The different sectors that this study deals with are as following:
Urban traffic/transport management
Civil safety and security
AI will be involved in every field, every sector in the coming years of global digitalization. As referenced in Forbes Technology Council Post , AI will soon revolutionize industries including Cybersecurity, DevOps, Manufacturing, Healthcare, Construction, Senior Care, Retail, Business Intelligence, City Planning, Mental Health Diagnosis and Treatment, Education, Fashion and Supply Chain Management. The application of AI is not limited to chatbots, Intelligence systems, digital-healthcare, etc. but will surely expand its capabilities and applications in the above-mentioned fields (Fig. 2).
Healthcare is a major field of AI application, especially since the COVID-19 scenario broke out. Even the developed nations have struggled to cope with the pandemic and to provide ample healthcare professionals, medicines, equipment, and machinery. There is a deluge on research activities due to COVID-19 and AI with impactful-ness is being proposed for establishing new and smart solutions for hospitals, clinics, and connected healthcare belts. Medical research is working on analyzing and identifying patterns in various complex large datasets faster and more efficiently than previously possible . For many researchers making predictions is the most demanding field of medical science, image classification for the prediction of root cause where the disease has started can also be seen .
Healthcare Management is also working under developing IOT Based AI Solutions, for which many authors have proposed their architecture and solutions. As Jochen  discussed the platform approach to the Smart Living for medical as well as a smart home living. With the advancement of the same, societies may need to face the challenges and the privacy issues which is being discussed by Ibrahim  with taking care of the challenges that IOT Sleep trackers and privacy & security concerns lied by sharing this data with companies holding the responsibility of serving resources. With that approach, Ibrahim discussed with the best approach from the privacy angle as to use the sensing approach (SA), Awareness and Security (AS) for End-Users, Cloud Environment.
With the most dealt sector from the AI perspective is this sector, where researchers as well as many organizations are building their custom solutions to serve the best solution for education purposes. So, research is also going on to build better AI Solutions to automate the process of allocation of subjects, lectures, and other related things in hand with the advancement of classrooms. Sutjarittham  contributes by undertaking the evaluation of several approaches that involve IoT as the core model, for measuring classroom occupancy and manipulating this based on terms including cost, accuracy, privacy, and ease of use.
With the need for smart tutors, researchers too proposed some concrete solutions from where AI Tutors can help them to allot resources so that they can grasp the concepts in-hand efficiently and effectively. Francisco  discussed the approach for their designed algorithm that helps them to articulate profiles of students based on their historical records and then recognizing the candidate that encounters to personalize their smart tutoring experience taking into consideration for the similar candidates.
When technological innovation is used for the advancement of the society for smart innovation, every sector comes into context, similarly, the Environment plays an important context when one is considering AI to be used in Smart Cities Strategy plan. This sector deals with the management of controlling and eliminating pollution including air, noise, water, etc. from the environment. For this context, the ways by which AI can be used and get linked with Pollution management is the main concern of understanding. Many environmentalists and researchers are working on many new innovative solutions that can understand the patterns and learn on their own to predict and eliminate the density of pollution around. For many purposes, electro-static precipitators are used commonly for filtering dust particles to control air pollution, but they are also challenged beyond their specifications, for which Bazzi  discussed the approach of using ANN and random forest approach to improve the quality of these precipitators.
The most needed resource in today’s digital era is becoming for sure is Networking and communication among surroundings. With that, dealing with a huge amount of data surely comes with the analysis of the same, where only AI plays an important role. Communication among humans, machines, vehicles or related smart node, comes with new challenges to use it efficiently, where AI modeled itself to introduce with new approaches on which network engineers and AI Engineers are working on these days. CISCO is the parent company for the networking field, introduces AI Solutions for the B2B approach for networking in business maintenance and communication with customers .
Power/Energy Management is the base demand for the digitalization for cities and this can be fulfilled mainly in contrast under the use of Artificial Intelligence. Power/Energy Management deals with the management and maintenance of Energy Equipment and monitoring these with full efficiency. But, due to human prone errors in the management, many encounters suffer from the incidents. For solving this problem, an artificially intelligent system is mandatorily required with a full-fledged solution for the demand as well as supply.
Covering the recent research works, many worked and implemented different AI Models to deploy smart solutions for the household energy storage systems, Microgrid Applications, Energy Buildings, and many more included. With the coming trend in the use of AI in Power/Energy Management, Energy Internet is the coined term used by most of the researchers across the globe, this term is the key research hotspot for many of the researchers with their meaning to Energy Internet. Pointing to this term, Cheng  explains briefly the term and its related equipment systems that can be integrated to serve the purpose of delivering smart and connected solutions of Energy Management. On the other hand, deploying AI in smart energy systems can also be integrated as Sun  discussed in their paperwork. Their paper proposed a smart classification method that combines the intelligence of an Independent Energy Unit with Interconnected Energy Unit which resulted in the development of an AI Technique for energy systems.
Urban Traffic/Transport Management
Traffic as one of the daily routines dealt with problems across the globe is reaching its heights and deploying of self-driven vehicles won’t help the living anymore if that’s fully automated and being learned to compare with the rapidness of human-operated vehicles. Traffic Management is one of the top concerns which can be controlled, even eliminated with the proper usage of AI. Many researchers are working on the advancement of the same concern including , which describes the application of AI in Transportation systems as planning, Designing and controlling Transportation Network Structures, Incident Detection, Predictive models, and throw light on the need of Intelligent Urban Mobility. Moreover, AI in Traffic needs to get integrated with other technologies that are included with Data, various works implemented the use of Cloud and AI to give a concrete application of AI to build smart solutions for managing traffic in their surroundings. As Telegra Inc., USA deployed their TCC 2.0 (Traffic Control Centres)  across Russia Regions . Techniques including Dynamic Lane Assignment (DLA) , Vehicle-to-Everything , Autonomous Vehicles by using Network Calculus (NC) , Deep Neural Network  and many more. The challenge for every country is to work on the development of a universal approach that Traffic and Transport Management can be solved.
Civil Safety and Security
Crime being the darker end of any country is to one of the sectors where AI can be applied and proper analytical with forecasting of crimes can be seen. Similarly, Security and Safety Management becomes an urgent need to be getting used by various Crime Investigation Departments/Smart Homes. Various applications of AI are being seen under this Sector including public safety video and image analysis, DNA Analysis, gunshot detection, and crime forecasting , mainly researched by the National Institute of Justice. Inclusion of Safety AI Cameras can be deployed to ensure the safety of one’s home members as worked together by.
Catastrophe Management (Surveillance and Rescue Missions)
This sector includes the management of climate change, disaster management, emergency conditions, and included factors. As the need for the AI in Emergency automation is required, similarly introduction of smart solutions is required in parallel. For the same, many researchers are working rigorously in this sector, Mohammad  discussed the approach of using Markov Decision Algorithm to obtain a fast and short pathway to the emergency spot by making AI-based Units namely, Emergency Management Unit (EMU). Similarly, communication becomes very necessary when disaster scenarios occurred and manual communication through humans becomes very less due to a stressed situation, for the same then required for an AI-based communicator is needed, Okamoto  discussed this approach through a solution for connecting networks from image classification surveillance system which depends upon the content inputted through camera and Information-centric network then communicates to the disaster management team to help with the situation. Then the need of making this more meaningful, researchers worked onto using external solutions provided by cloud vendors like IBM Cloud Service as reflected by Talley  who illustrates the fact of making awareness among connected societies so that make surrounding aware regarding the disaster occurred and sharing same with disaster rescue team using cognitive cloud approach.