Semantic Web Technologies in Construction Facility Management: A Bibliometric Analysis and Future Directions

Abstract

The Facility Management (FM) sector is often hampered by data fragmentation and poor interoperability, hindering operational efficiency. To overcome these challenges, Semantic Web Technologies (SWTs) offer a robust framework by enabling machine-readable data integration. However, the application of SWTs in FM is underexplored. Therefore, this study systematically analyzes the structure, evolution, and emerging trends of SWT applications in FM to provide a clear research roadmap. A systematic literature review and bibliometric analysis were conducted on a final dataset of 107 academic articles using co-citation and keyword co-occurrence analysis. The results reveal that research in this domain has experienced exponential growth since 2021, with publications concentrated in high-impact journals. While a core group of influential authors has emerged, international collaboration remains fragmented. Thematic analysis identified a clear evolutionary trajectory from foundational concepts like BIM and ontologies toward applied Digital Twins and, most recently, advanced automation using Knowledge Graphs. This study provides a comprehensive roadmap for future inquiry, highlighting the need to mature technology integration, advance applied digital twins, and develop domain-specific ontologies to create more intelligent facilities. Ultimately, this study provides managers and policy-makers with a data-driven reference for strategically prioritizing investments in digitalization to achieve sustainable facility operation.

1. Introduction

The Architecture, Engineering, and Construction (AEC) sector plays a pivotal role in driving the global economy through the design and construction of buildings and infrastructure projects. However, its economic impact is most sustained during the post-construction phase [1]. The FM phase represents the longest and most cost-intensive phase of a building’s lifecycle, accounting for up to an estimated 75% of the total lifecycle expenditures [2]. This highlights the critical role of FM in the AEC sector. FM is closely tied to the long-term sustainability and functional viability of projects [3]. The purpose of FM can be defined through maintaining, enhancing, and increasing the effectiveness of an infrastructure’s physical and operational performance [4]. Effective FM is critical for ensuring that assets operate sustainably, directly impacting occupant well-being and the overall value of the asset. Unfortunately, the FM domain faces persistent challenges rooted in the nature of the AEC/FM industry: information fragmentation, poor interoperability between management systems, and difficulty obtaining accurate, up-to-date building data [5]. These challenges create significant operational hurdles with considerable consequences. For example, information fragmentation compels facility managers to adopt a reactive approach rather than a proactive approach in the maintenance of projects [6]. Likewise, poor data management increases the total cost of owning and operating a building [7].

Fortunately, in response to these challenges, the digital transformation of the AEC/FM sector has accelerated [8], with technologies such as Building Information Modeling (BIM), Internet of Things (IoT), and Digital Twins (DTs) gaining significant importance. BIM provides a rich, data-centric digital repository of a building’s physical and functional characteristics [9], while IoT sensors enable the collection of real-time operational data [10], forming the basis for dynamic DTs. The integration of these technologies promises to revolutionize FM by enabling data-driven decision-making and predictive maintenance [9]. However, the full potential of this integration is frequently curtailed by a fundamental obstacle: semantic interoperability. Data from building management systems and various IoT devices are often generated in heterogeneous formats, making automated reasoning a complex task [11]. For example, when data cannot flow seamlessly from the design software to the O&M systems, critical information is often lost or compromised. This data handover failure forces facility managers to engage in costly and error-prone manual data processing, hindering the potential that these technologies offer.

To address this semantic challenge, SWTs offer a robust framework. SWTs leverage foundational industry standards, such as Industry Foundation Classes (IFC) for data sharing (ISO 16739-1:2024) [12], by utilizing the corresponding ifcOWL ontology [13] to enable machine-readable data interpretation. The core of SWTs, including the Resource Description Framework (RDF), the Web Ontology Language (OWL), and the SPARQL Protocol, provides a standardized, logic-based method for formally defining, linking and querying data from disparate sources [14]. These technologies enable the creation of an integrated, machine-readable web of data that significantly improves data exchange and interoperability among diverse stakeholders [15]. This capability provides the “semantic backbone” necessary for advanced applications for FM.

Existing Literature

The existing literature has contributed significantly to understanding and implementing SWTs in the AEC/FM sector. These studies emphasize knowledge management, semantic enrichment, interoperability and information integration.

Table 1. Existing Reviews.

(Authors, Year)	Title	Area of Focus	Period Covered
Shen et al., 2024	Knowledge-based semantic web technologies in the AEC sector	Semantic Web Technologies in the AEC sector	2017–2023
Farghaly et al., 2023	The evolution of ontology in AEC: A two-decade synthesis, application domains, and future directions	Ontology applications across the entire AEC sector	2000–2021
Deng et al., 2022	Transforming knowledge management in the construction industry through information and communications technology: A 15-year review	ICTs for knowledge management (KM) in the AEC industry	2008–2022
Pauwels et al., 2017	Semantic web technologies in AEC industry: A literature overview	Development and application of semantic web technologies in the AEC domains	Up to 2016
Jia et al., 2024	Integration of Industry Foundation Classes and Ontology: Data, Applications, Modes, Challenges, and Opportunities	Integration of Industry Foundation Classes (IFC) and ontology across the whole lifecycle in the AEC/FM industry.	2011–2023
Jiang et al., 2023	Semantic enrichment for BIM: Enabling technologies and applications	Semantic Enrichment (SE) for BIM across the facility life cycle	Up to July 2022
Bloch 2022	Connecting research on semantic enrichment of BIM—review of approaches, methods and possible applications	Defining and reviewing approaches, methods, and applications of Semantic Enrichment for BIM	Up to 2021
Donkers et al., 2022	Semantic Web Technologies for Indoor Environmental Quality: A Review and Ontology Design	Semantic web technologies related to building topology, static, and dynamic properties for Indoor Environmental Quality (IEQ)	2009–2021
Ozturk 2020	Interoperability in building information modeling for AECO/FM industry	Interoperability in BIM for the AECO/FM industry	2004–2019
Karabulut et al., 2024	Ontologies in digital twins: A systematic literature review	The use of ontologies within Digital Twins (DTs) across various domains	Up to April 2023
Lei et al. (2022)	Ontology-Based Information Integration: A State-of-the-Art Review in Road Asset Management	Ontology development and implementation in road asset management	Up to 2019
Pandithawatta et al., 2024	Systematic Literature Review on Knowledge-Driven Approaches for Construction Safety Analysis and Accident Prevention	Knowledge-driven approaches (e.g., ontologies, knowledge graphs) in construction safety management	2000–2023
Lygerakis et al., 2022	Knowledge Graphs’ Ontologies and Applications for Energy Efficiency in Buildings: A Review	Ontologies and applications of Knowledge Graphs (KGs) for energy efficiency in buildings	Not specified
Peng et al., 2024	Knowledge graph of building information modelling (BIM) for facilities management (FM)	BIM for Facilities Management (FM)	2003 to 2023
Dinis et al., 2022	BIM and Semantic Enrichment Methods and Applications: A Review of Recent Developments	BIM-based Semantic Enrichment (SE) systems and applications, classified by BIM Use	2010–2020
Gilani et al., 2020	A review of ontologies within the domain of smart and ongoing commissioning	Building data ontologies within the Smart and Ongoing Commissioning (SOCx) domain	2014–2019
Lovell et al., 2024	Building Information Modelling Facility Management (BIM-FM)	Current implementation of BIM-FM for existing assets	2018–2024
Abideen et al., 2022	A Systematic Review of the Extent to Which BIM Is Integrated into Operation and Maintenance	Application of BIM in the Operation and Maintenance (O&M) phase	Up to 2022

provides a summary of recent studies that have reviewed the literature focused on either SWTs or FM. Recent reviews, such as those by Shen et al. [16] and Farghaly et al. [17], highlighted a trend toward using ontologies and BIM for data integration but limited their analysis to the technological depth required for the operational phase. Similarly, Deng et al. [18] discussed information and communications technology for knowledge management, and Pauwels et al. [19] provided a strategic roadmap for the interoperability challenge across the entire AEC sector. While these studies comprehensively review the application of SWTs within the AEC domain, a specific focus on FM is largely absent.

Other reviews further illustrate the academic focus on technological integration within the AECO/FM industry. Several studies have employed bibliometric or scientometric analyses to map research on broad themes, such as BIM interoperability [20] and the integration of IFC with ontologies [21]. A significant sub-field, semantic enrichment for BIM, has also been systematically reviewed by Jiang et al. [22] and Bloch [23]. Similarly, others have examined the role of SWTs in niche applications like Indoor Environmental Quality [24]. The application of knowledge-driven technologies, such as ontologies and Knowledge Graphs, has also been the subject of numerous systematic reviews in related fields. For example, Karabulut et al. [25] presented a broad overview of ontology use within DTs across various domains. More focused reviews have explored specific applications, such as ontology implementation for road asset management [26], knowledge-based approaches for construction safety [27], and the use of KGs to enhance energy efficiency in buildings [28].

Furthermore, some studies have directly addressed the intersection of digital technologies and FM. The broader BIM-FM landscape has been examined through multiple lenses, including bibliometric analysis [29], semi-structured reviews focused on existing assets [30], and systematic reviews on integration in the O&M phase [31]. Concurrently, more technologically specific studies have investigated the application of semantic methods. These include semi-systematic reviews of BIM-based Semantic Enrichment [32] and critical reviews of ontologies within the specialized domain of smart and ongoing commissioning [33].

Collectively, these works address the AECO industry broadly, highlight the value of structured knowledge in specialized areas, or confirm the importance of BIM in the operational phase. However, not a single review provides a holistic review that specifically synthesizes the broad spectrum of SWTs to address the core challenges in FM remains a notable gap.

It is also evident that more attention has been paid to the O&M phase for the BIM semantic enrichment research [22] and knowledge management [18]. This creates a need for a comprehensive overview of Semantic Web-based initiatives to better understand their application and impact on FM. Therefore, a comprehensive review is essential to identify developments in SWTs within FM over the course of time. The objectives of the study are threefold: (1) to map the intellectual landscape and research trajectory of the field; (2) to identify the most influential publications, authors, and collaborative networks; and (3) to synthesize the primary research themes and emerging trends. Ultimately, this study provides a comprehensive overview of the present state-of-the-art and concludes by proposing a set of future directions to guide subsequent research in this rapidly evolving domain. This offers valuable insights for researchers, software developers, and facility managers to make informed decisions.

The paper is structured into six sections. Section 2 presents the methodology of the study, including the main search terms and screening methods. Section 3 provides the bibliometric results, while Section 4 discusses the findings and also outlines future research directions. Section 5 includes the limitations of the study, and Section 6 provides the conclusions.

2. Materials and Methods

Knowledge and research domains are often investigated through two approaches: manual literature reviews and scientific metric analysis. Unlike a manual literature review, this study employed scientometric analysis to systematically handle large volumes of literature and mitigate the potential biases inherent in manual synthesis. Scientometric analysis is a quantitative method for structuring and presenting extensive bibliographic data [34]. Crucially, it utilizes graphical network maps to visualize the intellectual landscape, making it possible to identify core research domains, emergent themes, and knowledge hotspots with greater objectivity. The validity of this approach is demonstrated by its successful application in various previous studies [35,36,37,38,39].

To investigate the application of SWTs in FM, this study employed a quantitative research design. A comprehensive set of metadata was compiled using a Systematic Literature Review (SLR), following established protocols [40]. The SLR process was structured into three primary stages, as outlined by Tranfield et al. [41]: (1) planning the review, (2) conducting the review, and (3) reporting the findings. The research methodology is illustrated in Figure 1.

The planning stage involved a preparatory assessment of the literature to articulate the final research objectives. The conducting stage was executed by systematically collecting an unbiased corpus of literature in accordance with PRISMA guidelines [42]. This corpus was then subjected to a quantitative bibliometric analysis. The bibliometric analysis was performed using VOSviewer software (v 1.6.20) [43] to map the intellectual structure of the field. Specifically, co-citation analysis was used to identify foundational literature, bibliographic coupling to reveal the current state-of-the-art, and co-word analysis to map conceptual structures and identify emerging trends and future research trajectories [34].

In the final reporting stage, the synthesized results from the analysis are presented. This stage discusses the key findings, identifies research gaps, and outlines future directions for SWT research in FM. The detailed procedure for literature gathering, analysis, and reporting is further elaborated in subsequent sections.

2.1. Systematic Literature Review (SLR)

The SLR is primarily based on the PRISMA guidelines [44] (see the PRISMA checklist in Appendix A) and consists of various steps of literature selection and analysis to minimize subjectivity and ensure replicability in the analysis. In this article, SLR was divided into three stages: Scoping, Identification, and Screening, as shown in Figure 2.

The scoping stage involved defining the search databases. The primary database used for the systematic review was Scopus. Scopus is already a well-acknowledged online scientific database, known for its extensive coverage of construction-related research compared to other databases [45] and frequently used for literature analyses [46]. To ensure comprehensiveness, the results were supplemented by searches in Google Scholar and Web of Science to manually identify any missing articles, but it did not yield any additional unique articles, validating the robustness of the Scopus search protocol for this domain.

In the identification stage, the researchers defined a comprehensive search query:

{(“semantic web technolog*” OR “semantic web” OR “ontolog*” OR “knowledge graph” OR “linked data”) AND (“operation and maintenance” OR “asset management” OR “facilit* management” OR “building operation” AND (“building” OR “AEC” OR “construction” OR “built environment”)}

The asterisk (*) was used as a wildcard to include variations such as tense and plural forms. These terms were selected to match the scope and purpose of this study. The search was performed in June 2025 without any timeframe restrictions. The initial search yielded a total of 188 records. The search was then refined by three filters: the subject area was confined to “Engineering”, the language was restricted to English, and review articles were excluded to prevent skewing bibliometric results toward existing syntheses. After the initial filtration, the selection was narrowed down to 149 articles.

In the screening stage, a rigorous process was used to identify the most relevant research articles. First, automated methods were used to identify and remove duplicate records (27), reducing the initial dataset to 122 articles. This deduplication was performed using the automated feature in Rayyan software (https://www.rayyan.ai/, accessed on 1 June 2025), which applies an algorithm to match bibliographic fields like title, author, and year. Following this, a detailed manual review of these 122 articles was undertaken, including reading the title, abstract, introduction, and conclusion of each paper. An article was excluded if its primary focus was not on the O&M/FM phase of construction. This resulted in a final dataset of 107 articles.

2.2. Bibliometric Analysis

This study employed a bibliometric methodology, a quantitative approach for evaluating and mapping published research. In contrast to manual literature reviews, which are inherently prone to subjective bias [47], a robust bibliometric analysis provides an objective and reproducible overview of a research field’s intellectual structure and emerging trends [34,48]. To execute this analysis, the VOSviewer software was utilized to construct and visualize bibliometric networks, including co-citation, bibliographic coupling, and co-word analyses [43,46,49]. Following the framework established by Donthu et al. [34], the findings are synthesized with the specific analytical procedures detailed in the subsequent sections. Items were filtered based on representable occurrence thresholds, and normalized association strengths were computed to build co-occurrence and citation networks using the VOSviewer software.

2.2.1. Co-Citation Analysis

Co-citation analysis operates on the principle that publications frequently cited together by other works share a thematic relationship [50]. It is therefore a valuable technique for mapping the intellectual structure of a research domain [51]. Acknowledging its specific focus on a “past” domain, the present study employed co-citation analysis to complement other metrics to identify the foundational studies and historical motivations driving advancements in SWTs within the FM sector.

2.2.2. Bibliographic Coupling

Bibliographic coupling links documents that cite one or more of the same references, operating on the premise that a shared bibliography indicates thematic similarity [52]. In contrast to the retrospective nature of co-citation, bibliographic coupling is a forward-looking technique. As its analytical value is not dependent on accruing citations, it is highly effective for identifying current research fronts and nascent clusters, thereby representing the “present” of a research field [34]. Accordingly, this method was employed in the current study to map the contemporary landscape of SWTs research in FM.

2.2.3. Co-Word Analysis and Keyword Clustering

Co-word analysis, also termed “keyword co-occurrence analysis”, is a content analysis technique employed to map the conceptual structure of a research domain. The method operates by examining the co-occurrence frequency of author-provided keywords, positing that their frequent co-occurrence indicates a strong thematic relationship between concepts [34]. This analysis delineates a field’s intellectual landscape by identifying its core research topics, outlining conceptual boundaries, and elucidating the interconnections between them [52,53,54]. Furthermore, by tracking the prominence of specific keywords over time, this method can function as a forecasting tool to highlight nascent themes and suggest future research trajectories [34]. To ensure the accuracy of the analysis, duplicated keywords (e.g., “BIM”, “Building Information Modeling (BIM)”, “Building Information Modeling (BIM)”, “Facility Management”, “Facilities Management”, and so on) were merged in this study, resulting in a final list of 1022 keywords. For the temporal evaluation of keywords, an evolutionary analysis utilizing overlay visualization was conducted in VOSviewer to elucidate the developmental trajectory and emerging trends of SWTs in FM.

3. Results

This section presents and discusses the key findings of the research. To ensure clarity, the following subsections are structured to correspond with the methodological steps described previously.

3.1. Results of the Bibliometric Analysis

3.1.1. Metadata Overview

Figure 3 illustrates the annual publication trend, cataloging the documents into conference papers (n = 51), journal articles (n = 51), and book chapters (n = 5). The publication timeline, starting in 2007, reveals a field characterized by nascent and sporadic interest for over a decade, with no publications found in 2009 and 2010. However, the data shows a distinct inflection point around 2021, followed by a period of exponential growth in academic output, culminating in a peak of 23 publications in 2023. This sharp increase indicates that research into SWTs in FM has recently become a significant and rapidly expanding area of academic research. This surge can be primarily attributed to technological and industrial factors specific to the AEC/FM domain, particularly the increasing adoption of BIM standards by the industry and the widespread push toward DT implementation post-2020. These trends created an acute need for the semantic interoperability solutions uniquely offered by SWTs, thereby driving the academic inquiry. The equal distribution between conference papers and journal articles suggests that the field is both actively generating emerging research and simultaneously working to establish a core body of validated knowledge. The decline in 2025 can be attributed to the timing of this review, conducted mid-year (June 2025), with numbers likely to increase as the year progresses.

The publication landscape for SWTs in FM, shown in Figure 4, is highly concentrated, with research predominantly appearing in a select group of high-impact journals and key specialized conferences. The presence of reputable, high-impact, and often application-focused journals signifies that this research is recognized as a core topic within the construction informatics domain and underlines a key practical driver for this research: the urgent need for a clear theory-to-practice pipeline fostering sustainable and efficient building operations. In technology-driven disciplines, key conferences often serve as the primary venues for debuting innovative methods and prototypes. The presence of many articles in conference proceedings highlights the field’s dynamic and fast-moving character.

Figure 5 highlights the 15 key organizations driving this research, identifying the most impactful contributors by total citation volume. The analysis reveals that high-impact research in this domain is driven by specialized international research organizations across Europe and North America. This distribution suggests that the central scientific discourse has been predominantly shaped by Western institutions.

3.1.2. Network of Author Collaboration

The analysis of author collaborations serves as a proxy to reflect publication output and contributions in the SWTs for the FM domain. The overall dataset includes 357 authors who have contributed 107 articles. For the collaboration network (Figure 6), a minimum of two publications was required for an author to be included. In Figure 6, node size denotes publication volume, node color indicates temporal distribution, where the color gradient effectively represents each scholar’s activity level over time, and connections represent collaborative relationships. Analysis of this network indicates that authors with a substantial body of work have not formed dense collaborative clusters, suggesting that research collaboration in SWTs for FM has not yet established large-scale, interconnected networks at a global scale.

To identify the core authors in the field, this study applied a threshold of a minimum of three publications per author. This cutoff is a pragmatic choice to eliminate isolated nodes and ensure the resulting visualization focuses only on authors with a sufficient volume of output. Following this criterion, a group of 16 core authors was identified, as presented in

Table 2. Top sixteen core authors.

Author	Affiliation	Nation	TP	TC	TC/TP	TLS
Parlikad, Ajith	University of Cambridge, Cambridge,	United Kingdom	5	101	20.2	15
Merino, Jorge	University of Cambridge, Cambridge,	United Kingdom	4	101	25.3	14
Moretti, Nicola	University of Cambridge, Cambridge,	United Kingdom	4	101	25.3	14
Xie, Xiang	Newcastle University	United Kingdom	4	97	24.3	13
Bjørnskov, Jakob	University of Southern Denmark	Denmark	3	38	12.7	3
Chang, Janet yoon	University of Cambridge, Cambridge,	United Kingdom	3	97	32.3	12
Cruz, Christophe	University of Bourgogne	France	3	91	30.3	1
Enge-rosenblatt, Olaf	Fraunhofer Institute for Integrated Circuits IIS	Germany	3	20	6.7	0
Hagedorn, Philipp	Ruhr University Bochum, Germany	Germany	3	69	23.0	7
Hajdin, Rade	University of Belgrade	Serbia	3	56	18.7	3
Jradi, Muhyiddine	University of Southern Denmark	Denmark	3	38	12.7	3
König, Markus	Ruhr University Bochum	Germany	3	69	23.0	7
Li, Haijiang	Cardiff University	United Kingdom	3	14	4.7	0
Liu, Liu	Ruhr University Bochum	Germany	3	69	23.0	7
Nicolle, Christophe	University of Bourgogne	France	3	179	59.7	1
Tang, Shu	Xi’an Jiaotong-Liverpool University	China	3	11	3.7	0

Note. TP = Total Publications; TC = Total Citations; TC/TP = Total Citations per Publication; TLS = Total Link Strength.

. Collectively, these 16 individuals have authored 53 publications, accounting for 49.5% of the total literature in this domain. This aligns closely with the 50% threshold described by Price’s Law [55], which states that half of the output in a group is produced by the square root of the total number of contributors. In this case, with 357 authors, the square root is approximately 19, close to the 16 major contributors identified (each with at least three publications), indicating the emergence of a distinct and influential subgroup of researchers.

Among these core individuals, it is possible to recognize an influential research cluster including four researchers from the University of Cambridge, while the most cited author is affiliated with a French university, having 179 citations from 3 publications.

The geographic distribution of the core authors reveals a strong concentration of research in Europe, particularly in the United Kingdom, Germany, Denmark and France, confirming the field’s strong European foundation. Individual researchers from Serbia and China complete the list of core authors.

3.1.3. Network of Nations or Areas

Research on SWTs in FM has been contributed by a collective of 32 nations or regions. To identify the most productive countries in co-authorship, this study set a minimum threshold of five documents per country, as also suggested in [56]. This threshold focuses the analysis on countries with significant scholarly contributions, which improves the clarity and interpretability of the resulting network. For this co-authorship analysis, Full Counting is used. In this method, a nation is associated with a publication if the institutional affiliation of any author on the document is listed within that country. This ensures that each country is fully credited with that publication and that a collaboration link is established between all pairs.

Figure 7 depicts the academic collaboration network between countries, based on co-authorship relationships. The thickness of the links is proportional to the number of joint publications, and the size of the nodes corresponds to the total number of published works.

Table 3. Top ten most productive regions.

R	Countries	TP	TC	TC/TP	TLS
1	China	21	193	9.2	4
2	United Kingdom	18	194	10.8	4
3	United States	13	321	24.7	4
4	Germany	12	129	10.8	2
5	Denmark	8	183	22.9	2
6	Canada	6	138	23.0	4
7	France	6	246	41.0	0
8	Australia	5	156	31.2	3
9	Netherlands	5	185	37.0	4
10	South Korea	5	149	29.8	1

also presents the top-performing countries, with their publication counts, total citations, average citations per document, and total link strength. Results suggest that the field’s leading edge is driven by a strategic Asia-Europe axis (China, UK and the Netherlands), likely leveraging complementary expertise and funding programs like HORIZON 2020. In contrast, the below-average collaboration from scientifically advanced countries like Norway, Switzerland, and South Korea may indicate a more specialized or niche research focus that is less reliant on broad international partnerships. Notably, France and Taiwan showed no collaboration with other countries within this dataset, suggesting a focus on domestic research or an isolated position in the international collaborative landscape for this topic.

3.1.4. Co-Citation of Articles

Co-citation analysis reveals the foundational knowledge base upon which SWT research is built in FM.

Table 4. Most co-cited articles.

Authors & Year	Type	Cited Reference	TC	TLS
Pauwels & Terkaj, 2016	Journal Article	Express to owl for construction industry: towards a recommendable and usable ifcowl ontology	18	30
Boje et al., 2020	Journal Article	Towards a semantic construction digital twin: directions for future research	11	17
Pauwels et al., 2017	Journal Article	Semantic web technologies in AEC industry: a literature overview, automation in construction	10	22
Becerik-Gerber et al., 2012	Journal Article	Application areas and data requirements for Bim-enabled facility management.	8	13
Corry et al., 2015	Journal Article	A performance assessment ontology for the environmental and energy management of buildings	5	11
Costa & Madrazo, 2015	Journal Article	Connecting building component catalogues with Bim models using semantic technologies: an application for precast concrete components	5	12
Kim et al., 2018	Journal Article	Integration of ifc objects and facility management work information using semantic web	5	10
Niknam & Karshenas, 2017	Journal Article	A shared ontology approach to semantic representation of Bim data	5	7
Noy & McGuinness, 2001	Technical Report	Ontology development 101: a guide to creating your first ontology	5	8
Teicholz, 2013	Book	BIM for facility managers	5	10

presents the most frequently co-cited articles, using a minimum threshold of five co-citations. For the sake of consistency, in this analysis, the threshold used in the previous one (Network of Nations, Section 3.1.3) was kept constant to filter out incidental co-citations and ensure that the analysis focuses only on documents with a strong thematic relationship.

According to the analysis, the development of ontologies and SWTs for the AEC/FM industry emerged as the most frequently co-cited topic, forming its theoretical backbone. This is highlighted by the significant citation counts for seminal works such as Pauwels & Terkaj [57] on ifcOWL ontology and Pauwels et al. [19] on semantic technologies in the AEC industry. This signifies that the central scientific challenge preoccupying this community is the creation of a common, machine-readable language for building data.

Building upon this foundational pursuit, the analysis highlights the key application-oriented research fronts that translate this theory into practice. Influential research streams, such as semantic digital twins [11] and BIM-enabled facility management [1], represent the logical evolution of this core mission, showing where these sophisticated data structures are being deployed.

Furthermore, foundational guides showed critical direction for the field. The technical report “Ontology development 101” by Noy & McGuinness [58] and the book “BIM for facility managers” by Teicholz [2] were frequently cited, indicating their role as essential resources for researchers integrating advanced data structures with facility management.

3.1.5. Co-Citation of Sources

Academic studies on SWTs in FM are published across a spectrum of specialized journals. To identify the core scholarly venues that form the intellectual foundation of this field, a co-citation threshold of a minimum of 20 citations was applied. This relatively high threshold was deliberately chosen to isolate the elite tier of journals with the most significant and consistent impact, ensuring the resulting network is both robust and clearly interpretable. This rigorous filtering process yielded ten core journals as the most influential sources within the dataset. Their co-citation network is shown in Figure 8. These selected journals encompass a range of interconnected research fields, including civil and construction engineering, building energy performance, construction automation and advanced engineering informatics. This observation signifies that research in this domain has a strong interdisciplinary foundation, merging traditional engineering with computational and energy sciences.

3.1.6. Bibliometric Coupling of Documents

The present state of the implementation of SWTs in FM was identified using the bibliographic coupling.

Table 5. Articles sharing more than 40 references.

Authors	Type	Title	TC	TLS
Wetzel (2015)	Journal Article	The use of a BIM-based framework to support safe facility management processes	189	5
Mignard (2014)	Journal Article	Merging BIM and GIS using ontologies application to urban facility management in ACTIVe3D	151	0
Ozturk (2020)	Journal Article	Interoperability in building information modeling for AECO/FM industry	123	17
Kim (2018)	Journal Article	Integration of ifc objects and facility management work information using Semantic Web	101	11
Hosamo (2023a)	Journal Article	Digital Twin framework for automated fault source detection and prediction for comfort performance evaluation of existing non-residential Norwegian buildings	74	35
Xie (2023b)	Journal Article	Digital twin-enabled fault detection and diagnosis process for building HVAC systems	73	15
Hosamo (2023b)	Journal Article	Improving building occupant comfort through a digital twin approach: A Bayesian network model and predictive maintenance method	65	38
Schevers (2007)	Journal Article	Towards digital facility modelling for Sydney Opera House using IFC and semantic web technology	56	0
Hagedorn (2023)	Journal Article	BIM-Enabled Infrastructure Asset Management Using Information Containers and Semantic Web	53	3
Jiang (2023)	Journal Article	Intelligent control of building fire protection system using digital twins and semantic web technologies	52	10
Hu (2021)	Journal Article	Building energy performance assessment using linked data and cross-domain semantic reasoning	52	15
Lee (2013)	Journal Article	An integrated approach to intelligent urban facilities management for real-time emergency response	45	4
Pilehchian (2015)	Journal Article	A conceptual approach to track design changes within a multi-disciplinary building information modeling environment	44	0
Arslan (2019)	Journal Article	Semantic trajectory insights for worker safety in dynamic environments	42	1
Xie (2023a)	Journal Article	Knowledge map and forecast of digital twin in the construction industry: State-of-the-art review using scientometric analysis	40	4

lists the 16 documents with over 40 shared citations, highlighting them as key references shaping future developments. The overarching theme among these studies is the integration of digital technologies, such as BIM, Digital Twins, and the Semantic Web, for the lifecycle management of the built environment.

According to the analysis, the studies are diversely distributed across several research streams. A significant focus is on BIM-based frameworks for facility management, with key papers exploring applications for safety [59] and real-time emergency response [60]. Another prominent theme is the utilization of Digital Twins in building operations, with studies addressing fault detection and diagnosis [61,62], occupant comfort [63], and intelligent fire protection systems [64]. Furthermore, a substantial body of work is dedicated to data integration and interoperability, with influential articles on integrating IFC and facility management data using the Semantic Web [65] and leveraging linked data for energy performance assessment [66].

3.1.7. Network of Co-Occurrence Keywords

Analyzing the co-occurrence of high-frequency keywords is a common method to identify a field’s primary research areas [67]. Using VOSviewer, a co-occurrence network (Figure 9) was generated with a minimum frequency threshold of five, as suggested in [56]. This threshold was also adopted because of the representable network of keywords. In this network, node size reflects keyword frequency, connecting lines indicate association strength, and different colors delineate thematic clusters.

To gain deeper insight into the specific topics driving this field,

Table 6. Top fifteen keywords.

R	Keywords	Frequency	TLS
1	Ontology	52	290
2	BIM	48	304
3	Architectural Design	41	302
4	Office Buildings	35	236
5	Information Management	31	197
6	Facility Management	28	201
7	Operation and Maintenance	21	100
8	Decision-Making	20	139
9	Digital Twin	19	110
10	Information Theory	19	163
11	Knowledge Graph	19	73
12	Life Cycle	19	141
13	Maintenance	17	105
14	Semantic Web	17	118
15	Semantics	17	97

lists the 15 high-frequency keywords that appear in more than 15 publications. Both the co-occurrence network (Figure 7) and Table 6 highlight that Ontology (52), BIM (48), Architectural Design (41), Office Buildings (35), Information Management (31), Facility Management (28), Operation and Maintenance (21), Decision-Making (20), Digital Twin (19), Information Theory (19), Knowledge Graph (19), Life Cycle (19), Maintenance (17), Semantic Web (17), and Semantics (17) are the pivotal terms within this research domain.

In addition, based on the analytical findings of the VOSviewer software, the keywords were categorized into four distinct clusters as presented in

Table 7. Keyword clusters of articles and their research themes.

Cluster	Keywords	Research Themes
Red Cluster	Architectural design; BIM; Building life cycle; Buildings; Construction industry; Facility management; Frequency modulation; Information modeling; Information theory; Interoperability; Knowledge-based system; Life cycle; Office buildings; Project management	Topic: Information Management and Interoperability across the Building Lifecycle Applications: Architectural design, project management, facility management Lifecycle phases: Design, Construction, Operations, End-of-life Tools: Building Information Modeling (BIM), Knowledge-Based Systems
Green Cluster	Automation; Building operations; Data integration; Digital twin; Electronic data interchange; Energy efficiency; Fault detection; Intelligent buildings; Internet of Things; Semantic web technology	Topic: Digital Twins for Automated and Efficient Building Operations Applications: Real-time monitoring, energy efficiency, predictive maintenance, fault detection Lifecycle phases: Building Operations Tools: Internet of Things (IoT), Digital Twin, Semantic Web Technologies
Blue Cluster	Data mining; Decision-making; Knowledge graph; Knowledge management; Maintenance; Operation and maintenance; Semantics	Topic: Knowledge-Based Decision-Making for Maintenance Applications: Optimized scheduling for operation and maintenance, equipment diagnostics, data-driven decision support Lifecycle stages: Operation and Maintenance Tools: Data Mining, Knowledge Graphs, Knowledge Management Systems
Yellow Cluster	Asset management; Asset management system; Information management; Infrastructure asset management; Linked data; Ontology; Semantic web	Topic: Semantic Web Technologies for Integrated Infrastructure Asset Management Applications: Unified information management, holistic asset tracking and analysis Lifecycle stages: Long-term Management and Operations Tools: Ontology, Linked Data, Semantic Web

. Then these clusters were further mapped into specific research themes with their application area, life cycle phase and tools required for research themes. The detailed discussion of the four main research themes is presented in subsequent sections.

Information Management and Interoperability Across the Building Lifecycle

Cluster #1, marked in red, is the largest cluster, comprising 14 keywords. The central theme revolves around Information Management and Interoperability across the Building Lifecycle. This is strongly indicated by the high frequency and connectivity of keywords such as “BIM”, “architectural design”, and “information management”. These keywords, along with “buildings” and “office buildings”, underscore the focus on applying structured information processes to building projects. The keyword “BIM” is particularly central, acting as a major hub that connects design, construction, and management phases, reflecting its role as a foundational digital tool in the modern construction industry [7].

Furthermore, the theme is reinforced by keywords that span the entire project timeline, such as “building life cycle” and “life cycle”. The inclusion of “facility management” and “project management” highlights the research community’s focus on extending information management beyond the design and construction phases into the operational life of assets. The keyword “interoperability” is critical, pointing to the persistent challenge and research interest in ensuring seamless data exchange between different systems and stakeholders throughout a project’s lifecycle [20].

Digital Twins for Automated and Efficient Building Operations

Cluster #2, marked in green, consists of 10 keywords and clearly centers on the theme of Digital Twins for Automated and Efficient Building Operations. This cluster represents the shift towards real-time, data-driven management of buildings. The most influential keywords are “digital twin” and “intelligent buildings”, which together define the core concept of creating virtual replicas of physical buildings to optimize performance. The keyword “Internet of Things” represents the enabling technology for this theme, providing the sensor networks required to feed live data into these digital models [68,69,70].

The practical applications of this theme are captured by keywords like “building operations”, “energy efficiency”, and “fault detection”. These terms reflect a strong research interest in leveraging digital twin and IoT technologies to automate monitoring and reduce energy consumption. The presence of “semantic web technology” and “data integration” further suggests that a key research challenge within this theme is the need to structure and integrate heterogeneous data streams from various IoT devices to create meaningful and actionable insights [65,71,72,73].

Knowledge-Based Decision-Making for Maintenance

Cluster #3, marked in blue, contains 7 keywords focused on the theme of Knowledge-Based Decision-Making for Maintenance. This cluster highlights the move towards more intelligent and data-informed strategies for maintenance. The prominence of keywords such as “operation and maintenance”, “decision-making”, and “maintenance” establishes the core application domain. These keywords are tightly interwoven, indicating that research in this area is primarily concerned with improving the quality and efficiency of maintenance-related decisions.

The methodologies driving this theme are represented by “knowledge management”, “knowledge graph”, and “data mining”. This suggests a focus on capturing expert knowledge and extracting patterns from historical and operational data to support maintenance activities. The use of knowledge graphs, in particular, points to an advanced approach where complex relationships between building components, maintenance procedures, and performance data are formally modeled to provide a deeper understanding for decision-makers [28,29]. The keyword “semantics” reinforces this, emphasizing the importance of shared meaning and context in creating robust knowledge-based systems for maintenance.

Semantic Web Technologies for Integrated Infrastructure Asset Management

Cluster #4, marked in yellow, is a highly specialized cluster of 7 keywords centered on the theme of SWTs for Integrated Infrastructure Asset Management. This theme is dominated by the keyword “ontology”, which has the highest occurrence count and total link strength across all keywords, signifying its foundational role in this research area. Ontology provides the formal, explicit specification of concepts and relationships needed to overcome data silos and achieve true integration in asset management [74,75,76,77].

The keywords “semantic web” and “linked data” are direct applications of ontological frameworks. They enable the creation of a web of data where information from disparate sources can be connected and queried in a standardized way. The application area is defined by “asset management”, “asset management systems”, and “infrastructure asset management”. This indicates that the research is focused on managing large-scale, complex infrastructure assets, where information integration is a significant challenge.

3.1.8. Temporal Evaluation of Keywords

Price [55] suggests that as new research results emerge, the relevant knowledge network will become increasingly dense. Keywords with a frequency of five or more were extracted from the 107 articles. An overlay co-occurrence map was then generated to reveal the distribution of research hotspots over time, as shown in Figure 10.

The analysis reveals that the research landscape’s origins, from approximately 2019 to 2020 (keywords in purple and dark blue), are rooted in foundational concepts of the built environment. This initial phase focused on establishing core digital frameworks through BIM, architectural design, ontology, and information theory, with a strong emphasis on managing the entire building life cycle and facility management.

As the field matured between 2020 and 2022 (keywords in cyan and green), the research focus pivoted towards digital integration and operational intelligence. This period saw the rise in concepts like linked data, intelligent buildings, and data integration, aimed at enhancing building operations such as fault detection. Concurrently, there was a growing emphasis on data-driven decision-making for operation and maintenance.

Since late 2022 (keywords in yellow), the most recent trend has shifted towards advanced data structuring and automation. The emergence of terms like knowledge graph, data mining, and automation indicates a move towards creating sophisticated, interconnected knowledge systems to automate complex processes. Overall, a clear trajectory can be observed, moving from establishing fundamental digital models to leveraging them for intelligent operations and, most recently, to automating decision-making through advanced knowledge systems.

3.2. Future Research Avenues

The primary aim of co-citation analysis is to identify the intellectual foundations and emerging research fronts in the research domain. This study employed VOSviewer to generate a co-citation analysis map for references published between 2020 and 2025. A threshold of at least three citations per cited reference was established, resulting in the selection of 25 papers for the analysis as presented in

Table 8. Top 25 articles for research directions.

Authors & Year	Cluster	Title	TC	TLS
Becerek-Gerber et al., 2012	Red	Application areas and data requirements for BIM-enabled facility management	5	18
Corry et al., 2015	Red	A performance assessment ontology for the environmental and energy management of buildings	5	21
Dave et al., 2018	Red	A framework for integrating BIM and IoT through open standards	4	21
Donkers et al., 2018	Red	Semantic web technologies for indoor environmental quality: A review and ontology design	3	14
Khajavi et al., 2019	Red	Digital twin: vision, benefits, boundaries, and creation for buildings	4	11
Succar, 2009	Red	BIM framework: a research and delivery foundation for industry stakeholders	3	14
Tang et al., 2019	Red	A review of BIM and the internet of things (IoT) devices integration: present status and future	4	21
Zhang et al., 2015	Red	Ontology-based semantic modeling of construction safety knowledge: towards automated safety planning	3	18
Boje et al., 2020	Green	Towards a semantic construction digital twin: directions for future research	11	36
Bortolini & Forcada, 2018	Green	Building Inspection System for Evaluating the Technical Performance of Existing Buildings	4	6
Grieves, 2015	Green	Digital twin: manufacturing excellence through virtual factory replication	3	4
Hosamo et al., 2022	Green	A digital twin predictive maintenance framework of air handling units	4	11
Lu et al., 2020	Green	Digital twin-enabled anomaly detection for built asset monitoring in operation and maintenance	4	8
Pauwels et al., 2017	Green	Semantic web technologies in AEC industry: a literature overview	8	22
Teicholz, 2013	Green	BIM for facility managers	3	11
Zhao et al., 2022	Green	Developing a conceptual framework for the application of digital twin technologies	3	8
Costa & Madrazo, 2015	Blue	Connecting building component catalogues with Bim models using semantic technologies	4	17
Kim et al., 2018	Blue	Integration of IFC objects and facility management work information using semantic web	4	7
Niknam & Karshenas, 2017	Blue	A shared ontology approach to semantic representation of Bim data	3	4
Noy & McGuinness, 2001	Blue	Ontology development 101: a guide to creating your first ontology	3	3
Zhong et al., 2018	Blue	Ontology-based framework for building environmental monitoring and compliance checking under BIM environment	3	6
Pauwels et al., 2017	Yellow	Enhancing the ifcowl ontology with an alternative representation for geometric data	4	14
Pauwels & Terkaj, 2016	Yellow	Express to owl for construction industry: towards a recommendable and usable ifcowl ontology	13	43
Ren et al., 2019	Yellow	Building an ontological knowledgebase for bridge maintenance	3	2

. The threshold, in line with the thresholds used in similar studies (e.g., [78]), is chosen to filter out the optimal number of studies for in-depth analysis. The final network of citations is illustrated in Figure 11, where the 25 most frequently cited references have been categorized into four distinct groups. These clusters signify promising research trends for implementing SWTs in FM.

3.2.1. Integrating Foundational Technologies for Smart Facility Management

Cluster #1 (Red) brings together foundational research on the integration of BIM with enabling technologies like the Internet of Things (IoT), Semantic Web, and early Digital Twin concepts. This cluster acts as a hub, connecting various technological threads. Key intellectual pillars established the basis for data-driven FM by defining specialized ontologies [79], outlining data requirements for BIM-enabled FM [1], and developing frameworks for integrating BIM and IoT using open standards [80]. Analytically, this foundational cluster defines the technological prerequisite for advanced FM, emphasizing that solving data convergence is the initial barrier to smart facility operation. Studies explored the use of SWTs for specific goals, such as automating safety planning and assessing energy performance in buildings [81,82]. The cluster also includes work that outlined the vision, benefits, and boundaries of Digital Twins [10]. This research directly validates the strategic requirement for FM managers to shift from stored data systems to an integrated Digital Twin foundation, enabling the critical transition from reactive maintenance schedules to proactive, condition-based maintenance strategies that drive operational efficiency.

3.2.2. Applied Digital Twins for Lifecycle Optimization

Cluster #2 (Green) is clearly centered on the development and practical application of Digital Twins for facility operations and maintenance. This cluster incorporates foundational concepts, widely adapted for the built environment. The research is highly application-oriented, focusing on critical FM functions. For example, studies focus on DT-based predictive maintenance for air handling units [83] and anomaly detection in asset monitoring [84]. Other research highlights the technology’s potential by developing a conceptual framework to revamp job hazard analysis, demonstrating its direct impact on site safety [85]. Analytically, this cluster demonstrates the maturity of the research domain in moving beyond foundational concepts to tangible, high-value FM applications. The inclusion of detailed literature on SWTs and BIM for FM [2] confirms the necessary technological convergence supporting these applications. The next research frontier in this area involves creating scalable and interoperable DT frameworks that can move beyond single-use cases (e.g., one HVAC system) to optimize entire building portfolios. For FM practitioners, the emphasis on predictive analytics provides a clear strategic directive: using DTs to maximize asset uptime and transition capital expenditure planning from reactive replacements to data-driven forecasting.

3.2.3. Semantic Web and ifcOWL for Data Structuring

Cluster #3 (Blue) focuses intensely on the use of formal ontologies to provide a structured, machine-readable representation of building information. The cornerstone of this cluster is the ifcOWL ontology and the methodology by Noy and McGuinness [58] for ontology development. Application-focused studies, such as Zhong et al. [86] on ontology-based compliance checking and Niknam and Karshenas [87] on semantic representation, demonstrate the practical value of this approach. Similarly, Kim [65] integrated the IFC objects with FM work using SWTs, and Costa [88] connected the component catalogs with BIM models using SWTs. Analytically, this focus confirms that achieving true semantic interoperability remains a central challenge and that formal knowledge representation is the primary research solution. Future work should focus on developing modular, extensible domain ontologies that can complement ifcOWL, specifically for complex FM tasks like automated fault detection and diagnosis (FDD) rule generation and dynamic maintenance scheduling. For FM practitioners, the output of this cluster is crucial, as it directly enables the creation of expert systems capable of reasoning about building data.

3.2.4. Ontologies for Specialized Infrastructure

Cluster #4 (Yellow) represents a highly specialized and emerging research front: the development of domain-specific ontologies for critical infrastructure. This cluster is defined by the work of Ren et al. [89], which focuses on building an ontological knowledge base specifically for bridge maintenance. While small, this cluster is significant, as it demonstrates a targeted deep dive, adapting the broader principles of semantic modeling to the unique requirements of a specific infrastructure asset. Analytically, the emergence of this cluster indicates that the generalized semantic solutions (like ifcOWL) are insufficient for complex, non-standard assets, requiring specialized knowledge representation to capture unique failure modes and maintenance requirements. This cluster highlights a significant opportunity to develop specialized ontologies for other critical infrastructure, such as tunnels, railways, and water treatment plants, enabling semantic integration with sensor networks for structural health monitoring and predictive lifecycle management. For FM practitioners managing these specialized assets, this research is crucial for advancing safety compliance and structural longevity by moving beyond standard maintenance schedules to data-driven, risk-based operational models.

4. Discussion

This research consolidates the fragmented knowledge of SWTs in the FM sector into a structured intellectual framework. Results presented so far allow the application of SWTs in FM to be systematically arranged in the comprehensive knowledge map, as illustrated in Figure 12. The subsequent sections discuss the study in detail.

4.1. State and Trajectory of SWT Research in FM

By examining the metadata, collaboration networks, and keyword co-occurrences, several key observations can be made about the current state and trajectory of the field:

(1)

The field is experiencing rapid and recent growth. As shown in Figure 3, the application of SWTs in FM was a niche topic for over a decade. However, a distinct inflection point around 2021 has led to a period of exponential growth in publications, signaling that the field has recently matured into a significant and rapidly expanding area of academic inquiry.

(2)

Research is published in high and impactful journals. The concentration of co-cited articles in top-tier journals (Figure 8) indicates that research on SWTs in FM has achieved a high standard of quality and is recognized by the broader AEC/FM community.

(3)

A core group of researchers is emerging, but global collaboration is still developing. While the analysis in Table 2 identifies a core group of 16 authors responsible for nearly 50% of the literature, the author collaboration network (Figure 6) suggests that large-scale, interconnected research teams have not yet formed. Similarly, the country-level analysis (Figure 7) shows strong international collaboration between leaders like China and the UK but also reveals that some productive nations like France appear to work in relative isolation, indicating that the collaborative potential of the field is still developing.

(4)

The research landscape is geographically concentrated but internationally impactful. While China leads in publication volume, institutions from the USA and the Nordic countries show the highest citation impact. This suggests that a cluster of highly influential foundational research originates from the USA and Europe, potentially due to well-established research ecosystems, funding availability and extensive international collaboration networks, which together help shape the direction of the field globally. It is worth noting that this concentration of academic influence contrasts with the landscape of large-scale implementation. This is because of leadership in Asian countries, particularly China and Japan, which invested in ambitious city-level projects characterized by strong government backing and the rapid integration of advanced technologies. A prime example is the Sino-Singapore Tianjin Eco-city, which showcases this model of development.

(5)

The intellectual focus has evolved from foundational models to intelligent applications. The temporal analysis of keywords (Figure 8) reveals a clear trajectory. Early research (2019–2020) focused on establishing fundamental concepts like BIM and ontology. The focus then shifted (2020–2022) to integration challenges and operational intelligence (e.g., linked data, fault detection). The most recent trend (2022 onwards) is towards advanced automation through knowledge graphs and data mining, showing a clear path from digital modeling to data-driven, intelligent systems.

4.2. Practical Implications of Semantic Web Technologies in FM

Compared to traditional FM approaches, beset by data fragmentation and information silos as outlined in the Introduction, SWTs offer significant practical implications. The research themes identified in the keyword analysis (Table 6) point to several key benefits:

4.2.1. Information Integration and Interoperability

SWTs directly address the core industry challenge of poor interoperability by creating a common, machine-readable language that breaks down data silos between disparate systems like BIM, building automation systems, and maintenance software. Rather than replacing existing workflows, SWTs act as a powerful semantic layer that enhances key industry standards. For instance, while ISO 19650 [90] defines the process for information management within a Common Data Environment (CDE) [91] and COBie facilitates the structured handover of asset data [92], these standards do not inherently make the data understandable to machines. SWTs bridge this critical gap by applying an ontology like the Brick Schema [93], and the data from these different sources is transformed from isolated information into an interconnected and queryable knowledge graph. This creates a truly unified web of building data, ensuring that critical information from the design and construction phases remains not just accessible but contextually rich and usable throughout the long operational lifecycle.

4.2.2. Automated and Efficient Building Operations

The integration of SWTs with DTs and IoT enables a paradigm shift from reactive to proactive and automated building operations. By providing a semantic layer to real-time sensor data, SWTs allow for sophisticated applications like automated Fault Detection and Diagnosis (FDD), optimized energy efficiency, and predictive maintenance scheduling. Ontologies like the Brick Schema [93] are a critical enabler in this process, providing the necessary context to interpret raw data and model the complex relationships between assets and systems [93,94]. The impact is quantifiable and significant; broader studies integrating semantic digital twins report reductions in maintenance costs by 25% and energy consumption by 20% [95]. Specific case studies further validate these benefits, such as a deployment at the UCSD campus that used Brick for plug load control and achieved energy savings of up to 86% [96]. This shift not only reduces operational costs but also fundamentally enhances building performance and occupant comfort.

4.2.3. Knowledge-Based Decision-Making

By structuring building data as a knowledge graph, SWTs elevate data from simple information into actionable knowledge, moving facility managers beyond the limitations of simple data dashboards. Using the SPARQL query language, managers can perform complex, pattern-based queries that fuse static asset data (e.g., from COBie) with dynamic operational data to diagnose root causes and inform strategic decisions. For instance, a single query could identify all chillers with expiring warranties, retrieve their complete maintenance history, and count recent high-priority fault alerts to create a data-driven case for capital replacement planning. Furthermore, federated query capabilities allow the internal building graph to be linked with external web data sources, such as supplier catalogs or weather services, enabling an unprecedented level of automated, context-aware decision support.

4.2.4. Holistic and Scalable Asset Management

The frameworks enabled by SWTs provide a scalable method for managing entire portfolios of infrastructure assets, moving beyond the single-building level. This scalability is formally underpinned by standards like ISO 19650, which establishes the process for creating a continuous digital thread from the Project Information Model (PIM) to the operational Asset Information Model (AIM) [91]. This lifecycle data forms the basis of a Semantic DT, an integrated knowledge base that serves as a comprehensive single source of truth. Crucially, because applications for analytics and control are built upon open, standardized ontologies, they become vendor-agnostic and portable. This allows organizations to gain a holistic view of their assets and track performance consistently.

4.3. Prospective Research Directions

The analysis of emerging research clusters (Figure 9) provides a clear roadmap for future inquiry. Researchers and practitioners should focus on the following promising areas to advance the application of SWTs in FM:

The first cluster highlights an ongoing need to strengthen the foundational links between BIM, IoT, and SWTs. Future work should move beyond conceptual frameworks to develop and validate robust, open-standard workflows for “live” data integration. This includes addressing long-standing challenges by creating specialized ontologies for specific FM tasks. Specifically, research should answer: How can open-standard workflows be validated to ensure real-time semantic consistency? And what core classes must be included in specialized ontologies to enable automated safety planning and performance assessment using live data? For the practitioners, this research directly enables a core day-to-day task: moving from manual scheduled inspections to automated maintenance alerts.

The second cluster shows that the Digital Twin concept is a major driver of innovation. While progress has been made in areas like predictive maintenance and anomaly detection, a significant opportunity remains to expand the use of digital twins across the entire asset lifecycle. Future research must focus on answering: What semantic digital twin frameworks can effectively optimize the long-term strategic decisions across entire building portfolios? And what predictive models are required to move digital twin maintenance applications from single-asset failure prediction to portfolio-wide resource optimization? In practice, this translates to optimizing resource allocation by predicting equipment failures, which reduces reactive work orders and allows FM staff to focus on strategic, high-value tasks.

The third cluster underscores the critical importance of formal data structures. The ifcOWL ontology is a cornerstone, but its full potential is unrealized. Future studies should focus on answering: How can ifcOWL be modularly extended with domain-specific sub-ontologies (e.g., for warranty tracking or space management) to support automated maintenance scheduling? And what technical architecture is required to develop intuitive “low-code” query tools that allow non-expert facility managers to utilize structured semantic data? This research is vital for the front-line manager, as it enables simple, natural language queries against the Digital Twin (e.g., “Show me all assets due for maintenance this week”) without needing specialized programming skills.

The fourth and most specialized cluster points towards a vital new frontier. The work on a bridge maintenance ontology serves as a powerful example. There is a pressing need for research dedicated to answering: How can detailed, standardized ontologies be developed for highly complex asset types (e.g., hospitals, airports, or utility networks) to model their unique operational and compliance requirements? And what semantic reasoning mechanisms are necessary to advance safety compliance and structural longevity in specialized infrastructure? The practical value here is in ensuring specialized assets are managed safely and compliantly; for example, a dedicated hospital ontology ensures that crucial medical equipment is always visible within the Digital Twin for rapid facility response.

5. Limitations of the Study

This study is intended to provide a comprehensive overview of the current research landscape, while acknowledging the limitations discussed below.

The scope of this analysis is defined by the data source. The study is based on English-language publications indexed in the Scopus database, which is selected for its comprehensive and reputable coverage of the international scientific literature. Consequently, research published in other languages or indexed exclusively in regional databases, such as the Chinese Science Citation Database (CSCD) and the Chinese Social Sciences Citation Index (CSSCI), falls outside the boundaries of this review.

Second, the findings are inherently constrained by the search query used for data retrieval. While the query was crafted to be comprehensive, it may not have captured all relevant articles that use different terminology to describe similar concepts.

Furthermore, the bibliometric methods themselves have inherent limitations. The keyword analysis relies on the quality and consistency of author-provided keywords, while co-citation analysis tends to favor older, more established publications, potentially overlooking more recent or niche research clusters [34]. Also, it focuses on mapping the intellectual structure of academic research, which may differ from the geography of practical implementation. For instance, while our findings highlight a concentration of foundational research in Europe, this academic landscape does not fully capture the significant, large-scale deployment of smart building projects in Asia.

Finally, it is worth noting that this study represents a snapshot of the research field up to June 2025, and given the rapid growth identified in the results, new publications will continue to shape the field’s trajectory.

6. Conclusions

The effective management of the built environment’s operational phase is frequently undermined by persistent issues of data fragmentation and a lack of interoperability. This study employed bibliometric methods to analyze the global research landscape of SWTs applied in the FM domain. The analysis aimed to (1) map the intellectual structure and evolutionary trajectory of the field; (2) identify the influential authors, publications, and collaborative networks that have shaped the discourse; and (3) synthesize the primary research themes and emerging trends, thereby revealing critical knowledge gaps and informing a clear agenda for future studies. By conducting a systematic bibliometric analysis of 107 articles, this study has revealed key contributors, the conceptual structure, and the evolutionary trajectory of SWTs in FM.

Results established that SWTs provide a robust framework to address the aforementioned challenges, enabling a shift from reactive to data-driven, intelligent facility management. The findings indicate that after a decade of nascent interest, research in this domain has been experiencing exponential growth since 2021, a trend underscored by its increasing appearance in high-impact journals.

The keyword co-occurrence and temporal evaluation trace a clear trajectory of research: an initial focus on foundational concepts like BIM and Ontology (2019–2020) shifted toward operational intelligence and integration challenges (2020–2022) and has recently pivoted to advanced automation through Knowledge Graphs and data mining (2022 onwards).

The study provides a quantitative roadmap of the domain, validating the critical importance of SWTs for interoperability in the next generation of asset management solutions. The synthesis of knowledge hotspots and emerging research clusters points toward several strategic future avenues: developing robust “live” data integration workflows between BIM, Digital Twins, and the IoT; scaling Digital Twin applications from single assets to entire building portfolio optimization; and designing specialized, modular domain ontologies to support complex infrastructure and advanced knowledge-based maintenance decision-making. These research fronts are directly linked to the operational demands of BIM-based facility management practice and offer a clear and actionable agenda for the academic and industrial community.

Finally, the strategic adoption of SWTs is critical for guiding future efforts to create safer, more efficient, and more sustainable facilities, and this research provides a vital roadmap for industry uptake to unlock the full potential of SWTs.

Future work could build upon this study by incorporating a wider range of databases and languages and by employing advanced methods like AI-driven natural language processing to analyze the full text of articles for a deeper thematic analysis.