IoT, AI, and Digital Twins in Smart Cities: A Systematic Review for a Thematic Mapping and Research Agenda
Abstract
Highlights
- This paper presents an SLR comprising the analysis of 64 studies on Urban Digital Twins, IoT, and AI in the development of smart cities using the PRISMA framework.
- The review identifies three thematic groups and key research points, while revealing gaps such as limited empirical evidence and interoperability challenges.
- This paper highlights the need for integrated, data-driven strategies to improve urban management and policies through Urban Digital Twins, IoT, and AI.
- The review allows proposing a research agenda to guide future innovations, supporting the development of more sustainable, resilient, and smart cities.
Abstract
1. Introduction
- Identification of key technological and methodological trends: This SLR examines the current state of the art in IoT and AI technologies applied to DTw for smart cities. It highlights the predominant concepts, trends, and methodological approaches, providing a basis for understanding technological developments in this area.
- Classification of practical urban applications: The SLR classifies DTw applications in key urban sectors, such as traffic management, urban planning, and environmental monitoring. This structured analysis demonstrates its role in improving urban sustainability and operational efficiency.
- Guidance for future research directions: Finally, the study identifies knowledge gaps and under-explored areas, such as AI-based real-time predictions and interoperability of IoT platforms. These findings provide a roadmap for future innovations and research in DTw for smart cities.
Background
2. Research Methodology
2.1. PSALSAR Phases
- Protocol:The development of a research protocol is a crucial step in conducting an SLR [48]. The protocol is designed to improve transparency, reproducibility, and transferability throughout the review process. This approach aims to minimize the risk of bias in identifying, selecting, and analyzing relevant literature.At this stage, the scope, objectives, and limitations of the review are clearly defined. This includes formulating specific research questions that align with the study’s overall goals and provide a structured framework for systematically identifying and categorizing primary studies.
- Search:This phase enables the identification of relevant documents for review through a strategic search. Carefully defining search strings and selecting relevant databases, such as IEEE Xplore, ACM Digital Library, Web of Science, and Scopus, are essential steps to ensure comprehensive coverage of the topic. Moreover, this phase is fundamental to ensuring transparency, transferability, and reproducibility so that other researchers can replicate the process. The search phase not only establishes the foundation for the methodological quality of the review but also reinforces its scientific rigor and robustness as a systematic strategy.
- Appraisal:This stage is a fundamental component of the quality assessment of the search. This evaluative process must be systematic and transparent, avoiding reliance on subjective or arbitrary judgments, incorporating theoretical and ethical considerations [49,50,51]. The Critical Appraisal Skills Programme (CASP) tool is commonly employed as a structured evaluative framework to ensure the rigor and consistency of this phase. This tool helps verify critical appraisal checklists, such as: Are the results of the study valid? What are the results? Will the results help locally? Is the study methodologically sound?
- Synthesis:The principal goal is to systematically extract and categorize data from the studies that passed the appraisal stage. This stage serves as the critical bridge between raw evidence and meaningful insights. It requires organizing the findings of the selected studies into coherent groups or thematic categories, facilitating both qualitative and quantitative content analysis. This phase involves organizing the extracted data to support subsequent analytical steps. This involves combining various findings into a unified evidence base to recognize trends, patterns, and gaps in knowledge.
- Analysis:A bibliometric analysis was conducted on the synthesized data to address the research questions. The analysis combined quantitative and qualitative information, yielding key conclusions and projections for future research. Data were organized according to information extracted from the documents, enabling effective comparison and analysis of the results.
- Report:In this phase, clear documentation and presentation of both the methodological procedures and the results of the SLR are essential. This final step ensures that the review is transparent, reproducible, and accessible to both the academic community and the broader public. Specifically, the authors in [25] highlight that the report should detail the specific steps taken, including search strategy, appraisal criteria, synthesis and analysis processes, and articulate the findings in a structured format consistent with recognized standards such as PRISMA.The report of the methodological procedure for applying PSALSAR in this SLR is detailed in the next Section 2.2, and the results of the SLR are reported in Section 3.
2.2. PSALSAR Applied to SLR on DTw, IoT and AI in Smart City Development
- Protocol:The mapping questions (MQs) in Table 1 were proposed as an integral part of the review protocol. These MQs aim to support the initial categorization of the literature and ensure comprehensive coverage of the topic of the thematic dimensions being investigated [50]. In the following, a description of each of the questions presented in Table 1 is provided.MQ1 aimed to identify publication trends in the field, focusing on the temporal evolution and thematic concentration observed in the literature. MQ2 permitted recognizing the thematic areas and domains of application in which Urban Digital Twins (UDTw), IoT, and AI have been implemented, such as energy, mobility, governance, and sustainability. MQ3 focused on examining enabling technologies, including IoT infrastructures, AI techniques, and interoperability middleware, which form the basis for the development of UDTw. MQ4 was designed to analyze reported use cases and practical implementations of technologies. It highlighted how these technologies were applied in real-world contexts and enabled discussing the resulting outcomes, limitations, and lessons learned from these experiences. This set of MQs ensured the review not only provided an overview of the current state of the art but also offered a deeper understanding of technological enablers and their practical applications, thereby establishing a comprehensive analytical framework.Subsequently, to determine the scope of the research, the Population, Intervention, Comparator, Outcomes, and Context (PICOC) framework presented by [52] was used, with the research question (RQ) according to the details of each parameter defined below:
- Population: Studies and projects that integrate DTw in smart cities, hereafter referred to as Urban Digital Twins (UDTw).
- Intervention: Utilization of IoT and AI technologies to enhance UDTw performance.
- Comparator: Studies without advanced integration of these technologies, or with partial integration.
- Outcomes: Improved urban efficiency, sustainability, decision-making, prediction, and monitoring.
- Context: Implementations in urban contexts of smart cities, both real and simulated.
Based on the definition of the PICOC parameters, the following research question was formulated: - Search:The databases selected for this review are presented in Table 2. As observed, the search was conducted in four relevant databases: IEEE Xplore, ACM Digital Library, Web of Science (WoS), and Scopus. These sources provide broad coverage of the field, ensuring the quality and significance of the selected literature. The combination of these sources ensured both technical depth in engineering and computer science (IEEE and ACM), as well as interdisciplinary coverage and indexing of high-impact articles in social sciences, urban planning, and sustainability (WoS and Scopus). This set of databases enabled a balanced representation of studies that address digital twins from technological, methodological, and applied perspectives in urban contexts.We decided to prioritize articles from peer-reviewed scientific journals over those from conference proceedings or grey literature (such as technical reports, theses, or institutional documents). Although the field of DTw has a strong tradition of conferences, most influential contributions are usually published later in expanded and consolidated versions in indexed journals. This approach helps minimize the risk of overlooking important works. The decision to exclude grey literature was made to maintain methodological consistency and ensure that quality standards are upheld. This approach reduces biases arising from the heterogeneous nature of such materials. As a result, the final collection was based on studies with greater traceability and scientific validity.The data for this SLR were collected on 24 May 2025. Initially, the researchers conducted a pilot search in the WoS and Scopus databases using broad search terms. The search terms were then refined and standardized through a bibliometric analysis. Following an iterative process, the final search equation was established for each database. For the Scopus database, the search equation encompassed the fields title, abstract (ABS), and keywords (KEY), as detailed in Listing 1. In the case of WoS, the search strategy included the fields title (TI), abstract (AB), author keywords (AK), and keywords plus (KP), as shown in Listing 2. For the IEEE Xplore database, the query was applied to All Metadata (Listing 3), whereas in the ACM database the search covered the field All, as presented in Listing 4.
Listing 1. Boolean Query for Scopus. ( TITLE-ABS-KEY ( "urban digital twins" ) AND TITLE-ABS-KEY ( "smart cities" ) AND TITLE-ABS-KEY ( "internet of things" OR "iot" ) AND TITLE-ABS-KEY ( "artificial intelligence" OR "AI" ) ) Listing 2. Boolean Query for WoS. (TI=("digital twins") AND TI=("smart cities") AND TI=("internet of things" OR "iot") AND TI=("artificial intelligence" OR "AI")) OR (AB=("digital twins") AND AB=("smart cities") AND AB=("internet of things" OR "iot") AND AB=("artificial intelligence" OR "AI")) OR (AK=("digital twins") AND AK=("smart cities") AND AK=("internet of things" OR "iot") AND AK=("artificial intelligence" OR "AI")) OR (KP=("digital twins") AND KP=("smart cities") AND KP=("internet of things" OR "iot") AND KP=("artificial intelligence" OR "AI")) Listing 3. Boolean Query for IEEE Xplore. ("All Metadata":"digital twins") AND ("All Metadata":"smart cities") AND (("All Metadata":"internet of things") OR ("All Metadata":"iot" )) AND (("All Metadata":"artificial intelligence") OR ("All Metadata": "AI")) Listing 4. Boolean Query for ACM. [All: "digital twins"] AND [All: "smart cities"] AND ([All: "internet of things"] OR [All: "iot"]) AND ([All: "artificial intelligence"] OR [All: "ai"]) - Appraisal:The Critical Appraisal Skills Programme (CASP) tool was tailored for engineering research, specifically focusing on studies related to DTw, IoT, and AI in the development of smart cities. The checklist items included:
- Clear research question: Presence of an explicit and well-defined research question or objective related to the integration of DTw, IoT, and AI in SCD.
- Internal validity: Assessment of the methodological rigor of the study design, including measures taken to minimize potential sources of bias.
- Accuracy of methods: Evaluation of the validity and appropriateness of the sensors, data acquisition systems, and diagnostic algorithms employed.
- Data analysis: Quality and robustness of data pre-processing procedures, as well as the rationale behind the selection of analytical techniques.
- Adequate design: Coherence between the research design and the stated objectives, ensuring methodological alignment with the intended outcomes.
- Clear results: Transparency and clarity in the presentation of findings, including the use of quantitative performance metrics such as precision, recall, F-score, or RMSE, when using ML approaches.
- Justified conclusions: Logical consistency between the empirical results and the conclusions drawn, avoiding overgeneralization or unsupported claims.
- Comparability: Inclusion of comparative analysis across different classification models or approaches to demonstrate relative performance.
- External validity: Evaluation of the generalization and applicability of the findings to other real-world settings or operational environments.
- Practical implications: Assessment of the practical relevance and feasibility of implementing the proposed models or systems in actual smart city contexts.
Considering the items above, each study received a score on the basis of a quantitative scale, described in Table 3. In this sense, to ensure transparency and methodological rigor, the criteria were applied independently by two reviewers. A calibration phase was carried out on a random 10% of the studies, achieving a substantial level of inter-rater agreement (Cohen’s Kappa = 0.82), in accordance with classical recommendations for the interpretation of reliability [53,54]. Discrepancies were resolved through adjudication by a third reviewer, ensuring that decisions reflect a reasoned consensus and not simple score averages.The evaluation outcomes of the analyzed articles are presented in Table A1 in Appendix A, where a maximum score of 10 could be achieved. It is worth noting that this table is provided as a representative sample of the evaluation framework, illustrating how the assessment criteria were systematically applied to the reviewed studies. The selection process for relevant literature is illustrated in Figure 1, where:- A total of 899 documents were found (159 in Scopus, 26 in Web of Science, 83 in IEEE, and 631 in ACM).
- In the next step, we eliminated 25 duplicate documents and 803 items classified as gray literature, which included books, book chapters, conference papers, and review articles.
- After reviewing the titles and abstracts, we removed 10 documents that did not align with the thematic area of the SLR. Additionally, five more documents were excluded due to inaccessibility.
- Finally, after reading the articles, analyzing their information in detail, and performing the quality assessment, 64 documents met the eligibility criteria for the SLR process.
- Synthesis:A thematic analysis using an inductive coding approach was used to classify the selected studies into four main areas. This process involved organizing similar or related themes into higher-level categories that emerged directly from the reviewed documents:
- Applications of DTw in urban management (e.g., infrastructure, transportation, planning).
- Synergies between IoT and DTw for real-time urban monitoring.
- Implementation of AI for predictive analytics and service automation.
- Technical and ethical challenges, including interoperability, privacy, security, and the design of integrated frameworks and architectures for Smart Cities.
The synthesis process was further supported by the use of R software (bibliometrix package), which enabled bibliometric analysis and the generation of thematic maps. Table 4 presents the fields used for data extraction from scientific articles, ensuring that the essential elements of each publication are clearly and systematically organized.During this phase, we systematically extracted and organized the most relevant data from the selected papers listed in Table A1 to facilitate in-depth analysis in the following stages of the review. The extraction process aimed to identify key variables that directly relate to the objectives of the systematic review. These variables were carefully compiled into a structured spreadsheet for further examination. The annual evolution of publications reveals a growing interest in the application of UDTw, IoT, and AI in urban environments. The most frequent keywords and factor analysis allowed the identification of thematic clusters, such as urban governance, energy efficiency, and smart mobility, which guided the organization of the findings, as depicted in Figure 2. - Analysis:The analysis of the selected documents pointed out the temporal trends, the studies into coherent thematic clusters, main sectors where DTw are being implemented, common challenges, and emerging thematic fields. The following specific findings were identified:
- Temporal trends: The scientific output has shown steady growth since 2019, with a notable increase in 2022 and 2023, indicating the maturation of the field. In addition, the thematic trend analysis reveals a shift from conceptual approaches to concrete applications in energy, mobility, and urban governance.
- Keyword co-citation and co-occurrence networks: Factor analysis allowed grouping the studies into coherent thematic clusters. These include: (i) digital governance and sustainable cities, (ii) energy efficiency through DTw, (iii) AI integration for urban decision making, and (iv) security and privacy in smart environments.
- Classification of studies by impacted urban sector: Five key sectors where DTw is utilized include energy, transportation, environmental management, urban planning, and health. Each sector was analyzed based on the type of technological application and the reported benefits.
- Critical analysis of research gaps and emerging controversies: Common challenges persist in the reviewed studies, such as platform interoperability, standardization of UDTw models, data protection, and scalability of solutions. Controversies also emerge about the ethical use of AI and the accessibility of technologies in diverse urban contexts.
Emerging thematic fields in the use of UDTw and AI were also identified, such as their application in risk prediction and urban energy sustainability. - Report:The narrative discussion was structured around the following thematic clusters:
- The most outstanding contributions in each thematic axis.
- The practical implications for the design and management of smart cities.
- Recommendations for future research, highlighting the need for standardization, data governance, and social impact studies.
Additionally, to emphasize the novelty of this study, Table 5 summarizes and contrasts our work with previous SLRs already included in the reference corpus. While prior studies have explored DTw in urban contexts [18,23,30], none have provided an integrated framework combining UDTw, IoT, and AI with emerging approaches such as GenAI, interoperability middleware, and testbeds. Such a comparison reveals the novel contribution introduced by this paper.To strengthen the comparison in Table 5, it is necessary to highlight that the inclusion and exclusion criteria applied in this study differ from those used in previous reviews. For example, in [30] the authors focused on data integration in UDTw, but did not employ a comprehensive approach to IoT and AI. Similarly, the authors in [18] addressed urban sustainability and resilience, although without an explicit methodology combining PRISMA with a structured thematic analysis. Finally, El-Agamy et al. [23] conducted a large-scale bibliometric analysis; however, the selection criteria used prioritized AI-oriented publications, leaving aside aspects of governance and interoperability middleware.
2.3. Rationale for PSALSAR
3. Results
3.1. Bibliometric Trends
- Keyword co-occurrence tree: Figure 4 illustrates the keyword co-occurrence network for the studies in Table A1. Larger nodes mark frequent terms (e.g., “AI”, “IoT”, “DTw”); edges indicate co-occurrence, forming coherent thematic clusters that align with our categorization. The smaller nodes indicate that the terms appear together in the same documents, highlighting coherent thematic clusters. The proximity and grouping of these nodes illustrate the synergy between the technologies and their applications, creating conceptual groups that support the thematic categorization proposed in this study.
- Publication trends and thematic evolution: Figure 5 illustrates the cumulative frequency of the most relevant keywords over the years of publication, from 2019 to 2025. Specifically, it reflects the trend of constant growth in the scientific production of the field, with a notable acceleration starting in 2022. The slope of the lines for terms such as “AI” and “DTw” indicates the growing interest and maturity of these topics in the literature. Finally, it allows establishing a temporal perspective of the evolution of the field of technologies that support UDTw.
- Trend topics and temporal evolutionFigure 6 represents the cumulative frequency of the most relevant keywords from 2019 to 2025. It reflects a consistent upward trend in scientific production within the fields of AI and DTw, particularly from 2023 onwards. This trend indicates an increasing interest in these areas, as well as in IoT and Smart Cities.Figure 7 depicts the evolution and change in research focus over the two main study periods: 2019–2022 and 2023–2025. The nodes on the left represent the dominant keywords in the first period, while those on the right represent the prominence of topics in the second period. The width of the arrows connecting the nodes indicates the “flow” of research, demonstrating how interest has shifted from foundational concepts (such as “AI” and “DTw”) to more applied and emerging topics (such as “big data” and “smart city”) in the most recent period.Finally, about the interpretation of inflection points in the revised literature since 2018, the proliferation of sensors, the exponential growth of urban data, and the evolution of AI algorithms have enabled the development of new generations of UDTw, now capable of not only reflecting but also anticipating and optimising urban processes [57,59]. Recent publication trends show a steady increase, particularly starting in 2019, with noticeable peaks in 2022 and 2023. These milestones can be attributed to several key developments, such as the deployment of 5G infrastructure and advancements in edge-fog computing. These innovations have enabled: (i) real-time experimentation with UDTw [60,61], (ii) the inclusion of UDTw in Horizon Europe’s Smart Climate-Neutral Cities Mission, which provided substantial research funding [58], and (iii) the launch of industrial DTw platforms by major companies like Siemens, NVIDIA, and Microsoft. This has fostered greater collaboration between academia and industry, thereby increasing the visibility of DT solutions. Collectively, these technological, political, and industrial factors account for the significant growth observed in recent years.
3.2. Integration of UDTw, IoT, and AI into the Development of Smart Cities
3.2.1. Key Technological and Methodological Trends
- (a)
- Emerging concepts and technologies
- IoT: The IoT constitutes the technological cornerstone for the implementation of UDTw, as it facilitates uninterrupted data acquisition, real-time information processing, and a robust interconnection between physical systems and their digital counterparts [65,66,67]. The proliferation of thousands of urban sensors is essential to capture a wide range of environmental, energy, and mobility variables [68]. In this context, integrating IoT sensors with metaverse platforms for smart building management enables real-time, multi-user collaboration, overcoming spatial constraints and enhancing situational awareness across distributed environments.Fatorachian et al. [69] also propose an approach based on IoT, predictive analytics, and cyber feedback for adaptive urban logistics. Likewise, to manage the complexity and volume of data generated by these massive networks, the creation of large-scale multimodal reference datasets, such as TUM2TWIN, is crucial [70]. This dataset, which covers approximately 100,000 m2 and contains 767 GB of data, integrates georeferenced 3D models with ground, mobile, airborne, and satellite observations, laying the groundwork for overcoming limitations in UDTw creation and validation.
- Edge computing and Cloud computing: Edge–cloud architectures optimize heterogeneous data processing and storage, particularly for applications that requirelow latency [71,72]. Platforms such as KTWIN [73], built on Kubernetes and following serverless principles, offer a unified, vendor-agnostic solution for deploying and operating UDTw components seamlessly from edge to cloud. This approach significantly reduces operational overhead and associated costs, making it vital for handling the growing data volume and real-time processing demands [72]. Similarly, the authors in [74,75] describe UMBRELLA as a Platform as a Service (PaaS) model with edge nodes and remote experimentation capabilities using DTw and IoT for smart cities.
- 3D visualization and augmented reality (AR): 3D visualization and AR have made advances that allow for more intuitive and immersive interaction with the DTw [76]. Recent innovations include the use of techniques such as Gaussian Splatting for extracting 3D meshes of buildings, and the integration of large language models (LLMs) to generate detailed visual and semantic descriptions from multi-view images [77]. The above not only improves the accuracy and detail of 3D city models but also opens up new avenues for interaction. Despite the significant potential of AR in UDTw for public participation and urban planning visualization, recent literature indicates that mature and detailed applications in this field are still limited [78].
- AI: The AI is a key enabler for data analysis, decision-making, and predictive modeling at UDTw, as described in [79]. In this regard, models such as Generative Adversarial Networks (GANs) [80], Variational Autoencoders (VAEs) [81], and Generative Pre-trained Transformers (GPT) are revolutionizing the autonomous creation of urban data, hypothetical scenarios, designs, and 3D models of cities [19]. In this sense, the GANs overcome limitations in data quality and availability by automating the creation of urban models and facilitating the development of UDTw. This democratizes its use, allowing professionals without technical experience to apply their domain knowledge and fostering greater adoption [66]. Additionally, in [64], the authors propose a Generative AI-powered Digital Twinning (GenTwin) framework to create DTw models with GenAI algorithms.
- Blockchain: The blockchain technology provides value by improving traceability, security, and interoperability among the multiple data sources that feed UDTw. Its integration with federated learning (FL) in multi-tier (fog/edge) computing systems is a promising strategy for addressing security and privacy concerns in industrial IoT environments, ensuring the integrity of model payloads and the identity of participants in a distributed ecosystem [72,82].
- (b)
- Methodological approaches
- Hybrid models with CPS: In [83], the authors emphasize the need for simulation in resilient urban planning using CPS and DTw. Specifically, these models integrate sensors, actuators, and dynamic simulations to optimize critical urban processes such as transportation, energy, or sanitation, incorporating data into AI, offering a robust and flexible architecture [66]. Recent research has focused on leveraging LLM-powered evolutionary algorithms, including HDTwinGen [84], to streamline the automated specification and performance optimization of hybrid DTw.This approach enables better generalization in data-sparse environments, efficient learning in samples, and greater flexibility for modular evolvability, which is vital in dynamic urban systems. Furthermore, DTw can offload computationally expensive calculations and provide fault-tolerance mechanisms from CPSs, resulting in significantly reduced operating costs and downtime [58].
- Interoperable middleware: Interoperability remains a fundamental challenge, and interoperable middleware solutions are key to overcoming it. Standards such as MQTT, OPC-UA, FIWARE, and NGSI-LD facilitate integration between disparate platforms, enabling the digital urban ecosystem [59,85]. Finally, the authors in [86] propose a verifiable and secure framework for data in UDTw, fostering interoperability.
- Testbed platforms: These controlled environments allow testing and scaling of UDTw prototypes in controlled environments before their actual deployment [74]. Specifically, platforms used for autonomous vehicle networks (AVNs) are essential for designing, deploying, and testing AI algorithms under safe conditions before their real-world implementation [87]. The existence and continued development of dedicated testbeds are a clear indicator of the technological maturity of the UDTw field.These environments enable rapid iteration and fault identification in a safe environment, as well as the optimization of AI algorithms and communication systems under controlled conditions, something that would be unfeasible or too costly in real-life urban settings [88]. Beyond technology validation, testbeds also function as training environments for personnel, such as first responders, improving their emergency preparedness and response capabilities. This creates a direct and effective connection between technological development and its practical effects on urban safety and efficiency. These testbeds allow for the integration of simulated wireless communication and real-time data from IoT sensors, enabling signal strength analysis and training in scenarios involving catastrophic incidents.
3.2.2. Classification of Practical Urban Applications
- Traffic management and smart mobility: AI and DTw are increasingly used to manage complex urban traffic networks and mobility [79,94]. In this sense, UDTw allows the integration of real-time data on vehicle gauging, Global Positioning System (GPS), cameras, and smart traffic lights to optimize urban traffic flow [79]. Cities like Barcelona have demonstrated their successful use to reduce congestion through AI that dynamically adjusts traffic lights and routes [60,79]. These systems not only dynamically adjust traffic lights and routes to reduce congestion but also prioritize electric mobility and public transport, improving their punctuality and efficiency [95,96]. The development of traffic DTw, such as the prototype in New York City, integrates object detection and tracking, resource allocation, edge and cloud computing, and communication for online traffic simulation, operation, control, and management, leveraging big data and AI tools. These systems also prioritize electric mobility and public transportation, improving their punctuality.
- Urban planning and citizen participation: UDTw facilitates data-driven decision-making and enables predictive simulations for the development of smart cities [59,89,97]. In addition, tools such as Virtual Singapore have developed 3D digital models that allow urban projects to be simulated before implementation, assisting planners in evaluating potential impacts on aspects such as urban ventilation, traffic flow, and public services [58,98]. Other initiatives, such as the case of Dublin, illustrate how UDTw can contribute to the democratization of urban planning by actively involving citizens in the decision-making process [99]. Furthermore, it is recognized that the practical and social value of UDTw is often unrealized because their implementations fail to address socio-technical complexities, such as disciplinary fragmentation and conflicts between actors [100].To overcome this limitation, the concept of “Augmented Urban Planning” (AUP) has been proposed, which conceives UDTw as an integrated collection of urban data and models with operationalization and contextualization in an interdisciplinary manner [65] A prominent example of AUP is the development of Civic Digital Twins (CDT), such as the one in Bologna [101]. These systems are designed to enhance citizen engagement in urban transformation by modeling infrastructure, human behavior, and social and environmental dynamics [102]. CDT enables scenario visualization and policy understanding, promoting a “citizen-in-the-loop” approach. By offering intuitive interfaces, UDTw empowers citizens to contribute to planning and strengthen trust with decision-makers. This democratization fosters more equitable and sustainable solutions, promoting collaborative governance. Projects such as GeoAI-supported DTw have driven global initiatives, such as the network of zero-emission universities in Latin America [97].Within the framework of governance and citizen trust, several recent studies have highlighted that UDTw go beyond being a simple technical representation; they are tools that co-produce new forms of knowledge and urban governance. For example, in [103], the authors have proposed a governance framework called “Dynamic Authorization Data Sandbox” that seeks to give citizens more effective control over their personal data. This model has demonstrated a significant improvement in the perception of data sovereignty (67%) and an increase in citizens’ willingness to participate in DTw activities (42%), which directly addresses the issue of trust. Likewise, in [104], a comparative study of UDTw projects in Boston, Namur, and Munich has shown how local political priorities influence the design of these UDTw, affecting how information is represented and how interaction with the public is enabled. This approach highlights the importance of considering local politics and culture in the design of UDTw to foster more meaningful participation and public trust.
- Environmental monitoring and climate resilience: UDTw enables the monitoring of air quality, urban temperature, noise levels, and waste management, among other parameters. Cases such as Lisbon, which utilize DTw to simulate floods, demonstrate their effectiveness in designing climate-resilience strategies based on scientific evidence [61,91]. A recent UDTw combines computational fluid dynamics (CFD) simulations with real-time meteorological data to analyze the dispersion of pollution such as particulate matter and nitrogen dioxide, enabling the identification of residential exposure hot-spots and supporting adaptive urban planning. Low-cost sensor networks are emerging as a viable support for UDTw in real-time air quality management, enabling the densification of observations and the integration of diverse information for dynamic and proactive decision-making [105]. Finally, the project integrating technology and urban resilience in Sydney was developed to become a resilient city through technology and digital twins based on real-time governance [106].
- Energy management and smart grids: UDTw contributes to resource optimization in smart energy cities through AI-based physical and virtual layered architectures [107,108]. Practical applications include public service management, energy systems, and critical infrastructure monitoring [59,89]. Thus, from the building level to distribution networks, UDTw applied to energy allows for predicting demand, detecting faults, balancing electrical grids, and optimizing energy consumption in real-time, as demonstrated by experiences in Vienna and China [109]. Finally, in [75], the authors describe a complete framework for using digital twins for energy monitoring in smart cities through virtual-real interaction.
- Public services and emergencies: Practical applications of UDTw include public service management, energy systems, and critical infrastructure monitoring, as described in [92,96,110]. In the case of water, waste, and public safety, UDTw have been key to anticipating leaks, optimizing collection routes, and coordinating emergencies through simulations in georeferenced 3D environments. In the context of smart grids, UDTw offer promising solutions to improve the monitoring, control, and optimization of electrical systems, especially with the increasing integration of renewable energy sources and the need to adapt to increasing energy demand [92]. A multifaceted smart grid prototype integrates environmental analysis, radio frequency identification (RFID)-based security, and IoT-based load management and energy monitoring. This holistic approach improves grid sustainability, security, and efficiency, enabling more balanced and resilient energy distribution.In solid waste management (SWM), UDTw are key to anticipating waste generation, optimizing collection routes, and reducing operating costs and emissions through waste generation simulations and optimized routing algorithms. For emergency management in civil infrastructure (EMCI), UDTw are applied in all stages of the disaster lifecycle: mitigation, preparation, response, and recovery [96]. A data-driven dynamic digital twin testbed is being developed to improve first responder training and communication by incorporating simulated wireless communication in a realistic virtual environment [111]. This type of testbed is crucial for preparing emergency response teams for catastrophic incidents safely and efficiently. Furthermore, in [70], the authors describe how a UDTw and Big Data model was integrated to monitor urban environments during the pandemic.
- Cross-Sectoral Enabling Layers and Re-Usable Patterns: Although the application of DTw in smart cities covers heterogeneous areas, several enabling layers emerge as cross-cutting and reusable across all sectors. Table 8 illustrates how the layers of IoT sensors, AI analytics, and decision support are implemented repeatedly in different urban sectors, highlighting reusable technological patterns.As shown in Table 8, IoT sensing is critical across the board, from smart meters in energy to wearables in healthcare. AI/ML analytics consistently deliver value by enabling prediction and optimization, whether in traffic routing or demand forecasting. Finally, decision support layers such as dashboards and 3D models provide actionable insights to stakeholders across multiple sectors. These recurring enabling layers indicate that UDTw implementations are not isolated silos but share modular and reusable components, opening up opportunities for interoperability and cross-domain scalability in future smart city ecosystems. Figure 10 shows this interconnectedness, revealing the density of technological integration between urban sectors and enabling layers. The heat map shows that IoT infrastructure and AI analytics are the most integrated components, confirming their role as fundamental elements in UDTw implementations.
3.2.3. Guidance for Future Research Directions and Research Gaps
- Real-time predictions and latency reduction: Despite advances, more lightweight AI models adaptable to edge computing architectures are required to react with minimal latency in critical applications such as emergencies or instantaneous traffic [15,64,66]. The Snap4City framework [113], for example, illustrates an approach for integrating multiple real-time data and publicly distributing UDTw, enabling operational and generative analytics. However, latency management remains a key challenge, especially for maintaining security in control systems and IoT while meeting real-time requirements [114]. Finding the balance between response speed and robustness of security measures is an active area of research.Based on the above, empirical validations remain limited to controlled or small-scale environments. The authors of [64,66] describe the need for adaptive AI models at the network edge, but their studies do not yet demonstrate implementations at the urban scale. Similarly, Ref. [113] describes how Snap4City integrates heterogeneous data flows, without addressing critical latency scenarios, such as emergency control. Finally, the authors of [114] highlight the problem of minimizing latency and maintaining cybersecurity, due to the lack of practical methods to balance both requirements. These examples confirm that achieving ultra-low latency in real urban conditions remains an unresolved gap for UDTw.
- Semantic interoperability and open standards: The interoperability and architectural challenges in large-scale DTw and IoT platforms are important, as described in [115,116]. However, barriers to integration between heterogeneous platforms persist. Progress is needed in open standards, common ontologies, and multiprotocol middleware, even integrating blockchain to ensure transparency and traceability [15,66,72,82]. Notable research gaps include the limited interoperability of IoT platforms and challenges in real-time predictive analytics [59,69,89].In the case of ontologies, these are fundamental for knowledge representation, semantic interoperability, and automatic reasoning in DTw, allowing physical components, actions, processes, and digital assets to be modeled in a structured and machine-understandable manner [44,66]. In this regard, more research is needed on the integration of ontologies into DTw knowledge bases and their use as “state graphs” combined with sensor data to improve reasoning processes and reduce data processing time [117].Moreover, a notable gap persists in the area of semantic standardization and large-scale interoperability. This is evident in [69], which presents a solution based on a specific framework for application. Similarly, in [89], which focuses on privacy in environments for the metaverse, the authors demonstrate the fragmented nature and lack of a unified approach to data and platform integration.
- Scalability, performance, and hierarchical architectures: Developing UDTw that can scale from buildings to megacities requires hierarchical approaches and flexible, microservices-based architectures. This new hierarchical DTw paradigm is proposed for efficient orchestration of 6G networks, adapting to real-time network situations through an adaptive attribute selection mechanism and scalable network modeling [118,119]. This enables efficient assessment of network situations at higher layers to identify target areas, while more detailed DTw are developed at lower layers for specific solutions. The creation of large-scale benchmark datasets [70], is a critical step in addressing the scalability challenges of UDTw creation and validation, providing the data needed to train and test models at an unprecedented scale. Finally, research should focus on generic models adaptable to diverse contexts.Despite the described advances, research gaps in standardization and interoperability still exist, which hinder the large-scale adoption of UDTw. An example is the model presented in the study by Bauer et al. [59], which describes an approach focused on a specific framework for urban service management rather than an open and universal standard. Similarly, although the authors in [60] analyze the real-time integration of UDTw with the 5G network, their approach is based solely on connectivity, rather than considering data model interoperability. This demonstrates a lack of holistic approaches that combine connectivity technology with the capacity for data exchange between heterogeneous platforms.
- Security, privacy, and ethical considerations: The massive incorporation of personal data and the potential for critical infrastructure control make it urgent to develop cybersecurity, anonymity, and ethical framework solutions specific to UDTw [59,89,120] Similarly, cybersecurity challenges are multifaceted, encompassing data transmission issues (interception, modification, denial of service attacks), system authentication (requiring Multi-factor authentication and biometrics), integration with other digital structures (cloud, IoT, internal systems), and emerging AI-driven threats. Vulnerabilities vary depending on the type of UDTw and industrial implementation, with risks ranging from medium to very high, especially in network-level systems, where a minor vulnerability can have severe consequences [121].A notable gap in the field of UDTw is the lack of a unified and validated security framework at the urban scale. For example, in [120], the authors focus on specific security protocols. In [89], the authors address privacy in the context of the metaverse, demonstrating the fragmented nature of existing solutions.
- Impact assessment: A persistent limitation in the field of UDTw is the lack of clear metrics to assess the Return on Investment (ROI), operational efficiency, and social benefits derived from their implementation. Recent literature highlights a significant gap between the ambitions of UDTw and their realized contributions. Mostly, implementations are limited to small-scale laboratory testing and technical integration, with minimal contribution to operationalization in real-world planning and decision-making processes. This reveals a disconnection between technological development and the ability of UDTw to generate tangible and measurable value in urban settings.The lack of clear evidence on ROI and social benefits can hinder investment and large-scale adoption by governments and stakeholders. Furthermore, if UDTw do not address socio-technical complexities, such as stakeholder dynamics and opaque planning processes, their real impact on improving urban life will be limited, as described in [122]. This gap makes urgent the research of more robust impact assessment methodologies and conduct studies that quantify long-term benefits [90,123].In [122], the authors emphasize the need for a holistic approach that not only focuses on the technology but also addresses the ethical risks of virtual environments and their regulation. A notable gap is in the holistic assessment of the impact of UDTw, as most studies focus on technical aspects without measuring tangible benefits. For example, the works [59,60,70,122] present functional and connectivity models, but do not provide a clear methodology for quantifying return on investment or societal benefits, illustrating a limitation in justifying large-scale implementation.
- Scalability and replicability: The application of DTw in smart cities presents key challenges in scalability, specifically the ability to handle an increasing number of IoT devices and urban services. It also involves replicability, understood as the ability to migrate or adapt successful solutions between different urban contexts without requiring major restructuring.In this regard, in [124], the authors present a hierarchical resource management system for IoT-enabled smart cities. The layered architecture presented allows for efficient management of the number of nodes and sensors, optimization of data flow, and control of energy consumption.Furthermore, in [125], the authors develop a scalable and intuitive framework that integrates IoT and DTw into domestic energy management. The proposal demonstrates that it is possible to start from homes and expand to communities while maintaining both computational efficiency and usability. Furthermore, recent studies point to inherent limitations of urban-scale UDTw.In [120], the authors analyse how certain models lose accuracy when the urban system becomes complex, suggesting that not all approaches scale adequately. Similarly, [23] provides evidence that a small proportion of the work explicitly addresses replicability, as many reports focus on point demonstrations rather than transferable models.Finally, to move forward, future work needs to integrate scalability and replicability principles into its design. Proposals include modular architectures, the use of containerised services (microservices), federated learning for distributed models, and open communication standards. This can make it easier for UDTws not only to scale within a city but also to be replicable across different urban ecosystems with diverse infrastructure and governance.
3.3. Testable Research Tasks
- ROI-oriented research task:
- −
- Task 1: Minimal Key Performance Indicators (KPIs) taxonomy and benchmarking.Goal: Define a cross-domain KPI set (operational, socio-economic, environmental) and a benchmarking protocol for UDTw/AI projects.Obstacle: Gap between ambition and demonstrated contribution.
- −
- Task 2: Policy uptake tracker.Goal: Measure how often DTw insights are incorporated into official decisions and what outcomes follow.Obstacle: Lab-style pilots poorly integrated into real planning cycles.
- −
- Task 3: Cost-to-serve at scale.Goal: Quantify unit cost per simulated scenario and per data flow when scaling from building → district → city.Obstacle: Data management complexity and performance degradation at scale.
- Scalability-oriented research task:
- −
- Task 1: Replicability playbook (building → district → city)Goal: Quantify re-engineering needed to port solutions across heterogeneous contexts.Obstacle: Limited attention to replicability and validated multi-scale pathways.
- −
- Task 2: Federated vs. centralized benchmarkGoal: Compare accuracy, latency, and privacy risk of federated learning versus centralized training.Obstacle: Need to reconcile scale with privacy and governance constraints.
- −
- Task 3: Semantic interoperability via reusable ontologies.Goal: Reduce integration time between platforms using shared ontologies and state graphs.Obstacle: Platform fragmentation; need for open standards and reusable semantics.
- −
- Task 4: City-scale reference datasets.Goal: Create open, large-scale DTw benchmark datasets (synthetic and real) for scalability evaluation.Obstacle: Lack of reference data to evaluate performance at scale.
- Ethics-oriented research task:
- −
- Task 1: Algorithmic impact assessment (AIA) with citizens.Goal: Identify and mitigate bias/inequity risks before deployment.Obstacle: Sociotechnical complexity; need for participatory governance.
- −
- Task 2: Traceability and auditability records.Goal: Record data provenance/transformations and model-access events for external audits.Obstacle: Need for cross-platform transparency and accountable data governance.
- −
- Task 3: Zero-Trust pilot with latency–security analysis.Goal: Implement Zero-Trust/defense-in-depth and quantify latency/resilience trade-offs.Obstacle: Tension between real-time constraints and robust cybersecurity.
4. Conclusions
- The lack of a consensus in terminology and the immaturity of certain key components in the built environment underscore the need for standardization and unified frameworks.
- Semantic interoperability across heterogeneous platforms remains a critical obstacle, requiring further development of ontologies and open standards.
- Scalability to replicate entire cities demands hierarchical architectures and robust performance solutions. Furthermore, massive data integration and control of critical infrastructure pose significant security, privacy, and ethical challenges, requiring advanced solutions and transparent governance frameworks.
Author Contributions
Funding
Data Availability Statement
Conflicts of Interest
Abbreviations
Acronym | Definition |
3D | Three-Dimensional |
5G | Fifth-Generation Network |
AERPAW | Aerial Experimentation and Research Platform for Advanced Wireless |
AI | Artificial Intelligence |
AIA | Algorithmic Impact Assessment |
API | Application Programming Interface |
AR | Augmented Reality |
AUP | Augmented Urban Planning |
CASP | Critical Appraisal Skills Programme |
CDT | Civic Digital Twins |
CFD | Computational Fluid Dynamics |
CPS | Cyber-Physical System |
DTC | Digital Twin City |
DTw | Digital Twin |
Edge | Edge Computing |
SDC | Smart City Development |
EMCI | Emergency Management in Civil Infrastructure |
FIWARE | Future Internet Ware |
FL | Federated Learning |
GANs | Generative Adversarial Networks |
GenAI | Generative Artificial Intelligence |
GIS | Geographic Information System |
GPT | Generative Pre-trained Transformers |
GPS | Global Positioning System |
IoT | Internet of Things |
KPI | Key Performance Indicators |
KTWIN | Serverless platforms based on Kubernetes |
LLM | Large Language Model |
ML | Machine Learning |
MQ | Mapping Question |
NGSI-LD | Next Generation Service Interface—Linked Data |
QoS | Quality of Service |
ROI | Return on Investment |
SDG | Sustainable Development Goal |
SWM | Solid Waste Management |
TRL | Technology Readiness Level |
UDTw | Urban Digital Twin |
VAEs | Variational Autoencoders |
VR | Virtual Reality |
XR | Extended Reality |
Appendix A
No. | Title | Score | Relevance | Year |
---|---|---|---|---|
1 | Optimizing Smart City Services… [108] | 10.0 | High | 2025 |
2 | Digital Twin: Enabling Technologies… [62] | 6.5 | Limited | 2020 |
3 | Special Issue on Digital Twin… [126] | 5.0 | Limited | 2023 |
4 | The Microverse a Task-Oriented… [61] | 10.0 | High | 2024 |
5 | Synergistic Integration of Digital Twins… [58]. | 6.5 | Limited | 2025 |
6 | Enhancing Smart City Logistics… [69] | 10.0 | High | 2025 |
7 | Urban Digital Twins—a FIWARE-Based Model [59] | 9.5 | High | 2021 |
8 | Gentwin Generative Ai-Powered Digital Twinning… [64] | 10.0 | High | 2025 |
9 | Advancing Smart City Sustainability with… [91] | 9.5 | High | 2024 |
10 | Unleashing the Potential of 5G for Smart Cities… [60] | 9.5 | High | 2025 |
11 | Excavating the Role of Digital Twins in Upgrading… [121] | 9.0 | High | 2023 |
12 | Implementation of Microgrid Digital Twin System… [109] | 8.0 | Useful | 2023 |
13 | Dtwin-Tec an Ai-Based Tec District Digital Twin… [96] | 10.0 | High | 2024 |
14 | Digital Twin Challenge Road Damage Detection… [92] | 10.0 | High | 2024 |
15 | A Case Study Making Decisions for Sustainable… [97] | 8.5 | High | 2023 |
16 | Artificial Intelligence Inspired Task Offloading… [79] | 10.0 | High | 2025 |
17 | Generative Digital Twins a Novel Approach in the… [127] | 9.5 | High | 2024 |
18 | Integrating Technology and Urban Resilience… [106] | 8.0 | Useful | 2024 |
19 | Trends and Challenges in Aiot Iiot Iot Implementation… [115] | 6.5 | Limited | 2023 |
20 | Smart City Research a Bibliometric and Main Path… [57] | 6.5 | Limited | 2022 |
21 | Digital twin-enabled decision support service… [93] | 10.0 | High | 2021 |
22 | Digital Twinning as an Approach to Promoting… [102] | 10.0 | High | 2023 |
23 | Ai-Based Physical and Virtual Platform with 5-L… [107] | 10.0 | High | 2019 |
24 | Digital Twin Perspective of Fourth Industrial and Healthcare… [111] | 7.5 | Useful | 2022 |
25 | A Survey on 6G Networks Vision, Requirements… [119] | 7.5 | Useful | 2024 |
26 | Desirable World with CPS and IoT [83] | 6.5 | Limited | 2023 |
27 | Digital Twins in Built Environments… [100] | 10.0 | High | 2023 |
28 | Amond: Area-Controlled Mobile Ad-Hoc Networking… [116] | 6.5 | Limited | 2024 |
29 | Federated Learning Enabled Digital Twins for Smart Cities… [66] | 10.0 | High | 2024 |
30 | Digital Twin in the IoT Context: a Survey… [68] | 7.5 | Useful | 2020 |
31 | Logical and Innovative Construction of Digital… [63] | 8.5 | High | 2021 |
32 | Exploiting Digital Twins as Enablers for Synthetic… [105] | 8.5 | High | 2024 |
33 | Platformization and the Metaverse Opportunities… [76] | 4.5 | Limited | 2023 |
34 | Situation Awareness of Energy Iot Systems… [75] | 8.0 | Useful | 2023 |
35 | Umbrella: a One-Stop Shop Bridging the Gap… [74] | 6.5 | Limited | 2024 |
36 | When Crowdsensing Meets Smart Cities: a Review… [24] | 4.5 | Limited | 2024 |
37 | Network Digital Twin Toward Networking, Telecom… [87] | 6.5 | Limited | 2024 |
38 | Digital Twins for Ports Derived from Smart Cities… [112] | 8.5 | High | 2024 |
39 | Digital Entity Management Methodology for Digital Twins… [44] | 8.5 | High | 2023 |
40 | An Urban Digital Twin Framework for Reference… [65] | 10.0 | High | 2024 |
41 | Signed: Smart City Digital Twin Verifiable Data… [86] | 10.0 | High | 2023 |
42 | Digital Twin-Based Healthcare System (Dths)… [42] | 9.5 | High | 2023 |
43 | Computational Intelligence in Security of Digital Twins… [123] | 10.0 | High | 2022 |
44 | A digital twin smart city for citizen feedback… [99] | 10.0 | High | 2021 |
45 | Uetopsis: a Data-Driven Intelligence Approach… [71] | 8.0 | Useful | 2024 |
46 | Dna Computing-Based Multi-Source Data Storage… [114] | 4.0 | Limited | 2023 |
47 | Smart City Construction and Management by Digital Twin… [70] | 10.0 | High | 2022 |
48 | Covid-19 Secure Healthcare Iot Networks [110] | 5.0 | Limited | 2023 |
49 | Will the Metaverse Be Out of Control?… [122] | 6.0 | Limited | 2023 |
50 | Excavating the Role of Digital Twins in Upgrading… [121] | 6.3 | Limited | 2023 |
51 | Digital Twin of Intelligent Small Surface Defect… [128] | 10.0 | High | 2023 |
52 | Edge Computing for Cyber-Physical Systems: a Review… [90] | 10.0 | High | 2023 |
53 | When Internet of Things Meets Metaverse Convergence of… [89] | 8.5 | High | 2023 |
54 | Platformization and the Metaverse: Opportunities and Challenges for U… [76] | 7.5 | Useful | 2023 |
55 | A Blockchain-based Digital Twin for IoT Deployments in Logistic… [72] | 7.0 | Useful | 2024 |
56 | AI and Digital Twin for Consumer Electronics… [95] | 6.5 | Limited | 2024 |
57 | Automatically Learning Hybrid Digital Twins of Dynamical Systems… [84] | 7.5 | Useful | 2024 |
58 | Adaptive Approaches to Software Testing… [88] | 8.5 | High | 2025 |
59 | Hierarchical Resources Management System… [124] | 8.5 | High | 2025 |
60 | A scalable and user-friendly framework… [125] | 8.0 | High | 2024 |
61 | Insurmountable limitations of city-scale digital twins?… [120] | 7.5 | Useful | 2025 |
62 | A Comprehensive analysis of digital twins in smart cities… [23] | 8.0 | High | 2024 |
63 | Governance Framework for Citizen Digital Twins in Smart Cities… [103] | 8.0 | High | 2025 |
64 | Urban futures in the mirror of technology? The politics [104] | 8.5 | High | 2025 |
References
- Echendu, A.J.; Okafor, P.C.C. Smart city technology: A potential solution to Africa’s growing population and rapid urbanization? Dev. Stud. Res. 2021, 8, 82–93. [Google Scholar] [CrossRef]
- Kapucu, N.; Ge, Y.; Rott, E.; Isgandar, H. Urban resilience: Multidimensional perspectives, challenges and prospects for future research. Urban Gov. 2024, 4, 162–179. [Google Scholar] [CrossRef]
- Al-Raeei, M. The smart future for sustainable development: Artificial intelligence solutions for sustainable urbanization. Sustain. Dev. 2024, 33, 508–517. [Google Scholar] [CrossRef]
- Singh, M.; Fuenmayor, E.; Hinchy, E.P.; Qiao, Y.; Murray, N.; Devine, D. Digital twin: Origin to future. Appl. Syst. Innov. 2021, 4, 36. [Google Scholar] [CrossRef]
- Mouha, R.A.R.A. Internet of things (IoT). J. Data Anal. Inf. Process. 2021, 9, 77. [Google Scholar] [CrossRef]
- Rejeb, A.; Rejeb, K.; Simske, S.; Treiblmaier, H.; Zailani, S. The big picture on the internet of things and the smart city: A review of what we know and what we need to know. Internet Things 2022, 19, 100565. [Google Scholar] [CrossRef]
- Quispe-Vilca, J.L.; Moreno-Cardenas, E.; Sacoto-Cabrera, E.J.; Moreno-Cardenas, Y. Integrating IoT and Image Processing for Crop Monitoring: A LoRa-Based Solution for Citrus Pest Detection. Electronics 2024, 13, 4863. [Google Scholar] [CrossRef]
- Singh, J.; Sajid, M.; Gupta, S.K.; Haidri, R.A. Artificial intelligence and blockchain technologies for smart city. In Intelligent Green Technologies for Sustainable Smart Cities; Wiley Online Library: Hoboken, NJ, USA, 2022; pp. 317–330. [Google Scholar] [CrossRef]
- Osman, A.M.S.; Elragal, A.A.; Ståhlbröst, A. Data-driven decisions in smart cities: A digital transformation case study. Appl. Sci. 2022, 12, 1732. [Google Scholar] [CrossRef]
- Sacoto-Cabrera, E.J.; Perez-Torres, A. Digital transformation: A review of enabling technologies, maturity models, and open research issues. In Proceedings of the 2023 IEEE Seventh Ecuador Technical Chapters Meeting (ECTM), Ambato, Ecuador, 10–13 October 2023; pp. 1–6. [Google Scholar]
- Deren, L.; Wenbo, Y.; Zhenfeng, S. Smart city based on digital twins. Comput. Urban Sci. 2021, 1, 1–11. [Google Scholar] [CrossRef]
- Carlsen, L.; Bruggemann, R. The 17 United Nations’ sustainable development goals: A status by 2020. Int. J. Sustain. Dev. World Ecol. 2022, 29, 219–229. [Google Scholar] [CrossRef]
- Mylonas, G.; Kalogeras, A.; Kalogeras, G.; Anagnostopoulos, C.; Alexakos, C.; Muñoz, L. Digital twins from smart manufacturing to smart cities: A survey. IEEE Access 2021, 9, 143222–143249. [Google Scholar] [CrossRef]
- Santos, R.; Piqueiro, H.; Dias, R.; Rocha, C.D. Transitioning trends into action: A simulation-based Digital Twin architecture for enhanced strategic and operational decision-making. Comput. Ind. Eng. 2024, 198, 110616. [Google Scholar] [CrossRef]
- Alnaser, A.A.; Maxi, M.; Elmousalami, H. AI-Powered Digital Twins and Internet of Things for Smart Cities and Sustainable Building Environment. Appl. Sci. 2024, 14, 12056. [Google Scholar] [CrossRef]
- Ati, M.; Basmaji, T. Framework for managing smart cities security and privacy applications. In Proceedings of the 2018 IEEE Symposium on Computer Applications & Industrial Electronics (ISCAIE), Penang Island, Malaysia, 28–29 April 2018; pp. 191–194. [Google Scholar] [CrossRef]
- Wang, K.; Zhao, Y.; Gangadhari, R.K.; Li, Z. Analyzing the Adoption Challenges of the Internet of Things (IoT) and Artificial Intelligence (AI) for Smart Cities in China. Sustainability 2021, 13, 983. [Google Scholar] [CrossRef]
- Mazzetto, L. A review of urban digital twins integration, challenges, and future directions in smart city development. Sustainability 2024, 16, 8337. [Google Scholar] [CrossRef]
- Ullah, Z.; Al-Turjman, F.; Mostarda, L.; Gagliardi, R. Applications of Artificial Intelligence and Machine learning in smart cities. Comput. Commun. 2020, 154, 313–323. [Google Scholar] [CrossRef]
- Syed Abdul Rahman, S.A.F.; Abdul Maulud, K.N.; Ujang, U.; Wan Mohd Jaafar, W.S.; Shaharuddin, S.; Ab Rahman, A.A. The digital landscape of smart cities and digital twins: A systematic literature review of digital terrain and 3D city models in enhancing decision-making. Sage Open 2024, 14, 21582440231220768. [Google Scholar] [CrossRef]
- Caprotti, F.; Cowley, R.; Datta, A.; Broto, V.C.; Gao, E.; Georgeson, L.; Herrick, C.; Odendaal, N.; Joss, S. The New Urban Agenda: Key opportunities and challenges for policy and practice. Urban Res. Pract. 2017, 10, 367–378. [Google Scholar] [CrossRef]
- Deng, T.; Zhang, K.; Shen, Z.J.M. A systematic review of a digital twin city: A new pattern of urban governance toward smart cities. J. Manag. Sci. Eng. 2021, 6, 125–134. [Google Scholar] [CrossRef]
- El-Agamy, R.F.; Sayed, H.A.; AL Akhatatneh, A.M.; Aljohani, M.; Elhosseini, M. Comprehensive analysis of digital twins in smart cities: A 4200-paper bibliometric study. Artif. Intell. Rev. 2024, 57, 154. [Google Scholar] [CrossRef]
- Wang, Z.; Cao, Y.; Jiang, K.; Zhou, H.; Kang, J.; Zhuang, Y.; Tian, D.; Leung, V.C. When crowdsensing meets smart cities: A comprehensive survey and new perspectives. IEEE Commun. Surv. Tutor. 2024, 27, 1101–1151. [Google Scholar] [CrossRef]
- Page, M.J.; McKenzie, J.E.; Bossuyt, P.M.; Boutron, I.; Hoffmann, T.C.; Mulrow, C.D.; Shamseer, L.; Tetzlaff, J.M.; Akl, E.A.; Brennan, S.E.; et al. The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ 2021, 372, 71. [Google Scholar] [CrossRef]
- Orlowski, A.; Romanowska, P. Smart cities concept: Smart mobility indicator. Cybern. Syst. 2019, 50, 118–131. [Google Scholar] [CrossRef]
- Nastjuk, I.; Trang, S.; Papageorgiou, E.I. Smart cities and smart governance models for future cities: Current research and future directions. Electron. Mark. 2022, 32, 1917–1924. [Google Scholar] [CrossRef]
- Addas, A. The concept of smart cities: A sustainability aspect for future urban development based on different cities. Front. Environ. Sci. 2023, 11, 1241593. [Google Scholar] [CrossRef]
- Gracias, J.S.; Parnell, G.S.; Specking, E.; Pohl, E.A.; Buchanan, R. Smart cities—A structured literature review. Smart Cities 2023, 6, 1719–1743. [Google Scholar] [CrossRef]
- Lei, B.; Janssen, P.; Stoter, J.; Biljecki, F. Challenges of urban digital twins: A systematic review and a Delphi expert survey. Autom. Constr. 2023, 147, 104716. [Google Scholar] [CrossRef]
- Van der Valk, H.; Haße, H.; Möller, F.; Arbter, M.; Henning, J.L.; Otto, B. A Taxonomy of Digital Twins. In Proceedings of the AMCIS 2020 Proceedings, Salt Lake City, UT, USA, 10–14 August 2020. [Google Scholar]
- González-Gerpe, S.; Poveda-Villalón, M.; García-Castro, R. DTAG: A Methodology for Aggregating Digital Twins Using the WoTDT Ontology. Appl. Sci. 2024, 14, 5960. [Google Scholar] [CrossRef]
- Vetrivel, S.; Sowmiya, K.; Sabareeshwari, V. Digital Twins: Revolutionizing Business in the Age of AI. In Harnessing AI and Digital Twin Technologies in Businesses; IGI Global: Hershey, PA, USA, 2024; pp. 111–131. [Google Scholar] [CrossRef]
- Rezaei, Z.; Vahidnia, M.H.; Aghamohammadi, H.; Azizi, Z.; Behzadi, S. Digital twins and 3D information modeling in a smart city for traffic controlling: A review. J. Geogr. Cartogr. 2023, 6, 1865. [Google Scholar] [CrossRef]
- Li, Y.; Fan, Y.; Nie, L. Making governance agile: Exploring the role of artificial intelligence in China’s local governance. Public Policy Adm. 2025, 40, 276–301. [Google Scholar] [CrossRef]
- Elshoukry, H. Enhancing Disaster Management by leveraging Smart Cities Technologies. In Proceedings of the 2024 Mediterranean Smart Cities Conference (MSCC), Tetuan, Morocco, 2–4 May 2024; pp. 1–8. [Google Scholar] [CrossRef]
- Matei, A.; Cocoșatu, M. Artificial Internet of Things, Sensor-Based Digital Twin Urban Computing Vision Algorithms, and Blockchain Cloud Networks in Sustainable Smart City Administration. Sustainability 2024, 16, 6749. [Google Scholar] [CrossRef]
- Nayak, R.; Karunakaran, S.; Yamunaa, P.; Kayalvili, S.; Tiwari, M.; Unni, M.V. Deep Learning and IoT Data in Smart City Applications. Urban Data Sci. 2023, 4, 16–27. [Google Scholar]
- Li, X.; Liu, H.; Wang, W.; Zheng, Y.; Lv, H.; Lv, Z. Big data analysis of the internet of things in the digital twins of smart city based on deep learning. Future Gener. Comput. Syst. 2022, 128, 167–177. [Google Scholar] [CrossRef]
- Nica, E.; Popescu, G.; Poliak, M.; Kliestik, T.; Sabie, O. Digital Twin Simulation Tools, Spatial Cognition Algorithms, and Multi-Sensor Fusion Technology in Sustainable Urban Governance Networks. Mathematics 2023, 11, 1981. [Google Scholar] [CrossRef]
- Santos, S.; Vilela, J.; Carvalho, T.; Rocha, T.; Candido, T.; Bezerra, V.; Silva, D. Artificial Intelligence in Sustainable Smart Cities: A Systematic Study on Applications, Benefits, Challenges, and Solutions. Sustainability 2024, 1, 644–655. [Google Scholar] [CrossRef]
- Abirami, L.; Karthikeyan, J. Digital Twin-Based Healthcare System (DTHS) for Earlier Parkinson Disease Identification and Diagnosis Using Optimized Fuzzy Based k-Nearest Neighbor Classifier Model. IEEE Access 2023, 11, 96661–96672. [Google Scholar] [CrossRef]
- Nalini, D. Securing Smart Cities: A Cybersecurity Perspective on Integrating IoT, AI, and Machine Learning for Digital Twin Creation. J. Electr. Syst. 2024, 20, 1420–1429. [Google Scholar] [CrossRef]
- Lee, Y.; Baek, M.S.; Yoon, K. Digital Entity Management Methodology for Digital Twin Implementation: Concept, Definition, and Examples. IEEE Trans. Broadcast. 2025, 71, 19–29. [Google Scholar] [CrossRef]
- Marzi, G.; Balzano, M.; Caputo, A.; Pellegrini, M.M. Guidelines for bibliometric-systematic literature reviews: 10 steps to combine analysis, synthesis and theory development. Int. J. Manag. Rev. 2025, 27, 81–103. [Google Scholar] [CrossRef]
- Mengist, W.; Soromessa, T.; Legese, G. Method for conducting systematic literature review and meta-analysis for environmental science research. MethodsX 2020, 7, 100777. [Google Scholar] [CrossRef] [PubMed]
- Haddaway, N.R.; Page, M.J.; Pritchard, C.C.; McGuinness, L.A. PRISMA2020: An R package and Shiny app for producing PRISMA 2020-compliant flow diagrams, with interactivity for optimised digital transparency and Open Synthesis. Campbell Syst. Rev. 2022, 18, e1230. [Google Scholar] [CrossRef]
- Cabrera, D.; Cabrera, L.L. The steps to doing a systems literature review (SLR). J. Syst. Think. Prepr. 2023, 34, 3–15. [Google Scholar]
- Tod, D.; Booth, A.; Smith, B. Critical appraisal. Int. Rev. Sport Exerc. Psychol. 2022, 15, 52–72. [Google Scholar] [CrossRef]
- Booth, A.; James, M.S.; Clowes, M.; Sutton, A. Systematic Approaches to a Successful Literature Review; SAGE Publications Ltd.: Thousand Oaks, CA, USA, 2021. [Google Scholar]
- Flyvbjerg, B.; Landman, T.; Schram, S. Real Social Science: Applied Phronesis; Cambridge University Press: Cambridge, UK; Taylor & Francis: Abingdon, UK, 2012. [Google Scholar] [CrossRef]
- Hosseini, M.S.; Jahanshahlou, F.; Akbarzadeh, M.A.; Zarei, M.; Vaez-Gharamaleki, Y. Formulating research questions for evidence-based studies. J. Med. Surg. Public Health 2024, 2, 100046. [Google Scholar] [CrossRef]
- Landis, J.R.; Koch, G.G. The measurement of observer agreement for categorical data. Biometrics 1977, 33, 159–174. [Google Scholar] [CrossRef]
- Kolaski, K.; Romeiser Logan, L.; Ioannidis, J.P. Guidance to best tools and practices for systematic reviews. J. Pediatr. Rehabil. Med. 2023, 16, 241–273. [Google Scholar]
- Moher, D.; Liberati, A.; Tetzlaff, J.; Altman, D.G. Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. PLoS Med. 2009, 6, e1000097. [Google Scholar] [CrossRef]
- Vom Brocke, J.; Simons, A.; Niehaves, B.; Riemer, K.; Plattfaut, R.; Cleven, A. Reconstructing the giant: On the importance of rigour in documenting the literature search process. In Proceedings of the ECIS 2009, Verona, Italy, 8–10 June 2009; Volume 372. Available online: https://aisel.aisnet.org/ecis2009/372 (accessed on 2 October 2025).
- Rejeb, A.; Rejeb, K.; Abdollahi, A.; Keogh, J.G.; Zailani, S.; Iranmanesh, M. Smart city research: A bibliometric and main path analysis. J. Data Inf. Manag. 2022, 4, 343–370. [Google Scholar] [CrossRef]
- Bibri, S.E.; Huang, J.; Omar, O.; Kenawy, I. Synergistic integration of digital twins and zero energy buildings for climate change mitigation in sustainable smart cities: A systematic review and novel framework. Energy Build. 2025, 333, 115484. [Google Scholar] [CrossRef]
- Bauer, M.; Cirillo, F.; Fürst, J.; Solmaz, G.; Kovacs, E. Urban Digital Twins—A FIWARE-based model. Automatisierungstechnik 2021, 69, 1106–1115. [Google Scholar] [CrossRef]
- Mahomed, A.S.; Saha, A.K. Unleashing the Potential of 5G for Smart Cities: A Focus on Real-Time Digital Twin Integration. Smart Cities 2025, 8, 70. [Google Scholar] [CrossRef]
- Qu, Q.; Hatami, M.; Xu, R.; Nagothu, D.; Chen, Y.; Li, X.; Blasch, E.; Ardiles-Cruz, E.; Chen, G. The Microverse: A Task-Oriented Edge-Scale Metaverse. Future Internet 2024, 16, 60. [Google Scholar] [CrossRef]
- Fuller, A.; Fan, Z.; Day, C.; Barlow, C. Digital Twin: Enabling Technologies, Challenges and Open Research. IEEE Access 2020, 8, 108952–108971. [Google Scholar] [CrossRef]
- Zhang, X.; Li, S.; Zhou, Q.; Sun, Y. Logical and Innovative Construction of Digital Twin City. J. Geod. Geoinf. Sci. 2021, 4, 113. [Google Scholar] [CrossRef]
- Duran, K.; Shin, H.; Duong, T.; Canberk, B. GenTwin: Generative AI-Powered Digital Twinning for Adaptive Management in IoT Networks. IEEE Trans. Cogn. Commun. Netw. 2025, 11, 1053–1066. [Google Scholar] [CrossRef]
- Supianto, A.A.; Nasar, W.; Aspen, D.M.; Hasan, A.; Karlsen, A.T.; Torres, R.D.S. An Urban Digital Twin Framework for Reference and Planning. IEEE Access 2024, 12, 152444–152465. [Google Scholar] [CrossRef]
- Ramu, S.P.; Boopalan, P.; Pham, Q.V.; Maddikunta, P.K.R.; Huynh-The, T.; Alazab, M.; Nguyen, T.T.; Gadekallu, T.R. Federated learning enabled digital twins for smart cities: Concepts, recent advances, and future directions. Sustain. Cities Soc. 2022, 79, 103663. [Google Scholar] [CrossRef]
- Zhang, Z.; Xiao, Y.; Ma, Z.; Xiao, M.; Ding, Z.; Lei, X.; Karagiannidis, G.K.; Fan, P. 6G wireless networks: Vision, requirements, architecture, and key technologies. IEEE Veh. Technol. Mag. 2019, 14, 28–41. [Google Scholar] [CrossRef]
- Minerva, R.; Lee, G.M.; Crespi, N. Digital twin in the IoT context: A survey on technical features, scenarios, and architectural models. Proc. IEEE 2020, 108, 1785–1824. [Google Scholar] [CrossRef]
- Fatorachian, H.; Kazemi, H.; Pawar, K. Enhancing Smart City Logistics Through IoT-Enabled Predictive Analytics: A Digital Twin and Cybernetic Feedback Approach. Smart Cities 2025, 8, 56. [Google Scholar] [CrossRef]
- Lv, Z.; Chen, D.; Lv, H. Smart city construction and management by digital twins and BIM big data in COVID-19 scenario. ACM Trans. Multimed. Comput. Commun. Appl. (TOMM) 2022, 18, 1–21. [Google Scholar] [CrossRef]
- Xiao, L.; Han, D.; Li, K.C.; Khan, M.K. UETOPSIS: A Data-Driven Intelligence Approach to Security Decisions for Edge Computing in Smart Cities. ACM Trans. Sens. Netw. 2024, 20, 2. [Google Scholar] [CrossRef]
- Negueroles, J.; Garcia-Valls, M. A Blockchain-based Digital Twin for IoT Deployments in Logistics and Transportation. Future Gener. Comput. Syst. 2024, 161, 81–84. [Google Scholar] [CrossRef]
- Wermann, A.G.; Wickboldt, J.A. KTWIN: A serverless Kubernetes-based Digital Twin platform. Comput. Netw. 2025, 259, 111095. [Google Scholar] [CrossRef]
- Mavromatis, G.; Karypidou, M.; Kassler, A. UMBRELLA: A One-Stop Shop Bridging the Gap From Lab to Real-World IoT Experimentation. Sensors 2024, 24, 112. [Google Scholar] [CrossRef]
- He, H.; Liu, Z.; Wu, X. Situation Awareness of Energy Internet of Things in Smart City Based on Digital Twin: From Digitization to Informatization. IEEE Internet Things J. 2023, 109, 7439–7458. [Google Scholar] [CrossRef]
- Martinez, C.; Gutierrez, P.; Huang, L. Platformization and the Metaverse: Opportunities and Challenges for Urban Sustainability and Economic Development. Cities 2023, 139, 104289. [Google Scholar] [CrossRef]
- Gao, K.; Lu, D.; He, H.; Xu, L.; Li, J.; Gong, Z. Enhanced 3D Urban Scene Reconstruction and Point Cloud Densification using Gaussian Splatting and Google Earth Imagery. IEEE Trans. Geosci. Remote Sens. 2025, 63, 1–14. [Google Scholar] [CrossRef]
- Boos, U.C.; Reichenbacher, T.; Kiefer, P.; Sailer, C. An augmented reality study for public participation in urban planning. J. Locat. Based Serv. 2023, 17, 48–77. [Google Scholar] [CrossRef]
- Rawlley, O.; Gupta, S.; Chandrakar, J.; Johnson, M.K. Artificial Intelligence Inspired Task Offloading and Resource Orchestration in Intelligent Transportation Systems. Cogn. Comput. 2024, 17, 18. [Google Scholar] [CrossRef]
- Megía, M.; Melero, F.J.; Chiachío, M.; Chiachío, J. Generative Adversarial Networks for Improved Model Training in the Context of the Digital Twin. Struct. Control Health Monit. 2024, 2024, 9997872. [Google Scholar] [CrossRef]
- Vishnu Shankar, S.; Naffees Gowsar, S.; Manjubala, M. Variational Autoencoders (VAEs). In Revolution with Generative AI: Trends and Techniques; Springer Nature: Geneva, Switzerland, 2025; Volume 50, pp. 43–64. [Google Scholar] [CrossRef]
- Garcia-Valls, M.; Negueroles, J. CoTwin: Collaborative improvement of digital twins enabled by blockchain in cyber-physical systems and IoT. Future Gener. Comput. Syst. 2024, 161, 81–84. [Google Scholar] [CrossRef]
- Yoshida, T.; Fujita, H. Desirable World with CPS and IoT: Smart City Design Concepts and Challenges. Sensors 2023, 23, 7890. [Google Scholar] [CrossRef]
- Holt, S.; Liu, T.; van der Schaar, M. Automatically learning hybrid digital twins of dynamical systems. Adv. Neural Inf. Process. Syst. 2024, 37, 72170–72218. [Google Scholar]
- Hakiri, A.; Yahia, S.B.; Gokhale, A.S.; Mellouli, N. Special Issue on Digital Twin for Future Networks and Emerging IoT Applications (DT4IoT). Future Gener. Comput. Syst. 2022, 161, 81–84. [Google Scholar] [CrossRef]
- Williams, J.; Stone, E. SIGNED: Smart cIty diGital twiN vErifiable Data Framework. Future Gener. Comput. Syst. 2023, 152, 20–33. [Google Scholar] [CrossRef]
- Poorzare, R.; Kanellopoulos, D.N.; Sharma, V.K.; Dalapati, P.; Waldhorst, O.P. Network Digital Twin Toward Networking, Telecommunications, and Traffic Engineering: A Survey. IEEE Access 2025, 13, 16489–16538. [Google Scholar] [CrossRef]
- Hunko, I. Adaptive Approaches to Software Testing with Embedded Artificial Intelligence in Dynamic Environments. Int. J. Curr. Sci. Res. Rev. 2025, 8, 2036–2051. [Google Scholar] [CrossRef]
- Li, K.; Cui, Y.; Li, W.; Lv, T.; Yuan, X.; Li, S.; Ni, W.; Simsek, M.; Dressler, F. When Internet of Things Meets Metaverse: Convergence of Physical and Cyber Worlds. IEEE Internet Things J. 2023, 10, 4148–4175. [Google Scholar] [CrossRef]
- Sánchez, J.M.G.; Jörgensen, N.; Törngren, M.; Inam, R.; Berezovskyi, A.; Feng, L.; Fersman, E.; Ramli, M.R.; Tan, K. Edge computing for cyber-physical systems: A systematic mapping study emphasizing trustworthiness. ACM Trans.-Cyber-Phys. Syst. (TCPS) 2022, 6, 1–28. [Google Scholar] [CrossRef]
- Shah, K.B.; Visalakshi, S.; Guragain, D.P.; Panigrahi, R. Advancing smart city sustainability with Internet of Things and artificial intelligence aided low-cost digital twin systems for waste management. Microsyst. Technol. 2024, 31, 9. [Google Scholar] [CrossRef]
- Mahmudah, H.; Musyafa, A.; Aisjah, A.S.; Arifin, S.; Prastyanto, C.A. Digital Twin: Challenge Road Damage Detection on Edge Device. Chem. Eng. Trans. 2024, 109, 601–606. [Google Scholar] [CrossRef]
- Meierhofer, J.; Schweiger, L.; Lu, J.; Züst, S.; West, S.; Stoll, O.; Kiritsis, D. Digital twin-enabled decision support services in industrial ecosystems. Appl. Sci. 2021, 11, 11418. [Google Scholar] [CrossRef]
- Wahba, M.; Hosny, H.; ElSayed, A. A Digital Twin Decision Support System for the Urban Facility Management Process. J. Urban Technol. 2023, 30, 89–104. [Google Scholar]
- Wang, T.; Tian, J.; Fang, K.; Gadekallu, T.R.; Wang, W. AI and Digital Twin for Consumer Electronics in Smart Cities. IEEE Consum. Electron. Mag. 2024, 8, 1–8. [Google Scholar] [CrossRef]
- Wajid, M.S.; Terashima-Marin, H.; Najafirad, P.; Pablos, S.E.C.; Wajid, M.A. DTwin-TEC: An AI-based TEC district digital twin and emulating security events by leveraging knowledge graph. J. Open Innov. Technol. Mark. Complex. 2024, 10, 100297. [Google Scholar] [CrossRef]
- Kumaş, E.; Aslan, D. A case study: Making decisions for sustainable university campus planning using GeoAI. Int. J. Eng. Geosci. 2025, 10, 22–35. [Google Scholar] [CrossRef]
- Hämäläinen, M. Urban development with dynamic digital twins in Helsinki city. IET Smart Cities 2021, 3, 201–210. [Google Scholar] [CrossRef]
- White, G.; Zink, A.; Codecá, L.; Clarke, S. A digital twin smart city for citizen feedback. Cities 2021, 110, 103064. [Google Scholar] [CrossRef]
- Garrido, C.; Lopez, A.; Martin, D. Digital Twins in Built Environments: An Investigation of the Characteristics, Applications, and Challenges. Buildings 2023, 13, 310. [Google Scholar] [CrossRef]
- Qanazi, S.; Leclerc, E.; Bosredon, P. Integrating social dimensions into urban digital twins: A review and proposed framework for social digital twins. Smart Cities 2025, 8, 23. [Google Scholar] [CrossRef]
- Elhamy, K.; Ahmed, D. Digital Twinning as an Approach to Promoting Real Estate Projects: Residential Unit at The Lake West, Sheikh Zayed City, Cairo, Egypt. Archit. Res. 2023, 10, 2. [Google Scholar] [CrossRef]
- DeVries, C. Governance Framework for Citizen Digital Twins in Smart Cities. Intell. Hum. Futur. 2025, 1. [Google Scholar] [CrossRef]
- Knopf, S.; Macq, H.; Wentland, A. Urban futures in the mirror of technology? The politics of urban digital twins. New Media Soc. 2025, 27, 4385–4401. [Google Scholar] [CrossRef]
- Minerva, R.; Awan, F.M.; Crespi, N. Exploiting digital twins as enablers for synthetic sensing. IEEE Internet Comput. 2021, 26, 61–67. [Google Scholar] [CrossRef]
- Hardy, J.; Nguyen, L.; Smith, A. Integrating Technology and Urban Resilience: A Comprehensive Analysis of Smart City Initiatives in Sydney. Urban Stud. 2024, 61, 320–339. [Google Scholar]
- Park, S.; Lee, S.; Park, S.; Park, S. AI-Based Physical and Virtual Platform with 5-Layered Architecture for Sustainable Smart Energy City Development. Sustainability 2019, 11, 4479. [Google Scholar] [CrossRef]
- Ullah, I.; Noor, A.; Abbas, M.; Garg, S.; Choi, B.J.; Hassan, M.M.; Bai, X. Optimizing smart city services by utilizing appropriate characteristics of digital twin for urban excellence. Alex. Eng. J. 2025, 122, 399–410. [Google Scholar] [CrossRef]
- Kumar, M.G.V.; Karunakaran, S.; Chandre, S.; Godi, R.K.; Manirajkumar, P.; Balaram, A. Implementation of Microgrid Digital Twin System for Unmanned Vehicles with Cloud Computing Techniques. SN Comput. Sci. 2023, 4, 566. [Google Scholar] [CrossRef]
- Adil, M.; Ali, J.; Jadoon, M.M.; Alotaibi, S.R.; Kumar, N.; Farouk, A.; Song, H. COVID-19: Secure healthcare Internet of Things networks, current trends and challenges with future research directions. ACM Trans. Sens. Netw. 2023, 19, 1–25. [Google Scholar] [CrossRef]
- Khan, F.; Anwar, H.; Rehman, N. Digital Twin Perspective of Fourth Industrial and Healthcare Revolution. Int. J. Emerg. Trends Eng. Res. 2022, 10, 45–55. [Google Scholar] [CrossRef]
- Klar, R.; Fredriksson, A.; Angelakis, V. Digital twins for ports: Derived from smart city and supply chain twinning experience. IEEE Access 2023, 11, 71777–71799. [Google Scholar] [CrossRef]
- Adreani, L.; Bellini, P.; Colombo, C.; Fanfani, M.; Nesi, P.; Pantaleo, G.; Pisanu, R. Implementing integrated digital twin modelling and representation into the Snap4City platform for smart city solutions. Multimed. Tools Appl. 2024, 83, 37121–37146. [Google Scholar] [CrossRef]
- Li, W.; Zhang, H.; Liu, Y. DNA Computing-Based Multi-Source Data Storage Model in Digital Twins. Future Gener. Comput. Syst. 2023, 138, 48–60. [Google Scholar] [CrossRef]
- Al-Turjman, A.; Hassan, M. Trends and Challenges in AIoT, IIoT and IoT Implementation. IEEE Internet Things Mag. 2023, 6, 24–30. [Google Scholar]
- Ono, S.; Yamazaki, T.; Miyoshi, T.; Taya, A.; Nishiyama, Y.; Sezaki, K. AMoND: Area-controlled mobile ad-hoc networking with digital twin. IEEE Access 2023, 11, 85224–85236. [Google Scholar] [CrossRef]
- Burgueño, L.; Di Ruscio, D.; Sahraoui, H.; WImmer, M. Automation in Model-Driven Engineering: A look back, and ahead. ACM Trans. Softw. Eng. Methodol. 2023, 34, 1–25. [Google Scholar] [CrossRef]
- Alkhateeb, A.; Jiang, S.; Charan, G. Real-time digital twins: Vision and research directions for 6G and beyond. IEEE Commun. Mag. 2023, 61, 128–134. [Google Scholar] [CrossRef]
- El Mettiti, A.; Oumsis, M. A survey on 6G networks: Vision, requirements, architecture, technologies and challenges. Ing. Des Syst. D’Inform. 2022, 27, 15487–15521. [Google Scholar] [CrossRef]
- Moroni, S. Insurmountable limitations of city-scale digital twins? On urban knowledge and planning. Comput. Urban Sci. 2025, 5, 17. [Google Scholar] [CrossRef]
- Ghosh, J.; Banerji, O. Excavating the Role of Digital Twins in Upgrading Cities and Homes amidst 21st Century: A Techno-Legal Persepctive. Indian J. Law Justice 2023, 14, 248. [Google Scholar]
- Cheng, Y.; Zhang, X.; Coeckelbergh, M. Will the Metaverse Be Out of Control? Addressing the Ethical and Governance Implications of a Developing Virtual Society. Technol. Forecast. Soc. Chang. 2023, 189, 122327. [Google Scholar] [CrossRef]
- Lv, Z.; Chen, D.; Feng, H.; Singh, A.K.; Wei, W.; Lv, H. Computational intelligence in security of digital twins big graphic data in cyber-physical systems of smart cities. ACM Trans. Manag. Inf. Syst. (TMIS) 2022, 13, 1–17. [Google Scholar] [CrossRef]
- Papaioannou, C.; Dimara, A.; Papaioannou, A.; Tzitzios, I.; Anagnostopoulos, C.N.; Krinidis, S. Hierarchical Resources Management System for Internet of Things-Enabled Smart Cities. Sensors 2025, 25, 616. [Google Scholar] [CrossRef]
- Stogia, M.; Naserentin, V.; Dimara, A.; Eleftheriou, O.; Tzitzios, I.; Papaioannou, C.; Pantusheva, M.; Papaioannou, A.; Spaias, G.; Anagnostopoulos, C.N.; et al. A scalable and user-friendly framework integrating IoT and digital twins for home energy management systems. Appl. Sci. 2024, 14, 11834. [Google Scholar] [CrossRef]
- Wang, Z.; Grieves, M.; Wen, Y.; Xue, F. Special Issue on Digital Twins. IEEE Internet Comput. 2025, 29, 5–7. [Google Scholar] [CrossRef]
- Rueda, C.; Maalej, Y.; Bchir, O. Generative Digital Twins: A Novel Approach in the IoT Edge-Cloud Continuum. IEEE Internet Things Mag. 2024, 8, 42–48. [Google Scholar] [CrossRef]
- Wang, L.; Zhao, F.; Tan, J. Digital Twin of Intelligent Small Surface Defect Detection with Cyber-manufacturing Systems. IEEE Trans. Ind. Inform. 2023, 19, 2057–2067. [Google Scholar] [CrossRef]
Code | Mapping Question |
---|---|
MQ1 | Which are the top scientific journals where Digital Twins, IoT and AI in smart cities are published? |
MQ2 | Which are the most influential authors and articles at the intersection of DTw, IoT and AI? |
MQ3 | What enabling technologies and methodological approaches most frequently support UDTw? |
MQ4 | Which use cases and practical implementations are most frequently reported, and with what outcomes and limitations? |
MQ5 | What are the most frequently used methodologies in smart city DTw studies? |
MQ6 | What are the main research gaps? |
Electronic Database | Type | URL |
---|---|---|
IEEE Xplore | Digital Library | https://ieeexplore.ieee.org, accessed on 4 July 2025 |
Web of Science (WoS) | Digital Library | https://www.webofscience.com, accessed on 4 July 2025 |
Scopus | Digital Library | https://www.scopus.com, accessed on 4 July 2025 |
Association for Computer Machinery (ACM) | Digital Library | https://dl.acm.org, accessed on 4 July 2025 |
Criterion | Yes (1) | Partial (0.5) | No (0) |
---|---|---|---|
1. Clarity of Research Question (Max. 2 points) | |||
Is there an explicit and well-defined research question or objective? | |||
Is the research question related to Digital Twins, IoT, and AI in smart cities? | |||
2. Internal Validity (Max. 2 points) | |||
Does the study design demonstrate methodological rigor? | |||
Are measures taken to minimize potential sources of bias? | |||
3. Accuracy of Methods (Max. 2 points) | |||
Are the sensors and data acquisition systems valid and appropriate? | |||
Are the diagnostic algorithms accurately applied? | |||
… Other criteria evaluated … | |||
4. Practical Implications (Max. 1 point) | |||
Are the practical applications in smart cities discussed? | |||
Is the impact of IoT and AI on these applications analyzed? | |||
Final Assessment | |||
Total Score (out of 10) | |||
Interpretation | 8.5–10: Highly relevant and high-quality article. 7–8.4: Useful article with some areas for improvement. 5–6.9: Article with significant limitations. <5: Low relevance or limited quality article. |
Field Number | Content |
---|---|
1 | Authors/Year |
2 | Article title |
3 | Main objective |
4 | Methodology |
5 | Main findings |
6 | Relation with DTw, IoT and AI in smart cities |
Reference | Scope/Focus | Limitations | Contribution of This Review |
---|---|---|---|
[30] | Digital twins in smart cities, focus on data integration | Limited IoT/AI coverage, no testbeds | Integrates IoT, AI, GenAI, and testbeds |
[18] | UDTw for sustainability and resilience | Lacked technical depth in IoT/AI | Deepens Edge, FL, AR, LLMs in urban contexts |
[23] | AI-driven digital twins | Neglected governance, interoperability, testbeds | Adds governance, middleware (FIWARE, NGSI-LD), scalability |
This SLR | UDTw + IoT + AI in smart cities | N/A | Integrates SLR, applications, and research agenda |
Technology/Concept | Key Advance/Feature | Impact on UDTw | Ref. |
---|---|---|---|
GenAI | Autonomous generation of data, hypothetical scenarios, urban 3D designs and models. | Greater autonomy in UDTw creation, reduced costs and entry barriers, enhanced predictive capabilities. | [64] |
Massive IoT | Integration of thousands of urban sensors with metaverse platforms; development of large multimodal datasets. | Real-time multi-user collaboration, large-scale UDTw validation, improved 3D model accuracy. | [89] |
Edge/Cloud computing | Serverless platforms based on Kubernetes (KTWIN) for unified and agnostic deployment. | Optimization of heterogeneous data processing and storage, reduced latency, lower operational costs. | [90] |
3D visualization and AR | Use of Gaussian Splatting for 3D mesh extraction; LLMs for visual and semantic descriptions. | Intuitive and immersive interaction, photorealistic 3D modeling, improved understanding of urban environments. | [61] |
Blockchain | Enhanced traceability, security, and interoperability; integration with Federated Learning (FL). | Increased trust in data, privacy protection, model integrity in distributed systems. | [91] |
Hybrid models with CPS | Combination of mechanical and neural components; LLM-driven evolutionary algorithms (e.g., HDTwinGen). | Better generalization under data scarcity, efficient learning, greater flexibility for evolvability, reduced CPS operational costs. | [58] |
Interoperable middleware | Solutions like FIWARE and NGSI-LD for integrating heterogeneous platforms. | Enabling cohesive digital urban ecosystems, global-scale standardization. | [59] |
Testbed platforms | Controlled environments for testing and scaling UDTw prototypes (e.g., Aerial Experimentation and Research Platform for Advanced Wireless (AERPAW), first responder training). | Rigorous validation of AI algorithms, rapid iteration, communication system optimization, safe training. | [92] |
Application Domain | City/Project | Specific UDTw Application | Key Benefits/Outcomes | References |
---|---|---|---|---|
Traffic management and smart mobility | Sydney | Traffic accident risk prediction, dynamic optimization of traffic lights and routes. | Reduced congestion, improved road safety, prioritization of electric mobility. | [106] |
Traffic management and smart mobility | New York | Real-time traffic management, simulation and operation optimization. | Improved traffic efficiency, integration of big data and AI. | [64] |
Urban planning and citizen participation | Virtual Singapore | Simulation of urban developments, impact assessment (ventilation, traffic). | More informed and efficient urban planning. | [58,98] |
Urban planning and citizen participation | Bologna | Participatory governance, scenario visualization for citizens. | Greater citizen participation, more equitable and acceptable urban solutions. | [101] |
Environmental monitoring and climate resilience | Lisbon | Flood simulation, design of climate resilience strategies. | Reduction of climate risks, evidence-based planning. | [91] |
Environmental monitoring and climate resilience | Generic (recent studies) | Interactive pollution dispersion prediction, use of low-cost sensors for air quality. | Improved air quality, identification of critical exposure points, dynamic decision-making. | [105] |
Energy management and smart grids | Generic (smart grids) | Demand forecasting, fault detection, electric grid balancing, consumption optimization. | Greater energy efficiency, integration of renewable, grid stability. | [107] |
Public services (Waste) | Generic (recent studies) | Waste generation simulation, optimization of collection routes. | Reduction of operational costs and emissions, improved waste management efficiency. | [96] |
Emergency management | Generic (civil infrastructure) | Application in disaster mitigation, preparedness, response, and recovery. | Improved community resilience, fast and effective emergency response. | |
Emergency management | First responder testbed | Enhanced training and simulated communication in photorealistic virtual environments. | Reduced training risks, optimized communication in critical incidents. | [92] |
Enabling Layer | Sectors | Illustrative Examples |
---|---|---|
IoT sensor networks | Energy, mobility, waste | Deployment of IoT-based infrastructures to monitor energy consumption, traffic flows, and waste bins in real-time [91,92,108]. |
AI and data analytics | Transport, healthcare, environment | Use of AI/ML algorithms for predictive maintenance, anomaly detection, and environmental monitoring across different domains [58,64,95]. |
Cloud/Fog/Edge computing | Energy, transport, public safety | Distributed computing frameworks enabling real-time digital twin operations and adaptive management [61,64,109]. |
5G/Next-Gen connectivity | transport, safety, smart grids | Ultra-low latency communications enabling real-time DTw integration in mobility systems and critical infrastructure [60,89]. |
DTw platforms | Built environment, utilities, governance | FIWARE-based UDTw models and sector-specific DTw platforms reused for facility management, water, and energy systems [59,96,108]. |
Challenge Category | Specific Challenge | Proposed Research Direction | References |
---|---|---|---|
Real-time predictions | Latency in critical applications; need for lightweight AI models. | AI models adaptable to edge computing; latency management for control system safety. | [58,66] |
Interoperability | Barriers between heterogeneous platforms; lack of common standards and ontologies. | Advancement of open standards; ontologies for knowledge representation and state graphs; ontology reuse. | [58] |
Scalability and performance | Development of twins for megacities; complexity of data management. | Hierarchical approaches; microservice-based architectures; large-scale reference datasets. | [66] |
Security and privacy | Data transmission risks; authentication; integration with other systems; AI threats; fidelity-security dilemma; ethics. | Defense-in-depth strategies; blockchain for traceability/privacy; federated learning; Zero-Trust architectures; uncertainty quantification; ethical frameworks. | [122] |
Impact assessment | Lack of clear ROI metrics, operational efficiency, and social benefits; gap between ambition and actual contribution. | Benchmarking methodologies and longitudinal studies; integration of social dimensions (AUP). | [58] |
Scalability and replicability | Ability to manage the growing number of IoT devices, services, and data streams in urban contexts; difficulty in transferring solutions across heterogeneous cities without major redesign. | Adoption of hierarchical resource management, modular and microservice-based architectures, federated learning, and open communication standards; development of frameworks validated from household to megacity level; benchmarking replicability across diverse governance models. | [124,125] |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Sacoto-Cabrera, E.J.; Perez-Torres, A.; Tello-Oquendo, L.; Cerrada, M. IoT, AI, and Digital Twins in Smart Cities: A Systematic Review for a Thematic Mapping and Research Agenda. Smart Cities 2025, 8, 175. https://doi.org/10.3390/smartcities8050175
Sacoto-Cabrera EJ, Perez-Torres A, Tello-Oquendo L, Cerrada M. IoT, AI, and Digital Twins in Smart Cities: A Systematic Review for a Thematic Mapping and Research Agenda. Smart Cities. 2025; 8(5):175. https://doi.org/10.3390/smartcities8050175
Chicago/Turabian StyleSacoto-Cabrera, Erwin J., Antonio Perez-Torres, Luis Tello-Oquendo, and Mariela Cerrada. 2025. "IoT, AI, and Digital Twins in Smart Cities: A Systematic Review for a Thematic Mapping and Research Agenda" Smart Cities 8, no. 5: 175. https://doi.org/10.3390/smartcities8050175
APA StyleSacoto-Cabrera, E. J., Perez-Torres, A., Tello-Oquendo, L., & Cerrada, M. (2025). IoT, AI, and Digital Twins in Smart Cities: A Systematic Review for a Thematic Mapping and Research Agenda. Smart Cities, 8(5), 175. https://doi.org/10.3390/smartcities8050175