Evolving Trends in Smart Building Research: A Scientometric Analysis

Abstract

Background: Smart building, as an emerging building concept, has been a key driving force for the transformation and upgrading of the building industry; Methods: To better understand the latest research progress and trends in the field of smart building, this study uses CiteSpace 6.2.R4 bibliometric software to visualize, analyze, and interpret the literature related to the field of “Smart Building” in the WoS database from 2014 to 2023; Results: As a cross-sectoral and multidisciplinary field, smart building has received significant attention in recent years, with a rapid growth in the number of publications. International cooperation is strong, with China, the United States, and South Korea leading in the number of publications, but there is still room for enhanced collaboration among institutions. Keyword analysis shows that technology and humanized design are both crucial, and emerging technology has become the current research hotspot. Conclusions: The field of smart building has gained global attention, and more breakthroughs will be made in improving building efficiency, reducing energy consumption, and enhancing the user experience. This development is moving towards a smarter and more sustainable direction that will bring greater benefits to human life and the environment.

1. Introduction

In an industry that has not yet fully embraced digitalization, smart building technology is quietly overturning the rules of the traditional building game, prompting us to rethink the meaning of “building”. The deep integration of architecture and digitalization has gradually made smart construction technology a core element of the construction industry [1,2,3]. The 21st century is the most favorable stage for the development of smart construction technology [4]. In 2022, the European Commission released the “Horizon Europe” work program, emphasizing that research on digital transformation is key to achieving green transformation. Similarly, various countries have introduced a series of policies to promote smart construction technology [5]. The National Institute of Standards and Technology (NIST) in the United States has issued the “Digitalization Roadmap for the Construction Industry”, supporting research and the application of smart construction; the British government has established the Digital Built Britain project to promote the implementation of digital construction; and Germany has promoted Industry 4.0 in the construction field, emphasizing smart buildings and automated construction, and has set up a technology innovation fund. The introduction of these policies and regulations not only provides specifications and a basis for applying smart construction technology [6] but also creates a favorable environment for industry innovation and sustainable development.

Smart building technology has revolutionized the building industry by integrating advanced scientific and technological methods and smart systems to enhance human comfort and energy efficiency [7,8]. From traditional building design and construction to intelligent management and operation, intelligent construction technology has injected new vitality and momentum into the construction industry with its outstanding performance and advantages. The introduction of intelligent construction technology not only helps architects to better optimize building structures and enhance building performance in the building design phase but also improves productivity, reduces costs, and enhances safety in the construction phase [9,10]. At the same time, in the building operation and management stage, smart building technology through monitoring systems and energy management systems achieves real-time monitoring and smart adjustment of various building indicators. This improves the operational efficiency of the building, extends the life cycle of the building [11], and promotes energy savings and emission reductions.

In recent years, many scholars have begun to conduct in-depth research in the field of smart buildings, with a wide range of research topics being explored. Plageras et al. [12] discussed how the Internet of Things (IoT) can provide new services to improve daily life and by combining big data, cloud computing, and monitoring technology, they explored their joint operations and functional integration in smart buildings to achieve beneficial application scenarios. Shi et al. [13] developed a smart floor monitoring system, integrating self-powered triboelectric floor mats with deep learning-based data analysis, which can establish a functional interface for a variety of applications in smart buildings. Yu et al. [14] proposed a new method based on deep convolutional neural networks for identifying and locating damage in building structures equipped with smart control devices, demonstrating that it has excellent generalization ability and higher recognition accuracy. Guan et al. [15] proposed a privacy-protecting and efficient data aggregation scheme, and the analysis shows that the scheme can meet security requirements and has better performance than other methods. However, there is currently no overall summary and overview of the field of smart building. This article aims to analyze the current research status and future development trends in the field of smart buildings, to enhance the public’s understanding of smart construction technology, to promote societal acceptance of smart buildings and sustainable lifestyles, and to provide useful references and guidance for the future development of smart buildings. This article is to visualize and interpret the literature related to “smart building” in the WoS database from 1 January 2014, to 31 December 2023, using CiteSpace bibliometrics software.

2. Data Sources and Research Methods

2.1. Data Sources and Search Strategy

In this research, we selected the representative and authoritative comprehensive academic information platform—Web of Science Database (WoS), which was created by Eugene Garfield in 1964. Compared to Scopus and PubMed, WoS has the highest proportion of funded articles (FAs) [16]. However, while Scopus may seem more accurate than WoS, it often omits entries in the cited article list [17], which may affect our subsequent manual analysis of cited articles and related citations. WoS has developed into one of the leading global platforms for scientific citation search, discovery, and analysis. It is not only an important research tool for academic libraries but also a rich dataset for large-scale data-intensive research across various academic fields [18]. Moreover, WoS adds thousands of new entries daily, continuously expanding the breadth and depth of its database. Therefore, WoS was chosen as the database for this study. In the advanced search, we opted for the Web of Science Core Collection, selecting the following editions: Science Citation Index Expanded (SCI-EXPANDED) 1999–present, Social Sciences Citation Index (SSCI) 1999–present, Current Chemical Reactions (CCR-EXPANDED) 1985–present and Index Chemicus (IC) 1993–present [19] with the topic of “intelligent construction” OR “intelligent building” OR “smart building” OR “smart construction”, excluding “Retracted Publication”. We set the search time range from 1 January 2014, to 31 December 2023. A total of 1531 bibliographic records were collected, and all records and references were exported.

2.2. Methods of Scientometric Analysis

CiteSpace software is a visual metrology tool developed by Prof. Chao-Mei Chen using Java language. It effectively illustrates the current research status and advancements in a specific field [20]. In this research, we utilized the bibliometric software CiteSpace 6.2.R4 (64-bit) Advanced version, operating in Java 17.0.2+8-LTS-86 (64-bit) environment for data processing. We tagged all documents retrieved from the Web of Science Core Collection database and exported the data in “Full Record and Cited References” format as a plain text file. In the CiteSpace software, articles of types “Article”, “Review”, “Export WoS” and “Export DOI” were selected for the research. After deduplicating all the articles, 1508 unique documents were obtained, with no duplicate data. These data were imported into the software for analysis, excluding search terms or high-frequency terms with minimal significance. Similar keywords were merged, followed by co-occurrence analysis and visualization of keywords, countries, and institutions. The period in CiteSpace software was set from 2014 to 2023, with a time slice of 1 year. Configure node types were “Author”, “Keyword”, “Institution”, and “Country”. “Pathfinder” and “Pruning sliced networks” were chosen as the pruning methods. The pathfinder networks algorithm recommended by Chen was used to prune the network to obtain clearer visualization results [21].

3. Results

3.1. Basic Description

3.1.1. Annual Publishing Trends

The number of publications is often considered one of the most important indicators when measuring research contributions. It is a guide to evaluating the contribution and impact of the academic field [22].

This article counts the literature retrieved from the WoS database from 2014 to 2023. A time-series distribution graph of the number of publications in the field of smart buildings from 2014 to 2023 is established with the year as the horizontal coordinate, the number of publications as the vertical coordinate, and Citations as the sub-coordinate as shown in Figure 1. As a whole, the number of publications shows an upward trend. The development history is divided into two phases: the first phase is from 2014 to 2017, where the growth rate of publication output in this research field is relatively flat, and the number of publications is low. In 2015, the number of journals only increased by six compared to 2016. During this period, as the field was still in its early stages of development; researchers and scholars had relatively low levels of interest and investment in smart construction. The technology may not have fully matured. As time progressed between 2017 and 2023, technological advancements and maturation in the field of smart buildings continued to occur. This surge in attention resulted in a sharp increase in the number of publications in the field, though there was a slight decline in 2023. Looking at it from a general point of view, the growth rate of the second phase is significantly higher than that of the first phase, while the citation frequency also shows a steep trend. This growing trend reflects the importance and potential of the smart building sector in technological innovation and social development. Meanwhile, the expansion of WoS in recent years and the emergence of additional academic publishing platforms for research publication may also lead to abnormal increases in the number of journal articles [23,24]. This factor is also an important reason for the abnormal growth problem observed in the second stage.

3.1.2. The Core Cited Journals

To gain a more comprehensive understanding of the current situation in the field of smart buildings, it is essential to conduct a detailed analysis of the key literature in the field and based on this, to have better control over future research trends. Therefore, the co-citation map of highly cited articles [25] related to the field of smart buildings from 2014 to 2023 is generated using CiteSpace software, as shown in Figure 2. Highly cited literature is defined as of March/April 2024, this highly cited paper received enough citations to place it in the top 1% of its academic field based on a highly cited threshold for the field and publication year. Data from Essential Science Indicators. The diagram illustrates the communication and collaboration within the academic community.

Through citations, researchers can establish connections and form a knowledge-sharing network. This network not only helps researchers stay informed about the latest developments in their field but also facilitates interdisciplinary collaboration. Each node in the diagram represents an article or a researcher, and the size of the node is typically proportional to the number of citations that the article has received. Larger nodes indicate that the article has been cited more frequently, reflecting its influence in the academic community. For instance, nodes like “Dong B (2019) [26]” and “Minoli D (2017) [27]” are larger, suggesting that these articles have been widely cited in their respective fields. These highly cited articles often serve as important references for researchers conducting new studies, providing theoretical foundations and data support, and are considered foundational or significant discoveries in the field. Additionally, the names of some scholars are associated with multiple related articles, indicating that they have produced a wealth of research in the field and have considerable influence. For example, “Jia MD (2019) [28]” and “Al Dakheel J (2020) [29]” may be leading figures in the area.

The diagram employs a color gradient to represent the passage of time. Deep red nodes typically signify more recent articles, while light yellow nodes represent older ones. Newer articles often cite earlier classic works, creating a lineage of knowledge. Based on this color coding, we can see that the earlier article (light yellow nodes) is often foundational or exploratory, lacking sufficient empirical data or application cases, which may lead to relatively lower citation counts initially. However, as time progresses, there is a noticeable surge in citation frequency for later works, indicating that we are currently in a phase of rapid development of literature related to intelligent construction. Moreover, for a work to be frequently cited, it also requires translators and introducers, as well as a favorable social environment to foster its development [30]. This is one of the reasons why early development may be slow, while later periods see significant growth.

The key literature with the top five citation frequencies in the WoS Core Ensemble database and the top five key documents by centrality were extracted for analysis, as shown in

Table 1. Top 5 highly cited key documents sorted by count/centrality.

Number	Count-Based			Centrality-Based
	Count	Centrality	Title	Count	Centrality	Title
1	37	0.06	IoT Considerations, Requirements, and Architectures for Smart Building-Energy Optimization and Next-Generation Building Management Systems	7	0.26	Energy intelligent building based on user activity: A survey
2	32	0.1	Adopting Internet of Things for the development of smart building: A review of enabling technologies and applications	5	0.25	Demand response and smart grids-A survey
3	26	0.01	A review of smart building sensing system for better indoor environment control	7	0.17	Theory and applications of HVAC control systems—A review of model predictive control (MPC)
4	24	0.07	A review of building information modeling (BIM) and the internet of things (IoT) devices integration: Present status and future trends	5	0.16	Smart Building: Use of the Artificial Neural Network Approach for Indoor Temperature Forecasting
5	22	0.02	Smart building features and key performance indicators: A review	8	0.13	Automatic HVAC control with real-time occupancy recognition and simulation-guided model predictive control in low-cost embedded system

.

The article “IoT Considerations, Requirements, and Architectures for Smart Building-Energy Optimization and Next-Generation Building Management Systems” [27] by Minoli, and the article “Adopting Internet of Things for the development of smart building: A review of enabling technologies and applications” [28] by Jia, both discuss the technical opportunities and challenges of the Internet of Things (IoT) in the field of smart buildings. Although these two articles have a relatively high citation frequency, their centrality is low, with values of 0.06 and 0.1, respectively. Among them, “IoT Considerations, Requirements, and Architectures for Smart Building-Energy Optimization and Next-Generation Building Management Systems” is the most cited article, covering a wide range of topics such as smart cities, smart grid, smart homes, physical security, e-health, asset management, and logistics, which contributes to its higher citation rate. Centrality is a metric in network analysis used to measure the importance or influence of nodes within a network. By analyzing the centrality of nodes, one can identify the position of key nodes in the network structure, which helps in understanding how the network operates and designing effective strategies [31]. When the centrality value in the network diagram is greater than or equal to 0.1, it indicates that the node has strong centrality and plays a pivotal role [32]. The highest centrality value in the text is 0.26, found in “Energy smart building based on user activity: A survey” [33] by Tuan, which has been cited 7 times. This article mainly focuses on energy smart building based on user activity, emphasizing the characteristics of user activity. Compared to the previous article, it delves deeper into the impact of user activity on smart buildings, exploring more specialized research areas and enhancing the depth and effectiveness of the article. While it has high centrality in its field, it has relatively low citation frequency in other disciplines. Secondarily, “Demand response and smart grids-A survey” [34] by Siano, with a citation frequency of 5 and a centrality of 0.25. This article focuses on reviewing and describing the potential and benefits of the smart grid side towards demand management and discusses actual industrial case studies and research projects. Lateral demand management refers to the intelligent management and optimal operation of power grids by monitoring, analyzing, and regulating electricity demand. In the smart grid, buildings are one of the most important carriers of electricity consumption. The combination of Lateral Demand Management and intelligent construction in the smart grid can realize the intelligent interconnection between buildings and the grid, providing support and guarantee for the efficient operation of the grid and buildings. At the same time, this combination also helps to promote the sustainable development of energy and realize energy saving and environmental protection. It can be seen that these two articles are more detailed and specific in their approach compared to highly cited articles.

3.1.3. The Major Countries and Institutions

Analyzing the publication venue and the affiliations of the articles can help to better locate relevant literature references [35]. We are using CiteSpace to analyze the number of articles published by different countries.

Table 2. Top 10 countries and institutions in the count.

Rank	Country			Institution
	Countries	Count	Centrality	Institution	Count	Centrality
1	China	558	0.1	Tsinghua University	34	0.18
2	USA	247	0.19	University of California System	27	0.07
3	Republic of Korea	103	0.1	University of Hong Kong	26	0.08
4	India	73	0.15	Hong Kong Polytechnic University	26	0.08
5	Spanish	72	0.27	Chinese Academy of Sciences	24	0.11
6	Australia	70	0.02	French National Center for Scientific Research (CNRS)	20	0.07
7	Italy	69	0.12	Southeast University—China	20	0.09
8	UK	62	0.15	Zhejiang University	19	0.02
9	Saudi Arabia	56	0.11	United States Department of Energy (DOE)	19	0.15
10	France	53	0.12	Aalborg University	19	0.11

displays the top 10 countries and institutions in terms of publication frequency. The United States ranks second in both publication volume and centrality, with a publication frequency of 247 and a centrality of 0.19, placing it second among the top 10 countries. China has the highest publication volume among countries but a relatively low centrality of 0.1. Spain has the highest centrality among the top ten countries but a lower publication count of 72. Among the top 10 journal publishers, there are seven universities, two academic institutions, and one government department, the United States Department of Energy (DOE), and the top four institutions in terms of influence are all universities. It shows that universities play a key role in the field of smart building. In addition, among the top 10 countries in terms of the number of published articles, China is the country with the highest number of articles, and among the top 10 institutions, six of them belong to China. This shows that many institutions in China have invested a lot of resources in the field of smart building and achieved some results [36]. Although some institutions have advantages in the number of publications, their centrality is low, and their influence needs to be improved. It also shows that there is no strict positive correlation between influence and the number of publications [37]. The institution with the highest number of publications and impact is Tsinghua University, but it lacks close ties with other institutions in the graph and lacks cooperation with other highly productive institutions.

Converting tables to images is an effective way to convey information. The country portion of the table has been expressed in more detail to make it easier for readers to capture key information quickly. Figure 3 illustrates the publication levels and collaborations among various countries, with each node representing a country. Purple circles denote centrality and signify the importance of nodes. Nodes with centrality exceeding 0.1 are encompassed by purple circles, with thicker circles indicating higher node significance [38]. The circle sizes represent the number of relevant publications from each country, while the connections between circles depict collaborative relationships among different countries.

In the bottom left corner of the diagram, there is a timeline displaying the years, showcasing the evolution of relationships between countries over time. By examining changes in node sizes and connections across different years, one can pinpoint shifts in global dynamics. A total of 84 countries/regions are engaged in research on “smart building”. Countries such as China, the United States, and South Korea lead the ranking of publications in the field of smart building. China stands out with large nodes and numerous connections, reflecting a substantial publication volume. However, its node is not enclosed by a purple circle, suggesting relatively low centrality.

In today’s era of globalization and information technology, academic cooperation is particularly important among institutions of higher learning. Cooperation among institutions can more effectively pool wisdom and resources and promote cultural exchange and understanding. Figure 4 depicts a co-occurrence network diagram of institutions, illustrating the relationships among various institutions in academic collaboration. The nodes in the diagram symbolize different institutions, with the node size reflecting the extent of collaboration of the institution. The node color indicates the duration of collaboration of the institutions, while the connections between nodes represent the collaborative ties between institutions. The purple circle indicates the centrality, which represents the importance of the node, with the same meaning as in Figure 3. Among them, Chinese Academy of Sciences, Tsinghua University and other organizations are surrounded by purple circles, which represent that these organizations have outstanding contributions in smart buildings. In the meantime, the figure highlights those institutions such as the Chinese Academy of Sciences, Tsinghua University, the University of Hong Kong, the Hong Kong Polytechnic University, and the University of California system, which have larger nodes, signifying significant influence. Most institutions in the diagram are from China. Concurrently, with the rapid development of emerging technologies such as information technology, the Internet of Things (IoT), and artificial intelligence (AI), the technological foundation for intelligent construction has been strengthened, facilitating the generation of research outcomes and applications. Smart building has garnered attention for its ability to enhance construction efficiency, reduce costs, and minimize resource waste. In this context, an increasing number of universities and research institutions have established relevant research projects and courses, promoting talent cultivation and research development in the field of intelligent construction, and thereby facilitating the writing and publication of related research. Some institutions in the diagram have established early collaborations, like the Chinese Academy of Sciences, Tsinghua University, and Southeast University, as shown by their nodes leaning toward the blue end of the color spectrum. Conversely, some institutions have more recent collaborations, such as Shenzhen University and the Hong Kong Polytechnic University, represented by nodes leaning towards the green end of the color spectrum. The top left corner shows a total of 322 nodes representing 322 global organizations involved in “smart building” related issues [39]. There are 376 connections, with a density of 0.0073, indicating relatively few connections between institutions. In particular, the weak links between large nodes indicate a lack of cooperation between institutions that publish a high number of articles.

3.1.4. The Main Journals and Funding Agencies

In the field of smart construction, the literature analysis dendrogram provides us with a visual perspective to observe and understand the level of activity of different research topics and funding organizations within the field [40]. From the two graphs in Figure 5 (the areas on the graph are not exactly proportional to the values of each entry), we can obtain some relevant information about the dynamics and funding trends in the field of smart buildings. Smart building, as an emerging field, has received high attention worldwide. From the data of fund grants and publications, the inputs and outputs of results in the field of smart buildings are remarkable.

First of all, we can see from the publication report graph that Energy tops the list, with 75 articles, which indicates that the energy sector plays an important role in smart buildings. This may involve aspects such as energy efficiency, renewable energy utilization, and intelligent energy management systems. This is closely followed by the journal Energy and Buildings, with 69 articles, which further emphasizes the importance of building energy efficiency and optimization of building environmental performance in the field of smart construction. The journal Sensors had the third highest number of articles, at 60, which reveals the wide range of applications of sensor technologies in smart construction, including but not limited to structural health monitoring, environmental monitoring, and automated construction processes. The National Natural Science Foundation of China (NSFC) topped the funding report with 243 grants, demonstrating the country’s significant investment in promoting the foundations of the smart building sector [41]. The European Union (EU) came in second place, with 65 grants, reflecting European research collaboration and financial support in the field. Korea’s 37 grants from the National Research Foundation (NRF) demonstrate Korea’s research activity in the field of smart construction. In addition, 36 grants from China’s National Key Research and Development Program (NKRDP) and 47 from the National Natural Science Foundation of China (NSFC) further highlight China’s strategic R&D investment in smart construction.

Synthesizing the contents of the two figures, we can see that the research hotspots in the field of intelligent construction mainly focus on energy efficiency, sensor technology, automated construction, and the performance of the building environment. These articles not only focus on improving the intelligence level of buildings but also aim at realizing the sustainable development of the construction industry. Meanwhile, the distribution of funding organizations shows the research investment and strategic focus of different countries and regions in intelligent construction, among which, China, the European Union, and South Korea are particularly prominent in research funding in this field.

China’s significant investment in fundamental research in smart building underscores its emphasis on this area. Chinese government departments have introduced a series of policies and plans to promote the transformation and upgrading of the construction industry. For instance, the “National Medium- and Long-Term Scientific and Technological Development Plan Outline (2006–2020)” emphasizes the application of technology in the construction sector. Additionally, the “Made in China 2025” plan issued by the State Council in 2015 highlights the deep integration of new-generation information technology with manufacturing, encouraging the application of intelligent technologies in the design, construction, and management of the construction industry. Furthermore, the “Opinions on Further Promoting the Development of the Construction Industry”, released by the Long Gang Municipal Committee of the Communist Party of China and the People’s Government Office in 2022, also emphasize the transformation and intelligent application of the construction sector. These policies provide a foundation and financial support for relevant research. Moreover, as international academic exchanges have increased, collaborations between Chinese researchers and their international counterparts have deepened, further promoting the sharing and dissemination of relevant research findings. Collectively, these factors have driven the rapid growth of research papers in the field of smart construction in China.

Article and funding trends in smart construction indicate that it is a fast-growing and widely influential field. As technology continues to advance and sustainable development goals are pursued globally, smart construction will continue to attract more research attention and funding. Future research is likely to focus more on interdisciplinary collaboration, innovative application of technology, and talent cultivation. This will promote the development and application of smart construction technologies. It will also contribute to the realization of a smarter, more efficient, and more sustainable built environment. At the same time, we cannot deny that the coverage of the WoS database has some limitations. Grant information in the original article may be lost during the extraction process, and grant data usually exists in unstructured form in the acknowledgment section, which is difficult to extract. In addition, although the WoS database has supported automated data extraction since August 2008, the heterogeneity of the field remains a barrier to system development [42]. Among all journal citation indexes, FA (Funding Acknowledgment) information is most comprehensively recorded for English publications, followed by Chinese articles, while WoS records in other languages are less likely to contain FA details [43], which may lead to incomplete data representation and bring a little impact on the results of this analysis.

3.2. Research Trends and Frontiers over Time

3.2.1. Keywords Co-Occurrence Analysis

Keywords play a significant role in an article, serving as a concrete manifestation of the central ideas within the literature [44]. In CiteSpace software, keywords are used as nodes to analyze literature data from the past decade, which consolidates keywords with similar meanings [45]. For instance, terms like “smart building”, “intelligent building”, “smart construction”, and “intelligent construction” can be simplified to “smart building”, while terms like “internet of thing”, “internet of things”, “things”, “internet”, and “IoT” can be simplified to “internet of things”. This consolidation of keywords with similar meanings helps streamline the analysis process and present the relationships and importance of keywords in the literature more clearly. The node size represents the frequency of keyword occurrences, while the node color signifies the average year of appearance of the keyword, as depicted in Figure 6. The centrality of a keyword indicates its importance in an overall keyword co-occurrence network, serving as a core element of research [46]. We extracted the top 15 keywords based on frequency and centrality, as shown in

Table 3. Top 15 keywords in co-citation count and centrality.

Number	Count	Centrality	Keywords	Count	Centrality	Keywords
1	587	0.04	smart building	40	0.13	behavior
2	276	0.05	energy efficiency	131	0.12	model
3	268	0.07	system	63	0.11	wireless sensor networks
4	204	0.09	internet of things	129	0.1	design
5	131	0.12	model	122	0.1	performance
6	129	0.06	management	204	0.09	internet of things
7	129	0.06	genetic algorithm	42	0.08	model predictive control
8	129	0.1	design	268	0.07	system
9	122	0.1	performance	17	0.07	artificial neural network
10	106	0.02	machine learning	15	0.07	recognition
11	105	0.06	demand response	129	0.06	management
12	95	0.02	optimization	129	0.06	genetic algorithm
13	77	0.03	thermal comfort	105	0.06	demand response

.

Smart building is an interdisciplinary and multidisciplinary emerging topic [47]. Figure 6 illustrates the research hotspots, key technologies, and development trends in the field of smart building in recent years. In terms of research focus, core concepts such as “energy efficiency”, “internet of things”, “modeling”, “management”, and “demand response” are highlighted, indicating the direction of the industry’s development. Smart buildings embody technological innovation and signify a comprehensive advancement in the construction industry concerning sustainable development, resource optimization, and management efficiency [48,49,50,51]. It is also closely related to national policies as well as market drivers. More and more countries and regions are introducing policies and standards to promote energy efficiency in the building sector. These strategies usually include support for smart building technologies and encourage the adoption of new technologies to improve the energy efficiency of building systems. The policy push has led to more active exploration and implementation of smart building solutions by relevant companies and research organizations. As the market demand for smart building solutions continues to grow, driven by the emphasis on the quality of the built environment and the escalating energy crisis, these core concepts reflect both research trends in academia and market demand within the industry.

Key technologies such as “wireless sensor networks”, “machine learning”, “artificial intelligence”, “digital twin”, “cloud computing”, and “building information modeling (BIM)” have emerged in optimizing design and operational strategies. These technologies play a crucial role in improving design and operational strategies [52,53,54]. Wireless sensor networks can be used to remotely monitor building facilities to identify and solve potential problems and improve operational efficiency; machine learning can be used to form intelligent control systems that adjust equipment operation in real time to achieve optimal performance and energy efficiency. AI can also analyze user behaviors and preferences to provide personalized building environments that enhance people’s comfort and satisfaction with the space; and through the simulation of digital twins, operators can predict equipment failures and conduct effective maintenance planning to reduce costs and downtime. Through cloud computing, stakeholders such as designers, engineers, and operations managers can share information in real time to improve communication efficiency and collaboration. BIM provides detailed information to support the construction process and subsequent operations management, which helps to improve project management efficiency, reduce costs, and enhance project management. Meanwhile, in the past 10 years, research has shifted its focus to consider the user’s needs, with living spaces primarily being human-centered [26]. “Thermal comfort” and “behavior” are becoming more prevalent. Simultaneously, smart buildings are encountering challenges such as data privacy, cybersecurity, and standardization [55,56].

The connections between keywords highlight the close relationship among different specializations, like “energy efficiency” and “smart building”, “internet of things”, and “model”, showcasing the interplay and mutual reinforcement of these fields. The keywords “design”, “behavior”, “model”, and “wireless sensor networks” are highlighted in purple on the periphery, signifying their importance as core concepts in the diagram. Design is the foundation of intelligent buildings, which determines the functionality, sustainability, and aesthetics of the building; User behavior and habits have a direct impact on building energy consumption, comfort, and safety; the model building can provide important visual information at all stages of design, construction, and operation; and wireless sensor networks are the core technology to achieve real-time monitoring and control in smart buildings, where sensors are used to collect environmental and equipment data to provide a basis for decision making. These four keywords are interrelated and jointly promote the design, operation, and management of intelligent buildings. Not only are they individually important, but they also play a key role in integrating smart building technologies, increasing building efficiency, and improving user experience.

From the top 15 keywords in co-citation count and centrality, we can see that energy efficiency, system, and Internet of Things have high frequencies of occurrence, with 276, 268, and 204, respectively. These three keywords are dominant in a broad field and often involved in interdisciplinary or cross-disciplinary applications. These three keywords also have horizontal connections with various other keywords. But the centrality of these three nodes is relatively low, at 0.05, 0.07, and 0.09, respectively. They exhibit fewer vertical connections in the overall keyword network, meaning they have fewer direct links to other keywords, resulting in their lower centrality. Among them, the keywords with high keyword centrality are “behavior”, “model”, and “wireless sensor networks”, with high keyword centrality of 0.13, 0.12 and 0.11, respectively. These words begin to focus on the technical support of smart buildings, user behavior, and other specific events. The frequency is not very low, with 40, 131, and 63, respectively. This indicates that smart buildings have begun to receive widespread attention and research into their actual application situation has commenced.

3.2.2. Keyword Timeline Analysis

Cluster analysis can summarize the interrelated keywords in the research field of “smart building” and identify the focus of research trends at a given time as well as the interconnections between different research trends [7]. The keywords are clustered, and the clustering blocks with less than 36 articles are removed. Six major clustering modules are presented: #0 smart building, #1 occupant behavior, #2 solar energy, #3 internet of things (IoT), #4 demand side management (DSM), and #5 fuzzy logic. These topics are prominent in the smart building domain. Keyword timeline maps and cluster maps complement each other. Through keyword time–occurrence mapping, we can not only clearly understand the relationship between the research topics in the field but also comprehend the evolution path of keywords in each cluster [57]. To draw the keyword time–occurrence maps, as shown in Figure 7. “Timeline view” was tilted by 15 degrees, and “Max Number of Node Labels per Year” was adjusted to three. The frequency of keyword occurrence and the pattern of change over time in clustered blocks were blended. A Q value greater than 0.3 in cluster analysis indicates a significant cluster structure, while an S value greater than 0.5 indicates a reasonable cluster structure [58]. In the present research, the modularity is Q = 0.542 and the silhouette is S = 0.7761; the cluster structure of this sample of literature is evident and compelling.

As depicted in Figure 7, “smart building” is the earliest and most frequent noun in the image, with the largest circle volume. The text mainly focuses on the theme of “smart construction”, with other related nouns. Therefore, the presentation of the graph is reasonable. Analysis can be done based on this. The keywords “internet of things”, “system”, and “energy efficiency” are displayed with equal font sizes, indicating they appeared around the same time and with similar frequencies. This suggests that 2014 was in the early research stage, focusing on fundamental technology overviews but also starting to consider practical applications of smart buildings. The “internet of things” refers to the concept of physical devices connecting and sharing data over the internet [59,60]. Driven by the idea of managing resources more effectively and efficiently using information and communication technology, smart environments have emerged in the form of smart cities, smart buildings, smart grids, smart water, and smart mobility, seamlessly embedded in everyday objects. This trend promotes interconnected devices, making our lives more intelligent and convenient. It also provides more possibilities for energy management, resource utilization, and environmental protection. A key driver of smart environment development is the integration of technologies such as the IoT and big data [61,62,63], which are propelling the advancement of smart environments and enabling us to achieve higher levels of energy efficiency, reduce waste, and lower energy consumption costs. These technologies are revolutionizing the way we interact with our surroundings. By leveraging IoT devices and data analytics, we can monitor and optimize energy usage in buildings, leading to significant cost savings and environmental benefits.

The later emergence of “machine learning” and the earlier “internet of things” are significant nodes, highlighting their importance in smart building technology and offering new possibilities for building intelligence and automation [64,65,66]. Keywords like “energy efficiency” and “renewable energy” hold crucial positions, emphasizing sustainable development as a key goal in the field of smart building. Moreover, there is a growing emphasis on human-related factors such as “occupant behavior”, “indoor air quality”, and “thermal comfort”, indicating that the design and operation of the smart building are moving towards a human-centered approach in smart building design and operation. The red circle signifies a turning point in this clustering module, suggesting a breakthrough in the field. The latest keywords related to turning points in modules #0, #1, and #5 include “energy saving”, “reinforcement learning”, and “control strategy”. From these keywords, we can verify the idea that smart building is transitioning towards practical applications rather than remaining theoretical.

3.2.3. Keywords Breakout Analysis

Keyword mutations reflect the developmental frontiers of a research field. By utilizing CiteSpace software and its burst detection function, we can identify keywords that have experienced a sudden and dramatic increase in frequency over a short period, as well as display the period between the emergence and decline of these keywords [67]. The higher the intensity of a keyword burst, the more significant its impact on the field of research [68]. Detecting the changes in burst words allows us to highlight the indicators of frontiers in the “smart building” domain, assisting scholars in understanding the academic focus within this area, and aiding scholars in grasping the academic concerns in this field. After adjusting the parameters following keyword co-linearization for all recorded documents, 20 keywords were selected for analysis, as illustrated in

Table 4. Top 20 keywords with strong citation bursts.

Phase Summaries	Number	Keywords	Strength	Begin	2014–2023
Early Research Focus	1	wireless sensor networks	8.51	2014
	2	smart grid	7.49	2014
	3	comfort management	3.09	2014
	4	indoor localization	1.58	2014
	5	building automation	1.3	2014
Intelligent Control	6	fuzzy logic	1.95	2015
	7	predictive control	1.54	2015
Occupant Experience	8	heat transfer	1.7	2017
	9	occupant behavior	1.35	2017
Building Management and Optimization	10	fault detection	1.92	2018
	11	building management system	1.74	2018
	12	facility management	1.23	2019
	13	building energy management system	1.45	2020
	14	mechanical property	3.24	2020
Application of Emerging Technologies	15	reinforcement learning	2.68	2020
	16	edge computing	1.72	2020
	17	energy saving	2.93	2021
	18	neural networks	1.48	2021
	19	data models	1.3	2021
	20	electrical conductivity	1.3	2021

where the red line segment portion of the period of emergence for each keyword [69]. From this, we can see those early focuses are on basic technologies, such as “wireless sensor networks” and “indoor localization”. Wireless sensor networks, a fundamental technology in smart buildings, can collect and transmit environmental data, providing support for subsequent intelligent management [70]. Indoor localization technology enables the tracking of personnel, equipment, and materials, offering technical support for safety management, asset management, and service provision [71,72]. As the theory matured and practical applications began, the terms “fuzzy logic” and “predictive control” emerged in 2015 as applied to the smart building field, reflecting an intention to handle uncertain information in the built environment through these techniques for smarter control [73,74].

After 2017, terms related to human-centric considerations such as “heat transfer” and “occupant behavior” began to emerge. Research on occupant behavior emerged, highlighting the recognition of its impact on building energy consumption and comfort, and the desire to optimize building management by analyzing this behavior [75,76,77]. With the overall maturity of the theory, the focus on optimization continued to evolve. Since the beginning of 2018, fault detection technology has become a research hotspot, showing a trend toward using smart technology to promptly identify faults in building systems to prevent accidents and ensure safe operation [70,78]. Concurrently, terms like “building management system”, “facility management”, “building energy management system”, and “mechanical property” have also gained recognition. The emergence of “patch” in this context is equivalent to providing “maintenance” for the system, aimed at preventing issues and ensuring long-term system stability.

Since the emergence of reinforcement learning, keywords in the last part of the table indicate hot topics in recent years, continuing to the present. Intelligent construction technology is a technology that combines advanced technology and energy-saving concepts, aiming to enhance the efficiency, safety, and sustainability of buildings. In this context, reinforcement learning can optimize building design, construction processes, and equipment management by enabling intelligent systems to interact with the environment, learn, and adjust continuously [79]. This, in turn, improves energy utilization efficiency. For example, by using reinforcement learning algorithms to optimize the operation strategy of a building’s air conditioning system, automatic adjustments can be made to temperature and airflow based on different periods and the user needs to strike a balance between energy savings and comfort. Edge computing enables rapid responses to real-time data demands, helping to improve the responsiveness and efficiency of energy systems [80]. Neural networks play a role in optimizing building design and fine-tuning energy management in intelligent construction technology [81]. Data models also play a crucial role in intelligent construction technology. They can help identify bottlenecks in building energy consumption and potential energy-saving opportunities [82,83], thereby optimizing the operation strategies of building systems to enhance energy utilization efficiency. Electrical conductivity mainly pertains to building materials [84,85]. In intelligent construction technology, choosing materials with lower electrical conductivity and integrating them with intelligent energy management systems can effectively reduce building energy consumption and achieve energy-saving goals. Thus, it seems that smart building technology is developing rapidly and has now entered the technology application stage, as can be seen from Table 4, the application of emerging technologies at this stage, such as reinforcement learning, edge computing, neural networks, data models, electrical conductivity are the current hot topics in the field of smart building.

4. Discussion

The general direction of smart construction is a cross-sectoral and multidisciplinary field of combined construction. The amount of the literature related to smart building has been growing significantly in recent years, indicating that smart building technology has attracted much attention in recent years. It can be seen in the highly cited articles that the rapid development of artificial intelligence, Internet of Things, big data and other technologies, which provide a new theoretical foundation and technical support for smart buildings [86], promote innovation and breakthroughs in the field of smart buildings, and also promote the publication and citation of related literature.

Smart building is a global research field, and international cooperation and exchanges have been continuously strengthened, with active exchanges and cooperation among scholars from various countries, which jointly promote the development of the field of smart building. Roach from Australia and Ben from Belgium [87] explored how different architectural features impact energy demands and quantified changes in energy consumption in various building configuration scenarios to achieve minimal energy consumption. Mustafa from Malaysia and Moreno-Rangel from the UK [88] evaluated the latest smart window technologies to contribute to reducing energy consumption and enhancing human comfort. Countries such as China, the United States, and South Korea lead in publications, with institutions like Tsinghua University and the University of California playing pivotal roles. However, the co-occurrence analysis reveals low network density and limited collaboration among high-output institutions, suggesting a need for improved partnerships.

Analyzing the trend of keyword hotspot changes in Figure 6, the image shows that the field of smart building is evolving, with a gradual shift in focus from fundamental technologies to practical applications, incorporating emerging technologies such as the IoT and machine learning to achieve smarter and more sustainable built environments. In addition, researchers are increasingly emphasizing the role of human factors in smart buildings to contribute to a more comfortable [89,90,91], healthy, and sustainable built environment. In the phase 2014–2023, the nodes in the clusters are mainly concentrated in 2014, where “management”, “internet of things” and “energy efficiency” are more prominent. The keyword “energy efficiency” is more prominent. Project management reform is the first consideration in reforming traditional building technology into smart building technology. Smart buildings, through the use of IoT technology to monitor and manage building equipment and systems, to achieve remote monitoring and smart control, can be a perfect smart building technology and improve the efficiency of building operations. Meanwhile, energy is the foundation and driving force of modernization [92]. The International Energy Agency (IEA) is urging governments towards the target of tripling global renewable energy capacity by 2030 and doubling energy efficiency every year. The building industry, as the primary force of energy saving and emission reduction [93], should respond to the national call to vigorously promote smart buildings to minimize energy waste.

After 2014, the large nodes started to become smaller, but the circles of small nodes became increasingly numerous and densely connected with complex crossovers, coinciding with the previously mentioned increase in citation frequency in the middle and late periods. The latest keywords on the turning point—pertaining to platform, energy saving, equipment, reinforcement learning and control strategy—show that smart building has started to move towards practical technology and is not limited to theoretical innovations. On the one hand, there is a technical upgrade, through the integration of new technologies and practical applications. Smart buildings not only improve building efficiency, but also effectively improve building quality and safety, and also provide a model and reference for innovation and application in more fields; on the other hand, the building industry is paying more and more attention to people’s comfort feelings in the built environment. Al Dakheel [29] pointed out that there are gaps in the existing key performance indicator (KPI) in terms of user needs and that KPIs should be further developed in this area.

For keywords with the strongest citation burst chart analysis, the early articles focused on the basic technology to provide data support for subsequent smart management. Since 2015, the focus has been on how to optimize the operation of smart buildings, seen in the increasing frequency of keywords such as “fuzzy logic”, “heat transfer”, “building management system”, “fault detection”, etc. These conceptual terms describe a “patching” of the system to improve the efficiency and stability of smart buildings. At the same time, more humane words also began to appear; the keywords “heat transfer” and “occupant behavior” appeared in 2017, indicating that people are aware of the impact of occupant behavior on building energy consumption and comfort. In 2020, reinforcement learning techniques began to be applied to the field of smart buildings, to optimize building system performance by automatically learning optimal control strategies through this method [94,95]. Edge computing allows for data collected from endpoints to be analyzed directly in local devices or networks close to the data’s source, thereby eliminating the need to transmit data to cloud-based processing centers and reducing latency in data transmission. In 2021, neural network technology and data modeling started to be integrated into smart building applications, with the hope of enhancing the precision of building management. This is achieved by creating effective data models for the storage, management, and analysis of smart building data, as well as for performing data analysis and predictions using neural networks. As can be seen from Table 4, the application of emerging technologies is the current hot topic in the field of smart building. Through the analysis of keywords, the evolution of research hotspots in the field of smart buildings in recent years has been demonstrated. This evolution has progressed from early attention to fundamental technologies, to the development of indoor positioning technologies, and then to a focus on occupant behavior and comfort, as well as building automation and intelligent control. The application of emerging technologies in recent years indicates that the field of smart buildings is continually deepening and expanding. In the future, this field will continue to develop towards a smarter and more sustainable direction.

5. Conclusions

In this research, 1531related articles from the literature were obtained from the WoS core database and imported into CiteSpace 6.2.R4 (64-bit) Advanced software. Through the analysis of the number of publications, highly cited articles, the ranking of the number of publications and countries, the research hotspots, and the latest research trends for in-depth discussion. It was found that smart building, as a cross-sectoral and multi-disciplinary combination of research fields, has attracted much attention in recent years. The number of papers showed a trend of rapid growth from 2014 onwards. In the analysis of highly cited articles, the breadth and depth involved in the literature have different impacts on the frequency and centrality of citations. For example, Daniel’s research is widely cited in terms of technological opportunities and challenges facing the smart building sector, while Nguyen TA’s research focuses on the impact of user activities on energy smart building.

There is strong international collaboration in the field of smart building globally, with China, the United States, and South Korea leading the ranking of publications. Institutions such as Tsinghua University, the University of Hong Kong, and the University of California system play an important role in the field. But the density of inter-agency networking is low, indicating a lack of close collaboration among most institutions. University institutions continue to play a key role in the field of smart building [96].

The trends and evolution of keywords in the field of smart buildings from 2014 to 2023 show that the focus of the initial articles was on basic technologies such as sensor networks and positioning technologies. Over time, the focus gradually shifted towards practical applications. Increasingly granular keywords are starting to appear more frequently, focusing on areas such as management, the Internet of Things (IoT), and energy efficiency. Additionally, there is an increasing focus on integrating emerging technologies such as IoT and machine learning to achieve a smarter and more sustainable building environment. At the same time, the keywords indicate that technological upgrading and humanized design are both important, reflecting the trend of smart building towards healthier, more comfortable and sustainable development. In addition, smart building technology has gradually entered the technology application stage, and emerging technologies such as reinforcement learning, edge computing, and neural networks have become research hotspots. The application of these technologies has led to the optimization of building design, building and facility management, and energy efficiency, reflecting the efforts of smart buildings to enhance building efficiency, safety, and sustainability.

This review, after the meticulous organization of the literature, clearly reveals that in recent years, intelligent construction has begun to focus on user experience and the sustainability of buildings, which are important trends for the future development of the construction industry. Sustainable buildings typically emphasize creating a healthy and comfortable indoor environment, which directly affects the user experience. At the same time, sustainable buildings reduce energy consumption through efficient energy management systems and energy-saving designs, which not only reduces environmental impact but also lowers energy costs for users, enhancing their economic experience. However, there is a lack of user-centered articles and user experience standards in the smart building field. Government agencies, industry associations and enterprises should encourage and fund user-centered research related to smart building to promote the development of knowledge and practical innovation in this field. In the future, we should encourage more interdisciplinary research, encouraging experts from architecture, engineering, psychology, human–computer interaction design, and other disciplines to collaborate to better understand and meet user needs and experiences. We should establish user experience standards and assessment policies for intelligent construction to provide guidance and norms for the industry. At the same time, we should develop and validate a set of user experience indicators to assess user satisfaction, comfort, convenience, accessibility, and health impacts of intelligent construction projects, contributing to the improvement of the construction system.

The limitations of this systematic review are manifold. First, this review focuses on summarizing the literature of the last decade and does not analyze the overall historical lineage of smart buildings in sufficient depth. Some classic studies and theories, although they may not have been cited much in the last decade, have still had a profound impact on the field of intelligent buildings. Ignoring this research may lead to an incomplete understanding of the issues. Secondly, this review only included articles for which systematic data were available, ignoring unpublished or difficult-to-obtain data, thus compromising the comprehensiveness of the conclusions. In addition, the available data are concentrated in the United States and China, limiting the diversity and generalizability of the findings, making the conclusions influenced by geographic, cultural, or socioeconomic factors, resulting in an incomplete review that may not be sufficient to demonstrate the breadth of the smart building field. Finally, due to the wide range of data, this research selectively discusses representative data. While the selection of representative data is an effective research methodology, too much reliance on these data may overlook the heterogeneity in the sample. This can lead to a lack of awareness of certain important features or phenomena of smart buildings, limiting a comprehensive understanding of a complex reality. Therefore, to ensure more extensive and reliable conclusions, future research should consider a wider range of studies and data sources to delve into the diversity and complexity of smart buildings.

The field of smart building has received extensive attention globally, with increasing international cooperation and exchanges toward smarter and more sustainable development. At the same time, researchers and industry insiders are increasingly emphasizing the role of human factors in smart building and committing to creating more comfortable, healthy, and sustainable built environments. In the future, smart building technologies will continue to drive innovation, bring greater benefits to human life and the environment, and make more breakthroughs in improving building efficiency, reducing energy consumption, and enhancing user experience.