Figure - uploaded by Kasun Herath
Content may be subject to copyright.
Source publication
Recently, the population density in cities has increased at a higher pace. According to the United Nations Population 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-effective, the c...
Citations
... These gaps highlight the need for frameworks that facilitate the seamless integration of AI and ML into health risk management systems. The Integration of AI in smart healthcare presented by Herath & Mittal, 2022, is shown in figure 2. Ethical and technical considerations also play a crucial role in the implementation of AI and ML in occupational health. Data privacy is a major concern, as wearable devices and sensors collect sensitive personal information that must be protected from misuse. ...
... Integration of AI in smart healthcare(Herath & Mittal, 2022). ...
Occupational diseases remain a significant challenge in high-risk industries, where hazardous working conditions expose employees to health risks that often go undetected until symptoms become severe. To address this, leveraging artificial intelligence (AI) and machine learning (ML) offers transformative potential for proactive health risk management by enabling predictive modeling, real-time monitoring, and data-driven decision-making. This study presents a conceptual framework for integrating AI and ML technologies to predict and mitigate occupational diseases in high-risk industries such as mining, construction, and manufacturing. The proposed framework encompasses three key components: data acquisition, predictive modeling, and intervention strategies. Data acquisition involves collecting real-time health and environmental data through wearable sensors, IoT-enabled devices, and workplace monitoring systems. Predictive modeling employs advanced ML algorithms, such as decision trees, neural networks, and support vector machines, to identify patterns and risk factors associated with occupational diseases. Intervention strategies leverage predictive insights to develop targeted prevention measures, such as redesigning work environments, optimizing workflows, and implementing personalized health interventions. A case study approach evaluates the framework's applicability, focusing on high-risk industries in Nigeria. Initial results demonstrate the feasibility of using AI-driven systems to identify early indicators of diseases such as respiratory disorders, musculoskeletal conditions, and noise-induced hearing loss. The findings also highlight the framework's potential to enhance workplace safety, reduce healthcare costs, and improve employee well-being by transitioning from reactive to proactive health management. The framework underscores the importance of cross-disciplinary collaboration among engineers, healthcare professionals, and policymakers to ensure effective implementation. Ethical considerations, such as data privacy and fairness, are also addressed to ensure equitable access and compliance with international health and safety standards. This conceptual framework lays the foundation for future research and policy development aimed at integrating AI and ML technologies into occupational health systems, particularly in resource-constrained settings, to foster safer and healthier work environments.
... In urban planning and infrastructure, there are also notable examples; by using data and knowledge acquired by AI, cities can shift to another level and have the potential to revolutionize city development, which will enable over 30% of smart city applications by 2025, including urban transportation solutions, significantly enhancing urban sustainability, social welfare, and vitality (Herath and Mittal, 2022). Furthermore, AI-enabled robots are deployed in the hospitality sector to provide personalized services and facilitate seamless guest experiences (Szpilko et al., 2023). ...
... Worldwide, non-renewable resources provide over eighty percent of the energy (2) . Additionally, cities are the residence of approximately 55% of the global population, a figure that is expected to increase to 68% by 2050 and continue to rise until it reaches 10.9 billion by 2100 (3) . There will be major ecological concerns brought on by the expected 2.5 billion people living in metropolitan areas by the year 2050 (4) . ...
Objectives: As conventional energy dwindles and environmental issues rise, solar energy gains traction for its sustainability. Rooftop potential, improved panels, affordability, and easy installation drive its urban adoption, necessitating accurate PV panel feasibility assessments. Methods: This study aimed at measuring rooftop solar PV potential and energy consumption and production using a GIS based tool within ArcGIS City Engine. Analysing the built footprint data enables the estimation of PV potential. Findings: Studying the Gandhinagar Smart City revealed the existence of 39,738 unique rooftops covering a total area of 56,976,756.77 square meters. Additionally, it was revealed that their annual PV potential was found to be 1,017.74 MWh, while the replacement potential dwarfs to 1,216.75 MWh every year. However, in terms of environmental benefits, it is estimated that a shift towards solar based energy systems in urban environments can mitigate AGCC effects by reducing annual CO2 emissions up to 10.40 million tons. Novelty: This study introduces a novel GIS-based approach within ArcGIS City Engine to accurately assess rooftop solar PV potential at an urban scale. The analysis of Gandhinagar Smart City provides a comprehensive evaluation of rooftop energy potential, distinguishing between generation capacity and replacement potential. Keywords: 3D City Modelling, CO2 Emission Reduction, Decentralized Energy Systems, GIS-Based Analysis, Rooftop Solar Potential
... For the degree of Architecture, IoT provides improvements in efficiency, sustainability, and user experience. With this technology, smart buildings can be managed, where connected sensors control systems, optimize energy efficiency [23], [24], enable structural monitoring, design based on data, manage waste by providing information to containers, and examine physical variables such as temperature, humidity, air quality, light, and sound [25], [26]. In architecture, the tools offered by IoT allow for the design of more user-centered efficient environments. ...
The Internet of Things (IoT) applications are pervasive across various sectors; however, there remains some resistance to its adoption. Education 4.0 promotes the full integration of new technologies, both as tools for learning and instruments for professional development. This work studies the influence of higher education on the willingness towards IoT adoption after hands-on learning experiences. The primary objective is to determine whether a correlation exists between IoT adoption and the education of university students from three distinct professional degrees. The methodology employed involves a practical class where students engage in developing applications for manual data collection. These applications are designed to send data to the Internet, which is then visualized through a web interface. Tailored to each respective degree, three similar applications are developed. For this research, M5 Stack Core2 kits are utilized, along with UIFLOW programming language and the ThingSpeak platform, operating under the MQTT protocol. Following the training, students complete a Technology Acceptance Model (TAM) survey for IoT. The analysis of the influence of higher education on IoT acceptance employs ANOVA to identify differences between group means. The results reveal statistically significant differences in IoT acceptance between students in Industrial and Architecture degrees.
... Examples of big data storage include MongoDB, Cassandra, and other types of NoSQL databases along with distributed file systems like Hadoop HDFS. These systems can store large amounts of data coming from sensors and fetch the data for efficient processing by an AI model [39][40][41]. ...
Background: The integration of artificial intelligence (AI) with the internet of things (IoTs) represents a significant advancement in pharmaceutical manufacturing and effectively bridges the gap between digital and physical worlds. With AI algorithms integrated into IoTs sensors, there is an improvement in the production process and quality control for better overall efficiency. This integration facilitates enabling machine learning and deep learning for real-time analysis, predictive maintenance, and automation—continuously monitoring key manufacturing parameters. Objective: This paper reviews the current applications and potential impacts of integrating AI and the IoTs in concert with key enabling technologies like cloud computing and data analytics, within the pharmaceutical sector. Results: Applications discussed herein focus on industrial predictive analytics and quality, underpinned by case studies showing improvements in product quality and reductions in downtime. Yet, many challenges remain, including data integration and the ethical implications of AI-driven decisions, and most of all, regulatory compliance. This review also discusses recent trends, such as AI in drug discovery and blockchain for data traceability, with the intent to outline the future of autonomous pharmaceutical manufacturing. Conclusions: In the end, this review points to basic frameworks and applications that illustrate ways to overcome existing barriers to production with increased efficiency, personalization, and sustainability.
... Frameworks have also been proposed for specific technical areas, such as sustainable hydropower plant operations [36], asthma attack prediction [37], machine learning on the edge [38], security, privacy and risks [39], and planning and management [40]. Focusing more closely on AI framework, several reviews present broad evidence supporting the effectiveness and practical value of AI frameworks [31,41,42]. In terms of applications, the most impact from AI is reported in healthcare, mobility, privacy, and energy, while the notion of an 'urban artificial intelligence' has also been proposed through developments in autonomous cars, robots, and the built environment. ...
Highlights
What are the main findings? Smart Cities as Hyper-Connected Digital Environments generate large and diverse data streams and repositories that do not consistently translate into insights and decisions.
A Responsible AI Hyper-Automation framework with Generative AI agents is developed and evaluated to address these complex challenges.
What are the implications of the main findings? The developed AI framework is effective when grounded on five core technical capabilities with an independent cognitive engine for hyper-automated agentic AI that feeds into human-in-the-loop processes.
The framework provides a prototypical setting for university cities of the future to provide direction, guidance, and standards for sustainable and safe smart cities of the future.
Abstract
Smart cities are Hyper-Connected Digital Environments (HCDEs) that transcend the boundaries of natural, human-made, social, virtual, and artificial environments. Human activities are no longer confined to a single environment as our presence and interactions are represented and interconnected across HCDEs. The data streams and repositories of HCDEs provide opportunities for the responsible application of Artificial Intelligence (AI) that generates unique insights into the constituent environments and the interplay across constituents. The translation of data into insights poses several complex challenges originating in data generation and then propagating through the computational layers to decision outcomes. To address these challenges, this article presents the design and development of a Hyper-Automated AI framework with Generative AI agents for sustainable smart cities. The framework is empirically evaluated in the living lab setting of a ‘University City of the Future’. The developed AI framework is grounded on the core capabilities of acquisition, preparation, orchestration, dissemination, and retrospection, with an independent cognitive engine for hyper-automation of these AI capabilities using Generative AI. Hyper-automation output feeds into a human-in-the-loop process prior to decision-making outcomes. More broadly, this framework aims to provide a validated pathway for university cities of the future to take up the role of prototypes that deliver evidence-based guidelines for the development and management of sustainable smart cities.
... Facial recognition 16 Classification performance of ML models 17 ...
Lighting Systems (LSs) play a fundamental role in almost every aspect of human activities. Since the advent of lights, both academia and industry have been engaged in raising the quality of the service offered by these systems. The advent of Light Emitting Diode (LED) lighting represented a giant step forward for such systems in terms of light quality and energy saving. To further raise the quality of the services offered by LSs, increase the range of services they offer, while at the same time consolidating their reliability and security, we see the need to explore the contribution that can be derived from the use of the Artificial Intelligence of Things (AIoT) emerging technology. This paper systematically reviews and compares the state-of-the-art with regard to the impact of the AIoT in the smart LS domain. The study reveals that the field is relatively new, in fact the first works date back to 2019. In addition to that, the review delves into recent research works focusing on the usage of Machine Learning (ML) algorithms in an edge Cloud-based computing architecture. Our findings reveal that this topic is almost unexplored. Finally, the survey sheds light on future research opportunities that can overcome the current gaps, with the final aim of guiding scholars and practitioners in advancing the field of smart LSs. The study is reported in full detail, so it can be replicated.
... های روش :2 سطح تحلیلی 18 )%29( Nikitas et al. 2020, Syed et al. 2021, Serban & Lytras 2020, Golubchikov & Thornbush 2020, Herath. & Mittal 2022, Omitaomu & Niu 2021, Verma 2022, Kuguoglu et al. 2021, Osamy et al. 2022, Zamponi & Barbierato 2022. Alahi et al. 2023, Bokhari & Myeong 2023, Lauri et al. 2023 خاص های نظریه :3 سطح 28 )%6 ( Allam & Dhunny 2019, Ullah et al. 2020, Cugurullo 2020, Chui et al. 2018, Yigitcanlar & Cugurullo 2020, Zhang et al. 2021, Guo et al. 2018 ...
Artificial intelligence offers highly suitable solutions for numerous challenges in urban transformation and development, such as ensuring an adequate water supply, energy management, waste management, and reducing traffic congestion, noise, and pollution. Given the social and technical nature of smart cities and the applications of artificial intelligence in this field, university research has seen a significant increase in recent years. Furthermore, an analysis of the popularity of keywords "smart city" and "artificial intelligence" on Google Trends indicates that these key terms have been increasingly popular from 2014 to the present. Therefore, this article systematically examines the current status and future directions of research on the application of artificial intelligence in smart cities through a sequential review method. To this end, the databases "Scopus" and "Google Scholar" were searched, identifying a total of 3384 articles. Following a systematic review and final screening process, 61 articles were selected for analysis. The findings show an increasing trend in research publications in this area from 2018 onwards. Additionally, an examination of the thematic scope of selected research indicates that the application of artificial intelligence in smart cities is predominantly focused on urban management and sustainable development (30%), smart living and intelligent infrastructures (28%), and intelligent environment (21%). The results also reveal that 64% of studies have employed qualitative methods, 21% quantitative methods, and 15% a combination of methods. As the application of artificial intelligence in smart cities is still in the conceptualization stage, the noticeable preference for qualitative methods among researchers in this
... AI borderline is transforming the smart city foundation by permissive honest-occasion accountability, lowering abeyance, optimizing frequency range exercise, and reconstructing solitude and security. This example change authorizes smart places to run more capably, behave swiftly in changeful environments, and enhance the overall characteristics of city life (Herath & Mittal, 2022). ...
The healthcare and industry fields are being significantly impacted by the dynamic synergy of edge computing and artificial intelligence (AI) algorithms. The complex AI algorithms at the center of this revolution are designed for edge computing. This chapter's goal is to provide an understanding of these algorithms' capabilities in practical applications by analyzing them in terms of accuracy, speed, and resource consumption. By using edge AI, AI calculations can now take place locally on devices instead of being limited to centralized cloud servers. This makes use of the complementary properties of edge computing and data processing power. It entails the creation and implementation of cutting-edge computer programs and algorithms that work directly on adjacent computing endpoints, Internet of Things (IoT) devices, and smartphones. This is different from the traditional use of cloud servers that are centralized. Due to the complicatedness of most AI models and the challenge of calculating conclusion results on devices accompanying restricted possessions, AI duties are currently achieved to process requests in cloud dossier centers. However, these "end cloud" architectures cannot meet the needs of palpable-period AI services to a degree of instant science of logical analysis and astute production. Therefore, bringing AI requests to the edge opens the potential for AI request synopsizes, especially in conditions of attaining reduced abeyance traits (Wang et al., 2020). Edge AI enhances the effectiveness of data processing. The system may manage data more effectively, maximizing resource utilization and guaranteeing that crucial insights are obtained quickly, by dividing computing responsibilities across local devices. This paradigm shift toward edge AI is in line with the changing demands of industry and presents a viable path forward for breakthroughs across a range of fields, including healthcare where secure and quick data processing is critical. Figure 8.1 displays the direct deployment of artificial intelligence algorithms into edge devices. Edge AI is independent of the cloud so it can process data more quickly and efficiently from its source. Edge AI enhances user experience and uses local processing to boost data protection. Since there is no requirement for data transfer, security precautions are increased in contrast to centralized cloud systems. From a technical perspective, less bandwidth needed can result in cheaper rates for Internet services that require subscriptions. One unique characteristic of edge technology gadgets is their self-contained design. These gadgets function independently and do not require continuous assistance from data scientists or AI developers, in contrast to typical setups. Monitoring becomes more efficient when graphical data flows are provisioned automatically (Raj et al., 2022).
... Through the use of AI, intelligent systems [7][8][9], virtual agents and assistants [10][11][12], and multiagent systems [13][14][15] can be created. Recent literature review studies have explored its use in various domains, such as education [16][17][18], industry [19][20][21], healthcare [22][23][24], business [25][26][27], smart cities [28][29][30], etc. The outcomes of these studies highlight the potential of AI to transform and enrich various sectors, which, in turn, reveals the need to further explore its capabilities to be used in combination with other novel technologies to further amplify its impact. ...
The research into artificial intelligence (AI), the metaverse, and extended reality (XR) technologies, such as augmented reality (AR), virtual reality (VR), and mixed reality (MR), has been expanding over the recent years. This study aims to provide an overview regarding the combination of AI with XR technologies and the metaverse through the examination of 880 articles using different approaches. The field has experienced a 91.29% increase in its annual growth rate, and although it is still in its infancy, the outcomes of this study highlight the potential of these technologies to be effectively combined and applied in various domains transforming and enriching them. Through content analysis and topic modeling, the main topics and areas in which this combination is mostly being researched and applied are as follows: (1) “Education/Learning/Training”, (2) “Healthcare and Medicine”, (3) “Generative artificial intelligence/Large language models”, (4) “Virtual worlds/Virtual avatars/Virtual assistants”, (5) “Human-computer interaction”, (6) “Machine learning/Deep learning/Neural networks”, (7) “Communication networks”, (8) “Industry”, (9) “Manufacturing”, (10) “E-commerce”, (11) “Entertainment”, (12) “Smart cities”, and (13) “New technologies” (e.g., digital twins, blockchain, internet of things, etc.). The study explores the documents through various dimensions and concludes by presenting the existing limitations, identifying key challenges, and providing suggestions for future research.
