SDI-Enabled Smart Governance: A Review (2015–2025) of IoT, AI and Geospatial Technologies—Applications and Challenges

Abstract

This paper presents a systematic, narrative review of 62 academic publications (2015–2025) that explore the integration of spatial data infrastructures (SDIs) with emerging smart city technologies to improve local governance. SDIs provide a structured framework for managing geospatial data and, in combination with IoT sensors, geospatial and 3D platforms, cloud computing and AI-powered analytics, enable real-time data-driven decision-making. The review identifies four key technology areas: IoT and sensor technologies, geospatial and 3D mapping platforms, cloud-based data infrastructures, and AI analytics that uniquely contribute to smart governance through improved monitoring, prediction, visualization, and automation. Opportunities include improved urban resilience, public service delivery, environmental monitoring and citizen engagement. However, challenges remain in terms of interoperability, data protection, institutional barriers and unequal access to technologies. To fully realize the potential of integrated SDIs in smart government, the report highlights the need for open standards, ethical frameworks, cross-sector collaboration and citizen-centric design. Ultimately, this synthesis provides a comprehensive basis for promoting inclusive, adaptive and accountable local governance systems through spatially enabled smart technologies.

1. Introduction

The rapid development of digital technologies is transforming urban governance and ushering in an era of data-driven decision-making and integrated governance systems. The rise of smart city paradigms driven by technologies such as the Internet of Things (IoT), artificial intelligence (AI), machine learning (ML) and cloud computing has challenged traditional governance models by enabling new forms of interaction, coordination and optimization at the local level [1,2,3]. In this context, spatial data infrastructures (SDIs), originally designed to promote the sharing, accessibility, and interoperability of geospatial data, have become important enablers for inclusive, transparent, and efficient smart governance systems [4,5].

SDIs serve as a basic framework for managing geospatial data assets, facilitating stakeholder coordination and strengthening the ability of public institutions to address complex urban challenges [6,7]. In conjunction with the Internet of Things, SDIs can process sensor data from urban infrastructure and the environment in real time, enabling the dynamic monitoring of traffic, weather, air quality and health indicators [8,9]. AI and ML algorithms enhance these infrastructures by uncovering patterns, forecasting trends, and generating actionable insights that can inform policy and planning processes [10,11]. Cloud technologies further support this integration by providing the scalability and interoperability needed to manage distributed data sets and complex analytics pipelines across administrative boundaries [12,13].

The strategic convergence of these technological pillars with SDIs enables smart local governance, a concept that refers to the digitally augmented capacity of municipal systems to respond to citizen needs, optimize resources, and manage change in real time [14,15]. This integration improves operational efficiency and promotes resilience, inclusivity, and sustainability in urban areas such as smart mobility, smart health, smart housing, and smart environment [16,17].

Despite the recognized benefits, the research landscape remains fragmented. While areas such as sensor-based environmental monitoring and AI-powered urban analytics have made rapid progress, other areas such as geospatial data interoperability, citizen engagement and cross-sector coordination continue to face technical and institutional barriers [18,19]. These inequalities hinder the equitable deployment of integrated smart systems [20,21].

Although the academic interest in SDIs and smart technologies has steadily increased over the past decade, there is still little comprehensive analysis of their convergence in the context of local governance. Most previous research either focuses on specific technologies (e.g., GeoAI, cloud platforms, IoT networks) or treats SDIs from a policy or infrastructure perspective, rarely systematically linking both dimensions [22,23,24]. As a result, there are still gaps in the knowledge about how integrated systems work in practice, which technical and organizational models are most effective, and how to mitigate regional inequalities in adoption.

The literature also reflects ongoing debates about the design and governance of systems. These include tensions between centralized and decentralized data infrastructures, trade-offs between data openness and privacy, and questions about the explainability and participatory nature of AI in civilian contexts [1,25,26]. In addition, some scholars argue for top-down institutional coordination as key to successful SDI integration, while others emphasize the role of community-led innovation and bottom-up data generation, especially in resource-constrained environments [27,28].

This review aims to address these gaps by systematically examining the literature from 2015 to 2025 on the integration of SDIs with emerging smart city technologies, focusing on their applications in smart local governance. Based on 62 peer-reviewed publications, it identifies the prevailing technological trends, assesses methodological developments and presents the evolution of the scientific discourse in this multidisciplinary field. The aim is to provide a comprehensive overview of how digital technologies are embedded in spatial infrastructures to support local government functions.

To achieve this, the review employs a systematic narrative methodology that combines thematic classification with bibliometric analysis. The selected studies are categorized into four main technological domains: (1) IoT and Sensor Technologies, (2) Geospatial Platforms and 3D Mapping, (3) Data Infrastructures and Cloud Technologies, and (4) Artificial Intelligence and Analytics. This classification enables the examination of each technology’s specific contributions to spatial governance, while also uncovering synergies, limitations, and future directions for research and practice.

The results show a strong emphasis on data infrastructures and cloud technologies, reflecting their foundational role in supporting scalable, secure, and interoperable systems. AI and analytics emerge as the second most addressed domain, with growing interest in predictive modeling, pattern recognition, and decision support systems. IoT and sensors are widely applied in environmental and infrastructural monitoring, while geospatial platforms and 3D mapping support visualization, urban simulations, and participatory planning. Notably, the literature highlights the increasing importance of standards-based interoperability, metadata quality, and real-time processing capabilities as enablers of integration [17,29,30].

Geographically, the research focuses on Europe and Asia, with significant contributions from global and institutional case studies. However, representation from Africa, Oceania, and the Americas remains limited, suggesting the need for greater inclusivity in global smart city research and SDI implementation efforts. The upward trajectory of publication trends indicates growing academic and policy interest in the topic, although critical gaps persist in the standardization of practices, citizen engagement mechanisms, and the evaluation of long-term impacts [15,31].

In summary, the integration of SDIs with smart technologies holds great promise for transforming local governance. However, to realize this potential, crucial technological, institutional and ethical challenges need to be addressed. By synthesizing a decade of academic literature into coherent thematic insights, this review provides a foundational resource for scholars, practitioners, and policymakers seeking to understand and advance spatial smart governance systems.

2. Materials and Methods

This study adopts a systematic narrative review approach to investigate the integration of emerging smart city technologies such as Artificial Intelligence (AI), Machine Learning (ML), and the Internet of Things (IoT) with Spatial Data Infrastructures (SDIs) in the context of smart local governance. The chosen methodology allows for the focused exploration of a multi-disciplinary and technically complex field, while preserving the systematic rigor and transparency of a structured review process.

2.1. Search and Selection Criteria

To ensure the scientific validity, thematic relevance, and comprehensiveness of the review, both the inclusion criteria and the search strategy were carefully designed and systematically applied. The following criteria were used to identify eligible studies:

• IC1: Language – Only publications written in English were included to ensure accessibility to internationally recognized scientific literature.
• IC2: Timeframe – The analysis covered the period from 2015 to 2025 in order to capture the most recent technological advancements and trends, as well as the evolutionary arc of SDI integration with emerging digital technologies.
• IC3: Content Focus and Thematic Scope—Studies were considered eligible only if they explicitly addressed the relationship between Spatial Data Infrastructures (SDIs) and smart city technologies such as Artificial Intelligence (AI), Machine Learning (ML), or the Internet of Things (IoT). Eligible studies included theoretical models, methodological frameworks, case studies, and practical implementations. Publications that focused solely on these technologies without any reference to SDIs were excluded.
• IC4: Geospatial Interoperability Standards – Although not a strict inclusion requirement, references to interoperability standards (e.g., OGC, INSPIRE, ISO 19100 [32]) were considered valuable for enriching the analysis of interoperability aspects.

The systematic literature search was conducted using the Scopus database, selected due to its extensive coverage of peer-reviewed publications in geospatial science, digital technologies, and governance. The search process was iterative and covered the period 2015–2025, aiming to reflect the evolution of the field. An interactive keyword refinement approach was adopted to optimize the relevance of the results. The search began with a broad query using the term “Spatial Data Infrastructures”, which yielded 2119 documents. This was followed by more targeted Boolean queries, “Spatial Data Infrastructures” AND (“Artificial Intelligence” OR “AI”) – 50 documents, “Spatial Data Infrastructures” AND (“Internet of Things” OR “IoT”) – 29 documents, “Spatial Data Infrastructures” AND (“Machine Learning” OR “ML”) – 23 documents. A final consolidated query was formulated as follows: (“Artificial Intelligence” OR “AI” OR “Internet of Things” OR “IoT” OR “Machine Learning” OR “ML”) AND (“Spatial Data Infrastructures”), which returned 94 documents.

Subsequently, the results were filtered based on publication year (2015–2025), and document type (article, conference paper, book chapter, or book), resulting in 66 documents. After the removal of duplicates, a total of 62 unique publications were retained for further evaluation. This final Boolean query was specifically designed to capture studies that explore the integration of digital technologies within SDIs in the context of smart governance, allowing for a comprehensive analysis of their contribution across various smart city dimensions (see Figure 1).

2.2. Study Selection

The initial pool of documents retrieved from Scopus underwent a rigorous multi-step filtering and selection process involving two independent reviewers. Duplicate records were first removed. Then, a two-phase evaluation was applied:

Phase 1: Screening of titles, abstracts, and keywords against the four inclusion criteria (IC1–IC4) to identify potentially relevant publications.

Phase 2: Full-text review of selected documents to verify thematic alignment and methodological quality. Discrepancies between reviewers were resolved through consensus, and where needed, a third reviewer was consulted to ensure objectivity.

Data from the selected studies were extracted by both reviewers independently using a standardized template designed to capture thematic domain, geographic scope, publication type, and keyword occurrence. This meticulous filtering process led to the selection of 62 publications, which were subsequently classified and analyzed. Given the narrative nature of this review and its focus on thematic and bibliometric trends rather than effect estimation, a formal risk of bias assessment was not conducted.

2.3. Descriptive Statistics

This comprehensive review is based on the analysis of 62 academic publications exploring the intersection of Spatial Data Infrastructures (SDIs) and emerging smart city technologies. Appendix A summarizes the highly cited articles in this field of study. Thematic classification of the literature reveals a pronounced focus on four main technological domains. The most extensively covered area is Data Infrastructures and Cloud Technologies, represented in 20 publications, reflecting the foundational role of scalable and interoperable data systems. This is followed by Artificial Intelligence and Analytics (19 publications), which highlights the increasing application of intelligent systems in spatial decision-making. The IoT and Sensor Technologies category accounts for 16 publications, indicating the value of real-time data acquisition in smart environments. Lastly, Geospatial Platforms and 3D Mapping appear in seven pubications, showcasing more specialized but essential spatial visualization technologies (Figure 2).

A keyword analysis further reinforces these findings. The term “Spatial Data Infrastructure” appears in 37 publications, underlining its centrality to the field. Artificial Intelligence is featured in 23 papers, while Internet of Things (IoT) and Machine Learning (ML) are present in 11 and 7, respectively. These patterns demonstrate the expanding integration of advanced computational technologies into spatial data ecosystems for urban innovation (Figure 3).

In terms of geographic distribution, the selected studies demonstrate a relatively balanced global interest. Europe ([9,15,16,17,22,23,24,25,28,31,33,34,35,36,37,38,39,40]), Asia ([8,11,12,13,14,20,41,42,43,44,45,46,47,48,49,50,51,52]) and global studies ([1,2,3,8,10,18,19,21,29,53,54,55,56,57,58,59,60,61]) lead with 18 publications each (

Table 1. Citation and link strength of included references in the co-citation analysis.

Author	Citations	Total Link Strength
Di Liping	33	300
Zhang Chao	18	233
Yue Peng	18	180
Rajabifard Abbas	39	74
Williamson Ian	17	67
Biljecki Filip	21	22

). Regional representation from America ([26,30,62,63]), Oceania ([4,5,7]), and particularly Africa ([27]) remains limited, suggesting an uneven research landscape and indicating a need for broader inclusion of perspectives from the Global South (Figure 4).

To respond more directly to this concern, we also performed a country-level analysis of the 62 selected studies. This revealed that 22 articles adopt a global or international perspective without reference to specific countries. Among those that do refer to national contexts, Russia (6 studies), India (5), Germany (4), and Australia (3) are the most frequently represented. Countries such as the United Kingdom, Turkey, Bosnia and Herzegovina, Italy, and Ecuador appear in two studies each. The remaining countries including China, Bangladesh, Brazil, Indonesia, France, Iran, Greece, South Korea, Kenya, Oman, Cyprus, and Spain are represented only once. This distribution supports the reviewer’s observation that countries from the Global North, especially in Europe, East Asia, and North America, dominate the dataset, while the Global South remains sparsely represented. These findings highlight the need for future reviews to adopt broader and more inclusive search strategies that can better reflect the diversity of global research contributions.

The temporal distribution of publications shows a generally increasing trend in scholarly output over the past decade. From 2015 to 2017, the number of publications was relatively low and stable, ranging from 1 to 3 per year. Starting in 2018, there was a significant rise, reaching eight publications, while 2020 marked the first peak with 10 publications. Although there was a drop in 2022 with only two publications, the following years saw a recovery, with 2024 recording the highest output at 13 publications. In 2025, despite limited data so far, two publications have already been noted, indicating that the upward trend may continue. Overall, this progression reflects growing research interest and output in the field (Figure 5).

Finally, the distribution of publication types shows that journal articles constitute the largest portion of the output, with a total of 31 publications. Conference papers follow closely, with 24 publications, also representing a significant share of the research activity. Book chapters and reviews have a smaller presence, with three publications each, while only one book has been recorded. This distribution suggests that the research primarily favors rapid knowledge dissemination through articles and conference presentations, with less emphasis on longer monographs or edited volumes (Figure 6).

In summary, the quantitative landscape of the reviewed literature validates the core premise of this study: that the strategic integration of SDIs with modern digital technologies including AI, ML, and IoT is increasingly essential for advancing smart local governance. These technologies collectively contribute to critical smart city domains such as Smart Society, Smart Living, Smart Governance, Smart Mobility, Smart Economy, Smart Health, and Smart Environment, reaffirming their transformative potential in shaping inclusive, data-informed urban futures.

Between 2015 and 2025, a total of 6624 authors contributed to the body of literature examining the integration of Spatial Data Infrastructures (SDIs) with cutting-edge technologies such as the Internet of Things (IoT), Artificial Intelligence (AI), and Machine Learning (ML) in the context of enhancing smart local governance. Within this scientific output, several authors emerged as particularly influential, including Rajabifard Abbas (39 citations), Di Liping (33 citations), Biljecki Filip (21 citations), Yue Peng (18 citations), Zhang Chao (18 citations), and Williamson Ian (17 citations). Their ranking is based on the number of citations received within the relevant scholarly production during the reviewed decade, reflecting their academic impact in the field.

To understand the relationships and interactions among these authors, co-citation analysis and bibliographic coupling were employed. Specifically, in the co-citation analysis, cited references were used as the unit of analysis, with a minimum citation threshold set at 16. Out of the 6624 total references identified, only six met this criterion and were included in the co-citation network. The visualization of the interconnections among these references is presented in Figure 7, where two major clusters can be observed. The red cluster holds a central and influential position within the co-citation network. It includes highly cited authors such as Rajabifard Abbas, Williamson Ian, and Biljecki Filip (Table 1). This cluster maintains strong linkages with other nodes in the network, indicating a combination of conceptual centrality and historical influence. Its position suggests a leading role in shaping the theoretical discourse around the integration of Spatial Data Infrastructures (SDIs) into smart governance systems, particularly through institutional and policy-oriented perspectives. In contrast, the green cluster is the second largest and demonstrates high levels of internal connectivity and thematic cohesion. It includes Di Liping, Zhang Chao, and Yue Peng, with Di Liping appearing to play a central role within the cluster. These authors are associated with more recent and technically focused contributions to the field, especially concerning the application of geospatial technologies, artificial intelligence, and data infrastructures in smart urban environments. To support the interpretation of the network structure, Table 1 presents the citation count and total link strength (TLS) for each of the six central references included in the co-citation network:

A co-occurrence analysis of keywords was conducted to explore the thematic structure and conceptual interlinkages within the selected body of literature. The unit of analysis included all keywords assigned to the selected scholarly articles, with a minimum occurrence threshold set at seven. Out of a total of 633 unique keywords, only nine met this criterion and were included in the analysis. As shown in

Table 2. Frequency of occurrence and total link strength (TLS) of the selected keywords.

Keyword	Occurrences	Total Link Strength
spatial data infrastructure	37	55
artificial intelligence	23	46
spatial data	21	38
GIS	13	26
remote sensing	9	19
interoperability	7	18
decision support systems	7	16
internet of things	11	11
machine learning	7	11

, the most frequently occurring keyword was spatial data infrastructure (37 occurrences, total link strength: 55), followed by artificial intelligence (23; TLS: 46), spatial data (21; TLS: 38), and GIS (13; TLS: 26). Other prominent terms included remote sensing (9; TLS: 19), internet of things (11; TLS: 11), machine learning (7; TLS: 11), interoperability (7; TLS: 18), and decision support systems (7; TLS: 16).

These findings highlight spatial data infrastructure as the most central and frequently connected concept within the dataset, confirming its role as a foundational theme in the literature (Figure 8). At the same time, the substantial presence and linkage of terms related to artificial intelligence, machine learning, and IoT underscore the increasing integration of advanced computational technologies into spatial governance frameworks. The prominence of terms such as interoperability and decision support systems further reflects a growing focus on system integration and practical application in decision-making contexts. Overall, the keyword co-occurrence analysis reveals two converging thematic trajectories: one centered on the enduring significance of spatial data infrastructures, and another emphasizing the dynamic influence of emerging digital technologies in shaping smart governance systems.

3. Results

The systematic review identified four major technological domains that intersect significantly with Spatial Data Infrastructures (SDIs) in the context of smart governance: (1) IoT and Sensor Technologies, (2) Geospatial Platforms and 3D Mapping, (3) Data Infrastructures and Cloud Technologies, and (4) Artificial Intelligence and Analytics. Each domain contributes distinct capabilities to SDI-driven systems, ranging from real-time data collection to semantic reasoning and predictive modeling. To provide a structured overview of the key findings, Table 3 provides a comparative summary of the key roles, applications, challenges, and representative studies for each domain. The results are presented analytically in the following subsections, organized by technology domain.

Table 3. Overview of technology domains and their role in SDI development.

Technology Domain	Role in SDI	Common Applications	Key Challenges	References
IoT and Sensor Technologies	Event-driven architecture integration with SDIs, real-time fusion of sensor data with SDI layers, integration of cloud–fog–edge architecture into SDI workflows, IoT and GIS integration for real-time emergency services	Real-time monitoring, IoT-based sensing, disaster management, smart mobility, environmental sensing, SDI automation	Standardization of event-handling, semantic interoperability, real-time event integration, sensor reliability and communication delays	[8,16,17,20,22,29,33,34,40,47,51,53,55,56,60,61]
Geospatial Platforms and 3D Mapping	Integration of real-time sensor platforms with SDIs, participatory urban platform development, SDI-based semantic 3D urban modeling, real-time geospatial feedback integration	Environmental monitoring, dynamic mapping, smart city planning, urban governance	Standardization of platforms, integration latency, citizen engagement, platform usability	[8,13,14,19,21,26,30]
Data Infrastructures and Cloud Technologies	Cloud-native or distributed approaches for SDI scaling and interoperability, development of foundational spatial data infrastructures, security framework for distributed SDIs, cloud-based SDI extensions for IoT integration	Land management, disaster response, public health, smart agriculture	Data security, metadata standards, institutional coordination, institutional collaboration	[1,2,3,5,9,16,18,19,31,35,37,38,40,42,44,46,48,58,59,63]
Artificial Intelligence and Analytics	Decision support and spatial reasoning, general AI integration in SDI, model training and data repositories, predictive modeling and classification	General smart governance applications, environmental monitoring, model training and benchmarking, agricultural monitoring	General AI deployment challenges, integration complexity, lack of standardization, semantic challenges	[4,10,11,15,23,24,25,28,36,39,40,41,49,50,51,52,54,61,62]

Table 3. Overview of technology domains and their role in SDI development.

Technology Domain	Role in SDI	Common Applications	Key Challenges	References
IoT and Sensor Technologies	Event-driven architecture integration with SDIs, real-time fusion of sensor data with SDI layers, integration of cloud–fog–edge architecture into SDI workflows, IoT and GIS integration for real-time emergency services	Real-time monitoring, IoT-based sensing, disaster management, smart mobility, environmental sensing, SDI automation	Standardization of event-handling, semantic interoperability, real-time event integration, sensor reliability and communication delays	[8,16,17,20,22,29,33,34,40,47,51,53,55,56,60,61]
Geospatial Platforms and 3D Mapping	Integration of real-time sensor platforms with SDIs, participatory urban platform development, SDI-based semantic 3D urban modeling, real-time geospatial feedback integration	Environmental monitoring, dynamic mapping, smart city planning, urban governance	Standardization of platforms, integration latency, citizen engagement, platform usability	[8,13,14,19,21,26,30]
Data Infrastructures and Cloud Technologies	Cloud-native or distributed approaches for SDI scaling and interoperability, development of foundational spatial data infrastructures, security framework for distributed SDIs, cloud-based SDI extensions for IoT integration	Land management, disaster response, public health, smart agriculture	Data security, metadata standards, institutional coordination, institutional collaboration	[1,2,3,5,9,16,18,19,31,35,37,38,40,42,44,46,48,58,59,63]
Artificial Intelligence and Analytics	Decision support and spatial reasoning, general AI integration in SDI, model training and data repositories, predictive modeling and classification	General smart governance applications, environmental monitoring, model training and benchmarking, agricultural monitoring	General AI deployment challenges, integration complexity, lack of standardization, semantic challenges	[4,10,11,15,23,24,25,28,36,39,40,41,49,50,51,52,54,61,62]

3.1. IoT and Sensor Technologies

The integration of Internet of Things (IoT) technologies and sensors into Spatial Data Infrastructures (SDIs) is transforming how cities manage the environment, resources, and crises. The shift from static or request/reply communication models to event-driven architectures using IoT technologies is recognized as a critical requirement for supporting smart, environmentally sensitive decision-making systems [8].

A central contribution of IoT to SDIs is its capacity to enable dynamic data integration. Through protocols such as MQTT and its spatial extension GeoMQTT, SDIs are capable of continuously ingesting sensor data from distributed sources in near real time [39]. This continuous streaming architecture supports live geoprocessing and situational responsiveness across sectors like agriculture, disaster response, and transportation. Complementing this, cloud–fog–edge frameworks allow computational tasks to be distributed across layers. At the edge and fog level, real-time analytics and filtering reduce latency and bandwidth usage, while cloud layers ensure scalability, redundancy, and broader analytical capabilities [16,42]. For instance, STROVE demonstrated a 24% increase in diagnostic accuracy and a 55% reduction in response time using this hybrid architecture. RESTful APIs and OGC-compliant microservices further enhance system interoperability and modularity. In parallel, participatory sensing initiatives, conceptualized through the “citizen as sensor” model, provide additional decentralized data flows, particularly useful in monitoring urban dynamics and emergencies [28]. These implementations confirm that IoT-enabled SDIs act not only as data transmission backbones but as intelligent, adaptive infrastructures capable of supporting real-time, multi-layered governance functions.

In agriculture, the case of Turkey highlights the value of sensor integration into rural SDIs. By deploying IoT in forest and agricultural lands, it becomes possible to monitor environmental parameters and support sustainable agriculture and food security programs [41]. Similar architectures, such as GeoPipes, extend the MQTT protocol with spatiotemporal capabilities (GeoMQTT), enabling continuous streams of geospatial data for real-time geoprocessing [39]. The SMACiSYS architecture and its SENSDI subsystem integrate sensors and open-source software (GeoNode version 2.6, QGIS version 2.18 ), offering automated data analysis and support for sustainable natural resource management [14]. During the COVID-19 pandemic, the Cloud–Fog–Edge model supported real-time decision-making, improving diagnosis accuracy by 24% and reducing transmission delays by 55% compared to cloud-only solutions [42].

Interoperability challenges are addressed through standards such as the Open Geospatial Consortium’s Discrete Global Grid Systems (DGGS), which enable multi-thematic data integration at scale [29]. In Newcastle, sensor and weather data analysis revealed the immediate impact of storms on traffic, leading to the development of interactive forecasting tools using GIS and Python [9]. In industrial settings, IoT deployment through the OLS3D model enables sensor data integration into BIM/CAD and CityGML, promoting transparency and ecological sustainability in Industry 4.0 [23]. Structural Health Monitoring (SHM) technologies using IoT improve damage prediction and automated risk assessment in infrastructure, while the concept of the “citizen as sensor” opens new avenues for public participation [28].

In disaster management, combining data from LoRa, GNSS, RS, social media, and sensors within SDIs enhances the capability for real-time analysis and response [56]. Additionally, IoT technologies support the optimization of prehospital care through geoportals and sensors that deliver critical information to medical teams [48].

The EarlyDike project demonstrates how GeoMQTT and OGC standards strengthen the protection of coastal dikes, connecting with national SDIs and enabling real-time functionality through geoportals [50]. The combination of satellite, in-situ, and airborne sensors with big data and machine learning enhances natural resource monitoring and sustainable governance [60].

Modern geoportals now function as decision-making centers. With long-range wireless communication and modular design, IoT devices become pivotal tools for monitoring and managing critical systems [17]. The continuous stream of sensory data into GIS infrastructures enables dynamic visualization of urban environments and supports improved decision-making [61]. Finally, the use of LoRaWAN and OLAP warehouses in integrated geoportals enhances forecasting and transparency in socio-environmental systems, highlighting the significance of IoT-SDI combinations for sustainable local governance [15].

3.2. Geospatial Platforms and 3D Mapping

Geospatial platforms and 3D mapping technologies are critical pillars for sustainable and intelligent urban governance. They act as integration hubs, combining data from infrastructure, environment, and administrative systems to support comprehensive and transparent local decision-making. However, while spatiotemporal data are continuously generated through IoT sensors, their utilization remains limited due to fragmented systems and a lack of interoperability—especially for citizens and non-expert users [31].

The transition from 2D to 3D spatial data infrastructures (SDIs) is considered pivotal. In the case of Oman, it was shown that existing 2D infrastructures are inadequate to support complex 3D/4D spatiotemporal applications, highlighting the need for institutional readiness, high-resolution data processing standards, and specialized 3D modeling workshops [45]. Similar needs are observed in European cities, where models like the Smart District Data Infrastructure (SDDI) integrate CityGML-based 3D models with IoT sensors, enabling crowd simulations, building energy monitoring, and event management through OGC standards [37].

Semantic 3D city models further enhance SDIs by enabling interoperability between heterogeneous datasets, BIM models, and temporal observations. The conversion of BIM into CityGML-compatible models using extensions such as the Dynamizer ADE allows for energy simulations within high-detail LoD4 3D models, with visualization supported via ArcGIS Pro (version 2.6) and RESTful APIs. Although challenges remain in integrating real-time sensor data, upcoming CityGML versions are expected to natively support dynamic data streams [24].

Geoportals provide intuitive environments that promote accessibility and transparency. The case of the Bosnia and Herzegovina Biomass Atlas illustrates how such platforms can connect operational databases with spatial data to support natural resource management and investment decisions [35]. Moreover, in cities like Paris and London, geoportals are enhanced with dashboards and domain-specific extensions that include utility infrastructure, offering a holistic view of urban function [37].

In the maritime and port sectors, the integration of GIS and SDIs is seen as vital for risk analysis, sustainable planning, and operational efficiency. Despite an 8.59% annual growth rate in related publications, significant gaps remain in areas such as interoperability and the development of Marine SDIs. The hierarchical visualization of data, the application of machine learning, and the user-friendliness of digital platforms across diverse stakeholder groups from port authorities to the general public are key challenges moving forward [19].

The integration of IoT networks with geoportals in metageosystem architectures offers a modular, low-cost framework for geospatial governance. Technologies such as LoRaWAN and the use of geoportals as operational decision-support systems enhance spatial intelligence in crisis management, infrastructure planning, and efficient resource allocation [50].

Altogether, geospatial platforms and 3D mapping technologies are radically transforming how cities are planned, monitored, and managed. From port systems to energy grids, and from climate resilience to participatory governance, they represent critical tools for smart, inclusive, and adaptive urban development.

3.3. Data Infrastructures and Cloud Technologies

The integration of Spatial Data Infrastructures (SDIs) with cloud technologies represents a critical foundation for enabling scalable, interoperable, and resilient smart city ecosystems. The evolution of geospatial infrastructures, from early GIS systems to contemporary visions such as the “Digital Earth,” highlights the need for a bold, forward-looking paradigm that incorporates open standards, data fusion, and provenance tracking within a cohesive data ecosystem [2].

Modern smart city requirements necessitate the incorporation of heterogeneous data sources through cloud-based architectures. Approaches such as the Smart District Data Infrastructure (SDDI), which relies on Open Geospatial Consortium (OGC) standards, demonstrate how IoT sensors, 3D city models, and meteorological data can be securely and semantically integrated [6,16]. The use of protocols like OAuth2, SAML, and OpenID Connect ensures GDPR compliance and user privacy, while RESTful APIs and publish–subscribe models facilitate interoperability and real-time communication.

The adoption of data cubes and Discrete Global Grid Systems (DGGS) further enhances the ability to manage and analyze large-scale geospatial data efficiently. These architectures support seamless integration of raster and vector data, foster machine learning applications, and enable advanced multi-level visualization, positioning data cubes as essential infrastructure for smart city analytics [3]. Similarly, the Urban Analytics Data Infrastructure (UADI) leverages open standards and ontology mapping to produce reliable, interoperable indicators, particularly for transport and sustainability metrics [7].

Cloud-based SDIs are particularly effective in risk management and emergency response. In China, the Real-time Disaster Decision Support System (RDDSS) integrates SDI and Earth Observation (EO) data to produce automated flood and typhoon impact assessments in under two hours, yielding substantial cost and time savings [12]. In agriculture, SDIs combined with FAIR (Findable, Accessible, Interoperable, Reusable) services and machine learning support food security and climate resilience [55].

Metadata management is also central to the efficiency of cloud-enabled infrastructures. An automated metadata system piloted by the Australian Urban Infrastructure Network (AURIN) provides continuous harvesting and updating capabilities, addressing inconsistencies between data and metadata while supporting federated multi-actor environments [4].

The rethinking of SDIs as spatial data supply chains through the Web reflects an emerging paradigm of user-centered data provision. Semantic Web technologies, spatial data provenance, and demand-driven architectures are expected to transform how data are discovered, validated, and applied [5]. Similarly, cloud-native SDIs like CloudGanga in India illustrate how environmental applications at scale can be supported using open-source platforms and machine learning tools [43].

National and corporate-level experiences also underscore the importance of governance and architectural clarity. Bangladesh’s NSDI project, despite substantial uptake of geospatial data, suffers from a lack of standardization and institutional coordination, highlighting the need for legal and organizational frameworks [20]. The SDI-Cemig initiative in Brazil applies formal architectural models to improve internal data discovery and reuse, demonstrating the adaptability of structured SDI design across contexts [30].

Innovations in metadata quality assessment further strengthen decision-making capabilities. A computational framework based on Hypergraphs and Topic Maps (HXTM) enables visual and quantitative evaluation of metadata completeness, using weighted algorithms validated against international standards such as INSPIRE and FGDC [13].

Finally, ecosystem-based approaches to SDI architecture suggest that future infrastructures must be collaborative, flexible, and self-sustaining. Agent-based simulations confirm the viability of modular and distributed models for digital spatial ecosystems [21]. In resource-constrained settings like Africa, cloud-native solutions built on open-source technologies (e.g., PostGIS, Docker, Kubernetes) and open data portals demonstrate strong potential for cost-effective, scalable, and innovation-driven geospatial services [27].

3.4. Artificial Intelligence and Analytics

The integration of Artificial Intelligence (AI), Machine Learning (ML), and advanced analytics into Spatial Data Infrastructures (SDIs) is transforming how smart cities collect, process, and utilize geospatial information to support evidence-based decision-making [1,10,46]. These technologies enhance capabilities in data analysis, forecasting, citizen engagement, resilience, and sustainability planning.

The application of FAIR (Findable, Accessible, Interoperable, Reusable) data principles to geospatial infrastructures tailored for AI enables standardized data access, documentation, and interoperability, forming a foundation for embedding AI into spatial computing environments [1]. Architectures that organize and store artificial neural networks (ANNs) within SDIs allow for efficient model selection and deployment for diverse spatial analytics tasks [44].

Computer vision techniques applied to richly annotated street view imagery, such as in the Global Streetscapes dataset, support high-quality benchmarking and urban environment analysis [53]. Similarly, the refinement of bare-earth elevation models using machine learning, as demonstrated by FABDEM, contributes to accurate terrain representation for environmental simulations and risk analysis [54].

AI is also used to power virtual assistants that enable citizens to access spatial information using natural language interfaces. These systems enhance participatory governance, although their reliance on simple vector data and the underutilization of complex sources like LiDAR or knowledge graphs reveal gaps in interoperability and sophistication [18,26].

For indoor spatial mapping, innovative approaches using BiLSTM networks and Bluetooth localization enable high-accuracy 3D cadastral data collection without architectural blueprints, reducing cost and fieldwork time [34]. Similarly, semantic methods based on ontologies and probabilistic reasoning support explainable landslide susceptibility mapping, leveraging interoperability standards such as INSPIRE [25].

Cloud-based platforms such as DIONE integrate AI and Earth Observation to monitor agricultural compliance with EU policy, leveraging open standards and Software-as-a-Service (SaaS) architecture to ensure scalability and semantic enrichment [22]. In national contexts, the systemic adoption of GeoAI, such as in Indonesia, enables data integration and redundancy reduction through initiatives like “One Map – One Data” [46].

Visual programming environments further support model transparency and reusability, making deep learning development more accessible to non-specialists [47]. In cases where theoretical understanding is limited, data mining techniques provide inductive solutions that support complex environmental or territorial planning through historical pattern recognition [58].

AI is also applied in heritage conservation platforms such as Odyssey SDI, which integrates ML, LiDAR, and GIS to manage large archaeological datasets and support automated feature detection [59].

In urban mobility, techniques such as Extreme Gradient Boosting (XGB) and SHAP value interpretation contribute to traffic volume prediction and infrastructure planning by revealing key spatial drivers [11]. Likewise, the identification of traffic blackspots through clustering and severity-based indicators provides actionable insights for road safety policy [52].

In agriculture, AI-driven early yield prediction models using vegetation indices and satellite imagery, such as those based on SVM and GEE, support investment planning and sustainability in olive oil production [15]. Broader applications of AI and Google Earth Engine in Latin America and the Caribbean highlight the importance of interoperability and international cooperation for tasks like crop mapping and deforestation monitoring [63].

Furthermore, a critical evolution observed across the reviewed studies is the transformation of SDIs into semantically enriched, AI-powered knowledge systems. Artificial Intelligence contributes significantly to the development of spatial data knowledge graphs, which enable the semantic organization, linkage, and reuse of heterogeneous datasets across domains. For example, Yue et al. [10] propose the AI Cube framework, a dynamic architecture that integrates GeoAI functions within spatial data cubes. It supports standard-compliant model registration, task-model matching, and real-time processing, thus laying the foundation for knowledge-centric SDIs.

Similarly, Roberti et al. [25] demonstrate how AI-based landslide susceptibility models can incorporate semantic transparency and ontological reasoning using INSPIRE interoperability standards. Their approach highlights the role of explainability in increasing institutional trust and promoting alignment with national geospatial strategies.

Moreover, Granell et al. [18] provide a systematic review on the use of AI in virtual geospatial assistants, showing how Explainable AI (XAI) techniques and natural language interfaces allow citizens to interact more meaningfully with spatial data. Their work emphasizes the need for deeper semantic integration and user-friendly AI tools in SDIs to foster inclusivity and trust.

Together, these studies illustrate that AI is not only a technical enhancer of SDIs but a structural enabler of knowledge-driven, transparent, and participatory governance. Through ontologies, XAI, and dynamic model orchestration, AI transforms SDIs into the intelligent backbones of smart cities.

Lastly, the AI Cube framework enables real-time integration of GeoAI functions within spatial data cubes, offering distributed processing, standard-based model registration, and automated task-model matching for dynamic and scalable analysis [10].

Together, AI and SDIs constitute a dynamic and multidimensional field that significantly expands the potential of smart cities across key dimensions such as governance, mobility, health, agriculture, environmental monitoring, and citizen participation. Sustainable advancement, however, depends on regulatory streamlining, shared standards, cross-disciplinary collaboration, and the institutional adoption of interoperable and explainable AI infrastructures.

4. Discussion

This review explored the integration of digital technologies into spatial data infrastructures (SDI), focusing on four key areas: Internet of Things (IoT), geospatial platforms and 3D mapping, cloud infrastructures and artificial intelligence (AI). In all these areas, SDIs are increasingly transforming from passive data repositories to active platforms for real-time decision making, participatory planning and responsive urban governance. These technologies are gradually transforming geospatial ecosystems and urban governance, providing novel mechanisms for informed decision-making, participatory urban planning and the development of smart services at the local level.

4.1. IoT and the Development of Event-Driven SDIs

The results show that the Internet of Things (IoT) and sensor networks significantly enhance SDIs by enabling a continuous flow of data in real time. Applications include hazard monitoring, precision agriculture and environmental assessment, as shown in studies from Turkey and India [14,41]. The shift from static to event-driven architectures enabled by protocols such as MQTT and GeoMQTT [39] has allowed SDIs to ingest streaming geospatial data from edge devices, supported by fog and cloud layers [16,42]. A recurring finding from the reviewed studies is that IoT technologies significantly enhance the dynamic integration capabilities of SDIs by enabling continuous, event-based data ingestion from distributed sources. In contrast to traditional static architectures, event-driven SDIs utilize lightweight protocols such as MQTT and GeoMQTT to support the real-time streaming of geospatial data from sensors and edge devices [39]. The design of edge–fog–cloud frameworks enables the intelligent sharing of computational workloads, while distributed cloud infrastructures ensure scalability, storage and deep learning-based processing [16,42].

Participatory data collection, exemplified by the “citizen as sensor” model [28], introduces a civic dimension, although challenges remain in terms of data quality, integration and digital literacy. These decentralized inputs are increasingly merged with institutional sensor networks and standardized via semantic interoperability layers (e.g., OGC and DGGS), creating integrated SDI ecosystems that can support both immediate decision-making and long-term urban planning.

Furthermore, technical heterogeneity and the lack of cross-platform standards continue to hinder integration, despite initiatives such as DGGS and OGC-based solutions. Although protocols such as GeoMQTT and Discrete Global Grid Systems (DGGS) provide pathways to standardized data communication [29,33], the actual implementation varies significantly, as many cities and sectors do not have consistent cross-platform strategies. Building on the real-time sensing capabilities enabled by IoT, the evolution of SDIs into multidimensional spatial platforms marks another significant change.

4.2. 3D Platforms and the Semantic Level of Governance

Spatial data platforms are evolving beyond 2D systems towards 3D and even 4D SDIs. Case studies from Oman and Europe [37,45] show how 3D city models, often based on CityGML and BIM integration via Dynamizer ADEs [24], support dynamic simulations, energy assessments and crowd analysis. Semantic 3D city models, especially those using CityGML and BIM conversion with ADEs such as Dynamizer, bridge the gap between models of the built environment and real-time data [24].

However, full interoperability is still limited, especially when integrating sensor data with legacy geoportals. It is expected that the upcoming versions of CityGML will solve this problem [24], but current practices rely on manual pre-processing or API-based workarounds.

Innovative dashboards in Paris and London [35,37] show a move towards user-centered decision support tools, even if citizen engagement is limited by user interface complexity and lack of knowledge. The biomass atlas of Bosnia and Herzegovina and the dashboards in Paris and London show that geoportals are evolving into user-centered, decision-support environments [35,37], even if citizen engagement is still limited by a lack of digital literacy and user interface complexity.

In the maritime domain, despite the increasing adoption of SDIs for the maritime domain, there is still fragmentation of standards and silos between stakeholders. Even though the use of SDIs in port management is growing by 8.59% annually [19], issues related to interoperability between stakeholders and standards for SDIs in the maritime domain point to a wider structural gap. Low-cost systems such as LoRaWAN [50] show promise for resource-constrained communities, but require governance models that support cross-sector coordination.

Low-cost, modular frameworks such as those based on LoRaWAN offer scalable solutions for underfunded municipalities [50], but require governance models that incentivize cross-sector coordination. While 3D modeling extends the semantic and spatial layers of governance, the effective management of such complex infrastructures increasingly depends on scalable and interoperable cloud systems.

4.3. Cloud Infrastructures and Institutional Constraints

Cloud-based SDIs such as SDDI [6], UADI [7] and RDDSS [12] demonstrate the scalability and interoperability required for emergency response and multi-source integration. However, institutional fragmentation often undermines these benefits. For example, the NSDI project in Bangladesh struggles with legal and organizational hurdles despite technical readiness [20], while Brazil’s SDI-Cemig shows how clear guidelines for data reuse can enable agile operations [30]. Together, the Smart District Data Infrastructure (SDDI) [6], the Urban Analytics Data Infrastructure (UADI) [7] and the Real-time Disaster Decision Support System (RDDSS) [12] illustrate how cloud architectures support the integration of different geospatial data sources and accelerate emergency response. In contrast, the SDI-Cemig project in Brazil shows how a clear architecture and internal protocols for data reuse can increase efficiency and adaptability [30]. It is notable that none of the 62 articles reviewed explicitly address serverless computing or federated learning, despite their relevance to SDI principles. This absence indicates a significant technology and research gap, as both approaches closely align with the basic principles of SDI, especially in terms of flexibility, security, and operational efficiency. Serverless architectures support dynamic resource allocation, ideal for variable spatial workloads such as disaster response, while federated learning enables AI training without centralizing sensitive data, supporting data privacy and regulatory compliance (e.g., GDPR). The ability to train models locally and aggregate them centrally ensures privacy and regulatory compliance, such as GDPR, which is critical in smart environments that manage spatial or citizen data. Therefore, the integration of these technologies is proposed as a strategic future direction for the development and operation of SDI-enabled platforms to improve the scalability, trust, and ethical use of geospatial data in public administration. Beyond infrastructure and scalability, a deeper transformation is underway with SDIs taking on operational and semantic roles within digital governance ecosystems.

4.4. SDIs as Semantic and Operational Infrastructures

A key finding of this research is that SDIs are no longer passive data repositories; they are becoming semantic operating systems for cities. They orchestrate different technologies via standard protocols, ontologies, and real-time processing services [2,16]. Edge-enabled SDIs based on OGC-compatible APIs [16], event-driven platforms using GeoMQTT [8] and smart district systems that integrate real-time flows with CityGML [24] are examples of this transformation. Semantic frameworks, and the embedding of domain knowledge into data infrastructures, improve spatial reasoning and decision making. As Dangermond and Goodchild [2] note, SDIs increasingly serve as infrastructures of meaning, not just data. AI and machine learning are now embedded in these platforms, enabling predictive modeling, automation, and citizen-centric tools such as the AI Cube [10] and spatial virtual assistants [18]. These semantic capabilities also reconfigure governance dynamics and place SDIs at the center of interactions between institutional design and participatory planning.

4.5. Governance Models and the Role of SDIs

SDIs mediate between top-down control and bottom-up participation by shaping data flows, architectural design and institutional dynamics. They serve not only as infrastructure but also as a coordinating framework for smart governance. Chaturvedi et al. [16] show how SDIs support decentralized data collection with centralized reasoning, while Herle [33] and Rieke [8] present event-driven decision-making systems that work in all urban settings. Nevertheless, there are still challenges in administration. Technical interoperability remains constrained by the uneven implementation of standards such as OGC and INSPIRE. Institutional silos hinder cross-sector collaboration and privacy protection remains acute—especially in urban environments where sensitive citizen data are managed. A solid legal and ethical framework is essential to support responsible data handling. Despite this progress, there are still systemic gaps in the global SDI landscape, both geographically and technologically.

4.6. Gaps, Inequalities and Strategic Directions

Two major overarching gaps are evident:

• Geographic inequality—the Global South is underrepresented in both academic discourse and practical implementation. Infrastructure deficits, low digital literacy and funding limitations constrain adoption. The Indonesian initiative “One Map—One Data” [46] is a rare example of the systematic integration of GeoAI in the global South. Broader adoption requires capacity building, open knowledge transfer and investment in local infrastructure.
• Techno-operational inequalities—promising technologies such as semantic web frameworks and explainable AI [25] remain underutilized in real-world systems, and many tools (e.g., visual programming, LiDAR integration) lack standardization or interoperability.

4.7. Towards Inclusive and Adaptive SDI-Driven Governance

Ultimately, this overview confirms that SDI-driven governance is more than just a metaphor; it is a concrete architectural paradigm. By embedding semantics, supporting distributed intelligence and enabling citizen interaction, SDIs will become the backbone and brain of smart urban systems.

However, realizing this vision requires overcoming complex technical (standardization, real-time integration), institutional (governance coordination, legal framework), social (literacy, equity, participation) and ethical (privacy, transparency) challenges. A multidisciplinary, equity-focused approach is essential, one that combines technological innovation with institutional reform and inclusive design. Strengthening cross-sector collaboration, investing in digital infrastructure, and empowering users will be crucial to achieving sustainable, participatory smart governance.

5. Conclusions

This study has shown that the integration of smart city technologies into spatial data infrastructures (SDIs) can significantly improve local governance in key smart city areas such as smart society, smart living, smart mobility and smart health. By analyzing four key technological areas: IoT and sensors, geospatial platforms, cloud infrastructures, and artificial intelligence, the study found that these tools support dynamic and participatory decision-making in real time.

However, the ongoing challenges of interoperability, data harmonization and institutional coordination continue to limit wider adoption. As mentioned above, inconsistent implementation of standards, outdated infrastructures and gaps in citizen participation limit the full potential of these systems.

Overcoming these limitations requires a holistic approach: introducing interoperable frameworks, fostering cross-sector collaboration, developing clear data management guidelines and enabling broad participation. Ultimately, realizing the promise of SDIs for sustainable and resilient cities depends not only on digital innovation but also on institutional maturity and citizen empowerment.

Importantly, this review confirms that SDIs are evolving beyond their original role as static geospatial repositories. Instead, they are increasingly functioning as “urban operating systems” that integrate heterogeneous data via standardized protocols and spatial semantics. In this model, technologies such as IoT, AI and ML are not external add-ons, but embedded layers that extend the SDI’s ability to orchestrate real-time data flows, automate public services and mediate between central and local governance needs.

Additionally, this review underlines that the success of SDI-enabled smart governance depends not only on technological maturity but also on the governance models adopted. The contrast between top-down and bottom-up approaches highlights the complementary strengths and weaknesses. Top-down models offer regulatory coherence and infrastructure standardization but may limit adaptability and civic participation. Conversely, bottom-up models promote local innovation, responsiveness and inclusivity, but are often not integrated into national frameworks. Therefore, hybrid governance strategies combining institutional coordination with decentralized innovation seem essential to realize the full transformative potential of SDIs. Such models enable SDIs to act not only as technical systems but as dynamic orchestrators of urban intelligence that promote both interoperability and democratic engagement.

Although this review underscores the transformative role of SDI-integrated smart technologies, several limitations and assumptions must be acknowledged. First, the analysis was based on literature retrieved from a specific database, potentially excluding relevant studies published in other repositories or non-English sources. Future studies should expand the scope of the literature search to include multiple databases and different publication types to ensure broader representation. In addition, significant geographical inequalities persist, particularly in the Global South, where limited infrastructure, funding and digital literacy continue to constrain SDI deployment. Addressing this requires more comprehensive capacity-building initiatives, and open knowledge transfer mechanisms are needed. Technologically, there are still gaps in the integration of emerging paradigms such as serverless computing and federated learning approaches that align well with SDI principles but are underrepresented in current practice. The standardization of semantic frameworks, LiDAR workflows and user-friendly tools such as visual programming environments also needs to be further developed. Finally, there is a need for more empirical studies that evaluate SDI implementations in real governance environments to assess performance, inclusion and long-term impact.

Ultimately, this review demonstrates that the concept of ’SDI-driven’ governance represents a paradigmatic shift in the way cities harness data and intelligence. SDIs are no longer passive data infrastructures, but are evolving into semantic operating systems for smart cities that integrate heterogeneous data sources via standardized interfaces, open protocols and spatial semantics. Using this architecture, SDIs orchestrate real-time sensor inputs, AI-powered analytics and policy automation to form the digital backbone of smart governance. Platforms such as SDDI and AI Cube are an example of this transformation, with SDIs serving as both data hubs and intelligent facilitators of urban processes. To fully realize this potential, technical standards and AI capabilities need to be further developed, in addition to ethical safeguards, participatory design and inclusive institutional frameworks that ensure SDIs remain transparent, accountable and citizen-centric.