Sage Journals: Discover world-class research

Abstract

Urban Digital Twins (UDTs) have emerged as integrated collections of urban data and urban models aspiring to enhance urban planning and decision-making processes. However, current UDTs often fail to connect siloed disciplines, represent diverse stakeholder views, or adapt to the dynamic nature of planning processes. Realizing UDTs potentials is hindered by these socio-technical challenges, we developed and validated FMU Ontology to address them. FMU Ontology provides a set of semantic representations that (1) promote interoperability and integration across disciplinary data and models, (2) enable developing and using a network of stakeholder-specific UDTs that facilitate engagement and consensus-building, and (3) embed these within planning processes to allow UDTs to adapt as stakeholders’ questions and priorities evolve. Furthermore, we validate the efficacy of FMU Ontology through consistency and competency tests. Lastly, in a case study on strategic urban densification in Eindhoven, the Netherlands, we demonstrate how FMU Ontology enables the adaptive and collaborative use of UDTs, addressing key challenges in urban planning and decision-making.

Keywords

Built environment participatory planning knowledge graph urban digital twin urban decision-making

Introduction

Urban Digital Twin (UDT) is a relatively new concept, and there are still no consensus on its definition (Weil et al., 2023; Wu and Guan, 2024). Nevertheless, the term ‘integrated’ is emphasized across the board with a spectrum of interpretations.

For some, UDTs are an integrated digital representation that provide real-time connection between cities and their digital replica (Ketzler et al., 2020). For others, UDTs are to provide an integrated platform for heterogeneous urban data (Hämäläinen, 2021). Some emphasize the importance of integrating various urban models (Simonsson et al., 2021). Others point to the necessary system integration for interoperability in the larger ecosystem of digital tools (Park and Yang, 2020). Generally, such technical views understand UDTs as a collection of interconnected urban data and models that provide integrated digital representations of the urban environment.

Nevertheless, more recent scholarly works increasingly underscore the overlooked social aspect in the development and use of UDTs (Azadi et al., 2025). Researchers highlight the prevailing technical optimism (Lei et al., 2023) and digital universalism (Charitonidou, 2022), warning that prioritizing technological solutions may overlook essential social and ethical issues, favouring economically efficient technical approaches instead (Calzati, 2023).

The emerging socio-technical perspective highlights key challenges in achieving effective integration of digital representations including UDTs. These include insufficient stakeholder engagement in development and use of UDTs, which limits the integration of diverse stakeholders’ views (Allan et al., 2024; Astarita et al., 2024), and the oversimplification through generic statistical methods that overlook the nuances of local context (Caprari et al., 2022; Wan et al., 2024). Additionally, UDTs often fail to support entire planning processes (Quek et al., 2023) and lack integration across decision-making stages (Chen et al., 2024). Other challenges involve multi-scalar integration (Charitonidou, 2022) and the need to integrate data, information, and knowledge from diverse disciplines (Peldon et al., 2024).

Here, we focus on three of these socio-technical gaps in integrating UDTs within urban planning processes (Azadi et al., 2025). The first gap is the challenge of ensuring that UDTs are interdisciplinarily integrated (II), that is, UDTs can bridge disciplinary silos and facilitate a holistic view of the urban environment by combining various urban models and data from different disciplines (Langenheim et al., 2022; Quek et al., 2023). The second gap is the unaddressed challenge of ensuring that UDTs are consensually contextualized (CC), that is, the views of different stakeholders about the shared urban environment are adequately represented to facilitate better communication and participatory decision-making toward consensus (Nochta et al., 2020; Oti-Sarpong et al., 2022; Ye et al., 2022; Yossef Ravid and Aharon-Gutman, 2023). Current UDTs often struggle with adequately representing the diverse and sometimes conflicting interests of stakeholders, leading to a gap in their ability to support inclusive and collaborative planning processes. The last gap is the unaddressed challenge of ensuring that UDTs are procedurally operationalized (PO), that is, they are embedded within urban decision-making or planning processes and can provide insight for various questions that arise in different phases of these processes (Calzati, 2023; Chen et al., 2024; Pedersen et al., 2021; Wang, 2021). Addressing these three gaps would bring us to an Augmented Urban Planning (AUP) process (Azadi et al., 2025): a planning process entwined with integrated, contextualized, and operationalized representations. The main purpose of FMU Ontology is an attempt to achieve the ideals of AUP by formally integrating urban digital twinning and planning processes in three phases of Formulating, Modelling, and Utilizing.

Within semantic technologies (ST), Ontologies are schemas behind Knowledge Graphs (KGs), specifying how data, information, and knowledge can be represented (Gruber, 1995). They can encapsulate conceptual and methodological frameworks into formal representations (Hoang et al., 2013). In this article we put forth an Ontology that structures the necessary semantic representations for three main digitally enabled planning practices of Formulating planning questions, Modelling UDTs to provide relevant insights for those planning questions, and Utilizing those specific UDTs for exploring and evaluating planning scenario alternatives for the original questions; hence we call it Formulating-Modelling-Utilizing (FMU) Ontology. FMU Ontology provides a semantic structure to address socio-technical complexities organized in four layers.

To address II, it provides (1) factor-specific representations to encapsulate urban data and (2) relation-specific representations to capture urban models. These layers provide a standardized set of digital modules that can be easily curated, configured and interpreted in the other layers. These concepts are based on common open standards.

To address CC, it provides (3) stakeholder-specific representations to incorporate the planning concerns, questions, alternatives, digital representations, and views of stakeholders, for example, stakeholder-specific UDTs. This layer is built on top of the previous layer as its concepts are defined as configuration of previous layers.

To address PO, it provides (4) process-specific representations to track the changes of the concepts from previous layers in complex dynamics of urban planning processes.

In Section Background we substantiate the socio-technical gaps and review how ST have contributed to UDTs and urban planning. Section Methodology outlines the methods used in development, validation, and demonstration of FMU Ontology. Section FMU Ontology goes into the details of the semantic concepts and semantic relations provided in each of the four layers of FMU Ontology. Section Validation & Demonstration explains the context of the test case, presents the result of validation tests, and goes through three demonstrative examples from participatory workshops for II, CC, and PO. Lastly, Section Discussion & Conclusion discusses the contributions and limitations of FMU Ontology.

Background

The history of using digital technologies in urban planning starts from the modelling of urban systems, GIS development, spatial planning and decision support systems (PSS and SDSS), City Information Modelling (CIM) tools, and UDTs mark the latest phase. Early work by Forrester (1970) introduced dynamic urban models aiming to digitally represent the dynamics of urban systems, later expanded by Batty (1971). GIS technologies enabled 3D digital representation, gathering, and analysis of urban environments (Tomlinson, 1969) which provided geospatial information in a structured manner for different phases of urban planning. PSS and SDSS aimed to bring urban models and GIS closer and provide tools and dashboards for analysis and visualization to support urban planning decision-making processes (Geertman and Stillwell, 2004; Stead, 2021). CIM integrates diverse data sources into standardized specifications (Gil, 2020) while benefiting from semantic technologies to resolve interoperability challenges (Yao et al., 2018). Most recently, UDTs aim to provide integrated digital representations consisting of heterogeneous urban data and models to facilitate urban planning (Von Richthofen et al., 2022; Batty, 2018).

This section provides the necessary background for FMU Ontology from two perspectives: firstly the socio-technical gaps urban digital twinning (UDTing) and secondly the contributions of semantic technologies to UDTing and urban planning.

Socio-technical view of UDTs

Earlier works on UDTs primarily focused on technical integration. Examples of such technical integration include creating realistic 3D models of urban environments, establishing real-time connections between virtual and real worlds, and achieving technical scalability (Al-Sehrawy et al., 2021; Ketzler et al., 2020; Marcucci et al., 2020). Accordingly, there has been an abundance of contributions in technical integration made in the last 8 years (Wu and Guan, 2024). Nevertheless, scholars are increasingly sceptic about tangible contributions of UDTs to urban planning and decision-making (Allan et al., 2024; Azadi et al., 2025; Weil et al., 2023).

More recent studies highlight the limitations of purely technical developments and propose ambitions for UDTs centred on socio-technical integration (Adade and De Vries, 2024; Wu and Guan, 2024; Yossef Ravid and Aharon-Gutman, 2022). These authors argue that UDTs should be transparent (Wang, 2021), ensure cybersecurity (Yang and Kim, 2021), provide privacy protection for citizens (Pang et al., 2021), address legal frameworks (Raes et al., 2021), incorporate in-depth local knowledge (Nochta et al., 2020), recognise citizens as individuals with agency (Wang, 2021), enable consensus among stakeholders (Pultrone, 2023), engagement throughout the planning process (Tzachor et al., 2022), and facilitate interdisciplinary collaborations (Lagap and Ghaffarian, 2024).

In the following, we focus on three particular socio-technical challenges of integrating UDTs within urban planning processes: interdisciplinary integration (II), consensual contextualization (CC), and procedural operationalization (PO) (Azadi et al., 2025). For each challenge, we review the conceptual underpinnings and latest UDT developments.

Gap of interdisciplinary integration

Disciplinary silos are not only a challenge of UDTs and city-related digital technologies (Bibri, 2021), but also an old fundamental issue in structuring any form of urban theory and planning practice (Harding and Blokland, 2014). Each urban discipline, such as Transportation Engineering and Planning, Urban Energy Management, or Environmental Planning, has traditionally developed and used theoretical frameworks, models, indicators, assessment methods, and other tools that are not necessarily commensurable with those of other disciplines. These intricate discrepancies between urban disciplines are an obstacle to developing and using integrated information systems such as UDTs (Hofmeister et al., 2024a; Wu and Guan, 2024). Nevertheless, UDTs’ ambition of providing holistic representations of urban environment spotlights this issue.

Various UDTs have managed to contribute here by attempting interdisciplinary integration for specific problems or use cases. A notable example is discipline agnostic structuring of data to ensure model results can be used for answering interdisciplinary questions (Langenheim et al., 2022). Another example is the work of Hofmeister et al. (2024b), which combines the dynamic control of district heating networks with embedded emission models for air pollution.

To address interdisciplinary integration, UDTs should bridge long-standing disciplinary silos in urban planning by providing modular and standard representations that can be dynamically configured for different intersections of disciplines. Establishing such standards to ensure interoperability and communicability requires explicit representation of the assumptions and scope of urban data and models (Quek et al., 2023).

Gap of consensual contextualization

Contextualization is a long-standing challenge of using quantitative models in urban planning. The core issue is the common assumption that knowledge, insights, and evidence about urban environments are easily transferable, thus we can inject them into planning processes irrespective of the social, organizational, and political complexities of context (Innes, 1995). Avoiding such reductionist simplifications demands us to recognise the social nature of knowledge, that is, any piece of information that describes or prescribes something in urban environments needs to be viewed and interpreted through local stakeholders’ lens. Therefore, stakeholders’ agency, their active participation, and ultimately, their consensus are crucial for the credibility of urban plans and decisions (Bai et al., 2020; Nourian et al., 2024; Thompson, 2022; Wan et al., 2024). However, it is not only the plans and decisions that are value-laden, the digital representations that are used in the process of reaching those plans and decisions also carry implicit values (Alexander, 2000; Davoudi, 2006; Throgmorton, 1996). In other words, the development and use of digital representations in urban planning processes are intertwined by sociopolitical questions that often remain obscure to developers of these digital representations Oti-Sarpong et al. (2022). Therefore, stakeholders’ involvement in the development and use of UDTs is needed to ensure their consensual contextualization.

Recent works introduce the notion of human-centric or citizen-centric UDTs to emphasize the potential of citizen science and citizen-sourced data for UDTs (Abdeen et al., 2023; Ye et al., 2022). Yet, they overlook stakeholders as active participants in planning processes. This is reflected in the theoretical contributions of Oti-Sarpong et al. (2022), which emphasize understanding UDTs and their larger process of delivery embedded within a multi-dimensional socio-technical network. For example, Nochta et al. (2020) developed a UDT for Cambridge EV policies, emphasizing stakeholders’ active participation in prioritizing and interpreting the relevant information, documents, and objectives that will be incorporated in the UDT. Similarly, Pedersen et al. (2021) developed a user-focused urban water UDT for multiple stakeholders, and considering their unique values and criteria.

We take a step further and emphasize explicit stakeholder-specific representations of planning questions, alternative intervention plans, and UDT to avoid a universalist approach (Charitonidou, 2022). Stakeholders have their own expertise, interests, and controls, which make their planning questions unique. These questions demand specific urban models and data to be included in their UDT to provide appropriate insights. Therefore, consensual contextualization demands a set of stakeholder-specific digital representations: (1) unique planning questions and (2) UDT configuration. These will systematically record, compare, and trace stakeholders's views throughout their active participation, enabling them to curate and interpret data and models, while making the implicit notion of consensus more explicit.

Consensual contextualization of digital representations remains a persistent gap as digital models often overlook the diverse social, political, and organizational complexities of urban environments. Particularly, UDTs lack stakeholder-specific adaptability to address unique planning questions, models, and configurations, as well as a systematic approach to compare alternatives, trace preferences, and build consensus in decision-making.

Gap of procedural operationalization

Earlier works emphasized the potential of UDTs to enhance urban planning, management, and governance process (Al-Sehrawy et al., 2021; Deng et al., 2021). Yet, tangible contributions with practical values for urban planning processes by UDTs have been scarce (Allan et al., 2024; Lei et al., 2023).

Researchers highlight various complexities of urban planning processes that have been neglected in the development and use of UDTs; namely implicit power dynamics (Calzati, 2023), complicated organizational structures (Adeagbo et al., 2024), opaque legal frameworks (Bibri et al., 2024), and sociopolitical conflicts (Wan et al., 2024). All of these, along with the inherent wickedness of the urban problems (Quek et al., 2023) result in complex iterations and dynamics within the process. This entails that the core planning question of all the stakeholders is dynamic.

Such dynamic processes, combined with the rigid and static nature of digital technologies, often limit the usability of UDTs. Therefore, a static UDT might answer one or more fixated planning questions in isolation. Nevertheless, when embedded in actual planning processes, the evolving question of stakeholders can easily evolve beyond the predefined scope of any static UDT.

Although many of the underlying sociopolitical complexities might require systemic remedies, flexibility and adaptability of UDTs can improve their usability for stakeholders with ever-changing questions. In this vein, Oti-Sarpong et al. (2022) emphasizes the importance of understanding UDTs as embedded in the delivery process (development and use). Pedersen et al. (2021) offer an outstanding example with the notion of living and prototyping UDTs, which engages stakeholders in ongoing UDT development to adapt to their evolving concerns. This view allows us to define procedural integration as bridging urban digital twinning and urban planning processes with explicit digital representations.

Procedural operationalization of UDTs entails their dynamic integration into urban planning processes. A key challenge is the lack of a systematic approach to represent and adapt to the dynamic iterations of planning activities and stakeholder needs, ensuring UDTs remain relevant and useful throughout the planning cycle.

Semantic technologies

Semantic technologies (ST) provide semantic representations of real entities that facilitate the communication and deduction of knowledge about those entities (Davis et al., 1993; Levesque, 1986). An important category of these technologies is Knowledge Graphs (KGs) that represent relationships among data as triples of (subject, predicate, object), which can be visualized as edges (semantic relationships) between nodes (data) in a mathematical graph. This approach allows for the aggregation of information into networks of interconnected representations (Ehrlinger and Wöß, 2016).

Computational Ontologies (hereafter ontologies) serve as the schemas behind KGs by establishing shared vocabulary, specifying possible relationships, and setting rules for modifying KGs. Described as representational artefacts that designate classes and certain relations among them (Arp et al., 2015), ontologies are essentially ‘an explicit specification of a conceptualization’ (Gruber, 1993). They are particularly useful for encapsulating conceptual frameworks and methodologies into formal representations (Hoang et al., 2013), facilitating interoperability and communication between humans and digital systems (Gruber, 1995). Therefore, ontologies offer significant potential for addressing the socio-technical complexities faced by digital representations such as UDTs.

Semantic technologies in UDTing

ST have a substantial track record of facilitating processing, structuring, and integrating data in the built environment (Pauwels et al., 2017). As such, ontologies and KGs have been suggested as the underlying system for UDTs to benefit from their potential for data integration and interoperability (Boje et al., 2020; Shi et al., 2023). Some suggest using ST for automated reasoning combined with other computational models in UDTs (Austin et al., 2020; Yu et al., 2021). Others propose dynamic KGs as the main UDT architecture combined with a set of computational agents executing all the computational models and methods (Akroyd et al., 2021).

Beside technical advancement, there are several UDT implementations that have extended data and model integration to achieve interdisciplinary integration. A notable example is Hofmeister et al. (2024b) which uses a dynamic graph to dynamically control a district heating network while simulating the air pollutant dispersion. Another instance is Hofmeister et al. (2024a) which demonstrates a UDT with the capability of assessing flooding risk and evaluating its impact on the population of Singapore.

In addition to bridging disciplinary gaps, UDTs need to facilitate stakeholder engagement (Nochta et al., 2020) and be integrated into broader processes (Oti-Sarpong et al., 2022). However, we could not find any semantic technology that could facilitate stakeholder engagement throughout the process of UDTing – that is, the development, use, and adaptation of digital twins. Moreover, a gap exists in ontologies that integrate UDTing with urban planning processes.

Semantic technologies in urban planning

Beyond the case of UDTs, ST have a history of contributing to urban planning and urban sciences (von Richthofen et al., 2022). Von Richthofen et al. (2022) structure these contributions into four types of planning meta-practices: Representational, for representing urban systems; Evaluative, for assessing properties of urban environment according to a set of goals; Projective, for specifying alternative visions for urban systems; and Synthetical, for managing and integrating data, information and knowledge pertaining to urban environments.

ST had a role in the development and management of 3D city models (Billen et al., 2012, 2014), and the establishment of necessary standards such as CityGML (Kolbe et al., 2021; Kutzner et al., 2020) and DCAT V3 (Albertoni et al., 2024) that facilitate data sharing, querying, collating, and interpretation. CityGML has enabled the creation of interoperable 3D geo-databases (Stadler et al., 2009), which are essential for managing, analyzing, and visualizing urban data (Yao et al., 2018). Additionally, they have enabled 3D city versioning and historical data management (Chaturvedi et al., 2016), and have been applied across various 3D city model applications (Biljecki et al., 2015). Moreover, ST have extensively contributed to the integration of sensor data and the Internet of Things in smart city technologies (Chaturvedi and Kolbe, 2016; Espinoza-Arias et al., 2018; Guo et al., 2018).

They have also been useful for integrating heterogeneous data in specific domains. For example, integrating Building Information Modelling (BIM) and Geographic Information Systems (GIS) is an active research area (Donkers et al., 2015; Wang et al., 2019); semantic graph databases have been utilized to enhance this integration (Hor et al., 2018), and CityGML has been extended for improved interoperability with IFC standards (Biljecki et al., 2021; Nguyen and Kolbe, 2020; Yao et al., 2018). In the domain of mobility, ST enabled the integration of transport data (Tempelmeier et al., 2019), interpretation of transport simulation outputs (Grisiute et al., 2022), as well as integration of land-use data, exemplified by the Kadaster knowledge graph (Ronzhin et al., 2019) and CityGML extension for civil infrastructures (Kumar et al., 2019).

Systematic data integration enabled by ST has unlocked the potential for more comprehensive assessments in the built environment. For example, integrating IFC and CityGML enabled microclimate analysis in neighbourhood development (Kardinal Jusuf et al., 2017). Ontology-based approaches have been used for sustainability assessments (Konys, 2018), carbon emission evaluation (Madrazo et al., 2012), econometric inference in the housing market (Malczewski and Jelokhani-Niaraki, 2012), and for city-scale building energy performance assessments (Corry et al., 2015). Additionally, more general frameworks have been developed for structuring and representing city indicators (Fox, 2015; Santos et al., 2017) and location-aware geographic question answering (Mai et al., 2020).

ST have some contributions to urban planning by providing systematic representations of planning concepts. For instance, ontologies for land-use planning have been developed to formalize and structure planning information (Montenegro et al., 2012), and system-theoretical representations offer frameworks for modelling urban systems Caglioni and Rabino (2007). More recent contributions focus on interdisciplinary data integration (Psyllidis, 2015), cross-domain scenario assessments (Eibeck et al., 2019), and scenario analysis within knowledge graphs (Eibeck et al., 2020).

Moreover, ST have been used to support complex planning practices as well. For example, Bojórquez-Tapia et al. (2011) use a goal-decision structure to address socio-environmental challenges, and Mostafa and El-Gohary (2015) propose stakeholder-conscious representations to aid planning and design discussions. Urban decision ontologies have been suggested to formalize planning decisions (Hopkins, 2001; Kaza and Hopkins, 2007). Additionally, multi-criteria decision analysis based on rule-based reasoning (Lee et al., 2019) and collaborative spatial decision analysis (Jelokhani-Niaraki, 2018) demonstrate the application of ST in complex decision-making processes. However, they fall short of providing a customized representation for each of the stakeholders involved.

Explicit representation of system configurations enhances the reusability and reproducibility of UDTs (Cranefield and Purvis, 1999; Guarino, 1998). Examples include the semantic representation of support system components like GIS and Multi-Criteria Decision Analysis (MCDA) (Jelokhani-Niaraki et al., 2018). Such a semantic framework for UDTs can enable adaptive stakeholder-specific UDTs.

Furthermore, some studies aim to explicitly represent the decision-making process itself. Wu et al. (2021) present a semantic model that depicts the project process as a set of interdependent actions, where each action can influence subsequent possibilities and specifications. Other works model the sequential logic of actions based on their dependencies to facilitate process navigation (El-Gohary and El-Diraby, 2010; Hepp and Roman, 2007). Neither of such procedural representations has been applied to urban planning processes.

Methodology

We propose that explicit yet transient digital representations can be a middle ground that can facilitate the ideals of Augmented Urban Planning (AUP) (Azadi et al., 2023): a planning process entwined with integrated, contextualized, and operationalized representations. The main purpose of FMU Ontology is an attempt to achieve the ideals of AUP through such explicit yet adaptive and stakeholder-specific digital representations.

Figure 1 illustrates the methodical flowchart of this paper.

Figure 1.

Flowchart of the methodology used for this Article. FMU Conceptual framework, competency questions, and details of FMU Ontology are included in Supplemental Materials.

Formalization

Formalization entails the explicit specification of concepts and their relations as well as the implementation of them in ST (Grau et al., 2008). Starting from a primary axiomatic draft of the FMU conceptual framework, we have iteratively expanded and edited FMU Ontology along the process of implementation. This iterative process was based on the Cyclic Modelling Process and Problem-Solving Methods as described by Studer et al. (1998). During this process, we have followed a demonstrative planning case. Using drafts of FMU Ontology, we iteratively constructed representative Knowledge Graphs (KGs) to test the inferential and expressive capabilities of ontology.

In the initial phases, we used HermiT reasoning engine (Glimm et al., 2010) based on its efficiency in complex ontologies. Nevertheless, as the ontology and test case KG grew in size, it was increasingly computationally expensive to make adjustments to the ontology. This is particularly important as we followed an iterative development process where we continuously refined and tested the ontology. Accordingly, we switched to the Pellet reasoning engine (Sirin et al., 2007) as it provides incremental reasoning that makes loading adjustment to large ontology and KGs more efficient.

We implemented FMU Ontology using the Python package of Owlready2 (Jean-Baptiste, 2021). This allowed us to provide a Python module including Python objects coupled with the ontology concepts, in addition to the customary OWL and TTL files: FMU Ontology repository. Reasoning rules were stated in SWRL (Semantic Web Rule Language), and queries were specified in SPARQL (SPARQL Protocol and RDF Query Language).

To ensure standardization and interoperability of the factor-specific and relation-specific semantic layers, we defined our concepts based on OntoCityGML to provide a bridge to CityGML standards for generic urban data Chadzynski et al. (2021). We also incorporated DCAT-AP-DONL, which is a DCAT Application Profile extension (European-Union, 2024) of Data Catalogue Vocabulary (DCAT) (Albertoni et al., 2024) for data.overheid.nl (De Nederlandse Overheid, 2022) used systematically by the Dutch government. This was important as it allow us to ensure interoperability with public data infrastructure in the Netherlands. Moreover, DCAT relies on many basic concepts from the Dublin Core terms (DCMI) (Baker, 2012) that are also integrated in FMU. By extension, it should be relatively easy to incorporate other standardized data infrastructures.

Structuring competency questions

To test the competency of FMU Ontology in addressing three socio-technical challenges. We reviewed the literature on these challenges and derived 15 competency questions (CQs) accordingly: four Interdisciplinary Integration (II), six Consensual Contextualization (CC), and five Procedural Operationalization (PO) questions. If FMU Ontology can provide sufficient answers to all of these 15 questions, we can claim its efficacy.

Section Competency Questions in Supplemental Materials provides a detailed description of how CQs are derived from socio-technical gaps. Furthermore, Table S4 in Supplementary Materials outlines the relation of CQs with semantic concepts and relations.

Validation

For validation of the efficacy of FMU Ontology, we assess its consistency and competency (Gruninger, 1995; Uschold and Gruninger, 1996). Consistency test assesses internal consistency of the axioms of ontology in information classification (Studer et al., 1998; Uschold and Gruninger, 1996). The competency test assesses whether the ontology and its subsequent knowledge graphs are capable of answering the questions that they were designed to answer; this ensures that the ontology delivers the functionalities that it promises (Gruninger, 1995).

After distilling the CQs, they are translated into SPARQL queries and implemented as Python test modules based on Owlready2 Python package (Jean-Baptiste, 2021). This allowed us to establish a continuous development and continuous integration pipeline which facilitated our iterative approach in development of the FMU Ontology.

This agile pipeline was crucial as we conducted the competency test intermittently during the demo case process. The Python test module would iteratively run the tests and compare query results with predefined outputs to verify correctness. These predefined outputs used for validation were specified collaboratively with expert participants of the workshop series.

Demonstration

To illustrate the capability of FMU Ontology, we use a demonstrative case study of strategic planning for the densification of Eindhoven, the Netherlands. For this case study, we conducted a series of participatory UDTing workshops with 12 expert stakeholders. The outcome of the workshops is a KG that represents the problem formulations, the specification of the UDT system, the investigated alternative intervention scenarios, and most importantly the planning process. The outline and results of the case study are presented in Section Validation & Demonstration.

FMU ontology

To achieve an Augmented Urban Planning (AUP) process, it is crucial to address the three socio-technical gaps, that is, integration of UDTs across disciplines (II), among stakeholders (CC), and throughout planning processes (PO). FMU Ontology is an applied ontology that offers semantic representations (concepts and relations) to achieve this. These semantic representations are structured in four layers: (1) factor-specific, (2) relation-specific, (3) stakeholder-specific, and (4) process-specific. Correspondingly, these layers contribute directly to the socio-technical gaps: layers (1, 2) assist interdisciplinary integration of UDTs (II), layer (3) helps consensual contextualization of UDTs (CC), and (4) facilitates procedural operationalization of UDTs (PO).

Figure 2 provides a visual map of all the layers with their corresponding concepts and relations. Factor-specific representations encapsulate heterogeneous urban data. Relation-specific representations capture various urban models or assessment methods. These representations build on various works that standardize urban data and models. Stakeholder-specific representations encapsulate each stakeholder's particular planning question, configuration of UDT building blocks, and alternative plans and scenarios. Lastly, Process-specific representations capture the dynamics, iterations, and interdependencies of the planning process itself.

Figure 2.

FMU Ontology concepts are introduced in four semantic layers, corresponding to four layers of complexity in wicked problems. Arrows indicate semantic relationships between concepts.

As it is evident in Figure 2, higher layers are defined and specified according to lower layers. Process-specific concepts define how stakeholder-specific concepts are dependent on each other and change during the process. Stakeholder-specific concepts specify how various relation-specific and factor-specific representations are curated, configured, and interpreted for each stakeholder. And lastly, relation-specific layer specifies how factor-specific representations relate to each other. This hierarchical structure of semantic layers allows for customizability and configurability of higher layers, while maintaining the standardization, modularity, and interoperability in lower layers.

In the following, we briefly describe each semantic layer: briefly describing their purpose, explaining the semantic details, and highlighting their implications. Further details about the semantic structure of FMU Ontology with more examples are included in the Supplemental Materials. FMU Ontology is available in conventional semantic formats such as OWL and TTL as well as with a python wrapper as an open-source package: FMU Ontology repository.

Factor-specific semantic layer

Factor-specific semantic layer standardizes urban data to ensure modularity and interoperability for UDT data.

Any piece of information about the urban environment is contained in a VariableComponent which is represented as an n dimensional tensor. A Dimension specifies a discrete indexing system for any piece of information in the process. As a general rule, every piece of information must be indexed at least with four main Dimension (4 ≤ n): Factor (what does it describe?); Stakeholders (in relation to whom?); Space (pertaining to where?); and Time (pertaining to when?). Custom dimensions can also be defined to specify Uncertainty (with how much confidence?), Scenario (in which scenario?), and Policy (in which policy alternative?).

Additionally, we incorporate existing standardizations for urban data, metadata, and data in Europe and the Netherlands to ensure compatibility and interoperability of FMU with existing infrastructure. Particularly, we rely on OntoCityGML to provide a bridge to CityGML standards for urban data Chadzynski et al. (2021). We use Dublin Core terms (DCMI) (Baker, 2012) and Data Catalogue Vocabulary (DCAT) (Albertoni et al., 2024) through DCAT-AP-DONL (De Nederlandse Overheid, 2022).

Detailed example in Supplemental Materials structures all of these concepts in Table S5.

This readily creates all the necessary preconditions to seamlessly integrate data from open and public data infrastructure of the Netherlands. Additionally, it provides an example of how other open data infrastructures can be integrated with FMU Ontology.

Relation-specific semantic layer

Relation-specific semantic layer standardizes urban models to ensure configurability and interoperability for UDT components.

Any information processing component is represented as a MethodComponent. Each MethodComponent, is_MethodicalMap_from it’s input VariableComponents and it is_MethodicalMap_to its output VariableComponents. With the extension of the integration of VariableComponents with external representations, FMU provides a systematic and modular structure for representing complex computational processes based on a set of MethodComponents.

Additionally, MethodComponent introduces a semantic implementation of a model card based on the systematic analysis by Liang et al. (2024). Model cards include background and context information about computational models to increase the transparency of their assumptions and biases (Mitchell et al., 2019). Table S6 in Supplementary Materials illustrates the detailed example of a MethodComponent.

Describing MethodComponents based on their inputs and outputs as VariableComponents establishes a shared representation for diverse methods of models from different disciplines. Moreover, this forms a modular basis for configuring various stakeholder-specific representations.

Stakeholder-specific semantic layer

Stakeholder-specific semantic layer allows each stakeholder to formulate their own planning questions (Formulating), curate their own set of data and components into a UDT (Modelling), and use that UDT to explore alternative plans and scenarios that might answer their particular planning question (Utilizing). The important feature of stakeholder-specific layer is that all of its concepts are configurations of factor-specific and relation-specific concepts. This allows a broad capacity for expression of the unique views and preferences of the stakeholder on the one hand, and on the other hand allows for their UDTs to be interconnected. An interconnected set of stakeholder-specific UDTs means that firstly these UDTs use the same building blocks, and secondly their content can be compared, shared, and mixed. For example, stakeholder can check out the alternative scenario of the other stakeholders, but they can also evaluate their own plans in the context of the scenario of other stakeholders.

The stakeholder-specific semantic layer has two levels of representations. The upper level consists of Problem, Plan, and Model. A Problem is a structured formulation of the planning questions of a stakeholder; a Plan is a particular answer to those questions; and a Model is a stakeholder-specific UDT that allows the stakeholder to explore and evaluate Plans according to a particular Problem. Problem structures planning questions as ‘what decisions can impact the state of affairs in such a way that we can reach certain goals?’ A Plan consists of a set of specific decisions, their impact on state of affairs, and how they measure up against specified goals. Therefore, there is a Class_Object_Relation between Problem and Plan, that is, a Problem identifies the boundaries of a large pool (solution space) of Plans and how they should be evaluated. This semantic structure shapes the isolated challenge of each stakeholder as the following: ‘what is the most appropriate Plan according to the Problem?’ On the other hand, the group of stakeholders faces a consensual challenge: ‘What combination of individual Plans amounts to acceptable results for everyone?’

The lower-level of representations are VariableFormulations, Variables, and Methods. They constitute the building blocks of the upper-level representations. In other words, the upper-level representations have a Modular_Relation with lower-level representations which allows for configurability of the upper level. A Plan consists of a particular set of three Variables: a configuration of Decisions with their impact on States and how they measure up in terms of Goals. A Problem consists of a particular configuration of three VariableFormulations: what is in control of the stakeholder as DecisionsFormulation, what is in the interest of the stakeholder as GoalsFormulation, and what does the stakeholder deem relevant which is not the previous two as StatesFormulation. A Model consists of a particular set of Methods. A Method is directional relationship between two Variables. Each Method can include a set of causal relationships, correlations, ad-hoc assessments, or any other type of information processing.

Lastly, there is another set of Modular_Relations between the lower level of stakeholder-specific representations with factor-specific and relation-specific representations: a Method is a configuration of MethodComponents, a Variable is a configuration of VariableComponents, and a VariableFormulation is a configuration of Dimensions. This set of Modular_Relations provides the configurability for stakeholders to curate and configure existing datasets and methods to their own preferences.

Supplemental Materials includes detailed explanation and examples for all introduced concepts, Modular_Relations, and Class_Object_Relations.

This semantic structure allows the stakeholders to define their own Problems, configure their own Models (UDTs), and explore and evaluate their own Plans. Yet all of these representations are defined based on a shared set of building blocks. This enables stakeholders to systematically compare, share, and combine Problems, Models, and Plans of different stakeholders. This is the foundation for better navigation of the planning process, building mutual understanding, exploring trade-offs, and ultimately reaching consensus.

Process-specific semantic layer

Process-specific semantic layer embeds the other three layers within the planning process to ensure that UDTs can adapt to the evolving questions of stakeholders.

FMU Process is a modular sequence of Actions. Each Action corresponds to a step in planning process and is modelled with certain inputs and outputs that are specified through In_Out Relations. The order of Actions can be arbitrary as long as their input requirements are satisfied. Actions can be performed by an individual or a group of stakeholders. By performing Actions, stakeholders can interact with all the concepts in the previous layers: define a Problem, simulate the impact of a Plan with the help of a Model, adjust a Method in a Model, mix & match a Plan from another stakeholder and assess it based on their own Goals, etc.

Depending on the type of the inputs and outputs, Actions are structured into three phases: FAction or Formulating actions, MAction or Modelling actions, and UAction or Utilizing actions. On a holistic level, FActions have no specific input and output Problem; MActions have Problem as input and output Model; UActions have Model and Variables as input and output Variables.

Navigation of these actions is enabled by the notion of dependency that is represented by Dependency_Relation. For each Action a_j that its inputs or method are the output of a_i, a_i is_dependency_of a_j.

Supplemental Materials includes detailed explanation and examples for all introduced concepts, In_Out_Relations, and Dependency_Relations.

Accordingly, this modular flexibility of Process can systematically represent the non-leaner iterations and interactions of stakeholders during the urban planning and decision-making processes.

Validation & demonstration

Here we present the context of the test case, report on the validation results, and present three demonstrative cases from the workshop series to showcase how FMU Ontology can be useful for each of the socio-technical gaps.

Case: Eindhoven

In the last century, Eindhoven has seen significant growth due to industrial and technological developments, prompting authorities to pursue urban densification. By 2040, the Eindhoven Urban Area aims to add 62,000 homes and 72,000 jobs, with 40,000 homes and many jobs within the city’s boundaries, including 21,000 homes inside the Ring (Eindhoven, 2012). This densification could enhance liveability through dense, diverse, and mixed-use urban patterns, but it also presents challenges in meeting the increased demand for housing, public transport, and other urban facilities (Bibri et al., 2020).

To illustrate FMU Ontology’s usefulness in urban planning, we organized a series of participatory digital twinning workshops with 12 expert participants representing stakeholders involved in the planning of central Eindhoven, focussing on urban densification. The stakeholders were invited through the Urban Development Initiative (UDI), which tackles complex urban challenges in the Eindhoven metropolitan area (Brainport-Eindhoven, 2021). The workshops were part of the Urban Planning Re-imagined (UPR) program of UDI (Oomen et al., 2023). During three sessions, we collected individual and group procedural data, recorded discussions and comments, and gathered questionnaires filled out by the participants.

The workshop series resulted in 12 stakeholder-specific UDTs with various intervention scenarios that stakeholders explored and evaluated using those UDTs. Over 3 months, stakeholders interacted with the FMU Ontology via a custom web-based interface, allowing them to define their problems and specify the data and methods for their tailored UDTs. Researchers developed prototypes based on these specifications between sessions. In subsequent sessions, stakeholders used their UDTs to specify, compare, and evaluate scenarios, while refining their planning questions and UDT configurations for the next session.

This study focuses on the results and processes involving three participants from the workshop series, all experts from Eindhoven Municipality addressing liveability challenges in their respective domains. Here we present the results and process of only three of the participants in workshop series, all experts from Eindhoven Municipality addressing liveability challenges in their respective domains. During the workshops, they discussed ongoing challenges and analyzed strategic documents, including the Eindhoven Mobility Masterplan 2050 (Eindhoven-Municipality, 2024a), the Green Policy Plan (Eindhoven-Municipality, 2018), and the Development Perspective of Eindhoven (Eindhoven-Municipality, 2024b), which detail ambitions and efforts in mobility, green development, and spatial planning.

They focused on one of the following challenges:

• Active Access Challenge: How can we improve the city’s infrastructure to support walking and cycling, aiming to enhance overall accessibility without reliance on motor vehicles?

• Public Parking Challenge: How can we optimize the distribution and availability of public parking to encourage the use of public and active transportation modes?

• Development Challenge: How can we ensure that the residents of the new housing projects have easy access to sufficient public spaces, without throttling the access of current residents of the surrounding areas?

The experts actively participated in defining their planning concerns, providing detailed input on preferred data and assessment methods, and discussing intervention alternatives. This collaborative process enabled the development of stakeholder-specific UDTs tailored to their needs. Using these UDTs, 158 combinations of intervention scenarios were explored, combined, evaluated, and compared.

The development and refining these UDTs followed an iterative process over three meetings, during which experts reviewed digital prototypes, saw their inputs integrated, and suggested refinements. This process ensured alignment with their visions. The interconnected UDTs allowed experts to access each other’s work, including problem formulations, assessment configurations, and evaluations of intervention scenarios. Firstly, this led the experts to reconsider, adjust, and adapt how they have structured their planning problem and UDT. Secondly, it allowed them to mix and match alternative intervention scenarios; that is, combine and evaluate their own intervention scenarios in the context of alternative intervention scenarios of other experts.

Feedback from the experts suggests that the structured approach and the semantic representations provided by FMU helped clarify the details of each other’s planning questions and how UDTs can be effectively used within the context of their daily responsibilities. This endorsement indicates that FMU not only facilitates detailed and targeted urban planning discussions but also enhances the effectiveness of collaborative urban decision-making. Below, we present the validation result and three demonstrative examples each corresponding to one of the socio-technical gaps in UDTs.

Validation

We have validated the consistency and coherency of all the concepts of FMU Ontology by Pellet reasoning engine (Sirin et al., 2007) through the interface of Owlready2 Python module (Jean-Baptiste, 2021). This confirms no internal inconsistencies.

To validate the competency of FMU Ontology, we used a portion of the larger KG resulted from the UDTing workshop series that pertains to the three stakeholders described earlier. The resultant KG captures the planning process, stakeholder-specific UDTs, and all the models and data in them. The competency test was carried out intermittently during the demo case process. The test module compares query results with predefined outputs to verify correctness. FMU Ontology passed all competency tests. Combined, these two tests establish the efficacy of FMU Ontology.

The detailed results and diagrams of all CQs with examples are included in Supplemental Materials section Validation Results and Figures S1–S13. For illustrative purposes, Figure 3 visualizes the results of competency question [OR1] which asks what planning actions have been taken during the process. The figure shows two major vertical iterations where stakeholders reconsidered their planning question (Problem) and updated their UDT (Model). Moreover, Parking expert has taken more vertical iterations switching between Formulating and Modelling actions. This indicates iterative revision of Problem and Model to ensure their consistency. Alternatively, ActiveAccess has taken far more horizontal iterations, that is, exploring and evaluating policies and scenarios.

Figure 3.

Visualization of the complete Process of the demonstration case as a series of Actions. This is also a sample result of the [OR1] query at the end. From left to right, it shows the sequence of Actions taken by stakeholders: Parking stakeholder is cyan, Active Access stakeholder is orange, Development stakeholder is purple. On the vertical axis different Actions are denoted and clustered to their corresponding phases: Formulating (planning questions), Modelling the UDT, Utilizing the UDT. The most important features are the Vertical and Horizontal iterations. Horizontal iterations denote the iterative steps of using the UDT to explore, change, compare, and evaluate plans. These are contained within the Utilizing phase. Vertical iterations denote the iterative cycle of updating the planning question and correspondingly updating the UDT. These span across all three phase.

Our validation tests indicate the efficacy of FMU Ontology in developing and using UDTs within planning processes. Using data from participatory workshops, we confirmed it has no inconsistencies and provides support for stakeholders in refining problems, updating stakeholder-specific UDTs, and exploring alternative policies.

Interdisciplinary integration in demo

The socio-technical challenge of integrating different disciplines (II) in UDTs stems from the lack of a modular and systematic framework to combine diverse urban data and models. To address this gap, FMU Ontology provides clear, specific representations of urban data and models. These act as building blocks for more comprehensive stakeholder-specific and process-specific representations. This approach allows for (1) a shared way to represent the detailed aspects of datasets and computational components, and (2) a systematic method to compare them.

The following is a demonstrative example that occurred during the workshop. The expert behind the Development Challenge initially focused on enhancing pedestrian access to public spaces such as parks and waterfronts. Exploring the results from traditional methods of measuring accessibility suggested a decline when the plan for new developments was taken into account. This is because more buildings mean more density, which would mean less green and blue spaces per capita.

To address this, the concept of accessibility elasticity was introduced. Accessibility elasticity measures how changes in the urban environment affect overall access to amenities, adjusted for any increase in local population density (Levine et al., 2017). This metric is crucial for understanding the true benefits of new developments for new residents who might not have had access to these amenities before moving to this area.

The Development Challenge expert updated their FMU representation of planning questions to include the new measure of accessibility elasticity. Next, they used FMU Ontology to identify suitable components for computing this metric. Furthermore, FMU confirmed that existing datasets in their UDTs already contained the necessary data for this new assessment, making it feasible to add the new metric without additional data sources. The middle row of Figure 4 shows their updated planning question (Problem) and UDT configuration (Model) for the Development Challenge. This illustrates how FMU Ontology’s modular representational system enables systematic integration of interdisciplinary resources. For implemented details see Section Demonstration of Supplemental Materials.

Figure 4.

Visualization of the latest stakeholder-specific representations of our three stakeholders’ Problem, that is, planning question, Model, that is, UDT, and Plan, that is, alternative intervention plans/scenarios. The middle column visualizes their Problem (planning question) and Model (UDT configuration) as a graph. The side columns visualize some examples of VariableComponents from particular Plans of stakeholders that was mentioned in the demo cases. The Problem is represented as a set of coloured circles, each of which indicates a VariableComponent that hold datasets. The Model is depicted as a set of squares with arrows to/from circles: squares are MethodComponents that hold computational processes; and arrows indicate the inputs and outputs of MethodComponents. The colour of circles indicate which Variable of Problem they belong to: red is Decisions, blue is States, and green is Goals. The planning question embedded in the Problem can be read with the following template: Given the current blue states, how should we configure the red decisions, in order to meet our target in green goals? As the Model is configured to provide insight with respect to this planning questions, it should assess the impact of alternative decisions on our stated goals. Therefore, it has a flow of data from left to right: red decisions as the starting point and green goals as final outputs.

To showcase the contribution of FMU Ontology to II, we explained a detailed instance from the workshop series. In this instance, FMU Ontology helps the stakeholder to update planning questions, identify suitable components for the new metric, and confirm the feasibility of using existing dataset in their UDT.

Consensual contextualization in demo

The socio-technical challenge of Consensual Contextualization (CC) highlights the need to better involve stakeholders in the development and use of UDTs. This requires ensuring that UDTs reflect the diverse social context and values of stakeholders involved in planning challenge. To address this, FMU provides stakeholder-specific representations for planning questions (Problems), UDTs (Models), and intervention plans (Plans). These modular representations allow stakeholders to systematically compare approaches, learn from how others frame problems, explore the components used in their UDTs, and combine plans with those of other stakeholders.

The following is a demonstrative example that occurred during the workshop. One of the themes that the expert behind the Active Access challenge focuses on is improving the city’s walking infrastructure by identifying and resolving missing links in the pedestrian network. One of their particular intervention plans, labelled as P3, involves constructing pedestrian bridges at strategic locations, such as between Gabriël Metsulaan and Tenierslaan in Eindhoven. This intervention aims to enhance the pedestrian access of inhabitants in Lakerlopen and Irisbuurt neighbourhoods which are divided by Kanaaldijk.

Using FMU Ontology, Active Access expert explored the scenarios and plans being considered by other stakeholders. Specifically, they found that the Development stakeholder had proposed a scenario called S3: Green Heart, which focuses on enhancing green spaces within Eindhoven’s ring (Groenehart). Since improving access to green spaces aligns with Active Access’s goals, incorporating this scenario into their planning could allow them to assess the combined effect and look for mutual benefits.

To evaluate the pedestrian bridge project in the context of Green Heart scenarios, the Active Access expert used a series of FMU queries. This allowed them identify the relevance between these interventions plans and combine them systematically (see details in Section Demonstration of Supplemental Materials). This enabled the expert to assess the bridge project’s impact (P3) within the enhanced Green Heart context (S3) and result in the appraisal (A32) which indicates improved active access to green. This approach highlights how FMU Ontology supports stakeholders in combining and evaluating mixed intervention planning scenarios.

To showcase the contribution of FMU Ontology to CC, we explained a detailed instance from the workshop series. FMU Ontology enabled the Active Access expert to integrate their pedestrian bridge plan with the Green Heart scenario proposed by another stakeholder. By combining and evaluating these plans systematically, FMU demonstrated its ability to align diverse stakeholder goals and assess the collective impact on green space accessibility.

Procedural operationalization in demo

The socio-technical challenge of Procedural Operationalization (PO) highlights the need for a better integration of UDTs into urban planning processes. This includes creating flexible digital tools that adapt to changing stakeholder needs and decision-making in participatory planning. FMU addresses this by offering process-specific representations, helping stakeholders understand how others’ choices have shaped their plans and how to adapt their UDTs to address evolving questions in the planning process.

The following is a demonstrative example that occurred during the workshop. FMU helped the expert behind the Parking Challenge to adapt their UDT as the need arises. Particularly, the Parking expert in the initial steps they were more focused on their individual set of alternative plans. Nevertheless, FMU allowed them to identify the overlap of spatial claims of their own planning alternative and others. This indicated some overlap with alternative interventions of Active Access expert as they explored the possibility of developing new bike paths. The overlap included two main aspects: (1) the potential for opening up space on the side of some streets to widen the bike path, and (2) the possibility of combining some public parking spaces into a Park-&-Bike point. To effectively incorporate the spatial requirements of the Active Access plan, Parking expert needed to adapt their UDT; which is not a straightforward process.

The process-specific representations in FMU Ontology provide a systematic way to facilitate and guide such a process. For this particular situation, we used a series of queries to identify areas of overlap or mismatch between the two stakeholders’ Plans, find the difference between their Problem and Model (UDT), and suggest a series of planning actions that would enable Parking expert to evaluate the Active Access plans within the context of their own question.

The Parking expert used a series of FMU queries to refine their representations: (1) identifying what datasets are missing, (2) compared assessment components, (3) retrieved the necessary planning steps to adjust the planning question and UDT configuration, and (4) ensured the coherence of the planning steps’ order (see Section Demonstration of Supplemental Materials for details). Based on these queries, FMU suggested a series of planning steps. As shown in the figure, the expert has initiated this iteration but diverged as they faced new concerns later. This iterative navigation capability of FMU, allows the planning process to remain agile and flexible and ensure effective operationalization of UDTs within the socio-organizational context.

To showcase the contribution of FMU Ontology to PO, we explained a detailed instance from the workshop series. In this instance, after the Parking Challenge expert used FMU to identify overlaps with the Active Access expert’s bike path plans. Then FMU guided him in how to adapt his UDT by helping the expert refine datasets, compare components, adjust planning steps, and ensure coherence. This makes digital representations flexible enough to support iterative and collaborative planning processes.

Discussion and conclusion

When viewed as a collection of urban data and models, UDTs can serve as valuable tools for urban planning and decision-making (Quek et al., 2023; Wu and Guan, 2024). Their usefulness, however, is hindered by three socio-technical challenges: integrating data and models from diverse urban disciplines (Interdisciplinary Integration or II), incorporating the views and preferences of various stakeholders (Consensual Contextualization or CC), and enabling UDTs to adapt to the dynamic iterations of the planning process (Procedural Operationalization or PO) (Azadi et al., 2025). Addressing these three challenges would bring us to an Augmented Urban Planning (AUP) process (Azadi et al., 2023): a planning process entwined with integrated, contextualized, and operationalized representations.

Formulating-Modelling-Utilizing (FMU) Ontology provides a formal framework for UDTs to achieve Interdisciplinary Integration (II), Consensual Contextualization (CC), and Procedural Operationalization (PO). The core idea is that UDTs should not aim to be boundless representations but must instead define their representational boundaries in a way that provides useful insights for specific urban planning and decision-making problems. Planning problems, therefore, prescribe the epistemological frame for UDTs, specifying the scope of their digital representations – what is included and what is excluded – to align with the needs of the problem.

Since planning problems are stakeholder-specific, UDTs must also be tailored to individual stakeholders accordingly. To facilitate this, we provide a modular and standardized set of urban data and models as building blocks (see Factor-specific Semantic Layer, Relation-specific Semantic Layer). Firstly, stakeholders formulate their planning questions and then curate, configure, and interpret these shared blocks into their specific UDTs. If needed, they can incorporate custom building blocks, as the system is based on open standards and open-source software. Then, these stakeholder-specific UDTs can deliver tailored insights for stakeholder-specific planning questions and support the exploration and evaluation of stakeholder-specific intervention plans and scenarios (see Stakeholder-specific Semantic Layer).

The modular and standardized building blocks shared among UDTs enables the creation of a network of interconnected stakeholder-specific UDTs. This allows stakeholders to systematically access, compare, and understand each other’s unique problems, UDTs, and plans. It also provides opportunities to learn from one another and explore the outcomes of joint decisions by combining alternative plans and scenarios.

As stakeholders engage in the participatory planning processes, their planning questions evolve, requiring their stakeholder-specific UDTs to adapt accordingly. UDTs must, therefore, remain flexible, continually aligning with the changing epistemological frames of their corresponding stakeholders to ensure relevancy to their views and concerns. This is done through tracing changes in the planning process and continuously comparing UDTs with the corresponding planning problem (see Process-specific Semantic Layer).

FMU Ontology provides a formalized framework implemented as an Ontology with a Python wrapper. We validated its efficacy and demonstrated its contributions to II, CC, and PO through concrete examples (see Validation & Demonstration). However, further testing is needed to assess its effectiveness and efficiency.

FMU Ontology needs further embedded testing in participatory processes involving expert stakeholders. Another important step is to evaluate its contributions in facilitating communication, collaboration, and engagement of non-experts in actual planning processes. These tests will provide a more realistic understanding of its capabilities and correspondingly advance its implementation and operationalization.

FMU Ontology addresses the three socio-technical challenges facing UDTs but falls short in others. While it takes initial steps to integrate with the Dutch data infrastructure, further research is needed to align it with legal frameworks, organizational structures, and governance primary processes. Additionally, FMU builds on open standards and open-source software to promote transparency and accessibility, but issues such as cybersecurity, privacy, and proprietary data ownership require further attention to how data and models are shared to enable broader integration.

Supplemental Material

Supplemental Material - Formalizing-modelling-utilizing ontology: A semantic framework for adaptive stakeholder-specific urban digital twins in urban planning processes

Supplemental Material for Formalizing-modelling-utilizing ontology: A semantic framework for adaptive stakeholder-specific urban digital twins in urban planning processes by Shervin Azadi, Dena Kasraian, Pirouz Nourian and Pieter van Wesemael in Environment and Planning B: Urban Analytics and City Science

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Regional Deal Brainport Eindhoven (Urban Development Initiative).

ORCID iDs

Shervin Azadi

Dena Kasraian

Pirouz Nourian

Data availability statement

Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.

Supplemental Material

Supplemental material for this article is available online.

Shervin Azadi is a researcher in Urbanism and Urban Architecture at Eindhoven University of Technology. He is interested in the socio-technical complexities of developing and using digital technologies in urban planning processes. His research focuses on developing a formal methodology for integrating, contextualizing, and operationalizing Urban Digital Twins within planning processes.

Dena Kasraian has studied and worked in the fields of architecture, urban planning and design, and transportation. She is interested in mixed-methods interdisciplinary research across these fields, exploring the interactions between the built environment, technology, and human behaviour, and their the public health impacts.

Pirouz Nourian is a computational design and planning technologist, specializing in applied computational mathematics for augmenting design and planning processes with intelligence for finding optimal or satisfactory (multi-criteria) configurations (topologies) and shapes (geometries). He has a hybrid educational background in Control Systems Engineering, Architecture, and Urban Design Informatics.

Pieter van Wesemael is full professor of Urbanism and Urban Architecture at Eindhoven University of Technology. He has researched and published widely on topics related to systemic urban design and planning, healthy cities, active mobility and place making. Focus in his research is on how behaviour is shaped by the interplay of spatial, socio-economic and digital factors as a base to design intervention strategies to achieve the UN sustainable development goals for an social inclusive, environmental healthy and economic resilient urban living environment.

References

Abdeen

Shirowzhan

Sepasgozar

(2023) Citizen-centric digital twin development with machine learning and interfaces for maintaining urban infrastructure. Telematics and Informatics 84: 102032.

Adade

de Vries

(2024) An extended toe framework for local government technology adoption for citizen participation: insights for city digital twins for collaborative planning. Transforming Government: People, Process and Policy 19: 53–73.

Adeagbo

Wang

(2024) Revamping structural health monitoring of advanced rail transit systems: a paradigmatic shift from digital shadows to digital twins. Advanced Engineering Informatics 61: 102450.

Akroyd

Mosbach

Bhave

, et al. (2021) Universal digital twin - a dynamic knowledge graph. Data-Centric Engineering 2: e14.

Al-Sehrawy

Kumar

Watson

(2021) A digital twin uses classification system for urban planning & city infrastructure management. Journal of Information Technology in Construction 26: 832–862.

Albertoni

Browning

Cox

, et al. (2024) Data catalog vocabulary (DCAT) - version 3. https://www.w3.org/TR/vocab-dcat-3

Alexander

(2000) Rationality revisited: planning paradigms in a post-postmodernist perspective. Journal of Planning Education and Research 19(3): 242–256.

Allan

Rajabifard

Foliente

(2024) Climate resilient urban regeneration and SDG 11 - stakeholders’ view on pathways and digital infrastructures. International Journal of Digital Earth 17(1).

Arp

Smith

Spear

(2015) Building Ontologies with Basic Formal Ontology. Cambridge: MIT Press. https://books.google.nl/books?id=AUxQCgAAQBAJ

10.

Astarita

Guido

Haghshenas

, et al. (2024) Risk reduction in transportation systems: the role of digital twins according to a bibliometric-based literature review. Sustainability 16(8): 3212.

11.

Austin

Delgoshaei

Coelho

, et al. (2020) Architecting smart city digital twins: combined semantic model and machine learning approach. Journal of Management in Engineering 36(4).

12.

Azadi

Kasraian

Nourian

, et al. (2023) Augmented urban planning: a framework for strategic urban planning. In: CUPUM 2023, the 18th International Conference on Computational Urban Planning and Urban Management, Montreal, Canada, 20–23-June 2023.

13.

Azadi

Kasraian

Nourian

, et al. (2025) What have urban digital twins contributed to urban planning and decision making? From a systematic literature review toward a socio-technical research and development agenda. Smart Cities 8(1): 32.

14.

Bai

Azadi

Nourian

, et al. (2020) Decision-making as a social choice game - gamifying an urban redevelopment process in search for consensus. In: Proceedings of the 38th International Conference on Education and Research in Computer Aided Architectural Design in Europe (eCAADe) [Volume 2], eCAADe 2020. volume 2, Berlin, Germany, 2020, pp. 555–564.

15.

Baker

(2012) Libraries, languages of description, and linked data: a dublin core perspective. Library Hi Tech 30(1): 116–133.

16.

Batty

(1971) Modelling cities as dynamic systems. Nature 231(5303): 425–428.

17.

Batty

(2018) Digital twins. Environment and Planning B: Urban Analytics and City Science 45(5): 817–820.

18.

Bibri

(2021) The core academic and scientific disciplines underlying data-driven smart sustainable urbanism: an interdisciplinary and transdisciplinary framework. Computational Urban Science 1(1): 1.

19.

Bibri

Krogstie

Kärrholm

(2020) Compact city planning and development: emerging practices and strategies for achieving the goals of sustainability. Developments in the Built Environment 4: 100021.

20.

Bibri

Huang

Jagatheesaperumal

, et al. (2024) The synergistic interplay of artificial intelligence and digital twin in environmentally planning sustainable smart cities: a comprehensive systematic review. Environmental Science and Ecotechnology 20: 100433.

21.

Biljecki

Stoter

Ledoux

, et al. (2015) Applications of 3D city models: state of the art review. ISPRS International Journal of Geo-Information 4(4): 2842–2889.

22.

Biljecki

Lim

Crawford

, et al. (2021) Extending CityGML for IFC-sourced 3D city models. Automation in Construction 121: 103440.

23.

Billen

Zaki

Servières

, et al. (2012) Developing an ontology of space: application to 3D city modeling. In: Leduc

Moreau

Billen

(eds). Usage, Usability, and Utility of 3D City Models - European COST Action TU0801, 3u3d2012. Les Ulis, France: EDP Sciences, 02007.

24.

Billen

Cutting-Decelle

Marina

, et al. (2014) 3D city models and urban information: current issues and perspectives: European cost action tu0801. In: Billen

Cutting-Decelle

Marina

, et al. (eds) 3D City Models and Urban Information: Current Issues and Perspectives - European COST Action TU0801, COST2013. Les Ulis, France: EDP Sciences.

25.

Boje

Guerriero

Kubicki

, et al. (2020) Towards a semantic construction digital twin: directions for future research. Directions for future research 114: 103179.

26.

Bojórquez-Tapia

Luna-González

Cruz-Bello

, et al. (2011) Regional environmental assessment for multiagency policy making: implementing an environmental ontology through GIS-MCDA. Environment and Planning B: Planning and Design 38(3): 539–563.

27.

Brainport-Eindhoven (2021) UDI. https://brainporteindhoven.com/udi/en/ (Accessed on 09/26/2024).

28.

Caglioni

Rabino

(2007) Theoretical approach to urban ontology: a contribution from urban system analysis. In: Ontologies for Urban Development. Berlin Heidelberg: Springer, 109–119.

29.

Calzati

(2023) No longer hype, not yet mainstream? Recalibrating city digital twins’ expectations and reality: a case study perspective. Frontiers in Big Data 6: 1236397.

30.

Caprari

Castelli

Montuori

, et al. (2022) Digital twin for urban planning in the green deal era: a state of the art and future perspectives. Sustainability 14(10): 6263.

31.

Chadzynski

Krdzavac

Farazi

, et al. (2021) Semantic 3D city database — an enabler for a dynamic geospatial knowledge graph. Energy and AI 6: 100106.

32.

Charitonidou

(2022) Urban scale digital twins in data-driven society: challenging digital universalism in urban planning decision-making. International Journal of Architectural Computing 20(2): 238–253.

33.

Chaturvedi

Kolbe

(2016) Integrating dynamic data and sensors with semantic 3D city models in the context of smart cities. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences IV-2/W1: 31–38.

34.

Chaturvedi

Smyth

Gesquiere

, et al. (2016) Managing Versions and History within Semantic 3D City Models for the Next Generation of CityGML. Göttingen, Germany: Springer International Publishing, 191–206.

35.

Chen

Alomari

Taffese

, et al. (2024) Multifunctional models in digital and physical twinning of the built environment—a university campus case study. Smart Cities 7(2): 836–858.

36.

Corry

Pauwels

, et al. (2015) A performance assessment ontology for the environmental and energy management of buildings. Automation in Construction 57: 249–259.

37.

Cranefield

Purvis

(1999) UML as an ontology modelling language.

38.

Davis

Shrobe

Szolovits

(1993) What is a knowledge representation? AI Magazine 14: 17–33.

39.

Davoudi

(2006) Evidence-based planning: rhetoric and reality. disP - The Planning Review 42(165): 14–24.

40.

de Nederlandse Overheid (2022) Dcat-ap-donl 1.1 — dcat-ap-donl documentation. https://dcat-ap-donl.readthedocs.io/en/latest/ (Accessed on 08/23/2024).

41.

Deng

Zhang

Shen

ZJM

(2021) A systematic review of a digital twin city: a new pattern of urban governance toward smart cities. Journal of Management Science and Engineering 6(2): 125–134.

42.

Donkers

Ledoux

Zhao

, et al. (2015) Automatic conversion of IFC datasets to geometrically and semantically correct citygml LOD3 buildings. Transactions in GIS 20(4): 547–569.

43.

Ehrlinger

Wöß

(2016) Towards a definition of knowledge graphs. In: International Conference on Semantic Systems.

44.

Eibeck

Lim

Kraft

(2019) J-park simulator: an ontology-based platform for cross-domain scenarios in process industry. Computers & Chemical Engineering 131: 106586.

45.

Eibeck

Chadzynski

Lim

, et al. (2020) A parallel world framework for scenario analysis in knowledge graphs. Data-Centric Engineering 1: e6.

46.

Eindhoven

(2012) Challenge mobility transition. https://www.stadslabeindhoven.nl/challenge-mobility-transition/ (Accessed on 09/10/2024).

47.

Eindhoven-Municipality (2018) Gemeenteblad 2018 251197 overheidnl officile bekendmakingen. https://zoek.officielebekendmakingen.nl/gmb-2018-251197.html (undefined 23/8/2024 16:58).

48.

Eindhoven-Municipality (2024a) Mobiliteits verslag en actieplan. https://raadsinformatie.eindhoven.nl/document/13655914/2 (undefined 23/8/2024 11:20).

49.

Eindhoven-Municipality (2024b) Ontwikkelperspectief 2040 centrum eindhoven. https://www.openeindhoven.nl/sites/default/files/2021-06/4.OntwikkelperspectiefCentrum200612_Hoofdrapportage_def%28DEF%29.pdf (undefined 23/8/2024 16:57).

50.

El-Gohary

El-Diraby

(2010) Domain ontology for processes in infrastructure and construction. Journal of Construction Engineering and Management 136(7): 730–744.

51.

Espinoza-Arias

Poveda-Villalón

García-Castro

, et al. (2018) Ontological representation of smart city data: from devices to cities. Applied Sciences 9(1): 32.

52.

European-Union (2024) Data catalogue vocabulary application profile (dcat-ap). https://github.com/SEMICeu/DCAT-AP

53.

Forrester

(1970) Urban dynamics. IMR; Industrial Management Review (pre-1986) 11(3): 67.

54.

Fox

(2015) The role of ontologies in publishing and analyzing city indicators. Computers, Environment and Urban Systems 54: 266–279.

55.

Geertman

Stillwell

(2004) Planning support systems: an inventory of current practice. Computers, Environment and Urban Systems 28(4): 291–310.

56.

Gil

(2020) City information modelling: a conceptual framework for research and practice in digital urban planning. Built Environment 46(4): 501–527.

57.

Glimm

Horrocks

Motik

(2010) Optimized description logic reasoning via core blocking. In: Giesl

Hähnle

(eds) Proc. Of the 5th Int. Joint Conf. on Automated Reasoning (IJCAR 2010), LNCS, Volume 6173. Edinburgh, UK: Springer, 457–471.

58.

Grau

Horrocks

Motik

, et al. (2008) OWL 2: the next step for OWL. Journal of Web Semantics 6(4): 309–322.

59.

Grisiute

Silvennoinen

, et al. (2022) Unlocking urban simulation data with a semantic city planning system - ontologically representing and integrating MATSim output data in a knowledge graph. eCAADe Proceedings. Ghent, Belgium: eCAADe.

60.

Gruber

(1993) A translation approach to portable ontology specifications. Knowledge Acquisition 5(2): 199–220.

61.

Gruber

(1995) Toward principles for the design of ontologies used for knowledge sharing? International Journal of Human-Computer Studies 43(5-6): 907–928.

62.

Gruninger

(1995) Methodology for the design and evaluation of ontologies. In: Proc. IJCAI’95, Workshop on Basic Ontological Issues in Knowledge Sharing, Montreal, 1995.

63.

Guarino

(1998) Formal Ontology in Information Systems: Proceedings of the First International Conference (FOIS’98), June 6-8, Trento, Italy, Volume 46. Amsterdam, the Netherlands: IOS press.

64.

Guo

Gao

, et al. (2018) Artificial intelligence-based semantic internet of things in a user-centric smart city. Sensors 18(5): 1341.

65.

Hämäläinen

(2021) Urban development with dynamic digital twins in Helsinki city. IET Smart Cities 3(4): 201–210.

66.

Harding

Blokland

(2014) Urban Theory: A Critical Introduction to Power, Cities and Urbanism in the 21st Century. London: Sage Publications.

67.

Hepp

Roman

(2007) An ontology framework for semantic business process management. Wirtschaftinformatik Proceedings 2007: 27.

68.

Hoang

Jung

Tran

(2013) Ontology-based approaches for cross-enterprise collaboration: a literature review on semantic business process management. Enterprise Information Systems 8(6): 648–664.

69.

Hofmeister

Brownbridge

Hillman

, et al. (2024a) Cross-domain flood risk assessment for smart cities using dynamic knowledge graphs. Sustainable Cities and Society 101: 105113.

70.

Hofmeister

Lee

Tsai

, et al. (2024b) Dynamic control of district heating networks with integrated emission modelling: a dynamic knowledge graph approach. Energy and AI 17: 100376.

71.

Hopkins

(2001) Urban Development: The Logic of Making Plans. Number v. 166 in Urban Development: The Logic of Making Plans. Washington, DC: Island Press. https://books.google.nl/books?id=Knzi2vTTRxYC

72.

Hor

Sohn

Claudio

, et al. (2018) A semantic graph database for BIM-GIS integrated information model for an intelligent urban mobility web application. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences IV(4): 89–96.

73.

Innes

(1995) Planning theory’s emerging paradigm: communicative action and interactive practice. Journal of Planning Education and Research 14(3): 183–189.

74.

Jean-Baptiste

(2021) Ontologies with Python: Programming OWL 2. 0 Ontologies with Python and Owlready2. Berkeley, CA: Apress.

75.

Jelokhani-Niaraki

(2018) Knowledge sharing in web-based collaborative multicriteria spatial decision analysis: an ontology-based multi-agent approach. Computers, Environment and Urban Systems 72: 104–123.

76.

Jelokhani-Niaraki

Sadeghi-Niaraki

Choi

(2018) Semantic interoperability of GIS and MCDA tools for environmental assessment and decision making. Environmental Modelling & Software 100: 104–122.

77.

Kardinal Jusuf

Mousseau

Godfroid

, et al. (2017) Integrated modeling of CityGML and IFC for city/neighborhood development for urban microclimates analysis. Energy Procedia 122: 145–150.

78.

Kaza

Hopkins

(2007) Ontology for land development decisions and plans. In: Ontologies for Urban Development. Berlin Heidelberg: Springer, 47–59.

79.

Ketzler

Naserentin

Latino

, et al. (2020) Digital twins for cities: a state of the art review. Built Environment 46(4): 547–573.

80.

Kolbe

Kutzner

Smyth

, et al. (2021) OGC city geography markup language (CityGML) version 3.0 Part 1: conceptual model standard. Open Geospatial Consortium. https://docs.ogc.org/is/20-010/20-010.html

81.

Konys

(2018) An ontology-based knowledge modelling for a sustainability assessment domain. Sustainability 10(2): 300.

82.

Kumar

Labetski

Ohori

, et al. (2019) Harmonising the OGC standards for the built environment: a CityGML extension for LandInfra. ISPRS International Journal of Geo-Information 8(6): 246.

83.

Kutzner

Chaturvedi

Kolbe

(2020) CityGML 3.0: new functions open up new applications. PFG – Journal of Photogrammetry, Remote Sensing and Geoinformation Science 88(1): 43–61.

84.

Lagap

Ghaffarian

(2024) Digital post-disaster risk management twinning: a review and improved conceptual framework. International Journal of Disaster Risk Reduction 110: 104629.

85.

Langenheim

Sabri

Chen

, et al. (2022) Adapting a digital twin to enable real-time water sensitive urban design decision-making. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XLVIII-4/W4-2022: 95–100.

86.

Lee

Tian

, et al. (2019) An efficient design support system based on automatic rule checking and case-based reasoning. KSCE Journal of Civil Engineering 23(5): 1952–1962.

87.

Lei

Janssen

Stoter

, et al. (2023) Challenges of urban digital twins: a systematic review and a Delphi expert survey. Automation in Construction 147: 104716.

88.

Levesque

(1986) Knowledge representation and reasoning. Annual Review of Computer Science 1(1): 255–287.

89.

Levine

Merlin

Grengs

(2017) Project-level accessibility analysis for land-use planning. Transport Policy 53: 107–119.

90.

Liang

Rajani

Yang

, et al. (2024) Systematic analysis of 32,111 AI model cards characterizes documentation practice in AI. Nature Machine Intelligence 6(7): 744–753.

91.

Madrazo

Sicilia

Gamboa

(2012) Semanco: semantic tools for carbon reduction in urban planning.

92.

Mai

Janowicz

Cai

, et al. (2020) SE-KGE: a location-aware knowledge graph embedding model for geographic question answering and spatial semantic lifting. Transactions in GIS 24(3): 623–655.

93.

Malczewski

Jelokhani-Niaraki

(2012) An ontology-based multicriteria spatial decision support system: a case study of house selection. Geo-spatial Information Science 15(3): 177–185.

94.

Marcucci

Gatta

Le Pira

, et al. (2020) Digital twins: a critical discussion on their potential for supporting policy-making and planning in urban logistics. Sustainability 12(24): 10623.

95.

Mitchell

Zaldivar

, et al. (2019) Model cards for model reporting. In: Proceedings of the Conference on Fairness, Accountability, and Transparency, FAT* ’19, Atlanta, GA, USA, January 29–31, 2019.

96.

Montenegro

Gomes

Urbano

, et al. (2012) A land use planning ontology: lbcs. Future Internet 4(1): 65–82.

97.

Mostafa

El-Gohary

(2015) Semantic system for stakeholder-conscious infrastructure project planning and design. Journal of Construction Engineering and Management 141(2).

98.

Nguyen

Kolbe

(2020) A multi perspective approach to interpreting spatio semantic changes of large 3D city models in CityGML using a graph database. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences VI-4/W1-2020: 143–150.

99.

Nochta

Wan

Schooling

, et al. (2020) A socio-technical perspective on urban analytics: the case of city-scale digital twins. Journal of Urban Technology 28(1-2): 263–287.

100.

Nourian

Azadi

Bai

, et al. (2024) Equicity game: a mathematical serious game for participatory design of spatial configurations. Scientific Reports 14(1): 10912.

101.

Oomen

Haase

Wiersma

, et al. (2023) Manifest of urban planning reimagined: a digitally enabled new way of working.

102.

Oti-Sarpong

Bastidas

Nochta

, et al. (2022) A social construction of technology view for understanding the delivery of city-scale digital twins. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences X-4/W3-2022: 205–212.

103.

Pang

Huang

Xie

, et al. (2021) Collaborative city digital twin for the covid-19 pandemic: a federated learning solution. Tsinghua Science and Technology 26(5): 759–771. https://www.scopus.com/inward/record.uri?eid=2-s2.0-85104573245&doi=10.26599%2fTST.2021.9010026&partnerID=40&md5=31967a2e58d29fe0427a123eaaf4cfac.CitedBy3

104.

Park

Yang

(2020) Gis-enabled digital twin system for sustainable evaluation of carbon emissions: a case study of Jeonju City, South Korea. Sustainability 12(21): 9186.

105.

Pauwels

Zhang

Lee

(2017) Semantic web technologies in AEC industry: a literature overview. Automation in Construction 73: 145–165.

106.

Pedersen

Borup

Brink-Kjær

, et al. (2021) Living and prototyping digital twins for urban water systems: towards multi-purpose value creation using models and sensors. Water 13(5): 592.

107.

Peldon

Banihashemi

LeNguyen

, et al. (2024) Navigating urban complexity: the transformative role of digital twins in smart city development. Sustainable Cities and Society 111: 105583.

108.

Psyllidis

(2015) Ontology-based data integration from heterogeneous urban systems: a knowledge representation framework for smart cities.

109.

Pultrone

(2023) The city challenges and the new frontiers of urban planning. TeMA - Journal of Land Use Mobility and Environment: Tools and Best Practices. https://www.serena.unina.it/index.php/tema/article/view/9392

110.

Quek

Sielker

Akroyd

, et al. (2023) The conundrum in smart city governance: interoperability and compatibility in an ever-growing ecosystem of digital twins. Data and Policy 5: e6.

111.

Raes

Michiels

Adolphi

, et al. (2021) DUET: a framework for building interoperable and trusted digital twins of smart cities. IEEE Internet Computing 26: 43–50. https://www.scopus.com/inward/record.uri?eid=2-s2.0-85101769197&doi=10.1109%2fMIC.2021.3060962&partnerID=40&md5=30144a08f54b85f9d6bd23090bb93674.CitedBy0

112.

Ronzhin

Folmer

Maria

, et al. (2019) Kadaster knowledge graph: beyond the fifth star of open data. Information 10(10): 310.

113.

Santos

Dantas

Furtado

, et al. (2017) From Data to City Indicators: A Knowledge Graph for Supporting Automatic Generation of Dashboards. Cham, Switzerland: Springer International Publishing, 94–108.

114.

Shi

Pan

Jiang

, et al. (2023) An ontology-based methodology to establish city information model of digital twin city by merging BIM, GIS and IOT. Advanced Engineering Informatics 57: 102114.

115.

Simonsson

Atta

Schweiger

, et al. (2021) Experiences from city-scale simulation of thermal grids. Resources 10(2): 10.

116.

Sirin

Parsia

Grau

, et al. (2007) Pellet: a practical OWL-DL reasoner. Web Semantics 5(2): 51–53.

117.

Stadler

Nagel

König

, et al. (2009) Making Interoperability Persistent: A 3D Geo Database Based on CityGML. Berlin: Springer Berlin Heidelberg, 175–192.

118.

Stead

(2021) Conceptualizing the policy tools of spatial planning. Journal of Planning Literature 36(3): 297–311.

119.

Studer

Benjamins

Fensel

(1998) Knowledge engineering: principles and methods. Data & Knowledge Engineering 25(1-2): 161–197.

120.

Tempelmeier

Rietz

Lishchuk

, et al. (2019) Data4urbanmobility: towards holistic data analytics for mobility applications in urban regions. In: Companion Proceedings of the 2019 World Wide Web Conference, WWW ’19, San Francisco, USA, 2019, pp. 137–145.

121.

Thompson

(2022) Escape from Model Land: How Mathematical Models Can Lead Us Astray and What We Can Do about it. New York: Basic Books.

122.

Throgmorton

(1996) Planning as Persuasive Storytelling: The Rhetorical Construction of Chicago’s Electric Future. Chicago: University of Chicago Press.

123.

Tomlinson

(1969) A geographic information system for regional planning. Journal of Geography 78(1): 45–48.

124.

Tzachor

Sabri

Richards

, et al. (2022) Potential and limitations of digital twins to achieve the sustainable development goals. Nature Sustainability 5(10): 822–829.

125.

Uschold

Gruninger

(1996) Ontologies: principles, methods and applications. The Knowledge Engineering Review 11(2): 93–136.

126.

von Richthofen

Herthogs

Kraft

, et al. (2022) Semantic city planning systems (SCPS): a literature review. Journal of Planning Literature 37(3): 415–432.

127.

Wan

Jin

Echenique

, et al. (2024) From urban modelling to city digital twins - reflections from the applied urban modelling (AUM) symposia. Environment and Planning B: Urban Analytics and City Science.

128.

Wang

(2021) The seductive smart city and the benevolent role of transparency. Interaction Design and Architecture(s) 48: 100–121.

129.

Wang

Pan

Luo

(2019) Integration of BIM and GIS in sustainable built environment: a review and bibliometric analysis. Automation in Construction 103: 41–52. https://www.sciencedirect.com/science/article/pii/S0926580518309828

130.

Weil

Bibri

Longchamp

, et al. (2023) Urban digital twin challenges: a systematic review and perspectives for sustainable smart cities. Sustainable Cities and Society 99: 104862.

131.

Guan

(2024) Advancing intra and inter-city urban digital twins: an updated review. Journal of Planning Education and Research.

132.

Wang

, et al. (2021) Ontological knowledge base for concrete bridge rehabilitation project management. Automation in Construction 121: 103428.

133.

Yang

Kim

(2021) Urban digital twin applications as a virtual platform of smart city. Sustainability 12: 17.

134.

Yao

Nagel

Kunde

, et al. (2018) 3DcityDB - a 3D geodatabase solution for the management, analysis, and visualization of semantic 3D city models based on CityGML. Open Geospatial Data, Software and Standards 3(1): 5.

135.

Han

, et al. (2022) Developing human-centered urban digital twins for community infrastructure resilience: a research agenda. Journal of Planning Literature 38(2): 187–199.

136.

Yossef Ravid

Aharon-Gutman

(2022) The social digital twin: the social turn in the field of smart cities. Environment and Planning B: Urban Analytics and City Science 50(6): 1455–1470.

137.

Wang

Mao

, et al. (2021) A digital twin-based decision analysis framework for operation and maintenance of tunnels. Tunnelling and Underground Space Technology 116: 104125.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

4.61 MB