. It depending on whether they want just to provide data to the
can be used remotely by any stakeholder within smart cities. facility, or to carry out an experiment.
Depending on the activity that they intend to do, users For novel experimenters and other stakeholders, OC has
can request different permissions, supporting the following developed the concept of the urban data observatory (UDO),
roles: 1) experimenters; 2) participants; 3) providers; 4) site a component that allows users to explore and visualize mixed
managers; and 5) facility administrators. Experimenters can static and real-time urban and social data streams, enabling the
implement their own solutions using a set of co-creation tools exploitation of such data. To this end, the UDO relies on the
and conduct the corresponding experiment. To validate the EaaS APIs [8] provided by the OC Platform. More precisely,
solutions, they can invite other participants within the scope it uses the asset discovery service and the data annotation ser-
of the experiment (e.g., using an application developed by vice, which provide support for crowdsourcing annotations
the experimenter). Providers are users who do not belong to of the urban data assets using the well-established social
any experiment, but want to contribute with crowdsourced media paradigm of tags. In more detail, through the UDO,
data to the facility (e.g., feed data from their own weather users can add the appropriate annotation tags to the assets they
stations). Facility administrators deal with the management of browse or view the tags other users have already assigned. In
the entire platform, being able to federate new urban ecosys- the background, the UDO is communicating via a well-defined
tems within the cities, assigning permissions, and roles to other API with the data annotation service to retrieve, add or update
users, configuring the facility parameters and monitoring its existing tags assigned to each asset. Based on that data. On
activity. Finally, site managers are users belonging to the fed- a second level, experimenters can identify real world events
erated cities that can configure the information related to the or generate further insights within the smart city by observing
urban services, the tags that can be used for annotating the the tags assigned to each asset.
data assets, and so on. Additionally, OC provides a set of co-creation tools [9],
The use of the facility and its tools within the urban such as SensiNact Studio, TinkerSpace, and Smartphone
ecosystems will guarantee that communities of users can grow Experimentation that facilitate the exploration of data within
around emerging experimental technologies. Experimenters the urban ecosystems, as well as the development of new ser-
and service providers can leverage active contributors and vices and applications using existing data sources. The tools
early technology adopters. These communities will not only support users from IoT developers to nontech-savvy activists,
inform and contribute to the design of technology, but also giving them the opportunity to define their application logic
672 IEEE INTERNET OF THINGS JOURNAL, VOL. 5, NO. 2, APRIL 2018

Fig. 3. High-level architecture of the OC facility.

as actions to be taken when specific events occur, or even cre- or providers, the OC experimenters site and the OC providers
ate their own applications from scratch. At the same time, it site, respectively, enable the injection of data into the platform.
helps them understand why open data, smart cities, IoT, crowd To ease this process, OC utilizes a set of hardware platforms,
sourcing, big data, and other related topics are important for such as TSmart, and DULradio that are already integrated,
the sustainability of cities. reducing the need for integration or development of custom
The core component of the facility is the OC platform that software. Additionally, the Smartphone Experimentation tool
can be found in the OC platform tier. Based on the use of is designed to allow participants to take part in data collec-
the Orion context broker generic enabler [10], it implements tion experiments inside the cities, feeding data collected by
what we call the Asset Directory, a catalogue that contains all the sensors integrated in smartphones, or other IoT devices
available data sources integrated in the platform. This compo- attached to them (e.g., wearables over Bluetooth LE).
nent integrates urban data streams of the federated cities, as
well as crowdsourced data generated by both providers and
experimenters, as data assets in OC. Besides the basic discov- IV. VALIDATION
ery functionalities, it also acts as a data exchange hub, which To illustrate how the different stakeholders can benefit from
allows other components of the facility to subscribe to updates the OC facility, two different experiments have been carried
of the available data and provides pointers to access histori- out. The first one has been conducted in collaboration with
cal data when and from, where they are available. Moreover, the officials of the Climate Change Office in Santander, Spain.
the OC platform has been designed to be easily configured They have studied the urban heat island (UHI) effect in the
through a dedicated Facility Management Portal that acts as city [11] during winter and summer based on the data collected
a central control point for administrators of OC and service by IoT devices installed in public vehicles, which is available
managers of the federated cities. in OC. The second one is based on the use of the Smartphone
Finally, within the OC site tier, different components enable Experimentation co-creation tool. In this experiment, the
the generation of data. Each OC city site integrates its participants have collected data with gas, humidity, and tem-
urban data streams generated within the IoT infrastructures, perature sensors attached to their smartphones. Through the
data available on open data platforms (e.g., comprehensive Smartphone Experimentation app installed in their devices, the
knowledge archive network) or any other existing service data retrieved in different parts of the city was fed to the OC
through their respective APIs (e.g., API provided by the Public Platform, aiming at conducting a similar study in the city of
Bus Company in Santander, Spain). To proceed with inte- Patras (Greece), in a scenario, where no related IoT city infras-
gration, every city willing to be federated within OC needs tructure is deployed. With these two experiments, we intend to
to deploy its own OC city site adaptors, so that the federa- showcase how flexible the OC framework is when developing
tion API of OC is able to consume the data in the central and validating distinct solutions under real conditions within
OC infrastructure. In the case of experimenters/participants smart cities.
GUTIÉRREZ et al.: EMPOWERING CITIZENS TOWARD CO-CREATION OF SUSTAINABLE CITIES 673

Fig. 4. Experiment information in the experimenter portal. Fig. 5. Santander zones used for the experiment analysis.

A. Analysis of the Urban Heat Island Effect

Many urban areas experience elevated temperatures com-
pared with their rural surroundings; this difference in temper-
ature being the basis for the formation of a UHI. Due to human
activity, the mean air temperature in the city is several degrees
warmer than the surroundings, this effect being less intense in
smaller cities.
To identify heat islands within the cities, different methods
and models can be applied. In this paper, we rely on data col-
lected by environmental sensors installed in nearly 150 urban
transportation buses and in park and garden vehicles mov-
ing around the city of Santander. The information collected
by these sensors when moving in the different city routes is
transmitted (relying on general packet radio service, infrastruc- Fig. 6. UHI effect comparison in Santander (in ◦ C).
ture) and fed to the OC facility by means of the corresponding
EaaS API.
The experiment consists in assessing the thermic load qual- EaaS APIs, historical data for different periods was retrieved.
itative model adopted by the Santander Local Climate Change Further data processing was done to obtain representative sam-
Office, which categorizes the city in different climate zones ples of temperature measurements in at least four cells for each
according to the probability of the UHI effect [12], [13] in one of the aforementioned climate zones.
these areas. Aiming at assessing the matching between the thermic load
The methodology applied is as follows. First, within the qualitative model and the collected measurements, the tem-
city area, a grid with cells of 500 × 250 m2 was created, perature variation per climate zone during one of the warmest
as depicted in Fig. 4. Taking into account several factors days in the summer period and during one of the coldest days
related to the vegetative cover and surface properties in in the winter period was obtained during the years 2014–2016.
each one of the cells, and according to the model adopted, This is shown in Fig. 6.
the cells were categorized into six climate zones: 1) rural As can be seen, the highest temperature difference between
(zone 0—null warm-up); parks (zone 1—residual warm-up); zone 5 (downtown) and zone 0 (rural zones located in the sur-
3) suburban residential (zone 2—low warm-up); 4) commer- roundings of Santander) is around 6 ◦ C during the summer
cial (zone 3—intermediate warm-up); 5) urban residential period, while in winter time this difference is 2 ◦ C. Hence,
(zone 4—intermediate/high warm-up); and 6) downtown (zone the results show the good matching between the thermic
5—high warm-up). Furthermore, multiple cells fitting the most load qualitative model and the collected measurements, which
commonly adopted city profile (multiple areas categorized as should facilitate the prediction of the UHI effect in different
zone 0-zone 2-zone 3-zone 5-zone 4-zone 1-zone 2-zone 0 city zones.
according to the climate zone model) are chosen [11]. Second,
to carry out the study, we created a new experiment in the B. Citizen Participation Through Smart Phones
experimenter portal, defining its scope within the Santander The experiment carried out in Patras, Greece showcases
area (see Fig. 5). Then, we selected the data assets used the smartphone crowdsourcing capabilities of OC, while
within the experiment; that is, those corresponding to the demonstrating how an experiment similar to the one carried out
ones collected by the sensors deployed in public buses and in Santander, a city with extensive IoT infrastructure, can be
vehicles, which provide temperature values as part of the envi- carried out in a location with no IoT infrastructure. The exper-
ronmental monitoring service in Santander. By means of the iment monitors nitrogen dioxide (NO2 ) concentration levels in
674 IEEE INTERNET OF THINGS JOURNAL, VOL. 5, NO. 2, APRIL 2018

Fig. 8. IoT box hardware used in Patras’ experiment.

the sensor box over BLE, and the bundling of measurements

Fig. 7. Measurement points in Patras—area approx. 14 km * 6 Km. as OC readings. The handling of the readings, i.e., formatting,
checking and uploading to OC’s servers is done automatically
various parts of the urban landscape of the city. This gas is by the Android app, thus the developers do not need to write
specifically related to diesel engine emissions, wood use in code for these tasks. The additional lines of code required in
fireplaces, and other similar human activity. this case are in the order of 40–50 lines in total.
To that end, we defined a new OC experiment to be car- Sensor Box Pseudocode
ried out in Patras through the experimentation portal. It offers Set Up BLE communication and Gas Sensors
additional capabilities provided for monitoring the progress of Power On Gas Sensor
the experiment, since these experiments are based on Android while(TRUE) {
software components provided by the project itself, in the if (“BLE Terminal is connected”) {
form of the Smartphone Experimentation portal (see Fig. 7). while (“BLE Terminal Connected”) {
Citizens that wish to contribute to the project can install the send “START”
OC Android app, which essentially provides both a sensing if message received send “FINISH”
instrument, through the device’s integrated sensors, and an else disconnect
IoT “proxy,” by forwarding measurements from connected }
devices. In this case, we use the second scenario, since there }
is no smartphone device available with an integrated NO2 sen- else Read sensors every 30 s
sor. The OC Android app [14] downloads the experiment to }
be executed, and, when the owner gives permission to the Read the sensor values:
app to use, e.g., Wi-Fi and Bluetooth, it transparently handles {
all aspects related to collecting measurements and forwarding Read Gas Sensor and power off
them to OC. Power On Temp-humid Sensor, read value, power off
In order to provide a sensing endpoint for the experiment, Power On Multichannel Gas Sensor
we constructed an IoT “box” (see Fig. 8), to be carried by }
citizens during their daily commute. It has an integrated battery BLE Smartphone Plugin Pseudocode
and communicates with an Android smartphone running the Every 2 min:
OC app over Bluetooth LE. The IoT box used is based on the Close previous connection if open and reset
following hardware parts. Scan for and create a list of discovered BLE devices
1) Blend micro [15], a popular microcontroller. while (Array List of BLE devices != empty)
2) Grove multichannel Gas Sensor and Grove Temperature {
and Humidity IR Sensor [16]. Connect with first BLE device on the list
3) Li-Po Rider and LiPo battery—2200 mAh 3.7 V. while (BLE Device connected)
4) A custom PCB board integrating the above. {
Delving into additional details regarding the implementa- Waiting for BLE device message
tion of the code executed on the smartphone and the sensor if (message==START)
box, the Android part of the code can be found online in the send message=Send a New Sensor Value
respective GitHub repository.1 In essence, developers utilize else if (message =New Sensor Value)
a provided code skeleton, extending Java classes and imple- update Sensor Values
menting a few methods for handling the communication with }
delete BLE device from list
1 Organicity GitHub repository, experiment plugins. [Online]. Available:
}
https://github.com/OrganicityEu/smartphone-experimentation-experiments
GUTIÉRREZ et al.: EMPOWERING CITIZENS TOWARD CO-CREATION OF SUSTAINABLE CITIES 675

• Zone 8, a suburban area with heavy traffic and a busy

train/bus stop.
• Zone 9, a city center area with a beachfront.
The measurements reveal some interesting trends; suburban
areas that are more open to wind currents or without traffic
produce similar readings. Vehicle traffic plays a major role
in nitrogen dioxide concentration; however, it is not the sole
producer. Zone 8 is greatly affected by the train and bus traf-
fic, even though it is suburban with no tall buildings in the
area. Zone 7 produced the highest readings in winter, proba-
bly due to the fact that in Greece many people have reverted
to using traditional wood fireplaces for heating in the win-
ter, combined with the high humidity in the area (frequently
over 60%). Zone 1 produces similar readings to city center
areas, although it is essentially a park area, due to heavy traf-
fic at certain times. In summertime, we see much lower NO2
levels overall, which seem to correlate with the perceived air
Fig. 9. Nitrogen dioxide concentration measurements in Patras in µg/m3 .
pollution levels as well.
Overall, we believe that these IoT experiments can poten-
tially offer great insights, e.g., into the current environmental
Regarding the experiment plugin, developers have to imple-
status inside urban areas. OC can be used as the substrate to
ment a “handleContextRequest” method, used by the frame-
design, organize, and implement such initiatives, even in the
work each time an updated sensor value is needed. This is
case of having no existing IoT infrastructure available.
usually done every 30–60 s, but times may vary based on the
experiment.
handleContextRequest Pseudocode V. C ONCLUSION
create an Array List for readings No two urban environments are the same, as they come
if (experiment exists) with various demographic, social, cultural, and geograph-
create new reading from values of BLE plugin ical contexts, resulting in a great diversity in smart city
add reading to the app internal database challenges. Dealing with such challenges will require holis-
wait for new event. tic management of the city by means of what we call
Over the course of 7 days in winter and 5 days in summer, urban platforms. The progressive adoption of IoT technolo-
we produced over 1400 measurements with the box, covering gies in urban services will enable the creation of the building
a large part of the city center. In this case, the usual method blocks for setting up the smart city paradigm. The mobi-
of producing gas concentration readings is to monitor the air lization and consolidation of innovation ecosystems will be
at specific points for a period of 30 min, 1 h or 24 h and fundamental as it permits us to take charge of tackling
then calculate average values. We opted to measure with the diverse city challenges, applying multidisciplinary strategies
IoT box while people were on the move, simulating real-life that can operate along with the traditional management of
conditions. the city.
Moreover, our sensors were calibrated with a rough estima- The instantiation of the OC facility, which has been
tion, and do not provide the same accuracy as more expensive designed to guarantee interoperability with the different urban
ones. We utilized air pollution monitoring station readings platforms, enables research and experimentation with smart
provided by the local government to make our calibration. city solutions and urban technology developments that have
However, our goal was to highlight that even with the use of been successfully adopted by citizens, local communities, and
simple equipment, we can reveal certain trends and that OC society as a whole, making the co-creation process more
can help, e.g., interested citizen groups and provide a good inclusive. Indeed, we believe that in the present digital trans-
foundation for such initiatives. Our results also generally formation phase, cities will have to make different facilities
correlate with other studies in the area [17]. available to the citizens, which will provide support to the
Overall, the experiment area covered is partitioned in EaaS paradigm as another urban service and with the same
nine zones. relevance as, for example, public transportation or waste man-
• Zone 1, a park with heavy vehicle (trucks) traffic. agement. Such an approach may become one of the enablers in
• Zone 2, a university campus, suburban area. the consolidation of open government policies or in the recon-
• Zone 3, a city area with medium levels of traffic. figuration of urban ecosystems from the business interaction
• Zone 4, the city center, with heavy traffic and activity. perspective.
• Zone 5, a suburban area with heavy traffic. Further to the experiments reported in the previous sec-
• Zone 6, a suburban area with medium traffic and lots of tions, in the near future, we plan to extend these measurements
houses using fireplaces for heating. and use-case scenarios to other parts in cities by utilizing the
• Zone 7, a central area with very light vehicle traffic. Smartphone Experimentation framework and experimentation
676 IEEE INTERNET OF THINGS JOURNAL, VOL. 5, NO. 2, APRIL 2018

management capabilities provided by OC, combined with Dimitrios Amaxilatis was born in Kavala, Greece,
incentive and reward techniques. in 1988. He received the B.S. degree in computer
engineering and M.S. degree in computer science
from the University of Patras, Patras, Greece, in
ACKNOWLEDGMENT 2011 and 2013, where he is currently pursuing the
Ph.D. degree.
The authors would like to thank the OrganiCity team. Since 2010, he has been a Researcher with the
Computer Technology Institute, Patras. He was also
a member of the founding teams of two tech-
R EFERENCES nological startups in the fields of microprocessor
[1] OrganiCity Project. [Online]. Available: http://www.organicity.eu programming (codebender.cc) and home automa-
[2] E. Theodoridis, G. Mylonas, V. Gutiérrez, and L. Muñoz, “Large-scale tion (Sensorflare), as well as the Patras City hackerspace (P-Space), the first
participatory sensing experimentation using smartphones within a smart hackerspace in Greece. His current research interests include distributed algo-
city,” in Proc. MobiQuitous, London, U.K., 2014, pp. 178–187. rithms, wireless sensor networks, home and building automation, Smart City,
[3] V. Gutiérrez et al., “Co-creating the cities of the future,” Sensors, vol. 16, and participatory sensing applications.
no. 11, p. 1971, 2016.
[4] D. Amaxilatis, E. Lagoudianakis, G. Mylonas, and E. Theodoridis,
“Managing smartphone crowdsensing campaigns through the OrganiCity
smart city platform,” in Proc. UbiComp Adjunct, Heidelberg, Germany,
2016, pp. 1460–1465.
[5] FIWARE NGSI Specification. [Online]. Available:
http://telefonicaid.github.io/fiware-orion/api/v2/stable/
[6] Microservices Pattern. [Online]. Available:
https://en.wikipedia.org/wiki/Microservices
[7] OrganiCity GitHub Repository. [Online]. Available: Georgios Mylonas received the Diploma, M.Sc.,
http://organicityeu.github.io/ and Ph.D. degrees from the Department of Computer
[8] OrganiCity Experimentation As a Service APIs. [Online]. Available: Engineering and Informatics, University of Patras,
http://api.organicity.eu/ Patras, Greece.
[9] OrganiCity Co-Creation Tools. [Online]. Available: He is a Senior Researcher with the Computer
http://organicity.eu/tools Technology Institute and Press “Diophantus,”
[10] Orion Context Broker FIWARE Generic Enabler. [Online]. Available: Patras. He has been involved in the AEOLUS,
https://fiware-orion.readthedocs.io/ WISEBED, Smartsantander, AUDIS, ABC4TRUST,
[11] Reducing Urban Heat Islands: Compendium of Strategies. Draft, and OrganiCity projects, focusing on algorithmic
U.S. Environ. Protect. Agency, Washington, DC, USA, 2008. [Online]. and software issues of wireless sensor networks. He
Available: https://www.epa.gov/heat-islands/heat-island-compendium is currently the Coordinator of the Green Awareness
[12] T. R. Oke, Boundary Layer Climates. Taylor & Francis, 2002. in Action H2020 project. His current research interests include wireless sensor
[13] I. Stewart and T. R. Oke, “Newly developed thermal climate zones for networks and distributed systems, Internet of Things, and pervasive games.
defining and measuring urban heat island magnitude in the canopy layer,”
in Proc. 8th Symp. Urban Environ. Org. Amer. Meteorol. Soc., Phoenix,
AZ, USA, Jan. 2009.
[14] OrganiCity Smartphone Experimentation on Google Play
Store. [Online]. Available: https://play.google.com/store/apps/
details?id=org.ambientdynamix.core
[15] Blend Micro Microcontroller. [Online]. Available:
http://redbearlab.com/blendmicro
[16] Grove Sensors. [Online]. Available: http://www.seeedstudio.com/depot/
Grove-Multichannel-Gas-Sensor-p-2502.html
[17] Report on University of Patras Campus Air Quality. Luis Muñoz (M’95–SM’15) was born in
[Online]. Available: http://green.upatras.gr/sites/green.upatras.gr/files/ Barcelona, Spain, in 1964. He received the
ReviewAtmosphere2015.pdf Telecommunications Engineering degree from the
Polytechnical University of Cataluña, Barcelona,
Spain, in 1990, the master’s degree in mathemat-
ics, statistics, and operation research from the
Verónica Gutiérrez was born in Santander, Spain, Licenciado en Ciencias Matemáticas, Universidad
in 1976. She received the Telecommunications Nacional de Educación a Distancia, Madrid, Spain,
Engineering degree from the University of and the Ph.D. degree from the Polytechnical
Cantabria, Santander, in 2000. University of Cataluña, in 1995.
She has developed most of her professional He is the Head of Network Planning and Mobile
career as a Senior Researcher with the Network Communications with the University of Cantabria, Santander, Spain. He has
Planning and Mobile Communications Laboratory, participated in several National and European research projects within the
University of Cantabria. She has been very active in 4th, 5th, 6th, and 7th Framework Program in which he was the Technical
different European projects framed under the Smart Manager of SmartSantander. He has authored or co-authored over 150 journal
City paradigm, such as SmartSantander, where and conference papers. His current research interests include advanced
she coordinated the development of the citizens’ data transmission techniques, heterogeneous wireless multihop networks,
services, the RADICAL project leveraging the Living Laboratory concept, Internet of Things, Smart Cities, and applied mathematical methods for
as well as OrganiCity, and focused on the design and implementation of the telecommunications.
Experimentation as a Service framework. Since 2017, she has been with Prof. Muñoz serves as an Editor for several journals. He has been invited
Telefonica, Madrid, Spain, where she is leading the project that deals with to participate in the Steering Committee and Technical Program Committee
the implantation of the Smart City platform and the transformation of the of the most relevant international conferences. In parallel to this activity, he
urban services in the city of Santander. Her current research interests include serves as a Consultant to the Spanish Government, as well as for different
development of Internet-of-Things-based services and applications for smart companies in Europe and USA. He has served as an Expert of the ETSI and
cities, middleware, experimentation platforms, and citizen participation. European Commission.