**4. Healthcare**

Janbi et al. [10] propose, implement, and evaluate Imtidad, a reference architecture that provides Distributed Artificial Intelligence (AI) as a Service (DAIaaS) over the cloud, fog, and edge. For this purpose, the authors develop a service catalog case study containing 22 AI skin disease diagnosis services. The services are divided into four service classes based on software platforms and are run on a variety of hardware platforms as well as four network types. Two standard Deep Neural Networks (DNNs) and two Tiny AI deep models were trained and tested using real-life dermatoscopic images to create the AI models for diagnosis. Several benchmarks were used to evaluate the services, including model service value, response time, energy consumption, and network transfer time. The services are intended to enable a variety of use cases, such as home patient diagnosis or sending diagnosis requests to traveling medical professionals via a fog device or cloud. This work contributed to the first, second, and fourth dimensions of smartness, sensing, data processing, and computing infrastructure.

### **5. Urban Infrastructure**

Yigitcanlar et al. [11] investigate whether artificially intelligent cities can protect humanity from natural disasters, pandemics, and other disasters. By charting the evolution of AI and the potential impacts of systematic AI adoption on cities and societies, the authors generate insights and identify prospective research questions. This viewpoint provides theoretical contributions to all four dimensions of smartness and puts forward directions for future research, as well as listing large number of critical research questions that need to be answered. In another work, Yigitcanlar et al. [12] derive insights from sensor city best practices by scrutinizing some well-known projects implemented by Huawei, Cisco, Google, Ericsson, Microsoft, and Alibaba. The authors highlight that platform urbanism is becoming a critical tool to support smart urban governance in an era of digitalization of urban services and processes. On the basis of the lessons learned from the best practices of leading innovation and technology companies, the study advocates the need for further research on the conceptualization and practice of the sensor city notion.

Feri et al. [13] propose a three-dimensional microstructure reconstruction framework based on a 3D improved Wasserstein Generative Adversarial Network (3D-IWGAN) with an enhanced gradient penalty. It is a computational system based on images for analyzing clogging in the permeable pavement. The physical property values extracted from their model are comparable to those obtained from real pavement samples. The authors are motivated by the fact that there is an increasing demand for research into how to improve the functionality of permeable pavement. Their proposed system starts with a two-dimensional image as input and extracts latent features from it. It generates a 3D microstructure image using their model's generative adversarial network. This work contributes to the data processing dimension of smartness.

Akram et al. [14] propose an architecture for designing and developing a customized sensor node and gateway based on LoRa (Long-Range radio) technology for solid waste management, specifically, to achieve the filling level of the bins while using the least amount of energy. The authors also include distinct evaluation metrics for the sensor node's longrange data rate, time on-air (ToA), LoRa sensitivity, link budget, and battery life. LoRa is a popular communication protocol that provides long-range transmission and low data rates while consuming little power. Only a small amount of data needs to be sent to the remote server in the context of solid waste management, hence the use of LoRa. This work contributes to the sensing dimension of smartness.

Jo et al. [15] examined the changes in particulate matter concentrations due to land use over time, as well as the spatial characteristics of the distribution of particulate matter concentrations in Daejeon, Korea, as measured by Private Air Quality Monitoring Smart Sensors (PAQMSSs). According to the primary land use around the 650 m sensor radius, land uses were classified into residential, commercial, industrial, and green groups. The results show that particulate matter concentrations in Daejeon decreased in the order of industrial, housing, commercial, and green groups overall; however, the concentrations of the commercial group were higher than those of the residential group between 21:00 and 23:00, reflecting the commercial group's vital night-time lifestyle in Korea. The study contributes to the data processing dimension of smartness. Janbi et al. [16] propose a framework for Distributed AI as a Service (DAIaaS) provisioning for the Internet of Everything and 6G environments with the aim to help standardize the mass production of technologies for smarter environments. To investigate the design choices and performance bottlenecks of DAIaaS, multiple DAIaaS provisioning configurations for distributed training and inference are proposed, including three case studies (a smart airport, a smart district, and distributed AI provisioning) with eight scenarios, nine applications and AI delivery models, and 50 distinct sensor and software modules. This work contributes to all four smartness dimensions.

### **6. Concluding Remarks**

The smartness that underpins smart cities and societies is defined by our ability to engage with our environments, analyze them, and make decisions, all in a timely manner. The IoT has been the focus of this Special Issue, and its concern has been to bring "smartness" to the IoT and other system layers using emerging technologies. The articles included in this issue cover a wide range of applications, including image analysis, permeable pavements, solid waste management, air quality monitoring, thermal anomalies and smart helmets in industrial environments, smart airports, smart districts, and smart travel choices.

The field of smartness is exciting, and while a lot has been achieved, the future possibilities with technologies such as Deep Learning, Edge Computing, Virtual Reality, and more are endless. There are many works that are complementary to the research presented in this Special Issue, such as deep journalism [17], smartization [18], smart families and homes [19], data-driven smart governance [20,21], responsible innovation [22], and green AI [23]. This is an exciting time for disruptive technologies, and this Special Issue is expected to clarify the concept of smartness, helping more researchers to contribute to this area and lead to the development of truly smart environments.

**Funding:** The authors would like to thank for and acknowledge the technical and financial support from the Deanship of Scientific Research (DSR) at the King Abdulaziz University (KAU), Jeddah, Saudi Arabia, under Grant No. RG-11-611-38.

**Acknowledgments:** The work carried out in this paper is supported by the HPC Center at the King Abdulaziz University.

**Conflicts of Interest:** The authors declare no conflict of interest.
