Content uploaded by Kasun Herath
Author content
All content in this area was uploaded by Kasun Herath on May 08, 2022
Content may be subject to copyright.
International Journal of Information Management Data Insights 2 (2022) 100076
Contents lists available at ScienceDirect
International Journal of Information Management Data
Insights
journal homepage: www.elsevier.com/locate/jjimei
Adoption of articial intelligence in smart cities: A comprehensive review
H.M.K.K.M.B. Herath
a , ∗
, Mamta Mittal
b
a
Faculty of Computing and IT, Sri Lanka Technological Campus, Padukka, Sri Lanka
b
Delhi Skill and Entrepreneurship University, New Delhi, India
Keywords:
Articial intelligence (AI)
Digital cities
Intelligent interaction
Internet of Things (IoT)
Smart cities
Recently, the population density in cities has increased at a higher pace. According to the United Nations Pop-
ulation Fund, cities accommodated 3.3 billion people (54%) of the global population in 2014. By 2050, around
5 billion people (68%) will be residing in cities. In order to make lifestyles in cities more comfortable and cost-
eective, the city must be smart and intelligent. It is mainly accomplished through an intelligent decision-making
process using computational intelligence-based technologies. This paper explored how articial intelligence (AI)
is being used in the smart city concept. From 2014 to 2021, we examined 133 articles ( 97% of Scopus and 73% of
WoS ) in healthcare, education, environment and waste management, agriculture, mobility and smart transporta-
tion, risk management, and security. Moreover, we observed that the healthcare (23% impact), mobility (19%
impact), privacy and security (11% impact), and energy sectors (10% impact) have a more signicant inuence
on AI adoption in smart cities. Since the epidemic hit cities in 2019, the healthcare industry has intensied its
AI-based advances by 60%. According to the analysis, AI algorithms such as ANN, RNN/LSTM, CNN/R-CNN,
DNN, and SVM/LS-SVM have a higher impact on the various smart city domains.
1. Introduction
A city is identied as a densely populated area. It is a permanent
and heavily populated region with ocially dened limits, with mem-
bers primarily engaged in non-agricultural tasks ( Goodall, 1987 ). The
density of human developments such as residences, commercial proper-
ties, roadways, bridges, and railroads is high in urban areas, indicating
highly developed. According to the United Nations Department of Eco-
nomic and Social Aairs, 55% of the global population resides in urban
areas, and the percentage will be expected to rise to 68% by 2050. There
were approximately 512 urban areas with a population of one million
people and around 31 megacities with a population of more than ten mil-
lion people in 2016. By 2030, it is expected to be approximately 662 city
areas and 41 megacities, the vast majority of which will be in emerging
regions ( Nations, 2018 ). This rate of urbanization will have a signicant
inuence on the environment, management, healthcare, energy, educa-
tion, and security of cities. Smart cities will then become the standard in
the country’s major urban areas. Modern smart cities use various tech-
nologies and support advancements that can assist cities in achieving
long-term socioeconomic goals and prospects. A diversity of smart-city
projects is being implemented across diverse geographical places, as ev-
idenced by many studies, generating a rich tapestry of urban visions
( Cowley et al., 2018 ; Dowling et al., 2019 ; Fernandez-Anez et al., 2018 ;
Pinna et al., 2017 ).
∗ Corresponding author.
E-mail addresses: kasunkh@sltc.ac.lk (H.M.K.K.M.B. Herath), mittalmamta79@gmail.com (M. Mittal) .
A scientometric analysis in ( Ingwersen & Serrano-López, 2018 )
shows that AI has been used in smart city research since 2008. Fur-
thermore, it has been connected to global sustainable developments, no-
tably by underdeveloped countries (e.g., ( Adunadepo & Sunday, 2016 )),
which are using AI to advance the UN’s Sustainable Development Goals
(SDG). Articial intelligence to enable smart city solutions has several
advantages, including more adequate water supply, energy manage-
ment, waste management, and reduced trac congestion, noise, and
pollution. Most smart city activities and technologies have focused on
creating data and obtaining new information about a city’s complexity
and dynamics ( Kapoor et al., 2021 ; Kar et al., 2017 ; Kar et al., 2017 ).
AI takes cities to the next level by allowing them to use that data and
knowledge to aid decision-making. AI will allow over 30% of smart city
applications by 2025, including urban transportation solutions, consid-
erably adding to urban life’s resilience, sustainability, social welfare,
and vitality ( Cugurullo, 2020 ). Thanks to the rapid growth of AI-based
smart city initiatives, researchers, government ocials, and practition-
ers are always seeking new information and approaches to making cities
smart.
As mentioned in ( Kassens-Noor & Hintze, 2020 ; Verma et al.,
2021 ), articial intelligence, like many other transformative technolo-
gies throughout history, will signicantly inuence society. Many urban
policy circles and discussions regarding smart city transformation have
centred on AI, particularly amongst urban policymakers and planners
seeking technocentric solutions to signicant urbanization concerns.
https://doi.org/10.1016/j.jjimei.2022.100076
Received 31 December 2021; Received in revised form 29 April 2022; Accepted 30 April 2022
2667-0968/© 2022 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
( http://creativecommons.org/licenses/by-nc-nd/4.0/ )
H.M.K.K.M.B. Herath and M. Mittal International Journal of Information Management Data Insights 2 (2022) 100076
Table 1
Dierent denition of term “Smart City ”.
Reference Denition
IBM A smart city, according to IBM, makes the most use of all the linked data available today to understand better and regulate its
operations while also maximizing the use of limited resources.
( Chourabi et al., 2012 ) Using the city’s collective intelligence through connecting the city’s physical infrastructure, information technology infrastructure,
social infrastructure, and commercial infrastructure.
( Aguilera et al., 2013 ) A smart city is a comprehensive notion that encompasses physical infrastructure and human and social issues.
( Khan et al., 2013 ) A city that engages in ICT enhanced governance and participatory procedures to dene suitable public service and transportation
investments that may provide improved quality of life, intelligent resource management, and sustainable socio-economic growth.
( Sterbenz, 2017 ) The phrase “Smart City ” refers to new industries that use ICT, as well as urban functions and environments.
( Mustafa & Kar, 2017 ) The concept of “Smart City ”has emerged from “Intelligent Cities ”. The basic idea of smart cities is to use the existing resources in a
“Smarter ”way.
Moreover, ( Zhou & Kankanhalli, 2021 ) stated that while smart city AI
applications provide benets such as automation and eciency, they
also pose regulatory issues, such as discrimination in service delivery,
privacy, legal, and ethical issues. Since various researchers have pre-
sented diverse points of view, it’s crucial to investigate the strengths,
downsides, and inuence of AI algorithms on smart cities. This review
explores how articial intelligence is being used in the modern smart
city to address the problems. Furthermore, we found no viable review
papers on AI adoption in smart cities after thoroughly examining rele-
vant literature.
The rest of this paper is divided into six sections and is organized
as follows. The literature review is presented in Section 2. Section 3
discusses the research methodology. Sect. 4 reviewed selected studies
during the previous eight years related to the notion of AI in smart cities.
Sect. 5 has covered all potential outstanding concerns, obstacles, and
future research paths. Section 6 is the last section of the review paper,
and it summarizes the main points.
2. Literature review
This section is composed of two subsections. Section 2.1 discusses
the topic of emerging smart cities. The denition of the term smart city,
major domains in the smart city, and top-ranked smart cities in the world
have been discussed here. Section 2.2 discusses articial intelligence and
the categorization of dierent AI technologies in smart cities.
2.1. Emerging smart cities
In the domain of information and communications technology (ICT),
there is no uniform understanding of what “Smart ” literally implies
( Cellary, 2013 ). Despite its popularity, this phrase is now widely used as
a synonym for nearly everything deemed contemporary and intelligent.
Perceptive, astute, shrewd, and fast are just a few synonyms for smart in
simply denitional words ( Gil-Garcia et al., 2014 ). Table 1 lists several
denitions of “smart city ”from various sources.
Fig. 1 depicts the trending of keywords ‘ Smart City ’ and ‘ Artificial
Intelligence ’ in Google Trend since 2014. The illustration shows that the
interest in articial intelligence and smart cities has increased over time.
Technology plays a crucial role in smart cities, according to
( Kar et al., 2019 ), and creative technological techniques eectively as-
sist cities in becoming smarter. Smart cities use ICT to automate pro-
cesses and improve the quality of people’s lives in urban regions. More-
over, it employs integrated intelligence technologies to improve mu-
nicipal infrastructure and enable responsive governance to involve in-
habitants in city administration. Various modern technologies and ap-
proaches allow smart service models to improve eciency and opera-
tions in healthcare, transportation, energy, education, and many other
areas ( Herath, Karunasena & Herath, 2021a ). It is well recognized that
cities can be classied as “Smart ”if they meet the criteria outlined in
Fig. 2 . Table 2 discusses the involvement of AI in the dierent domains
in smart cities described in Fig. 2 .
The 2021 smart city index ( Smart City Observatory Web, 2021 )
was presented by the Institute for Management Development (IMD)
in collaboration with Singapore University for Technology and Design
(SUTD), including important ndings on how the fast proliferation of
COVID-19 across urban populations has forced city leaders to take on
new responsibilities. In July 2021, hundreds of individuals from 118
cities were polled on their city’s technology oerings in ve major areas:
(1). Health and safety, (2). Activities, (3). Mobility, (4). Opportunities,
and (5). Governance. Table 3 depicts the top-ranked ten smart cities of
the world according to the Smart City Index 2021 record.
Global internet users have gradually increased over the last decades
( Fig. 3 ). With the development of internet technology and many other
types of networks and communication strategies, the Internet of Things
(IoTs) has become one of the most crucial forms of infrastructure in
smart cities. Furthermore, IoT technology is considered the most im-
portant and critical component of modern smart city initiatives, which
generate a large amount of data ( Al-Turjman, 2017 ; Chatterjee et al.,
2021 ). The term “Internet of Things ”can be described as a set of tech-
nologies that allow users to access data generated by various devices
through wireless and wired internet networks ( Gubbi et al., 2013 ). Over
1.6 billion IoT components and devices were used in smart cities in
2017, increasing 39% from 2015. In 2018, around 3.3 billion IoT devices
and components were estimably used ( Park et al., 2018 ). Furthermore,
IoT components and devices were expanded by 42% and 43% in 2017
and 2018, respectively. The growth of smart devices and components
in smart cities has aided the expansion of IoT-based solutions ( Khan &
Salah, 2018 ). For example, in a smart home setting, data created by
electronic household equipment, such as refrigerators, are exchanged
and saved to provide user-customized services ( Geneiatakis et al., 2017 ;
Samuel, 2016 ). It might be dicult to pinpoint the most precise and
ecient actions when dealing with enormous amounts of complex
data. AI, DRL, and ML are examples of sophisticated methodologies
that may be utilized to analyse massive data to get the best possible
conclusion.
2.2. Artificial intelligence (AI)
Articial intelligence ( Bostrom, 2017 ) was coined in 1979 by John
McCarthy, a computer scientist, who dened it as “The science and en-
gineering of making intelligent machines ”( Mathur & Modani, 2016 ).
Articial intelligence can be dened as training computers to mimic
thinking processes and even simulate human behavior ( Tecuci, 2012 ).
Moreover, it is a branch of computer science dedicated to simulating
human intelligence processes and a data-driven system that enables a
computer or software to execute tasks or make judgments. Brynjolfs-
son and McAfee have contended in their book “The Race Against the
Machine ”( Brynjolfsson & Andrew, 2012 ), as a result of signicant im-
provements in AI and ML, modern civilization will no longer be able to
sustain anything resembling full employment.
Fig. 4 depicts the six major AI technologies used in modern smart
cities.
2
H.M.K.K.M.B. Herath and M. Mittal International Journal of Information Management Data Insights 2 (2022) 100076
Fig. 1. The popularity of the keywords “Smart City ”and “Artificial Intelligence ”since 2014, (Source: Google Trend) .
Fig. 2. Major domains of smart city.
Smart city initiatives in major countries are now focusing on AI-
based smart applications. According to statistics ( New York Smart
Schools Commission Report 2018 ), the AI industry will be worth $190
billion by 2025. By 2021, global investment in cognitive and AI systems
will total $57.6 billion, with AI being deployed in 75% of enterprise
apps. China owns the largest number of smart cities in the world (Insti-
tute, 2019), which manage its cities and public spaces using metering de-
vices, cameras, embedded sensors, and other monitoring technologies,
as well as data mining, processing, and AI-based analysis techniques
( Caprotti & Liu, 2020 ; Dameri et al., 2019 ). China is said to have about
800 smart city pilot programs active or planned, accounting for more
than half of all smart city pilot programs worldwide. China’s AI has
progressed swiftly since 2015, and the Chinese government has taken
several steps to stimulate AI research, catapulting Chinese AI into a new
era ( Guo 2021 ). In recent years, AI has been used in smart city studies
and practices, with academics observing how it is becoming an increas-
ingly signicant aspect of smart urbanism. AI may promote eciency
and improve the quality of life for the smart cities of the future by draw-
ing on vast resource pools and combining contemporary machine vision,
NLP, ML, robotics and other technologies. Successful AI algorithms rely
on large amounts of data to perform useful tasks. This information may
be obtained through the use of digital and mechanical technologies that
3
H.M.K.K.M.B. Herath and M. Mittal International Journal of Information Management Data Insights 2 (2022) 100076
Table 2
AI involvement in dierent domains of smart city, Adopted from ( Bellini et al., 2022 ; Herath, Karunasena & Herath et al., 2021a ).
Smart city domain Description Involvements
Smart Mobility A smart mobility network is a network of intelligent
transportation and mobility
Trac management, Autonomous and sustainable
mobility, Supply chain resiliency, Smart routing and
parking
Education Smart education is a learning paradigm tailored to the needs of
emerging digital native generations
Smart classroom, Virtual reality-based learning
platforms, Student tracking management, Learning tools
for special needs students. Smart library
Healthcare Smart healthcare is a healthcare delivery method that uses
wearable tech, the IoTs, and mobile internet to constantly
access data and connect people, resources, and institutions
Smart hospital, Telemedicine, Telenursing, Smart
healthcare tracking, E-health record, Patient
monitoring, Pandemic predictions
Environment The concept of creating an environment with integrated
sensors, displays, and computer devices to let people
comprehend better and manage their environment
Air quality monitoring, Weather monitoring, Waste
management, Water management, Smart irrigation,
Photovoltaics
Governance Smart governance is the application of technology and
innovation to improve decision-making and planning in
governing organizations
E-governance, Decision-making policies, Disaster
prevention and management, Urban planning
Living & Infrastructures Smart living is an approach that uses the city’s infrastructure
while simultaneously enhancing people’s quality of life
Smart building, Smart home, Smart education, Smart
tourism, Smart policing
Economy A smart economy is an economy centred on technical
innovation, sustainability, a high level of social wellbeing, and
resource eciency as factors for success
Smart business, E-commerce, Retail, Smart shopping,
Peer-to-peer marketplace, Peer-to-peer labour services,
Smart supply chains, Smart sharing services
Table 3
Top ranked smart cities of the world in 2021, (Source: Smart city index 2021 ( Smart City Observatory Web, 2021 ).
City Rank in 2021 Rating 2021 Structure 2021 Technology 2021 Rank in 2020
Singapore 1 AAA AAA AAA 1
Zurich 2 AA AAA A 3
Oslo 3 AA AAA A 5
Taipei City 4 A A A 8
Lausanne 5 A AAA A New
Helsinki 6 A AA A 2
Copenhagen 7 A AA A 6
Geneva 8 A AA A 7
Auckland 9 A A A 4
Bilbao 10 BBB A BBB 24
Fig. 3. Global internet users since 2005, ( Source –ITU statistics) .
transfer and store essential data and process it to give a satisfying solu-
tion to a problem.
As shown in Fig. 4 , smart city AI technologies can be categorized
into machine learning (ML), natural language processing (NLP), speech,
vision, expert systems, and robotics. Machine learning (ML) is a sub-
set of AI that enables software programs to grow increasingly eec-
tive at predicting outcomes without explicitly programming them to
do so. Machine learning algorithms estimate new output values by us-
ing past data as input. Natural-language processing (NLP) is a branch
of AI that studies how computers interact with human languages. The
objective is to create a computer program or algorithm that can un-
derstand the contents of papers, including the nuances of language
in context. Speech recognition is an interdisciplinary topic of com-
puter science and computational linguistics that uses AI to generate re-
sults. Speech recognition, voice recognition, speech to text, and text to
speech are some of the other names. NLP and automated speech recog-
nition are two AI areas that are related yet distinct. Computer vision
( Aggarwal et al., 2021 ) is a branch of AI that allows computers and
systems to extract useful information from digital photos, videos, and
other visual inputs and then act or make suggestions based on that data.
An expert system is a computer program that employs articial intel-
ligence approaches to handle issues in a specic subject that would
normally need human skill. Robotics is the production of robots that
can-do tasks without human involvement, whereas AI is the process
of systems imitating the human mind in terms of decision-making and
learning.
4
H.M.K.K.M.B. Herath and M. Mittal International Journal of Information Management Data Insights 2 (2022) 100076
Fig. 4. Categorization of AI technologies in modern smart cities.
Fig. 5. Conceptual smart city-AI framework.
The next section of this paper discusses the proposed research
methodology.
3. Research methodology
This review aims to explore how articial intelligence is being used
in the modern smart city. We found no viable review paper on AI adop-
tion in smart cities after thoroughly examining relevant literature. This
section discusses the study’s methodology by proposing a conceptual
framework for AI uses in smart cities. Fig. 5 depicts the conceptual
framework of AI used in smart cities. As shown in the illustration, we
selected common smart city applications for the study.
We used keywords and keyword combinations such as Articial
Intelligence/AI, Machine Learning/ML, Internet of Things/IoT, Smart
Healthcare, Smart City/Smart Cities, Smart Education, Smart Infrastruc-
ture, Smart Living, Smart Security, and Smart Governance to search for
published papers in peer-reviewed journals, international conferences,
and books in electronic bibliographical sources such as Web of Science
5
H.M.K.K.M.B. Herath and M. Mittal International Journal of Information Management Data Insights 2 (2022) 100076
Fig. 6. Proposed research methodology.
(WoS) and Scopus. The initial search yielded 540 publications using the
above keywords. We studied all 249 related main content and selected
133 publications (Scopus: 97%, WoS: 73%) that are extremely relevant
to our study (publications between 2014 to 2021). The publications in-
cluded computational and analytical approaches, conceptual models,
design science, and case studies. Fig. 6 depicts the proposed research
methodology for the source selection.
The following are the inclusion criteria utilized to select the publi-
cations for the study:
1. Studies on the smart city notion have been undertaken. They ad-
dressed AI’s role in several sectors, including education, health-
care, energy, IT/IoT, mobility, security, localization, disaster and
city management, environment, and agriculture.
2. The research works appeared in a peer-reviewed English journal,
book, or conference paper.
3. The publication period is between 2014 to 2021.
4. Review of AI adoption in smart cities
Although smart cities were conceived more than twenty years ago,
fast iterations and advancements have resulted in many challenges. Af-
uence, healthier citizens, unparalleled social mobility, stronger and
more interconnected communities, and positive growth will dene cities
( Radu, 2020 ). This section identies and analyses contributing aspects
from a comprehensive review of published papers and studies. We were
mainly focused on the titles of adoption of AI in education, mobility and
smart transportation, agriculture, healthcare, environment and waste
management, privacy, security and risk management.
Allam & Dhunny (2019) have studied the urban potential of AI and
proposed a new system that binds AI and cities while assuring the in-
corporation of essential aspects such as governance, metabolism, and
also culture, all of which are recognized to be important in the eective
integration of smart cities in order to meet the sustainable development
goals (SDGs) as well as the new urban agenda. According to the PwC re-
search on the uses of Blockchain (BC) in the notion of smart city, it will
signicantly improve urban administration when combined with ML,
IoT, and AI. Robotics, NLP (machine translation, content subtraction,
text generation, question answering, and categorization), speech recog-
nition, expert systems, vision systems, and so on are only a few of the
subelds of AI. Big Data technologies enable collecting massive amounts
of data from many networks ( Kushwaha et al., 2021 ). The infrastructure
includes a diverse set of devices and sensors. AI-based technologies are
used to conduct advanced investigations on the information recorded.
Big Data and IoT will harvest more useful information for citizens, ac-
cording to ( Guevara & Auat Cheein, 2020 ), due to 5 G technology. The
automobile industry is an example of a sector that has reaped signi-
cant gains as AI and IoT have progressed. In recent years, monitoring,
analysing, and predicting environmental phenomena, particularly air
pollution levels, have become a more essential priority for public insti-
tutions. Researchers have been working on improving air quality (AQ)
prediction tools for a long time to provide consumers and governments
with more precise information about how healthy or unhealthy the air
around them is. With the advancement of articial intelligence, smart
cities are becoming increasingly capable of detecting air quality. Fur-
thermore, several AI-based algorithms are being developed to anticipate
and monitor environmental hazards such as oods and wildres.
An overview of current AI-based applications deployed in dierent
domains of smart cities is depicted in Table 4 . The articles have been
categorized into several major domains in smart cities, including educa-
tion, healthcare, energy, mobility, security, and agriculture. The table
demonstrates that interest in energy, healthcare, mobility, security and
privacy has increased over the previous eight years. This section dis-
cusses how AI has been used in smart cities in the last eight years.
4.1. AI in smart education
In recent years, the education sector has received a lot of attention
since AI applications have played a big role in a wide variety of educa-
tional elds. One of the key developments in smart education is the use
of IT technology and its AI-based applications. This section briefs the
AI-based developments for the education domain in smart cities.
Initially, we focused on the history of smart education and smart ed-
ucation initiatives in dierent countries. Malaysia’s smart school imple-
mentation plan, which began in 1997 with the smart education project,
was the rst ( Chan, 2002 ). Smart schools were adopted in this project
to strengthen the educational system to address the diculties of the
21st century. In 2006, Singapore launched its smart country master plan
( Pipe, 2010 ), which included smart schooling backed by cutting-edge
6
H.M.K.K.M.B. Herath and M. Mittal International Journal of Information Management Data Insights 2 (2022) 100076
Table 4
An overview of current AI-based applications deployed in dierent domains of smart city, (
∗
Studies related to novel COVID-19 ).
Year Education Healthcare Energy IT and IoT
Mobility and
transportation
Big data and
computing
Privacy and
security
Localization,
disaster and city
management
Environment,
waste, and
hazard
management
Agriculture and
irrigations
2014 ( Jain et al., 2014 ;
Fan et al., 2014 ;
Chou & Bui, 2014 )
( Zanella et al., 2014 ) ( Shamshiry et al.,
2014
;
Abbasi et al.,
2014
)
2015 ( Platon et al., 2015 ;
Massana i Raurich
et al., 2015
)
( Ahmad &
Mehmood, 2015
)
( Ahmad &
Mehmood, 2015
)
( Al-Ali et al.,
2015
; Anand
et al., 2015
)
2016 ( Ahmad &
Mehmood, 2016
;
Lo’ai et al., 2016 )
( Ahmad &
Mehmood, 2016
)
2017 ( Jemni &
Khribi, 2017
)
( Arfat et al., 2017 ;
Suma et al., 2017 ;
Zhao et al., 2017 )
( Song et al.,
2017
)
( dela Cruz et al.,
2017
)
2018 ( Pacheco et al.,
2018
; Bajaj &
Sharma, 2018
;
Kim et al., 2018 )
( Muhammed et al.,
2018
)
( Muhammed et al.,
2018
; Alam et al.,
2016
)
( Alam et al., 2016 ;
Arfat et al., 2018 ;
Suma et al., 2017 ;
Soomro, Miraz,
Prasanth, &
Abdullah, 2018
;
Khanna et al., 2018 )
( Suma et al.,
2017
)
( Al-
Dhubhani et al.,
2017
;
Chackravarthy et al.,
2018
;
Bappee et al.,
2018
;
Selvaganapathy et al.,
2018
)
( Lin et al., 2018 )
2019 ( Salem &
Nikitaeva, 2019
)
( Massaro et al.,
2019
; Ngiam &
Khor, 2019
;
Bruzelius et al.,
2019
)
( Hernández-
Ocaña et al.,
2019
)
( Impedovo et al.,
2019
; Qin et al.,
2019
; Liu et al.,
2019
; Hernández-
Jiménez et al., 2019
;
Celaya-Padilla et al.,
2019
;
Perez-Murueta et al.,
2019
)
( Estrada et al.,
2019
)
( Romero &
Salamea, 2019
;
Li et al., 2019 ;
Ullah et al.,
2019
)
( Crivellari &
Beinat, 2019
;
Al-Khaleefa et al.,
2019
; Kaur et al.,
2019
;
Elbaz et al.,
2019
)
( Golbaz et al.,
2019
;
Rajamanikam &
Solihin, 2019
)
( Alreshidi, 2019 ;
Shadrin et al.,
2019
;
Vincent et al.,
2019
)
( continued on next page )
7
H.M.K.K.M.B. Herath and M. Mittal International Journal of Information Management Data Insights 2 (2022) 100076
Table 4 ( continued )
Year Education Healthcare Energy IT and IoT Mobility and
transportation
Big data and
computing
Privacy and
security
Localization,
disaster and city
management
Environment,
waste, and
hazard
management
Agriculture and
irrigations
2020 ( Chang, 2020 )
∗
,
(
Patronov et al.,
2022
)
∗
,
(
Alsayed et al.,
2020
)
∗
,
(
Mollalo et al.,
2020
)
∗
,
(
Shahid et al.,
2020
)
∗
,
(
Ribeiro et al.,
2020
)
∗
, (
Ke et al.,
2020
)
∗
, (
Mei et al.,
2020
)
∗
,
(
Al-Humairi et al.,
2020
; Singh et al.,
2021
; Tuli et al.,
2020
)
∗
( Lu et al., 2020 ;
ElHusseini et al.,
2020
; Kumar et al.,
2020
; Ş erban &
Lytras, 2020
)
( Cugurullo, 2020 ;
Ge et al., 2020 ;
Noh et al., 2020 ;
Shin et al., 2020 ;
Lv et al., 2020 ;
Thanh et al., 2020 )
( Alotaibi et al.,
2020
)
( S. Singh et al.,
2020
;
Chakrabarty &
Engels, 2020
;
Rahman et al.,
2020
; S. K.
Singh, Jeong &
Park, 2020
;
Zhang et al.,
2020
)
( Barletta et al.,
2020
; Jung et al.,
2020
)
( Ye et al., 2020 ) ( Ciruela-
Lorenzo et al.,
2020
;
Ragavi et al.,
2020
)
2021 ( Ahmad et al., 2021 ;
Juanatey et al.,
2021
;
Alshmrany, 2021 )
( Herath, Karunasena
& Herath, 2021a
)
∗
,
(
Mansour et al.,
2021
; Juyal et al.,
2021
; Rathi et al.,
2021
; Muhammad &
Alhussein, 2021
;
Herath et al.,
2021b
)
∗
,
(
Herath, 2021 )
∗
,
(
Kaur et al., 2021 )
∗
,
(
Ezugwu et al.,
2021
)
∗
,
(
Ngabo et al.,
2021
)
∗
,
(
Mittal et al., 2021 )
( Selim et al., 2021 ) ( Iyer, 2021 ;
Garg et al., 2021 )
( Nasir et al.,
2021
; Chauhan &
Palivela, 2021
)
( Gondal et al.,
2021
;
Goyal et al.,
2021
)
( Sunny et al.,
2021
;
Ighalo et al.,
2021
;
Shaikh et al.,
2021
;
Zhang et al.,
2021
)
8
H.M.K.K.M.B. Herath and M. Mittal International Journal of Information Management Data Insights 2 (2022) 100076
technologies, including IoT and AI ( Hua, 2012 ). The concept called for
the establishment of eight smart schools that would focus on establish-
ing various learning settings. In 2011, Finland launched a smart educa-
tion initiative that included a system for continuous learning (SysTech).
Using user-driven motivating learning solutions, the initiative was in-
tended to enhance education for the twenty-rst century ( Mäkelä et al.,
2018 ). Australia cooperated with IBM in 2012 to develop and imple-
ment a smart, multidisciplinary education system that will connect all
educational institutions in the country ( IBM, 2012 ). The government of
South Korea has also established a smart education initiative ( Choi &
Lee, 2012 ). In 2014, the United States began the smart school program
to integrate the newest IT technologies into the classroom ( New York
Smart Schools Commission, 2014 ).
Moreover, we reviewed studies ( Ahmad et al., 2021 ;
Alshmrany, 2021 ; Bajaj & Sharma, 2018 ; Jemni & Khribi, 2017 ;
Juanatey et al., 2021 ; Kim et al., 2018 ; Pacheco et al., 2018 ; Salem
& Nikitaeva, 2019 ) related to the adoption of AI in smart education.
Some researchers have discussed the adoption of AI and computational
intelligence to improve the quality of the education system. For in-
stance, the function of Articial Intelligence Applications (AIA) in smart
education has been studied by ( Ahmad et al., 2021 ). To develop smart
education and learning systems, ( Jemni & Khribi, 2017 ) and ( Salem &
Nikitaeva, 2019 ) have discussed computational intelligence, knowledge
engineering paradigms, and AI-based smart learning frameworks.
The advent of smart devices based on intelligent technologies is con-
nected to the development of smart learning. Juanatey et al. (2021) have
engaged with long-term robotics AI education. The Robobo SmartCity
educational framework has been presented in particular. It has built
around two primary components: 1) Robobo, a smartphone-controlled
robot, and 2) a real-life smart city model. Technology is not just con-
nected to other elements of life; it may also be employed in smart class-
rooms to help students learn. Pacheco et al. (2018) have developed a
DL and Osmotic IoT computing-based smart classroom paradigm. An
emotionally aware AI smart classroom paradigm has been presented by
( Kim et al., 2018 ).
One of the most crucial parts of any learning environment
is the learning style that focuses on individual learning. Bajaj &
Sharma (2018) have proposed a framework for an education tool that
considers numerous learning models and AI approaches to assess stu-
dents’ learning styles. The proposed tool would allow users to compare
learning models to identify which is best for a given situation. This
tool has been advised to be used in a cloud setting to give a scalable
approach for quickly and easily determining learning styles. Based on
their learning styles, each learner has own way of comprehending, re-
taining, processing, and interpreting new knowledge. The capacity of an
e-learning system to assess a student’s learning style has become more
important. The growth of e-learning platforms has provided students
with more chances for online learning events. Alshmrany (2021) has
suggested a CNN-LFD method for learning style prediction. The pro-
posed e-learning system has been divided into two sections: automatic
learning style prediction and classication based on the number of learn-
ing styles included.
4.2. AI in smart energy
Modern societies rely on energy to function. Due to population
growth and increasing comfort requirements, the world’s energy con-
sumption and accompanying CO2 emissions have expanded substan-
tially. The inclusion of articial intelligence into the traditional energy
industry resulted in the development of smart energy models. Various
studies in ( Ahmad et al., 2014 ; Bourhnane et al., 2020 ; Daut et al., 2017 ;
Dong et al., 2016 ; ElHusseini et al., 2020 ; Hernández-Ocaña et al., 2019 ;
Kumar et al., 2020 ; Lilliu et al., 2019 ; Lu et al., 2020 ; Sadeghi et al.,
2020 ; Selim et al., 2021 ; Ş erban & Lytras, 2020 ; Zhong et al., 2019 )
have examined AI developments of energy eciency in sustainable
smart cities. Furthermore, AI-based energy forecasting ( Ahmad et al.,
2014 ; Bourhnane et al., 2020 ; Daut et al., 2017 ; Dong et al., 2016 ;
Sadeghi et al., 2020 ; Selim et al., 2021 ; Zhong et al., 2019 ), AI-
based cost optimization and price plan selection ( Lu et al., 2020 ), AI-
based Blockchain for smart grid ( ElHusseini et al., 2020 ; Kumar et al.,
2020 ), AI-based energy optimization ( Hernández-Ocaña et al., 2019 ;
Lilliu et al., 2019 ), and AI-based renewable energy solutions ( Ş erban
& Lytras, 2020 ) have all been addressed in this domain.
The development of fully automated smart energy grids necessitates
an accurate electric load predicting system. Both DL and gradient tree
boosting approaches have been provided by ( Selim et al., 2021 ) for as-
sessing the uncertainty in short-term electrical demand estimates. In
( Selim et al., 2021 ), the authors have trained Bayesian Deep Learning
(BDL) and Gradient boosting models that use real-world electric demand
data, demonstrating that an uncertainty estimate can be provided along-
side the forecast with minimal accuracy loss. The signicance of build-
ing energy performance grows in tandem with the degree of greenhouse
gas emissions. Sadeghi et al. (2020) have used DNNs to predict heating
load (HL) and cooling load (CL) for a range of structures. The evalua-
tions of ve dierent ML algorithms such as (1). ANN, (2). GPR, (3).
LS-SVM, (4). SVR, and (5). GMM, has presented in ( Dong et al., 2016 ),
which they have applied to four residential data sets that included smart
meters used for the electricity consumption forecasting. Ahmad et al.
(2014) have reviewed AI approaches such as SVM and ANN in creat-
ing electrical energy forecasts. The authors of ( Daut et al., 2017 ) have
examined the approaches for predicting building electrical energy use,
including traditional and AI methodologies. A new vector eld-based
SVR for predicting the energy consumption of the building has sug-
gested by ( Zhong et al., 2019 ). The ndings have shown that the sug-
gested technique performed better than generally used methods in terms
of accuracy, robustness, and generalization ability. Bourhnane et al.
(2020) have investigated various models (energy consumption predic-
tion) before settling on machine learning. They have accomplished it by
integrating ANN with genetic algorithms (GA).
Moreover, AI-based energy forecasting approaches are crucial since
most smart city utilizations require higher energy. To train an energy
consumption prediction model, a machine learning method is required.
Decision trees, ANN, SVM, and other statistical algorithms have been
used to forecast building energy usage in previous research. Some stud-
ies ( Chou & Bui, 2014 ; Fan et al., 2014 ; Jain et al., 2014 ; Massana i
Raurich et al., 2015 ; Platon et al., 2015 ) have analysed the performance
of various algorithms in predicting energy use. Table 5 depicts the com-
parison of dierent AI algorithms used in studies.
End users with smart meters and controllers may now optimize their
consumption cost portfolios by choosing from a variety of price plans of-
fered by dierent retail energy estimation systems, thanks to the growth
of AI technologies. Lu et al. (2020) have proposed a reinforcement
learning-based decision system for supporting the selection of electric-
ity price plans, which has the potential to reduce individual smart grid
end-user power payment and consumption dissatisfaction. Some authors
have integrated AI methodologies with Blockchain technologies in the
context of the smart city energy paradigm. In ( ElHusseini et al., 2020 ),
the authors have recognized two major electric vehicles (EV) challenges:
1) Susceptible charging stations and electric vehicles, 2) Non-optimal
charging schedules. In response to the challenges raised, the authors
have evaluated the integration of Blockchain and AI with the EV charg-
ing infrastructure. Kumar et al. (2020) have discussed distributed energy
resources and how IoT, AI, and Blockchain can be used in smart grids.
Moreover, they have stated that AI-based analytics, energy internet ar-
chitecture, Blockchain assistance, and IoT components have all played
a part in boosting smart grid services like dependability, stability, avail-
ability, security, resilience, and sustainability.
4.3. AI in mobility and smart transportation
Transportation, trac, and logistical issues aect most large cities
across the world. This is due to the fast-growing human population
9
H.M.K.K.M.B. Herath and M. Mittal International Journal of Information Management Data Insights 2 (2022) 100076
Table 5
Comparison of dierent AI algorithms used in ( Chou & Bui, 2014 ; Fan et al., 2014 ; Jain et al., 2014 ; Massana i Raurich
et al., 2015 ; Platon et al., 2015 ).
Ref Year
AI algorithms
ANN CBR SVM MLR ARIMA RF MLP BT MARS kNN
( Chou & Bui, 2014 ) 2014 X X
( Jain et al., 2014 ) 2014 X X
( Fan et al., 2014 ) 2014 X X X X X X X X
( Platon et al., 2015 ) 2015 X X
( Massana i Raurich et al., 2015 ) 2015 X X X
and the growing number of automobiles on the road. In designing
and managing a sustainable transportation system, technology might
be immensely benecial. We reviewed studies ( Bai et al., 2019 ; Celaya-
Padilla et al., 2019 ; Cugurullo, 2020 ; C. Englund, 2020 ; C. Englund,
2020 ; Englund et al., 2021 ; Garg et al., 2021 ; Ge et al., 2020 ; Hernández-
Jiménez et al., 2019 ; Huang et al., 2019 ; Impedovo et al., 2019 ;
Iyer, 2021 ; Khanna et al., 2018 ; Liu et al., 2019 ; Lv et al., 2020 ;
Noh et al., 2020 ; Perez-Murueta et al., 2019 ; Qin et al., 2019 ; Shin et al.,
2020 ; Soomro, Miraz, Prasanth, & Abdullah, 2018 ; Thanh et al., 2020 ;
Yi et al., 2019 ; Zhao et al., 2017 ) based on eective trac manage-
ment that include the safe integration of AI-based decision-making
( Bai et al., 2019 ; C. Englund, 2020 ; C. Englund, 2020 ; Huang et al.,
2019 ; Noh et al., 2020 ), trac monitoring/forecasting ( Englund et al.,
2021 ; Ge et al., 2020 ; Impedovo et al., 2019 ; Iyer, 2021 ; Khanna et al.,
2018 ; Liu et al., 2019 ; Qin et al., 2019 ; Soomro, Miraz, Prasanth, &
Abdullah, 2018 ; Yi et al., 2019 ; Zhao et al., 2017 ), routing ( Celaya-
Padilla et al., 2019 ; Hernández-Jiménez et al., 2019 ; Huang et al., 2019 ;
Perez-Murueta et al., 2019 ; Shin et al., 2020 ), transportation network
services ( Garg et al., 2021 ; Iyer, 2021 ), and other mobility optimization
technologies ( Lv et al., 2020 ; Thanh et al., 2020 ). Moreover, modern
AI-based smart transportation systems such as Personal Rapid Transit
(PRT) systems ( Cugurullo, 2020 ) have been discussed.
The intelligent transportation system, also known as ITS, is a con-
vergence of control systems, sensors, actuators, and Information and
Communication Technologies (ICTs) that generates massive amounts of
data and signicantly inuences the next generation of transportation in
the modern smart city. For eectively monitoring and estimating real-
time data related to the road trac ow in an urban setting, which
is a vital component of a future of smart transportation systems, ML,
AI, and DRL techniques are important. AI’s involvement in intelligent
public transit, trac control, manufacturing, safety management, and
logistics has been underlined by ( Iyer, 2021 ). Garg et al. (2021) have
used deep NLP concepts to develop a novel strategy for gaining insight
into employee engagement in a logistics organization.
AI-based vehicle trac prediction, driving and routing applications
have been discussed in articles published in a special edition of a journal.
Ge et al. (2020) have proposed the GSTGCN model, a unique DL-based
algorithm for predicting urban trac speeds. The model has consisted
of three identically structured spatial-temporal components and an ex-
ternal component. Predicting vehicular trac ow for a given date and
time period of the week is useful knowledge. In ( Impedovo et al., 2019 ),
Impedovo et al. have suggested ‘TracWave’, a time series analysis sys-
tem built with a novel generative deep learning architecture. Zhao et al.
(2017) have suggested an LSTM network-based trac prediction model.
Moreover, ( Englund et al., 2021 ) have presented an overview of AI-
based trac technologies that allow for road vehicle automation and
intelligent trac control. The authors of ( Yi et al., 2019 ) have used
a DRL-based method to forecast short-term trac ow on a roadway.
On the Gyeongbu Expressway in South Korea, they deployed the deep
LSTM-RNN for data analysis and congestion prediction.
A heavy trac state analysis of vehicular surveillance video is one
of the most vital and complex research topics of the modern intelligent
transportation system. Since the rapid advancement of transportation
networks, the prolonged growth of surveillance facilities on the street
creates huge trac data and is essential for achieving analysis objec-
tives. Qin et al. (2019) have proposed a Grassmann manifold-based NN
architecture to evaluate trac surveillance video data. They have stated
that, compared to numerous old approaches, the accuracy of trac
congestion has improved. After studying bus trac ow and its par-
ticular scenario patterns, ( Liu et al., 2019 ) have built better Spatio-
temporal residual networks to anticipate bus trac ow. Fully con-
nected NN have been utilized to capture bus scenario patterns and en-
hanced residual networks to capture bus trac ow Spatio-temporal
correlation in these networks. Perez-Murueta et al. (2019) have devel-
oped a vehicle redirection system that employs a deep learning model
to estimate the trac network’s future state in order to avoid trac
congestion.
The use of AI for decision-making or routing for mobility has in-
creased over the past decades. Englund (2020) ; Englund, (2020) has
estimated automotive and bicycle intentions at an impending road exit
using real-world trajectories. Englund has used AI techniques such as
MLP, SVM, and RF. In ( Bai et al., 2019 ), researchers have suggested a
DRL-based improved driving behavior decision-making approach in a
heterogeneous trac environment. The authors of ( Huang et al., 2019 )
have researched how to eectively use GPS trajectories data from taxis
in a given region for passenger searching. Hernández-Jiménez et al.
(2019) have employed a hybrid approach of AI to handle a frequent
routine problem, focusing on both the best next message to duplicate
and the best next hop in its path.
Some researchers have developed AI-based models to identify safety
issues such as cell phone-distracted drivers, pedestrian risky events,
human-centred threats, etc. Celaya-Padilla et al. (2019) have proposed
a novel method for detecting cell phone-distracted drivers. A ceiling-
mounted wide-angle camera and DL-CNN have been used to identify
such inattentive drivers. Noh et al. (2020) have suggested a novel
methodology for analysing potential pedestrian risky events (PPREs),
based on video data collected by trac security cameras previously de-
ployed at such crossings. Their technology has detected automobiles and
people by automatically calculating trajectories and extracting behav-
ioral information at the frame level. Shin et al. (2020) have developed
DL-based human-centred risk assessments for use in autonomous ve-
hicles. Manual driving features have been examined using real-world
driving test data to produce a naturalistic driver model that feels natu-
ral while being safe for a driver. Neural Architecture Search (NAS) and
learning from sequential data have been used to design a DNN and RNN
algorithm.
The PRT, also known as the Personal Rapid Transit system, is a new
mode of transportation designed to meet the needs of passengers. It is
also a system of self-driving automobiles that operates at a city’s un-
derground level and employs powerful AIs. Although this PRT system
has been presented for a long time, it has not received much attention
owing to technological restrictions. However, with the development of
AI into transportation systems, the personal rapid transit system has
developed to the next level. The development of the PRT system in Mas-
dar city began in 2009, shortly after the Emirati project was announced
( Cugurullo, 2020 ). Several PRT systems have been launched recently,
including CyberCab (UAE), ULTra PRT (UK, and China), and Skycube
(South Korea).
10
H.M.K.K.M.B. Herath and M. Mittal International Journal of Information Management Data Insights 2 (2022) 100076
4.4. AI in smart agriculture and irrigations
We have witnessed that agriculture has a big economic impact on
cities and countries. Agriculture automation is a big source of con-
cern and a trendy topic all around the globe. As the world’s popu-
lation grows, so does the demand for food and jobs. Farmers’ tradi-
tional practices were insucient to achieve these objectives. New au-
tomated techniques have been developed and revolutionized agricul-
ture with the support of AI. These techniques have safeguarded agri-
cultural yields against various factors such as labour challenges, cli-
mate changes, food security issues, and population increases. Multiple
researchers ( Al-Ali et al., 2015 ; Alreshidi, 2019 ; Anand et al., 2015 ;
Ciruela-Lorenzo et al., 2020 ; dela Cruz et al., 2017 ; Ragavi et al., 2020 ;
Shadrin et al., 2019 ; Vincent et al., 2019 ) have worked on AI-based ap-
plications concerning smart agriculture.
Smart agriculture is revolutionising the agricultural industry in terms
of social, economic, and environmental sustainability. Some researchers
have explored ways to combine digital technology with agriculture and
AI to development of smart, long-term agricultural systems. In ( Ciruela-
Lorenzo et al., 2020 ), Ciruela-Lorenzo and his team have presented a
detailed overview of the emergence of smart digital technologies such
as robotics, IoT, Big Data, AI, and Blockchain in agriculture. The re-
searchers in ( Alreshidi, 2019 ) have looked at existing IoT/AI technolo-
gies that have been used for smart, sustainable agriculture (SSA) and
then identied IoT/AI technological architecture that can support the
development of SSA platforms.
AI-based smart irrigation system developments are discussed in the
( Al-Ali et al., 2015 ; Anand et al., 2015 ; dela Cruz et al., 2017 ). Al-
Ali et al. (2015) have presented a smart home garden irrigation sys-
tem that can be linked with existing smart house control systems. The
master station contains an inbuilt fuzzy logic (FL) irrigation algorithm
that follows the rules to irrigate the grass and trees. The authors of
( dela Cruz et al., 2017 ) have employed neural networks to optimize wa-
ter consumption in smart farms by integrating them into the proposed
automated irrigation system and establishing an expert system. Drip ir-
rigation, a mobile network, and a fuzzy technique were employed by
( Anand et al., 2015 ). The fuzzy controller analysed the real-time data
and calculated the amount of water required.
The Internet of Things (IoT) has proven itself as a viable technique
for agriculture automation and decision-making, thanks to the rapid
growth of Wireless Sensor Networks (WSNs). Furthermore, works in
( Ragavi et al., 2020 ; Shadrin et al., 2019 ; Vincent et al., 2019 ) have
investigated sensor-based AI techniques in smart agriculture. For deter-
mining the appropriateness of agricultural land, ( Vincent et al., 2019 )
have proposed an expert system that integrates sensor networks with
AI algorithms such as ANN and MLP. This assessment is based on data
gathered from the various sensor devices used to train the system.
Shadrin et al. (2019) have proposed an AI-enhanced embedded sens-
ing system that allows for continuous analysis and in situ prediction of
growth dynamics of plant leaves. In the proposed methodology, they
have employed an RNN named LSTM as the core of AI. Agrobots can
be deployed by planting natural seeds in the soil, boosting overall agri-
cultural output performance. Using Agrobot, ( Ragavi et al., 2020 ) have
demonstrated an AI sensor-based smart agriculture approach.
4.5. AI in smart healthcare
Many traditional cities are attempting to imitate the notion of smart
city healthcare by implementing traditional technologies and devices
by merging medical resources with AI-integrated solutions. Since smart
health is connected with the ICT infrastructure of the smart city, it may
be classied as a subset of e-health. AI-integrated IoT has beneted
healthcare systems substantially. Reliability, network latency, and band-
width are just a few of the issues preventing next-generation healthcare
from becoming a reality. Rathi et al. (2021) have proposed a responsive,
reliable, low-latency, and scalable AI-enabled Internet of Things and
edge computing-based healthcare system. A neural network was utilized
to estimate transmission latency to assess system performance in a real-
world setting. Muhammed et al. (2018) have presented UbeHealth, a
ubiquitous healthcare system that addresses the next-generation health-
care challenges by combining edge computing, DL, Big Data, high-
performance computing, and the IoT.
The integration of AI into the healthcare industry is depicted in
Fig. 7 clearly and straightforwardly. The ML is a branch of AI that
uses statistical approaches to learn with or without being explicitly pro-
grammed. With the adoption of AI into the healthcare system in smart
cities, various researchers have made new developments. The authors
have discussed AI-based models to predict/detect diseases, infections,
and injury in ( Alotaibi et al., 2020 ; Juyal et al., 2021 ; Mansour et al.,
2021 ; Massaro et al., 2019 ; Muhammad & Alhussein, 2021 ; Singh et al.,
2021 ; Tuli et al., 2020 ), immunology in ( Andrés-Rodríguez et al., 2019 ;
Zhang et al., 2017 ), drug discovery in ( Ngiam & Khor, 2019 ), patient
health status prediction in ( Massaro et al., 2019 ), predicting hospitals
readmissions in ( Chaki et al., 2020 ; Raj et al., 2020 ; Uddin & Syed-
Abdul, 2020 ).
Using AI and IoT convergence approaches, ( Mansour et al., 2021 )
have developed a disease detection system for diabetes and heart-related
diseases. For illness diagnosis, the suggested technique employs a CSO-
LSTM model. Using healthcare data, the CSO-LSTM model’s perfor-
mance was validated. During testing, the proposed CSO-LSTM model
achieved maximum accuracies of 96.16 % and 97.26% in detecting
heart disease and diabetes. Juyal et al. (2021) have focused on im-
proving skin disease diagnosis and bridging the gap between diagnosis
and cure. They have also proposed AI and cloud-based IoT, with CNN
analysing medical imagery and making disease predictions. The authors
of ( Singh et al., 2021 ) have developed a reliable platform for early thy-
roid infection identication by combining fog computing, articial in-
telligence, and smart health. Massaro et al. (2019) have used an LSTM
neural network to predict patient health status, emphasising diabetes.
In ( Tuli et al., 2020 ), the authors have discussed HealthFog, a novel
framework for assessing heart diseases eectively and autonomously.
The research demonstrates a practical implementation of the LSTM tech-
nique in Decision Support Systems (DSSs) for homecare assistance and
de-hospitalization procedures.
AI in healthcare must give real-time, actionable, and tailored in-
formation to patients and clinicians to aid treatment decisions. Predic-
tive analysis using machine learning-based telemedicine services assists
hospitals or medical institutions in tracking and discharging patients
more quickly, particularly during pandemics. As shown in the studies
( Chaki et al., 2020 ; Raj et al., 2020 ; Uddin & Syed-Abdul, 2020 ), vi-
tal signs and clinical variables were benecial in determining hospital
readmissions. ( Muhammad & Alhussein, 2021 ) have proposed a vocal
pathology detection system as a part of a smart healthcare framework.
AI’s role in immunology has been discussed in ( Andrés-Rodríguez et al.,
2019 ; Zhang et al., 2017 ) studies. In ( Ngiam & Khor, 2019 ), the au-
thors have discussed how using AI for drug discovery may change cur-
rent pharmaceutical industry research and development operations. To
properly oer services across wide regions, community health systems in
remote areas require precise information about where people reside. In
hospitals, ( Mittal et al., 2021 ) have presented a web-based chatbot for
frequently asked queries (FAQ). Several machine learning techniques,
such as gradient descent and NLP algorithms, have been included in the
bot engine.
Some researchers have used imagery data and social media data to
develop systems for healthcare support. For instance, the authors of
( Bruzelius et al., 2019 ) have presented ML and DL algorithms to analyse
satellite pictures and maps of remote villages for improved healthcare
and planning. Social media has the potential to provide a constant and
ubiquitous engagement amongst healthcare stakeholders, resulting in
improved public health. Alotaibi et al. (2020) have proposed a Big Data
analytics platform by leveraging Twitter data in Arabic for the health-
care industry.
11
H.M.K.K.M.B. Herath and M. Mittal International Journal of Information Management Data Insights 2 (2022) 100076
Fig. 7. Integration of AI in smart healthcare, Adopted from ( Mak
& Pichika, 2019 ).
4.5.1. AI adoption in response to the novel COVID-19 pandemic
Coronavirus disease 2019, also known as COVID-19, caused by the
SARS-CoV-2, was discovered in December 2019 and rapidly spread
to most cities and nations worldwide ( Herath, Karunasena & Herath,
2021a ). The global spread of the COVID-19 epidemic has posed sig-
nicant obstacles to countries and urban governments ( Herath et al.,
2021b ). It is debatable if smart city initiatives play a substantial
part in the COVID-19 prevention and control procedure. AI-based
COVID-19 patient detection, pandemic prediction systems have pro-
posed in ( Al-Humairi et al., 2020 ; Alsayed et al., 2020 ; Ezugwu et al.,
2021 ; Herath, 2021 ; Herath, Karunasena & Herath, 2021a ; Kaur et al.,
2021 ; Mollalo et al., 2020 ; Ngabo et al., 2021 ; Ribeiro et al., 2020 ;
Shahid et al., 2020 ). Researchers in ( Ezugwu et al., 2021 ; Herath, 2021 ;
Herath, Karunasena & Herath, 2021a ) have engaged with the IoT and
AI-based concept or model implementations in response to COVID-
19. Researchers in ( Chang, 2020 ; Ke et al., 2020 ; Mei et al., 2020 ;
Patronov et al., 2022 ) have proposed AI-based systems to COVID-19
patient diagnosis and treatments.
Alsayed et al. (2020) have employed ML and the Susceptible–
Exposed–Infectious–Recovered (SEIR) model to forecast the epi-
demic’s progression and estimate the unreported number of infections.
Mollalo et al. (2020) have investigated the use of MLP neural networks
in modelling COVID-19 cumulative incidence at the county level across
the continental United States. To predict cases of conrmed, deaths, and
recoveries in ten major countries aected by COVID-19, ( Shahid et al.,
2020 ) have proposed prediction algorithms using SVR, LSTM, and other
AI methodologies. Ribeiro et al. (2020) have projected the total number
of conrmed COVID-19 cases in eleven Brazilian states with a high daily
incidence and rated the models based on their accuracy. In ( Kaur et al.,
2021 ), authors have engaged with ML-based system development to pre-
dict the impact of quarantine. Ngabo et al. (2021) have oered an AI-
based algorithm that accurately predicts the likely survival rate of coro-
navirus suspected patients based on a healthy immune system, regular
exercise, and age quantiles. Al-Humairi et al. (2020) have oered a con-
ceptual design solution for a smart AI helmet in response to the novel
COVID-19. In ( Herath, 2021 ; Herath, Karunasena & Herath, 2021a ),
The authors have discussed and engaged with the IoT-based AI imple-
mentations for the smart city model to mitigate the novel COVID-19.
Ezugwu et al. (2021) have established a novel framework for combating
COVID-19 in smart cities by intelligently combining machine learning
models and Internet of Things (IoT) technology. Their proposed study
aimed to improve the interoperability of ML algorithms with IoT by in-
teracting with a population and its surroundings to avert the COVID-19
pandemic.
Chang (2020) have provided a clear discussion of the function of
AI from several angles, ranging from diagnosis to therapy. The authors
have explored the use of AI in three areas: epidemiology, diagnostics,
and treatment. Patronov et al. (2022) have proposed AI-based genera-
tive models for drug design to combat COVID-19. Ke et al. (2020) have
employed AI to nd 13 drugs with anti-FIP coronavirus activity, and
further research demonstrated their anti-SARS-CoV-2 activity in clini-
cal applications. To promptly diagnose COVID-19 patients, ( Mei et al.,
2020 ) have developed a combined CNN model that incorporates chest
CT data with clinical symptoms, exposure history, and laboratory tests.
Nowadays, countries have paid attention to developing AI-based de-
velopments to mitigate the impact of COVID-19 in cities. In response to
the COVID-19 outbreak, the US, Canada, China, South Korea, and Singa-
pore launched AI-based solutions in recent years. Since pandemics have
such a signicant inuence on a country’s economy and people’s health,
governments are looking for AI initiatives that may assist in minimiz-
ing the pandemic while also increasing the economy. Several countries
have increased their spending on smart healthcare programs to address
the disease. The case studies presented in Table 6 demonstrate how ML
might help in the battle against the COVID-19 epidemic. As a result,
other countries may use machine learning tools to share their knowl-
edge in combating COVID-19.
12
H.M.K.K.M.B. Herath and M. Mittal International Journal of Information Management Data Insights 2 (2022) 100076
Table 6
AI-based smart healthcare solution for COVID-19 pandemic.
Ref Country Description
( Obeidat, 2020 ) Canada COVID-19 spread was tracked, recognized, and reported using ML and NLP by BlueDot, a Canadian company. They performed the responses
faster than the World Health Organization (WHO) or the US Centers for Disease Control and Prevention (CDCP).
( Dingli, 2020 ) China In China, 200 million security cameras have been installed. At the entrances of housing complexes, biometric scanners were also placed. All
of the data is saved in a central database, where ML algorithms analyse it to estimate the person’s potential social connections after they
leave the residence.
( Strickland, 2020 ) USA A machine learning system has been designed to give clinical decisions that aid in patient triage through training. This method is currently
being deployed at hospitals in New York to help with clinical ndings.
( Strickland, 2020 ) USA Intelligent systems are now being modied in order to foresee some COVID-19 outcomes, such as the need for incubation. Hundreds of
thousands of patient records are supplied into intelligent systems as training data, helping them to recognize disease trends.
( Park et al., 2020 ) South Korea The South Korean government recently employed articial intelligence to battle the COVID-19 outbreak by promoting proactive information
exchange, supporting individuals in recognizing the problem, and applying newly released safety regulations.
4.6. AI for security, rescue and environment in smart cities
4.6.1. AI adoption for security, rescue and hazards
Security concerns may become more prevalent as the population
of the smart city grows. Citizens have been provided with protec-
tion as a result of the enabeling AI in smart cities. In ( Bappee et al.,
2018 ; Baughman, Eggenberger, Martin, Stoessel, & Trim, 2019 ;
Chackravarthy et al., 2018 ; Romero & Salamea, 2019 ; Ullah et al.,
2019 ), authors have engaged with AI-based systems to predict or de-
tect crime, violence, and other security incidents in smart cities. AI-
based smart city solutions for cyber security have been provided by
( Chakrabarty & Engels, 2020 ; Nasir et al., 2021 ; Rahman et al., 2020 ;
Selvaganapathy et al., 2018 ; S. K. Singh et al., 2020 ; Zhang et al., 2020 ).
Some researchers such as ( Domingo, 2021 ; Huszár & Adhikarla, 2021 ;
Jung et al., 2020 ; Li et al., 2019 ; Lin et al., 2018 ; Zhang et al., 2021 ) have
presented AI-based safety precaution procedures for the early detection
of hazards, disasters, and environmental causes.
In ( Chackravarthy et al., 2018 ), the authors have suggested a smart
city crime identication system based on DRL and ANN to recognize
and investigate any illegal behavior accurately. Similarly, the authors
of ( Baughman, Eggenberger, Martin, Stoessel, & Trim, 2019 ) have pre-
sented an ML-based architecture for predicting incidents and producing
responses before they occur. In ( Ullah et al., 2019 ), Ullah et al. have
introduced a triple-staged end-to-end DL violence recognition system
based on the lightweight CNN model. Bappee et al. (2018) have de-
veloped a machine learning model for predicting crime based on geo-
graphical information for various types of crimes. Convolutional models
for detecting rearms in surveillance footage have been suggested by
( Romero & Salamea, 2019 ). They have utilized multiple CNN architec-
tures to examine the performance of the rearm identication system.
Chakrabarty & Engels (2020) have provided a methodology for se-
curing IoT-enabled smart cities utilizing Black Networks and AI to pro-
tect against existing and future cyber assaults. This structure provides
AI-enabled management solutions to be deployed both near the IoT edge
and in Big Data collections, leveraging localized data. For the resource-
constrained IoT, ( Rahman et al., 2020 ) have presented a machine
learning-based intrusion detection system. Singh et al. (2020) have pro-
posed an IoT-orientated architecture for smart cities. A cloud application
layer based on DL oers cost-eective and high-performance computing
resources for a secure smart city, utilizing Blockchain, Software-Dened
Networking (SDN), and the LSTM network. Selvaganapathy et al.
(2018) have developed a DBN and DNN-based malicious URL detec-
tion and classication system, with DBN conducting feature selection
and DNN performing binary classication. Zhang et al. (2020) have
proposed a DBN and SVM-based real-time cyber threat detection sys-
tem for smart cities. This technique has outperformed standard machine
learning methods regarding real-time detection eciency. Nasir et al.
(2021) have proposed a novel hybrid DL model that combines CNN-RNN
for fake news classication. Chauhan & Palivela (2021) have employed a
deep learning-based strategy to distinguish bogus news from true news.
The proposed model has been built using an LSTM neural network.
With the rise in environmental hazards, AI is increasingly being used
to detect threats early in smart cities. For smart cities, Big Data and AI-
based early risk warning systems for re hazards have been described by
in ( Zhang et al. (2021) . Lin et al. (2018) have developed and computed
the quantitative potential re risk using fuzzy inference and Big Data
analysis techniques. Jung et al. (2020) have proposed a novel concep-
tual framework for intelligent disaster management systems (IDMSs),
focusing on wildres and cold/heat waves. To assist decision-makers
in making quicker and more accurate judgments, the IDMSs have
leveraged massive data collected through articial intelligence with
open application programming interface (API) algorithms. Beach at-
tendance forecasting is an important surveillance tool for coastal man-
agers to provide security, rescue, health, and environmental aid services.
Domingo (2021) has developed experimental research that predicts the
number of visitors at Castelldefels beach (Barcelona, Spain) using IoT
data and DL. Huszár & Adhikarla (2021) have presented a method for
detecting live spoong in autonomous human activity identication
systems. To predict security concerns by semantic attributes recogni-
tion is crucial in crowded areas. In response to such problems, Li et al.
(2019) have presented a DN with dynamic weights and a joint loss func-
tion for pedestrian semantic attributes detection in smart surveillance.
4.6.2. AI adoption for environment and waste management
The ability of smart cities to solve environmental problems broadly
and waste management specically is an important subject that needs
to be addressed. Increasing urbanization, rising population, and eco-
nomic growth have boosted trash generation in countries across the
globe. According to the most recent data published in (World Bank,
2018) ( Waste, 2018 ), 2.0 billion tons of municipal solid waste (MSW)
were produced in 2016, with that gure expected to rise to 3.4 bil-
lion tons by 2050. The authors of ( Abbasi et al., 2014 ; Golbaz et al.,
2019 ; Gondal et al., 2021 ; Ighalo et al., 2021 ; Rajamanikam & Soli-
hin, 2019 ; Shaikh et al., 2021 ; Shamshiry et al., 2014 ; Song et al., 2017 ;
Sunny et al., 2021 ; Ye et al., 2020 ; Yoo et al., 2021 ) have investigated
the function of AI-based applications in the environment and waste man-
agement in smart cities.
The studies by ( Sunny et al., 2021 ) and ( Shaikh et al., 2021 ) have
proposed articial intelligence-based environmental monitoring sys-
tems. ( Ye et al., 2020 ) have discussed the utilization of AI to ad-
dress environmental challenges such as pollution control. The authors
of ( Ighalo et al., 2021 ) have explored articial intelligence for moni-
toring and assessing surface water quality. Some researchers have en-
gaged with AI-based developments for waste management in smart
cities. Yoo et al. (2021) have proposed a dual image-based CNN ensem-
ble model for reverse vending machine garbage categorisation. Gabor
Wavelet Transformation (GWT) has been utilized by ( Rajamanikam &
Solihin, 2019 ) to classify solid waste by convoluting an image with Ga-
bor Wavelet kernels of various sizes and orientations. The image training
database’s characteristics were used to create a supervised ANN with
the bin level grades. In ( Gondal et al., 2021 ), the authors have sug-
gested a real-time smart trash classication algorithm that categorizes
13
H.M.K.K.M.B. Herath and M. Mittal International Journal of Information Management Data Insights 2 (2022) 100076
Fig. 8. Analysis of AI-based deployments in ma-
jor smart city domains.
garbage into many categories using a hybrid method. Moreover, ML
models have been implemented, including the MLP and the ML-CNN al-
gorithms. Golbaz et al. (2019) have engaged in the study of predicting
hospital solid waste generation using MLP. Shamshiry et al. (2014) have
developed a model to estimate solid waste generation and lower the
cost of solid trash collection and transportation. Song et al. (2017) have
provided a construction and demolition waste forecasting system that
can predict the amount of each component in such waste in China.
Abbasi et al. (2014) have used SVM and hybrid Wavelet Transform-
Support Vector Machine (WT-SVM) models for solid waste generation
forecasting.
5. Discussion
The responsible AI initiative of the future society aims to provide
the knowledge, skills, and practical tools. Moreover, it will enable poli-
cymakers and key stakeholders to drive AI adoption for inclusive and
sustainable social and economic developments. Nowadays, cities are
dealing with various diculties, such as transportation, ageing infras-
tructure, environmental pollution, healthcare, privacy and security, en-
ergy consumption, etc. After carefully reviewing the selected papers, we
summarised dierent AI algorithms and their uses with major smart city
domains.
Fig. 8 illustrates the AI deployments in major smart city domains.
As described, the healthcare, mobility and transportation, privacy and
security domains have a greater impact on AI-based deployments in
smart cities. Furthermore, 23% (n = 25) of AI-based developments have
invaded the healthcare sector, while 19% (n = 21) have entered the mo-
bility sector. Smart energy management represents 10% (n = 11) of AI-
based developments in smart cities. The analysis shows that environ-
ment, waste, and hazard management only accounted for 10% (n = 11)
of AI deployments in cities. Security concerns may arise as the city’s
population grows. 11% (n = 12) of AI-based approaches have been im-
plemented for privacy and security concerns in modern smart cities.
Moreover, smart education accounts for 8% (n = 8) of AI involvement in
smart cities. AI is now used for the agriculture sector in smart cities at
7% (n = 8) . Recently 5% of AI systems have been used for Big Data and
computing in smart cities.
Table 7 depicts the use of AI methods in reviewed sources. After con-
ducting the review, we observed that AI algorithms such as ANN (20%),
RNN/LSTM (15%), CNN/R-CNN (17%), DNN (13%), and SVM/LS-SVM
(11%) are more popular in smart cities. Based on the data extracted
from Table 7 , we summarized AI algorithms used in dierent smart city
domains. Fig. 9 depicts the various AI algorithms employed in major
smart city domains.
The energy industry has grown since the urbanization of modern
cities and their society, which has a higher demand for energy require-
ments. The application of AI gives precise estimations for developing
energy maps that may be utilized for energy modelling and planning.
27% of SVM/LS-SVM and 28% of ANN-based developments have inu-
enced the energy sector.
The intelligent methods for tutoring, communication, analysis, as-
sessment, and evaluation of the student or learner, as well as monitor-
ing, process control, and optimization, play a vital role in AI’s function
in education. Recently, teacher’s and student’s tasks are performed by
AI-based systems. We observed that these technologies must be able to
communicate with the students so that the language and other elements
are understandable and make sense. After 2018, the implementation of
smart education systems in cities (mostly high populated cities) is ex-
pected to increase. The likelihood of more AI-based education partici-
pation might be related to the worldwide pandemic. Remote education
systems have been enhanced with AI-based intelligent systems since dis-
tance learning is a vital component during the pandemic. The Massive
Open Online Courses (MOOCs) concept, which sprang from open edu-
cational materials and e-learning, is now one of the newest and most
recent trends in online learning. Various researchers frequently used
ANN, RNN/LSTM, CNN/R-CNN, DNN, MLP methodologies in response
to education challenges in the notion of the smart city. 43% of CNN/R-
CNN-based solutions have been identied in the smart education sector.
Since 2019, nations have been aected by the novel COVID-19, and
cities have been challenged with a new issue in addition to their current
problems. It has also altered several aspects of smart health, logistics,
population monitoring, data security, crisis management, etc. The epi-
14
H.M.K.K.M.B. Herath and M. Mittal International Journal of Information Management Data Insights 2 (2022) 100076
Table 7
Various AI algorithms used by previous researchers ( ANN: artificial neural network, RNN: recurrent neural network CNN: convolutional neural network, DNN: deep neural network, MLP: multilayer
perceptron, LS-SVM: least squares-support vector machine, SVR: support vector regression, SVM: support vector machine, NB: naive bayes, DRL: deep reinforcement learning, LR: logistic regression,
DBN: deep belief network, FL: fuzzy logics, LSTM: long short-term memory, RL: reinforcement learning ), (
∗
Studies related to the novel COVID-19) .
AI related topic Ref Year ANN RNN/ LSTM CNN/R-CNN DNN MLP SVR SVM/LS-SVM RL/ DRL LR DBN FL NB
Smart energy management
(forecasting, planning)
( Ahmad et al., 2014 ) 2014 X X
( Dong et al., 2016 ) 2016 X X X
( Daut et al., 2017 ) 2017 X
( Zhong et al., 2019 ) 2019 X
( Sadeghi et al., 2020 ) 2020 X
( Lu et al., 2020 ) 2020 X
( Bourhnane et al., 2020 ) 2020 X
( Selim et al., 2021 ) 2021 X
Smart risk management
(warning prediction and
monitoring)
( Lin et al., 2018 ) 2018 X
( Jung et al., 2020 ) 2020 X X X
( Zhang et al., 2021 ) 2021 X
Smart environment,
geolocalization, and waste
management
( Shamshiry et al., 2014 ) 2014 X
( Abbasi et al., 2014 ) 2014 X
( Song et al., 2017 ) 2017 X
( Estrada et al., 2019 ) 2019 X
( Golbaz et al., 2019 ) 2019 X X
( Rajamanikam & Solihin, 2019 ) 2019 X
( Yoo et al., 2021 ) 2021 X
( Gondal et al., 2021 ) 2021 X
Smart security, rescue and
activity recognition
( Chackravarthy et al., 2018 ) 2018 X X
( Bappee et al., 2018 ) 2018 X
( Selvaganapathy et al., 2018 ) 2018 X X
( Ullah et al., 2019 ) 2019 X
( Romero & Salamea, 2019 ) 2019 X
( Li et al., 2019 ) 2019 X
( Zhang et al., 2020 ) 2020 X X
( Rahman et al., 2020 ) 2020 X
( S. K. Singh et al., 2020 ) 2020 X
( Domingo, 2021 ) 2021 X
( Chauhan & Palivela, 2021 ) 2021 X
( Huszár & Adhikarla, 2021 ) 2021 X
Smart building management
and life indexing
( Zhang et al., 2018 ) 2018 X
( Kaur et al., 2019 ) 2019 X X
Smart education ( Pacheco et al., 2018 ) 2018 X X
( Bajaj & Sharma, 2018 ) 2018 X
( Kim et al., 2018 ) 2018 X X
( Juanatey et al., 2021 ) 2021 X
( Alshmrany, 2021 ) 2021 X
( continued on next page )
15
H.M.K.K.M.B. Herath and M. Mittal International Journal of Information Management Data Insights 2 (2022) 100076
Table 7 ( continued )
AI related topic Ref Year ANN RNN/ LSTM CNN/R-CNN DNN MLP SVR SVM/LS-SVM RL/ DRL LR DBN FL NB
Smart healthcare ( Muhammed et al., 2018 ) 2018 X X
( Massaro et al., 2019 ) 2019 X
( Ngiam & Khor, 2019 ) 2019 X X
( Bruzelius et al., 2019 ) 2019 X
( Tuli et al., 2020 ) 2020 X
( Patronov et al., 2022 )
∗ 2020 X
( Alotaibi et al., 2020 ) 2020 X X
( Mollalo et al., 2020 )
∗ 2020 X
( Shahid et al., 2020 )
∗ 2020 X X
( Ribeiro et al., 2020 )
∗ 2020 X
( Ke et al., 2020 )
∗ 2020 X
( Mei et al., 2020 )
∗ 2020 X
( Muhammad & Alhussein, 2021 ) 2021 X
( Rathi et al., 2021 ) 2021 X
( Juyal et al., 2021 ) 2021 X
( Herath, 2021 )
∗ 2021 X
( Herath, Karunasena & Herath, 2021a )
∗ 2021 X
( Ngabo et al., 2021 )
∗ 2021 X X
( Mansour et al., 2021 ) 2021 X
Smart mobility (trac
monitoring, trac forecasting,
routing, etc.)
( Zhao et al., 2017 ) 2017 X
( Yi et al., 2019 ) 2019 X
( Impedovo et al., 2019 ) 2019 X X
( Qin et al., 2019 ) 2019 X
( Liu et al., 2019 ) 2019 X
( Hernández-Jiménez et al., 2019 ) 2019 X X
( Perez-Murueta et al., 2019 ) 2019 X X
( Celaya-Padilla et al., 2019 ) 2019 X
( Crivellari & Beinat, 2019 ) 2019 X
( Noh et al., 2020 ) 2020 X
( Shin et al., 2020 ) 2020 X X
( C. Englund, 2020 ) 2020 X X
( C. Englund, 2020 ) 2020 X X
Smart agriculture and
irrigation automation
( Al-Ali et al., 2015 ) 2015 X
( Anand et al., 2015 ) 2015 X
( dela Cruz et al., 2017 ) 2017 X
( Vincent et al., 2019 ) 2019 X X
( Shadrin et al., 2019 ) 2019 X
16
H.M.K.K.M.B. Herath and M. Mittal International Journal of Information Management Data Insights 2 (2022) 100076
Fig. 9. Analysis of dierent AI algorithm
used in major smart city domains.
17
H.M.K.K.M.B. Herath and M. Mittal International Journal of Information Management Data Insights 2 (2022) 100076
demic has only increased and justied the need for smart cities. Many
smart city applications utilized AI technologies for machine vision ap-
plications in response to the current pandemic. 21% of RNN/LSTM and
17% of ANN, DNN, and CNN/R-CNN-based deployments had the great-
est inuence on the city in the smart healthcare sector. Some research
used AI for making decisions using data collected by sensors and de-
vices. Previous research performed by Herath, Karunasena & Herath
(2021a) has found that smart cities had a lower impact rate (COVID-
19 impact rate < 2%), but conventional cities had a greater mortality
rate. He also stated that the usage of AI in smart cities has resulted in a
decrease in human-to-human interaction since AI-based services make
citizens’ lives safer.
Most smart cities are now dealing with a rise in trac ows, and con-
trolling these ows necessitates physical infrastructure as well as new
ways of thinking and modern technology. In response to such problems,
many trac management systems have been developed. Predict data
to minimize the number of challenges, including parking issues, trac
jams, and ensure intelligent routing through trip planning prediction.
ANN (16%), CNN/R-CNN (16%), and DNN (16%) algorithms have a
48% usage in the smart mobility sector while RNN/LSTM-based imple-
mentations inltrated the smart city in 26% of cases.
Because technology may easily intrude into people’s lives, the sub-
ject of risk management, privacy and security in smart cities is becoming
more popular. The smart security sector accounted for 22% of CNN/R-
CNN and 22% of DNN based implementations, while 40% of Fuzzy logic-
based implementations were employed to develop smart risk manage-
ment systems.
Cities can benet from smart agriculture by substituting limited
labour and lowering risk in the production and distribution process.
Demand-driven production and distribution are also possible with smart
agriculture. Through remote sensing, smart agricultural systems de-
crease waste, boost output, and enable the management of a larger num-
ber of resources. In the agriculture and irrigations sector, 33% of ANN
and 33% of fuzzy logic-based developments were found. Moreover, we
observed that 17% MLP and 17% RNN/LSTM were very popular in smart
agriculture and irrigation.
While AI has many advantages in many smart city services, it also
has several drawbacks, including data collection and sharing barriers,
an ethical framework for explaining AI ability, algorithms complexity,
poor design of AI systems. Due to the fragility of IoT systems and the po-
tential eects of their hacking on urban infrastructures and services, se-
curity and privacy are critical considerations in IoT deployment in smart
city applications. To guarantee a natural progression, citizens must be
informed about the advantages and disadvantages of using ICT. Psycho-
logical eect studies must be conducted, and responsibilities for both
ICT developers and users must be established.
5.1. Future direction
AI technology is being used to develop smart cities to alleviate the
strain on local resources and enhance governance and services. More
recently, the modern educational system has shifted to e-learning plat-
forms which now everyone can learn through the internet. Mobility and
transportation systems have also evolved, and researchers are still con-
ducting numerous studies to manage trac in cities. AI is now enter-
ing the security sector to safeguard citizens due to the improvement
of many IoT-based systems. With the advent of IoT-based platforms,
intelligence systems for cyber security are evolving rapidly. Countries
worldwide have faced an unprecedented health crisis since the out-
break began, raising complex social, economic, and ethical challenges.
To counteract the spread of COVID-19 and transition out of absolute
lockdown, countries must implement or are considering using arti-
cial intelligence to address the pandemic. Furthermore, AI, Blockchain,
Virtual reality/Augmented reality, IoT, and 3D printing are ve major
disruptive technologies propelling the smart city forward. In terms of
practical implications, AI-based smart city initiatives can assist govern-
ments and service providers in better dening strategies for optimizing
data exchanges and information ow throughout the service provision
process, monitoring and improving the emergence of co-creation in real-
time, and increasing service eectiveness by addressing how Big Data,
IoT, and ICTs facilitate value co-creation and advancement in smart sys-
tems.
In the future, we will engage on AI adoption in traditional cities and
how traditional city plans transformed into sustainable smart city mod-
els that collaborate with AI and Big Data. Future research will focus on
further disruptive technologies based on the AI technologies covered in
this article, such as gamication, virtual and augmented reality, wear-
able tech, 3D printing, social robots, etc.
6. Conclusion
Cities have changed dramatically due to the invention and imple-
mentation of multiple concepts such as resilient cities, sustainable cities,
and inclusive cities, to name a few. AI and the IoT are two important
technologies that have the potential to turn cities into sustainable smart
cities. This study assessed 133 papers published between 2014 and 2021
(97% indexed in Scopus and 73% in WoS). We discussed AI adoptions
in the major domains of smart cities such as healthcare, education, en-
vironment and waste management, mobility and smart transportation,
agriculture, risk management, and security. It is concluded that cities
can benet from incorporating AI into smart cities by automating opera-
tions, reducing human error, making eective data-driven decisions, im-
proving the environment through dierent systems, implementing new
commercial possibilities, and automating ecient urban management.
On the other hand, they also pose regulatory challenges, such as dis-
crimination in service delivery, privacy, legal and ethical considerations.
Furthermore, data availability, lack of qualied professionals, cost and
duration of AI initiatives, and a high unemployment rate have all been
recognized as risks and barriers to AI implementation in smart cities.
The healthcare (23%), mobility (19%), privacy and security (11%), and
energy (10%) sectors have a greater impact on AI adoption in smart
cities, according to our ndings. The healthcare sector has increased
its AI-based breakthroughs by 60% since the outbreak reached cities in
2019. AI algorithms such as ANN, RNN/LSTM, CNN/R-CNN, DNN, and
SVM/LS-SVM have a greater inuence on the various smart city sectors,
according to the analysis.
Declaration of Competing Interest
The authors declare that they have no conict of interest.
References
Abbasi, M., Abduli, M. A., Omidvar, B., & Baghvand, A. (2014). Results uncertainty of sup-
port vector machine and hybrid of wavelet transform-support vector machine models
for solid waste generation forecasting. Environmental Progress & Sustainable Energy,
33 (1), 220–228
.
Adunadepo, A. M. D., & Sunday, O. (2016,. February). Articial intelligence for sustainable
development of intelligent buildings. In Proceedings of the 9th CIDB Postgraduate
Conference, Cape Town, South Africa (pp. 1–4).
Aggarwal, A., Mittal, M., & Battineni, G. (2021). Generative adversarial network: An
overview of theory and applications. International Journal of Information Management
Data Insights, 1 (1), Article 100004
.
Aguilera, G., Galán, J. L., Campos, J. C., & Rodríguez, P. (2013). An accelerated-time
simulation for trac ow in a smart city. FEMTEC, 2013 , 26
.
Ahmad, A. S., Hassan, M. Y., Abdullah, M. P., Rahman, H. A., Hussin, F., Abdullah, H.,
& Saidur, R. (2014). A review on applications of ANN and SVM for building elec-
trical energy consumption forecasting. Renewable and Sustainable Energy Reviews, 33 ,
102–109
.
Ahmad, N., & Mehmood, R. (2015). Enterprise systems: Are we ready for future sustainable
cities. Supply Chain Management: An International Journal
.
Ahmad, N., & Mehmood, R. (2016). Enterprise systems and performance of future city
logistics. Production Planning & Control, 27 (6), 500–513
.
Ahmad, S. F., Rahmat, M. K., Mubarik, M. S., Alam, M. M. &, & Hyder, S. I. (2021).
Articial Intelligence and Its Role in Education. Sustainabilit, 13 , 12902
.
Al-Ali, A., Qasaimeh, M., Al-Mardini, M., Radder, S., & Zualkernan, I. A. (2015). Zig-
Bee-based irrigation system for home gardens. In 2015 International Conference on
Communications, Signal Processing, and their Applications (ICCSPA’15) (pp. 1–5). IEEE
.
18
H.M.K.K.M.B. Herath and M. Mittal International Journal of Information Management Data Insights 2 (2022) 100076
Alam, F., Mehmood, R., Katib, I., & Albeshri, A. (2016). Analysis of eight data mining al-
gorithms for smarter Internet of Things (IoT). Procedia Computer Science, 98 , 437–442
.
Al-Dhubhani, R., Mehmood, R., Katib, I., & Algarni, A. (2017). Location privacy in smart
cities era. In International Conference on Smart Cities, Infrastructure, Technologies and
Applications (pp. 123–138). Cham: Springer
.
Al-Humairi, S. N. S., Zainol, M. H., Razalli, H., Raya, L., & Irsyad, M. (2020). Conceptual
design: A novel COVID-19 smart AI helmet. Int J Emerg Technol, 11 (5), 389–396
.
Al-Khaleefa, A. S., Ahmad, M. R., Isa, A. A. M., Esa, M. R. M., Al-Saar, A., & Has-
san, M. H. (2019). Feature adaptive and cyclic dynamic learning based on innite
term memory extreme learning machine. Applied Sciences, 9 (5), 895
.
Allam, Z., & Dhunny, Z. A. (2019). On big data, articial intelligence and smart cities.
Cities, 89 , 80–91
.
Alotaibi, S., Mehmood, R., Katib, I., Rana, O., & Albeshri, A. (2020). Sehaa: A big data
analytics tool for healthcare symptoms and diseases detection using Twitter. Apache
Spark, and Machine Learning. Applied Sciences, 10 (4), 1398
.
Alreshidi, E. (2019). Smart Sustainable Agriculture (SSA) Solution Underpinned by Inter-
net of Things (IoT) and Articial Intelligence (AI). International Journal of Advanced
Computer Science and Applications, 10 (5)
.
Alsayed, A., Sadir, H., Kamil, R., & Sari, H. (2020). Prediction of epidemic peak and in-
fected cases for COVID-19 disease in Malaysia. 2020. International journal of environ-
mental research and public health, 17 (11), 4076
.
Alshmrany, S. (2021). Adaptive learning style prediction in e-learning environment using
levy ight distribution based CNN model. Cluster Computing , 1–14
.
Al-Turjman, F. M. (2017). Information-centric sensor networks for cognitive IoT: An
overview. Annals of Telecommunications, 72 (1-2), 3–18
.
Anand, K., Jayakumar, C., Muthu, M., & Amirneni, S. (2015). Automatic drip irrigation
system using fuzzy logic and mobile technology. In 2015 IEEE technological innovation
in ict for agriculture and rural development (TIAR) (pp. 54–58). IEEE
.
Andrés-Rodríguez, L., Borràs, X., Feliu-Soler, A., Pérez-Aranda, A., Rozadilla-Sacanell, A.,
Arranz, B., & Luciano, J. V. (2019). Machine learning to understand the immune-in-
ammatory pathways in bromyalgia. International journal of molecular sciences,
20 (17), 4231
.
Arfat, Y., Aqib, M., Mehmood, R., Albeshri, A., Katib, I., Albogami, N., &
Alzahrani, A. (2017). Enabling smarter societies through mobile big data fogs and
clouds. Procedia Computer Science, 109 , 1128–1133
.
Arfat, Y., Mehmood, R., & Albeshri, A. (2018). Parallel shortest path graph computations of
United States road network data on apache spark. In International Conference on Smart
Cities, Infrastructure, Technologies and Applications (pp. 323–336). Cham: Springer
.
Bai, Z., Shangguan, W., Cai, B., & Chai, L. (2019). Deep reinforcement learning based
high-level driving behavior decision-making model in heterogeneous trac. In 2019
Chinese Control Conference (CCC) (pp. 8600–8605). IEEE
.
Bajaj, R., & Sharma, V. (2018). Smart Education with articial intelligence based deter-
mination of learning styles. Procedia computer science, 132 , 834–842
.
Bappee, F. K., Junior, A. S., & Matwin, S. (2018). Predicting crime using spatial features.
In Canadian Conference on Artificial Intelligence (pp. 367–373). Cham: Springer
.
Barletta, V. S., Caivano, D., Dimauro, G., Nannavecchia, A., & Scalera, M. (2020). Man-
aging a Smart City Integrated Model through Smart Program Management. Applied
Sciences, 10 (2), 714
.
Bellini, P., Nesi, P., & Pantaleo, G. (2022). IoT-enabled smart cities: A review of concepts,
frameworks and key technologies. Applied Sciences, 12 (3), 1607
.
Bostrom, N. (2017). Superintelligence. Oxford . Oxford University Press .
Bourhnane, S., Abid, M. R., Lghoul, R., Zine-Dine, K., Elkamoun, N., & Benhad-
dou, D. (2020). Machine learning for energy consumption prediction and scheduling
in smart buildings. SN Applied Sciences, 2 (2), 1–10
.
Bruzelius, E., Le, M., Kenny, A., Downey, J., Danieletto, M., Baum, A., & Singh, P. (2019).
Satellite images and machine learning can identify remote communities to facili-
tate access to health services. Journal of the American Medical Informatics Association,
26 (8-9), 806–812
.
Soomro, S., Miraz, M. H., Prasanth, A., & Abdullah, M. (2018, April). Articial intelligence
enabled IoT: Trac congestion reduction in smart cities. In Smart Cities Symposium
2018 (pp. 1-6). IET.
Baughman, A.K., Eggenberger, C., Martin, A.I., Stoessel, D.S. and Trim, C.M., International
Business Machines Corp, 2019. Incident prediction and response using deep learning tech-
niques and multimodal data . U.S. Patent 10,289,949.
Brynjolfsson, Erik, & Andrew, McAfee (2012). Race against the Machine: How the Digital
Revolution Is Accelerating Innovation, Driving Productivity, and Irreversibly Trans-
forming Employment and the Economy. Digital Frontier Press
.
Caprotti, F., & Liu, D. (2020). Platform urbanism and the Chinese smart city: The co-pro-
duction and territorialisation of Hangzhou City Brain. GeoJournal , 1–15
.
Celaya-Padilla, J. M., Galván-Tejada, C. E., Lozano-Aguilar, J. S. A., Zanella-Calzada, L. A.,
Luna-García, H., Galván-Tejada, J. I., & Gamboa-Rosales, H. (2019). Texting & Driv-
ing ” detection using deep convolutional neural networks. Applied Sciences, 9 (15),
2962 .
Cellary, W. (2013). Smart governance for smart industries. In Proceedings of the 7th Inter-
national Conference on theory and practice of electronic governance (pp. 91–93)
.
Chackravarthy, S., Schmitt, S., & Yang, L. (2018). Intelligent crime anomaly detection in
smart cities using deep learning. In 2018 IEEE 4th International Conference on Collab-
oration and Internet Computing (CIC) (pp. 399–404). IEEE
.
Chaki, J., Ganesh, S. T., Cidham, S. K., & Theertan, S. A. (2020). Machine learning and
articial intelligence based Diabetes Mellitus detection and self-management: A sys-
tematic review. Journal of King Saud University-Computer and Information Sciences
.
Chakrabarty, S., & Engels, D. W. (2020). Secure smart cities framework using IoT and AI.
In 2020 IEEE Global Conference on Artificial Intelligence and Internet of Things (GCAIoT)
(pp. 1–6). IEEE
.
Chan, F. M. (2002). ICT in Malaysian schools: Policy and strategies. In In a Workshop on
the Promotion of ICT in Education to Narrow the Digital Divide (pp. 15–22)
.
Chang, A. C. (2020). Articial intelligence and COVID-19: Present state and future vision.
Intelligence-Based Medicine, 3 , Article 100012
.
Chatterjee, S., Kar, A. K., & Mustafa, S. Z. (2021). Securing IoT devices in smart cities
of India: From ethical and enterprise information system management perspective.
Enterprise Information Systems, 15 (4), 585–615
.
Chauhan, T., & Palivela, H. (2021). Optimization and improvement of fake news detection
using deep learning approaches for societal benet. International Journal of Information
Management Data Insights, 1 (2), Article 100051
.
Choi, J., & Lee, Y. (2012). The status of SMART education in KOREA. EdMedia + Inno-
vate Learning (pp. 175-178). Association for the Advancement of Computing in Education
(AACE)
.
Chou, J. S., & Bui, D. K. (2014). Modeling heating and cooling loads by articial intelli-
gence for energy-ecient building design. Energy and Buildings, 82 , 437–446
.
Chourabi, H., Nam, T., Walker, S., Gil-Garcia, J. R., Mellouli, S., Nahon, K., &
Scholl, H. J. (2012). Understanding smart cities: An integrative framework. In In 2012
45th Hawaii international conference on system sciences (pp. 2289–2297). IEEE
.
Ciruela-Lorenzo, A. M., Del-Aguila-Obra, A. R., Padilla-Meléndez, A., & Plaza-An-
gulo, J. J. (2020). Digitalization of agri-cooperatives in the smart agriculture context.
proposal of a digital diagnosis tool. Sustainability, 12 (4), 1325
.
Cowley, R., Joss, S., & Dayot, Y. (2018). In The smart city and its publics: insights from across
six uk cities: 11 (pp. 53–77). Urban Research & Practice
.
Crivellari, A., & Beinat, E. (2019). Identifying foreign tourists’ nationality from mobility
traces via LSTM neural network and location embeddings. Applied Sciences, 9 (14),
2861 .
Cugurullo, F. (2020). Urban articial intelligence: From automation to autonomy in the
smart city. Frontiers in Sustainable Cities, 2 , 38
.
Dameri, R. P., Benevolo, C., Veglianti, E., & Li, Y. (2019). Understanding smart cities as a
glocal strategy: A comparison between Italy and China. Technological Forecasting and
Social Change, 142 , 26–41
.
Daut, M. A. M., Hassan, M. Y., Abdullah, H., Rahman, H. A., Abdullah, M. P., &
Hussin, F. (2017). Building electrical energy consumption forecasting analysis using
conventional and articial intelligence methods: A review. Renewable and Sustainable
Energy Reviews, 70 , 1108–1118
.
dela Cruz, J. R., Baldovino, R. G., Bandala, A. A., & Dadios, E. P. (2017). Water usage
optimization of Smart Farm Automated Irrigation System using articial neural net-
work. In 2017 5th International Conference on Information and Communication Technol-
ogy (ICoIC7) (pp. 1–5). IEEE
.
Dingli, A. (2020). How technology is helping China crush the Coronavirus. Business Today .
Domingo, M. C. (2021). Deep learning and internet of things for beach monitoring: An
experimental study of beach attendance prediction at Castelldefels Beach. Applied Sci-
ences, 11 (22), 10735
.
Dong, B., Li, Z., Rahman, S. M., & Vega, R. (2016). A hybrid model approach for forecasting
future residential electricity consumption. Energy and Buildings, 117 , 341–351
.
Dowling, R., McGuirk, P., & Gillon, C. (2019). Strategic or piecemeal? Smart city initiatives
in Sydney and Melbourne. Urban Policy and Research, 37 (4), 429–441
.
Elbaz, K., Shen, S. L., Zhou, A., Yuan, D. J., & Xu, Y. S. (2019). Optimization of EPB shield
performance with adaptive neuro-fuzzy inference system and genetic algorithm. Ap-
plied Sciences, 9 (4), 780
.
ElHusseini, H., Assi, C., Moussa, B., Attallah, R., & Ghrayeb, A. (2020). Blockchain, AI and
smart grids: The three musketeers to a decentralized EV charging infrastructure. IEEE
Internet of Things Magazine, 3 (2), 24–29
.
Englund, C. (2020a). Action intention recognition of cars and bicycles in intersections.
International Journal of Vehicle Design, 83 (2-4), 103–121
.
Englund, C. (2020b). Aware and intelligent infrastructure for action intention recognition
of cars and bicycles. In VEHITS (pp. 281–288)
.
Englund, C., Aksoy, E. E., Alonso-Fernandez, F., Cooney, M. D., Pashami, S., & Ås-
trand, B. (2021). AI perspectives in Smart Cities and Communities to enable road
vehicle automation and smart trac control. Smart Cities,, 4 (2), 783–802
.
Estrada, E., Martinez Vargas, M. P., Gómez, J., Peña Pérez Negron, A., López, G. L., & Ma-
ciel, R. (2019). Smart cities big data algorithms for sensors location. Applied Sciences,
9 (19), 4196
.
Ezugwu, A. E., Hashem, I. A. T., Oyelade, O. N., Almutari, M., Al-Garadi, M. A., Abdul-
lahi, I. N., & Chiroma, H. (2021). A novel smart city-based framework on perspectives
for application of machine learning in combating COVID-19 (p. 2021). BioMed Research
International
.
Fan, C., Xiao, F., & Wang, S. (2014). Development of prediction models for next-day build-
ing energy consumption and peak power demand using data mining techniques. Ap-
plied Energy, 127 , 1–10
.
Fernandez-Anez, V., Fernández-Güell, J. M., & Ginger, R. (2018). Smart city implemen-
tation and discourses: An integrated conceptual model. The case of Vienna. Cities, 78 ,
4–16
.
Garg, R., Kiwelekar, A. W., Netak, L. D., & Ghodake, A. (2021). i-Pulse: A NLP based novel
approach for employee engagement in logistics organization. International Journal of
Information Management Data Insights, 1 (1), Article 100011
.
Ge, L., Li, S., Wang, Y., Chang, F., & Wu, K. (2020). Global spatial-temporal graph
convolutional network for urban trac speed prediction. Applied Sciences, 10 (4),
1509
.
Geneiatakis, D., Kounelis, I., Neisse, R., Nai-Fovino, I., Steri, G., & Baldini, G. (2017). Se-
curity and privacy issues for an IoT based smart home. In 2017 40th International Con-
vention on Information and Communication Technology, Electronics and Microelectronics
(MIPRO) (pp. 1292–1297). IEEE
.
Gil-Garcia, J. R., Helbig, N., & Ojo, A. (2014). Being smart: Emerging technologies and
innovation in the public sector. Government information quarterly, 31 , I1–I8
.
Golbaz, S., Nabizadeh, R., & Sajadi, H. S. (2019). Comparative study of predicting hospi-
tal solid waste generation using multiple linear regression and articial intelligence.
Journal of Environmental Health Science and Engineering, 17 (1), 41–51
.
19
H.M.K.K.M.B. Herath and M. Mittal International Journal of Information Management Data Insights 2 (2022) 100076
Gondal, A. U., Sadiq, M. I., Ali, T., Irfan, M., Shaf, A., Aamir, M., & Kantoch, E. (2021). Real
time multipurpose smart waste classication model for ecient recycling in smart
cities using multilayer convolutional neural network and perceptron. Sensors, 21 (14),
4916
.
Goodall, B. (1987). The Penguin dictionary of human geography . Pun Books .
Goyal, H. R., Ghanshala, K. K., & Sharma, S. (2021). Post ood management system based
on smart IoT devices using AI approach. Materials Today: Proceedings
.
Gubbi, J., Buyya, R., Marusic, S., & Palaniswami, M. (2013). Internet of Things (IoT): A vi-
sion, architectural elements, and future directions. Future generation computer systems,
29 (7), 1645–1660
.
Guevara, L., & Auat Cheein, F. (2020). The role of 5 G technologies: Challenges in smart
cities and intelligent transportation systems. Sustainability, 12 (16), 6469
.
Guo. (2021). Artificial intelligence and the mediation of social needs in smart-city initiatives:
A critical analysis Presentation, Dublin
.
Herath, H.M.K.K.M.B. (2021). Internet of Things (IoT) enable designs for identify and
control the COVID-19 pandemic. In Artificial intelligence for COVID-19 (pp. 423–436).
Cham: Springer
.
Herath, H. M. K. K. M. B., Karunasena, G. M. K. B., & Herath, H. M. W. T. (2021a). De-
velopment of an IoT based systems to mitigate the impact of COVID-19 pandemic in
smart cities. In Machine intelligence and data analytics for sustainable future smart cities
(pp. 287–309). Cham: Springer .
Herath, H. M. K. K. M. B., Karunasena, G. M. K. B., Madhusanka, B. G. D. A., &
Priyankara, H. D. N. S. (2021b). Internet of Medical Things (IoMT) enabled TeleCOVID
system for diagnosis of COVID-19 patients. In Sustainability measures for COVID-19
pandemic (pp. 253–274). Singapore: Springer
.
Hernández-Jiménez, R., Cardenas, C., & Munoz Rodriguez, D. (2019). Modeling and solu-
tion of the routing problem in vehicular delay-tolerant networks: A dual, deep learning
perspective. Applied Sciences, 9 (23), 5254
.
Hernández-Ocaña, B., Hernández-Torruco, J., Chávez-Bosquez, O., Calva-Yáñez, M. B., &
Portilla-Flores, E. A. (2019). Bacterial foraging-based algorithm for optimizing the
power generation of an isolated microgrid. Applied Sciences, 9 (6), 1261
.
Hua, M. T. A. (2012). Promises and threats: IN2015 Masterplan to pervasive computing
in Singapore. Science, Technology and Society,, 17 (1), 37–56
.
Huang, Z., Shan, G., Cheng, J., & Sun, J. (2019). TRec: An ecient recommendation sys-
tem for hunting passengers with deep neural networks. Neural Computing and Appli-
cations, 31 (1), 209–222
.
Huszár, V. D., & Adhikarla, V. K. (2021). Live spoong detection for automatic human
activity recognition applications. Sensors, 21 (21), 7339
.
IBM, Smart Education (2012). https://www.ibm.com/smarterplanet/ .
Ighalo, J. O., Adeniyi, A. G., & Marques, G. (2021). Articial intelligence for surface water
quality monitoring and assessment: A systematic literature analysis. Modeling Earth
Systems and Environment, 7 (2), 669–681
.
Impedovo, D., Dentamaro, V., Pirlo, G., & Sarcinella, L. (2019). TracWave: Generative
deep learning architecture for vehicular trac ow prediction. Applied Sciences, 9 (24),
5504
.
Ingwersen, P., & Serrano-López, A. E. (2018). Smart city research 1990–2016. Scientomet-
rics, 117 (2), 1205–1236
.
Iyer, L. S. (2021). AI enabled applications towards intelligent transportation. Transporta-
tion Engineering, 5 , Article 100083
.
Jain, R. K., Damoulas, T., & Kontokosta, C. E. (2014). Towards data-driven energy con-
sumption forecasting of multi-family residential buildings: Feature selection via the
lasso. In Computing in civil and building engineering: 2014 (pp. 1675–1682)
.
Jemni, M., & Khribi, M. K. (2017). The alecso smart learning framework. In Innovations in
smart learning (pp. 91–101). Singapore: Springer
.
Juanatey, D., Naya, M., Baamonde, T., & Bellas, F. (2021). Developing a Simulation Model
for Autonomous Driving Education in the Robobo SmartCity Framework. Engineering
Proceedings, 7 (1), 49
.
Jung, D., Tran Tuan, V., Dai Tran, Q., Park, M., & Park, S. (2020). Conceptual framework
of an intelligent decision support system for smart city disaster management. Applied
Sciences, 10 (2), 666
.
Juyal, S., Sharma, S., & Shukla, A. S. (2021). Smart skin health monitoring using AI-en-
abled cloud-based IoT. In Materials Today: Proceedings
.
Kapoor, N., Ahmad, N., Nayak, S. K., Singh, S. P., Ilavarasan, P. V., & Ramamoor-
thy, P. (2021). Identifying infrastructural gap areas for smart and sustainable tribal
village development: A data science approach from India. International Journal of In-
formation Management Data Insights, 1 (2), Article 100041 .
Kar, A. K., Gupta, M. P., Ilavarasan, P. V., & Dwivedi, Y. K. (2017a). Advances in smart
cities: smarter people, governance, and solutions (Eds.). CRC Press
.
Kar, A. K., Ilavarasan, P. V., Gupta, M. P., Dwivedi, Y. K., Mäntymäki, M., Janssen, M.,
& Al-Sharhan, S. (2017b). Digital Nations–Smart Cities, Innovation, and Sustainability:
16th IFIP WG 6.11 Conference on e-Business, e-Services, and e-Society, I3E 2017, Delhi,
India, November 21–23, 2017, Proceedings (Vol. 10595) (Eds.). Springer
.
Kar, A. K., Ilavarasan, V., Gupta, M. P., Janssen, M., & Kothari, R. (2019). Moving beyond
smart cities: Digital nations for social innovation & sustainability. Information Systems
Frontiers, 21 (3), 495–501
.
Kassens-Noor, E., & Hintze, A. (2020). Cities of the future? The potential impact of artificial
intelligence, 1 (2), 192–197 AI
.
Kaur, A., Mittal, N., Khosla, P. K., & Mittal, M. (2021). Machine learning tools to predict
the impact of quarantine. In Predictive and preventive measures for covid-19 pandemic
(pp. 307–323). Singapore: Springer
.
Kaur, M., Dhalaria, M., Sharma, P. K., & Park, J. H. (2019). Supervised machine-
learning predictive analytics for national quality of life scoring. Applied Sciences, 9 (8),
1613
.
Ke, Y. Y., Peng, T. T., Yeh, T. K., Huang, W. Z., Chang, S. E., Wu, S. H., & Chen, C. T. (2020).
Articial intelligence approach ghting COVID-19 with repurposing drugs. Biomedical
journal, 43 (4), 355–362
.
Khan, M. A., & Salah, K. (2018). IoT security: Review, blockchain solutions, and open
challenges. Future generation computer systems, 82 , 395–411
.
Khan, Z., Anjum, A., & Kiani, S. L. (2013). Cloud based big data analytics for smart future
cities. In 2013 IEEE/ACM 6th International Conference on Utility and Cloud Computing
(pp. 381–386). IEEE
.
Khanna, A., Goyal, R., Verma, M., & Joshi, D. (2018). Intelligent trac management sys-
tem for smart cities. In International Conference on Futuristic Trends in Network and
Communication Technologies (pp. 152–164). Singapore: Springer
.
Kim, Y., Soyata, T., & Behnagh, R. F. (2018). Towards emotionally aware AI smart class-
room: Current issues and directions for engineering and education. IEEE Access, 6 ,
5308–5331
.
Kumar, N. M., Chand, A. A., Malvoni, M., Prasad, K. A., Mamun, K. A., Islam, F. R., &
Chopra, S. S. (2020). Distributed energy resources and the application of AI, IoT, and
blockchain in smart grids. Energies, 13 (21), 5739
.
Kushwaha, A. K., Kar, A. K., & Dwivedi, Y. K. (2021). Applications of big data in emerging
management disciplines: A literature review using text mining. International Journal
of Information Management Data Insights, 1 (2), Article 100017
.
Li, Y., Tong, G., Li, X., Wang, Y., Zou, B., & Liu, Y. (2019). PARNet: A joint loss function
and dynamic weights network for pedestrian semantic attributes recognition of smart
surveillance image. Applied Sciences, 9 (10), 2027
.
Lilliu, F., Loi, A., Recupero, D. R., Sisinni, M., & Vinyals, M. (2019). An uncertainty-aware
optimization approach for exible loads of smart grid prosumers: A use case on the
Cardi energy grid. Sustainable Energy, Grids and Networks,, 20 , Article 100272
.
Lin, H., Liu, X., Wang, X., & Liu, Y. (2018). A fuzzy inference and big data analysis algo-
rithm for the prediction of forest re based on rechargeable wireless sensor networks.
Sustainable Computing: Informatics and Systems, 18 , 101–111
.
Liu, P., Zhang, Y., Kong, D., & Yin, B. (2019). Improved spatio-temporal residual networks
for bus trac ow prediction. Applied Sciences, 9 (4), 615
.
Lo’ai, A. T., Basalamah, A., Mehmood, R., & Tawalbeh, H. (2016). Greener and smarter
phones for future cities: Characterizing the impact of GPS signal strength on power
consumption. IEEE Access, 4 , 858–868
.
Lu, T., Chen, X., McElroy, M. B., Nielsen, C. P., Wu, Q., & Ai, Q. (2020). A reinforcement
learning-based decision system for electricity pricing plan selection by smart grid end
users. IEEE Transactions on Smart Grid, 12 (3), 2176–2187
.
Lv, Z., Lou, R., & Singh, A. K. (2020). AI empowered communication systems for intelligent
transportation systems. IEEE Transactions on Intelligent Transportation Systems
.
Mak, K. K., & Pichika, M. R. (2019). Articial intelligence in drug development: Present
status and future prospects. Drug discovery today, 24 (3), 773–780
.
Mäkelä, T., Helfenstein, S., Lerkkanen, M. K., & Poikkeus, A. M. (2018). Student partic-
ipation in learning environment improvement: Analysis of a co-design project in a
Finnish upper secondary school. Learning Environments Research, 21 (1), 19–41
.
Mansour, R. F., El Amraoui, A., Nouaouri, I., Díaz, V. G., Gupta, D., & Kumar, S. (2021).
In Artificial intelligence and internet of things enabled disease diagnosis model for smart
healthcare systems: 9 (pp. 45137–45146). IEEE Access
.
Massana i Raurich, J., Pous i Sabadí, C., Burgas Nadal, L., Meléndez i Frigola, J., & Colomer
Llinàs, J. (2015). Short-term load forecasting in a non-residential building contrasting
models and attributes. Energy and Buildings, 92 , 322–330 2015
.
Massaro, A., Maritati, V., Giannone, D., Convertini, D., & Galiano, A. (2019). LSTM DSS
automatism and dataset optimization for diabetes prediction. Applied Sciences, 9 (17),
3532
.
Mathur, S., & Modani, U. S. (2016). Smart City-a gateway for articial intelligence in
India. In 2016 IEEE Students’ Conference on Electrical, Electronics and Computer Science
(SCEECS) (pp. 1–3). IEEE
.
Mei, X., Lee, H. C., Diao, K. Y., Huang, M., Lin, B., Liu, C., & Yang, Y. (2020). Articial in-
telligence–enabled rapid diagnosis of patients with COVID-19. Nature medicine, 26 (8),
1224–1228
.
Mittal, M., Battineni, G., Singh, D., Nagarwal, T., & Yadav, P. (2021). Web-based chatbot
for Frequently Asked Queries (FAQ) in Hospitals. Journal of Taibah University Medical
Sciences, 16 (5), 740–746
.
Mollalo, A., Rivera, K. M., & Vahedi, B. (2020). Articial neural network modeling of
novel coronavirus (COVID-19) incidence rates across the continental United States.
International journal of environmental research and public health, 17 (12), 4204
.
Muhammad, G., & Alhussein, M. (2021). In Convergence of artificial intelligence and
internet of things in smart healthcare: a case study of voice pathology detection: 9
(pp. 89198–89209). Ieee Access
.
Muhammed, T., Mehmood, R., Albeshri, A., & Katib, I. (2018). UbeHealth: A personalized
ubiquitous cloud and edge-enabled networked healthcare system for smart cities. IEEE
Access, 6 , 32258–32285
.
Mustafa, S. Z., & Kar, A. K. (2017). Evaluating multi-dimensional risk for digital
Services in Smart Cities. In Conference on e-Business, e-Services and e-Society
(pp. 23–32). Springer
.
Nasir, J. A., Khan, O. S., & Varlamis, I. (2021). Fake news detection: A hybrid CNN-RNN
based deep learning approach. International Journal of Information Management Data
Insights, 1 (1), Article 100007
.
Nations, U. (2018). The world’s cities in 2018. Department of economic and social aairs.
Population Division, World Urbanization Prospects , 1–34
.
New York Smart Schools Commission, Report. https://www.governor.ny.gov/sites/
governor.ny.gov
.
New York Smart Schools Commission Report 2018. (2018). Articial Intelligence Mar-
ket - 2025 | MarketsandMarkets. Marketsandmarkets.com .
www.marketsandmarkets
.com/Market-Reports/articial-intelligence-market-74851580.html
.
Ngabo, D., Dong, W., Ibeke, E., Iwendi, C., & Masabo, E. (2021). Tackling pandemics in
smart cities using machine learning architecture. Mathematical biosciences and engi-
neering, 18 (6)
.
Ngiam, K. Y., & Khor, W. (2019). Big data and machine learning algorithms for health-care
delivery. The Lancet Oncology, 20 (5), e262–e273
.
20
H.M.K.K.M.B. Herath and M. Mittal International Journal of Information Management Data Insights 2 (2022) 100076
Noh, B., No, W., Lee, J., & Lee, D. (2020). Vision-based potential pedestrian risk analysis on
unsignalized crosswalk using data mining techniques. Applied Sciences, 10 (3), 1057
.
Obeidat, S. (2020). How articial intelligence is helping ght the COVID-19 pandemic.
Business Connect , Article 348368 Available at https://www. entrepreneur.com/article
.
Pacheco, A., Cano, P., Flores, E., Trujillo, E., & Marquez, P. (2018). A smart classroom
based on deep learning and osmotic IoT computing. In 2018 Congreso internacional de
innovación y tendencias en ingeniería (CONIITI) (pp. 1–5). IEEE
.
Park, E., Del Pobil, A. P., & Kwon, S. J. (2018). The role of Internet of Things (IoT) in
smart cities: Technology roadmap-oriented approaches. Sustainability, 10 (5), 1388
.
Park, Y. J., Choe, Y. J., Park, O., Park, S. Y., Kim, Y. M., Kim, J., & Jeong, E. K. (2020).
Contact tracing during coronavirus disease outbreak, South Korea, 2020. Emerging
infectious diseases, 26 (10), 2465
.
Patronov, A., Papadopoulos, K., & Engkvist, O. (2022). Has articial intelligence impacted
drug discovery? In In artificial intelligence in drug design (pp. 153–176). New York, NY:
Humana
.
Perez-Murueta, P., Gómez-Espinosa, A., Cardenas, C., & Gonzalez-Mendoza, M. (2019).
Deep learning system for vehicular re-routing and congestion avoidance. Applied Sci-
ences, 9 (13), 2717
.
Pinna, F., Masala, F., & Garau, C. (2017). Urban policies and mobility trends in Italian
smart cities. Sustainability, 9 (4), 494
.
Pipe, R. (2010). Singapore advances intelligent nation master plan. Journal of E-Gover-
nance, 33 (177), 225–236
.
Platon, R., Dehkordi, V. R., & Martel, J. (2015). Hourly prediction of a building’s electric-
ity consumption using case-based reasoning, articial neural networks and principal
component analysis. Energy and Buildings, 92 , 10–18
.
Qin, P., Zhang, Y., Wang, B., & Hu, Y. (2019). Grassmann manifold based state analysis
method of trac surveillance video. Applied Sciences, 9 (7), 1319
.
Radu, L. D. (2020). Disruptive technologies in smart cities: A survey on current trends and
challenges. Smart Cities, 3 (3), 1022–1038
.
Ragavi, B., Pavithra, L., Sandhiyadevi, P., Mohanapriya, G. K., & Harikirubha, S. (2020).
Smart agriculture with AI Sensor by using agrobot. In 2020 Fourth International Con-
ference on Computing Methodologies and Communication (ICCMC) (pp. 1–4). IEEE
.
Rahman, M. A., Asyhari, A. T., Leong, L. S., Satrya, G. B., Tao, M. H., &
Zolkipli, M. F. (2020). Scalable machine learning-based intrusion detection system
for IoT-enabled smart cities. Sustainable Cities and Society, 61 , Article 102324
.
Raj, P., Chatterjee, J. M., Kumar, A., & Balamurugan, B. (2020). Internet of things use cases
for the healthcare industry (Eds.). Springer Nature
.
Rajamanikam, A., & Solihin, M. I. J. I. J. I. T. E. E. (2019). Solid waste bin classica-
tion using Gabor wavelet transform. International Journal of Innovative Technology and
Exploring Engineering, 8 , 114–117
.
Rathi, V. K., Rajput, N. K., Mishra, S., Grover, B. A., Tiwari, P., Jaiswal, A. K., & Hos-
sain, M. S. (2021). An edge AI-enabled IoT healthcare monitoring system for smart
cities. Computers & Electrical Engineering, 96 , Article 107524
.
Ribeiro, M. H. D. M., da Silva, R. G., Mariani, V. C., & dos Santos Coelho, L. (2020).
Short-term forecasting COVID-19 cumulative conrmed cases: Perspectives for Brazil.
Chaos, Solitons & Fractals, 135 , Article 109853
.
Romero, D., & Salamea, C. (2019). Convolutional models for the detection of rearms in
surveillance videos. Applied Sciences, 9 (15), 2965
.
Sadeghi, A., Younes Sinaki, R., Young, W. A., & Weckman, G. R. (2020). An intelligent
model to predict energy performances of residential buildings based on deep neural
networks. Energies, 13 (3), 571
.
Salem, A. B. M., & Nikitaeva, A. Y. (2019). Knowledge engineering paradigms for smart ed-
ucation and learning systems. In 2019 42nd international convention on information and
communication technology, electronics and microelectronics (MIPRO) (pp. 1571–1574).
IEEE
.
Samuel, S. S. I. (2016). A review of connectivity challenges in IoT-smart home. In 2016
3rd MEC International conference on big data and smart city (ICBDSC) (pp. 1–4). IEEE
.
Selim, M., Zhou, R., Feng, W., & Quinsey, P. (2021). Estimating energy forecasting uncer-
tainty for reliable AI autonomous smart grid design. Energies, 14 (1), 247
.
Selvaganapathy, S., Nivaashini, M., & Natarajan, H. (2018). Deep belief network based
detection and categorization of malicious URLs. Information Security Journal: A Global
Perspective, 27 (3), 145–161
.
Ş erban, A. C., & Lytras, M. D. (2020). Articial intelligence for smart renewable energy
sector in europe —smart energy infrastructures for next generation smart cities. IEEE
Access, 8 , 77364–77377
.
Shadrin, D., Menshchikov, A., Somov, A., Bornemann, G., Hauslage, J., & Fe-
dorov, M. (2019). Enabling precision agriculture through embedded sensing with
articial intelligence. IEEE Transactions on Instrumentation and Measurement, 69 (7),
4103–4113
.
Shahid, F., Zameer, A., & Muneeb, M. (2020). Predictions for COVID-19 with deep learning
models of LSTM, GRU and Bi-LSTM. Chaos, Solitons & Fractals, 140 , Article 110212
.
Shaikh, R. A. J., Naidu, H., & Kokate, P. A. (2021). Next-generation wsn for environmental
monitoring employing big data analytics, machine learning and articial intelligence.
In Evolutionary computing and mobile sustainable networks (pp. 181–196). Singapore:
Springer
.
Shamshiry, E., Mokhtar, M., Abdulai, A. M., Komoo, I., & Yahaya, N. (2014). Combining
articial neural network-genetic algorithm and response surface method to predict
waste generation and optimize cost of solid waste collection and transportation pro-
cess in Langkawi Island, Malaysia. Malaysian Journal of Science, 33 (2), 118–140
.
Shin, D., Kim, H. G., Park, K. M., & Yi, K. (2020). Development of deep learning based hu-
man-centered threat assessment for application to automated driving vehicle. Applied
Sciences, 10 (1), 253
.
Singh, P. D., Dhiman, G., & Sharma, R. (2021). Internet of Things for sustaining a smart
and secure healthcare system. Sustainable computing: informatics and systems
.
Singh, S. K., Jeong, Y. S., & Park, J. H. (2020a). A deep learning-based IoT-oriented in-
frastructure for secure smart city. Sustainable Cities and Society, 60 , Article 102252
.
Singh, S., Sharma, P. K., Yoon, B., Shojafar, M., Cho, G. H., & Ra, I. H. (2020b). Conver-
gence of blockchain and articial intelligence in IoT network for the sustainable smart
city. Sustainable Cities and Society, 63 , Article 102364
.
Smart City Observatory Web, Page.. IMD business school . https://www.imd.org/smart-city-
observatory/Home/
.
Song, Y., Wang, Y., Liu, F., & Zhang, Y. (2017). Development of a hybrid model to predict
construction and demolition waste: China as a case study. Waste Management, 59 ,
350–361
.
Sterbenz, J. P. (2017). Smart city and IoT resilience, survivability, and disruption toler-
ance: Challenges, modelling, and a survey of research opportunities. In 2017 9th in-
ternational workshop on resilient networks design and modeling (RNDM) (pp. 1–6). IEEE
.
Strickland, E. (2020). AI can help hospitals triage COVID-19 patients-IEEE spectrum. IEEE
Spectrum: Technology, Engineering, and Science News
.
Suma, S., Mehmood, R., & Albeshri, A. (2017a). Automatic event detection in smart cities
using big data analytics. In International Conference on Smart Cities, Infrastructure, Tech-
nologies and Applications (pp. 111–122). Cham: Springer
.
Suma, S., Mehmood, R., Albugami, N., Katib, I., & Albeshri, A. (2017b). Enabling next
generation logistics and planning for smarter societies. Procedia Computer Science, 109 ,
1122–1127
.
Sunny, A. I., Zhao, A., Li, L., & Kanteh Sakiliba, S. (2021). Low-cost IoT-based sensor
system: A case study on harsh environmental monitoring. Sensors, 21 (1), 214
.
Tecuci, G. (2012). Articial intelligence. Wiley Interdisciplinary Reviews: Computational
Statistics, 4 , 168–180
.
Thanh, T. T. M., Ly, H. B., & Pham, B. T. (2020). A possibility of AI application on mode–
choice prediction of transport users in Hanoi. In CIGOS 2019, innovation for sustainable
infrastructure (pp. 1179–1184). Singapore: Springer
.
Tuli, S., Basumatary, N., Gill, S. S., Kahani, M., Arya, R. C., Wander, G. S., &
Buyya, R. (2020). HealthFog: An ensemble deep learning based Smart Healthcare Sys-
tem for Automatic Diagnosis of Heart Diseases in integrated IoT and fog computing
environments. Future Generation Computer Systems, 104 , 187–200
.
Uddin, M., & Syed-Abdul, S. (2020). Data analytics and applications of the wearable sen-
sors in healthcare: An overview. Sensors, 20 (5), 1379
.
Ullah, F. U. M., Ullah, A., Muhammad, K., Haq, I. U., & Baik, S. W. (2019). Violence de-
tection using spatiotemporal features with 3D convolutional neural network. Sensors,
19 (11), 2472
.
Verma, S., Sharma, R., Deb, S., & Maitra, D. (2021). Articial intelligence in marketing:
Systematic review and future research direction. International Journal of Information
Management Data Insights, 1 (1), Article 100002
.
Vincent, D. R., Deepa, N., Elavarasan, D., Srinivasan, K., Chauhdary, S. H., &
Iwendi, C. (2019). Sensors driven AI-based agriculture recommendation model for
assessing land suitability. Sensors, 19 (17), 3667
.
a Waste, W. (2018). 2.0: A global snapshot of solid waste management to 2050/Silpa Kaza.
Lisa Yao, Perinaz Bhada-Tata and Frank Van Woerden. World banc group
.
Ye, Z., Yang, J., Zhong, N., Tu, X., Jia, J., & Wang, J. (2020). Tackling environmental
challenges in pollution controls using articial intelligence: A review. Science of the
Total Environment, 699 , Article 134279
.
Yi, H., Bui, K. H. N., & Jung, H. (2019). Implementing a deep learning framework for short
term trac ow prediction. In Proceedings of the 9th international conference on web
intelligence, mining and semantics (pp. 1–8)
.
Yoo, T., Lee, S., & Kim, T. (2021). Dual image-based CNN ensemble model for waste
classication in reverse vending machine. Applied Sciences, 11 (22), 11051
.
Zanella, A., Bui, N., Castellani, A., Vangelista, L., & Zorzi, M. (2014). Internet of things
for smart cities. IEEE Internet of Things journal, 1 (1), 22–32
.
Zhang, H., Li, Y., Lv, Z., Sangaiah, A. K., & Huang, T. (2020). A real-time and ubiquitous
network attack detection based on deep belief network and support vector machine.
IEEE/CAA Journal of Automatica Sinica, 7 (3), 790–799
.
Zhang, J., Friberg, I. M., Kift-Morgan, A., Parekh, G., Morgan, M. P., Liuzzi, A. R., &
Eberl, M. (2017). Machine-learning algorithms dene pathogen-specic local immune
ngerprints in peritoneal dialysis patients with bacterial infections. Kidney interna-
tional, 92 (1), 179–191
.
Zhang, W., Hu, W., & Wen, Y. (2018). Thermal comfort modeling for smart buildings: A
ne-grained deep learning approach. IEEE Internet of Things Journal, 6 (2), 2540–2549
.
Zhang, Y., Geng, P., Sivaparthipan, C. B., & Muthu, B. A. (2021). Big data and articial
intelligence based early risk warning system of re hazard for smart cities. Sustainable
Energy Technologies and Assessments, 45 , Article 100986
.
Zhao, Z., Chen, W., Wu, X., Chen, P. C., & Liu, J. (2017). LSTM network: A deep learning
approach for short-term trac forecast. IET Intelligent Transport Systems, 11 (2), 68–75
.
Zhong, H., Wang, J., Jia, H., Mu, Y., & Lv, S. (2019). Vector eld-based support vec-
tor regression for building energy consumption prediction. Applied Energy, 242 , 403–
414
.
Zhou, Y., & Kankanhalli, A. (2021). AI regulation for smart cities: Challenges and princi-
ples. In Smart cities and smart governance (pp. 101–118). Cham: Springer
.
21
