Exploring Knowledge Domain of Intelligent Safety and Security Studies by Bibliometric Analysis

Abstract

Intelligent safety and security is significant for preventing risks, ensuring information security and promoting sustainable social development, making it an indispensable part of modern society. Current research primarily focuses on the knowledge base and research hotspots in the field of intelligent safety and security. However, a comprehensive mapping of its overall knowledge structure remains lacking. A total of 1400 publications from the Web of Science Core Collection (2013–2023) are analyzed using VOSviewer and CiteSpace, through which co-occurrence analysis, keyword burst detection, and co-citation analysis are conducted. Through this approach, this analysis systematically uncovers the core themes, evolutionary trajectories, and emerging trends in intelligent safety and security research. Unlike previous bibliometric studies, this study is the first to integrate multiple visualization techniques to construct a holistic framework of the intelligent safety and security knowledge system. Additionally, it offers an in-depth analysis of key topics such as IoT security, intelligent transportation systems, smart cities, and smart grids, providing quantitative insights to guide future research directions. The results show that the most significant number of publications are from China; the top position on the list of papers published by related institutions is occupied by King Saud University from Saudi Arabia. Renewable and Sustainable Energy Reviews, Sustainable Cities and Society, and IEEE Transactions on Intelligent Transportation Systems are identified as the leading publications in this field. The decentralization of blockchain technology, the security and challenges of the Internet of Things (IoT), and research on intelligent cities and smart homes have formed the knowledge base for innovative security research. The four key directions of intelligent safety and security research mainly comprise IoT security, intelligent transportation systems, traffic safety and its far-reaching impact, and the utilization of smart grids and renewable energy. Research on IoT technology, security, and limitations is at the forefront of interest in this area.

1. Introduction

Intelligent safety and security result from the deep integration of security theories, technologies, and next-generation information technologies—such as the industrial internet, big data, cloud computing, blockchain, artificial intelligence, and digital twins—within critical public safety domains, including industrial safety, disaster prevention and mitigation, and accident prevention. As an interdisciplinary and multi-technology research field, its primary objective is to enhance safety protection and emergency management capabilities through intelligent technologies. Its framework encompasses a comprehensive safety system spanning monitoring and perception, data analysis, early warning and response, protective control, and emergency management. The evaluation criteria focus on key performance indicators such as accuracy, real-time responsiveness, robustness, and scalability. In the field of industrial internet, intelligent safety and security, when combined with blockchain technology, provides a decentralized, tamper-proof record of transactions between devices and services, ensuring the authenticity and integrity of data [1]. In addition, the security of industrial internet networks is also improved by the convergence of blockchain and deep learning, enabling the dynamic classification of devices and trust-based protection, thereby enhancing the autonomy and protection of the network [2]. In the field of big data, intelligent safety and security utilizes AI technology to achieve anomaly detection and predictive maintenance, enabling a prompt response to potential threats and ensuring the security of data processing [3]. At the same time, when combined with blockchain technology, the immutability and transparency of data are further guaranteed. Intelligent safety and security in the cloud computing environment relies on AI and blockchain to improve the security and privacy of services. AI monitors cloud platforms in real time and detects threats, while blockchain secures data storage and transmission [4]. Security vulnerabilities in digital twins can be detected and fixed in advance through the application of artificial intelligence, which simulates physical systems for security testing and evaluation in a virtual environment.

In the top ten development trends of network security in 2024, released by the Computer Security Committee of the China Computer Federation, it is pointed out that intelligent safety and security technology has become a focus of research. Scholars at home and abroad have conducted detailed research on different sub-fields. For example, Li et al. [5] studied false data injection attacks using static estimators. In addition, many scholars have also investigated various types of specific attack detection, such as spoofing attacks and false data attacks [6,7]. Scholars have also proposed intelligent safety and security monitoring solutions for specific industries, such as coal mining and the electric power sector. Chen et al. [8] designed an intelligent safety and security monitoring system for coal mines and discussed the feasibility of using wireless sensor networks. Since Toffler first coined the concept of “big data” in 1980, the rapid development of computers and the internet has further refined its meaning, triggering a revolution in scientific thinking and methodologies in the field of security [9]. Ayhan and Tokdemir [10] used actual data from construction sites, combined with latent class clustering and artificial neural networks, to predict possible accidents during construction. Su et al. [11] used convolutional neural network (CNN) image recognition technology to develop an automated fire detection and alarm system, thereby improving the safety of construction sites. These studies demonstrate the wide application and great potential of big data in security.

Our investigation reveals that, despite extensive research on various technologies and application scenarios within intelligent safety and security, there are several research gaps as follows: (1) current studies tend to focus on specific domains or technologies, lacking a macro-level analysis of the overall knowledge structure in intelligent safety and security research; (2) the identification of research hotspots and cutting-edge trends relies predominantly on qualitative analysis, lacking systematic quantitative evaluation; and (3) traditional literature review methods rely heavily on researchers’ personal experience, leading to subjective outcomes and a lack of quantitative analysis, making it difficult for readers to accurately assess the depth and breadth of the research, as well as the connections and distinctions among different studies. With the digitization of the scientific and technological literature and the advancement of literature analysis technology, bibliometrics has become essential. It provides a quantitative approach for the in-depth investigation of the literature in a particular field, offering a clear picture of the field’s trajectory [12]. In recent years, with the development of computer technology, bibliometric software such as VOSviewer [13], Citespace [14], and SCImago [15] have emerged, enabling researchers to conduct bibliometric analysis more efficiently and explore the knowledge structure and development trends in the research field. Many scholars have used these tools for in-depth research. For example, Liu et al. [16] used bibliometric methods to visualize the knowledge structure and research trends in emergency evacuation. Lang et al. [17] used VOSviewer and Citespace to perform a bibliometric analysis of fire safety-related papers. Bibliometric analysis is utilized in this study to identify the core research directions, hotspots, and evolutionary trends within the field of intelligent safety and security, with an investigation into future development pathways also being conducted. Additionally, it analyzes the characteristics of interdisciplinary integration among different subjects to foster communication and collaboration between disciplines, thereby addressing the deficiencies in current research. The remainder of this article is organized as follows: Section 2 briefly describes the data sources and research methods. Section 3 presents the study results in detail, followed by a discussion in Section 4. Finally, Section 5 summarizes the key conclusions of the research in intelligent safety and security

2. Data and Methodology

2.1. Data Collection

The WoS database is recognized as an authoritative academic database covering multidisciplinary research and is considered a powerful tool for literature retrieval and bibliometric analysis [18]. The Science Citation Index Expanded (SCI-E) and Social Science Citation Index (SSCI) databases within the Web of Science Core Collection were utilized in this study. A topic-based search was performed in the titles, abstracts, and keywords using the terms “intelligent safety”, “smart safety”, “intelligent security”, and “smart security” to ensure the comprehensive coverage of core research in the field of intelligent safety and security. Publications from January 2013 to December 2023 were selected to provide a comprehensive overview of the field’s development over the past decade. An initial screening was conducted based on the titles, abstracts, and keywords to ensure that each document contained at least one of the specified terms. This step filtered out studies unrelated to intelligent safety and security, such as those focusing solely on general information security, cryptography, or other topics beyond the scope of intelligent safety and security research. Subsequently, two researchers independently conducted a manual review to further eliminate studies unrelated to studies and remove duplicate records. Since intelligent safety and security spans multiple disciplines, including computer science, engineering, and safety management, keyword co-occurrence analysis was employed to ensure the comprehensive coverage of research across various fields. To enhance the comprehensiveness of the dataset, publications from both high-impact and low-impact journals were considered. In total, 1400 publications were retained, as shown in

Table 1. Types of literature in the field of intelligent safety and security studies.

NO.	Type of Literature	Total Publications	Percentage/%
1	Article	1157	82.64
2	Review	212	15.14
3	Early Access	16	1.14
4	Others	15	1.07

.

In this table, the most numerous literature type was article (1157 articles), accounting for 82.64% of the total literature, while the second highest number of articles was review (212 articles), accounting for 15.14%. The total number of Early Access papers was 16, accounting for 1.14%, and other types of literature included Proceedings Papers, Book Chapters, and Edition Materials.

2.2. Research Methods and Tools

In the field of bibliometrics, mathematical and statistical tools are usually employed to provide in-depth analyses of multidimensional feature information in the literature. Among the many bibliometric analysis methods, knowledge maps uniquely reveal the evolution of scientific knowledge and its internal structural relationships, similar to knowledge navigation [19]. In this paper, VOSviewer and Citespace were used to analyze the literature knowledge map. VOSviewer, developed by Van Eck and Waltman of Leiden University in the Netherlands, is based on VOS technology and has significant advantages in presenting and mapping knowledge, particularly in cluster analysis. Citespace, another commonly used information visualization tool, was developed by Professor Chaomei Chen of Drexel University in the United States [20]. The software focuses on exploring and analyzing the changing trends of the frontiers of disciplines and the connections between research frontiers and their knowledge roots. Moreover, it can reveal the internal connections between different research frontiers. Therefore, many researchers have adopted the knowledge map method to gain insight into the research status of their field and predict future trends [21,22,23,24].

This study primarily focuses on two core aspects: an in-depth analysis of the current research status and the precise identification of research hotspots and frontier dynamics, as depicted in Figure 1. VOSviewer was used to analyze the cooperative relationship of countries, institutions, and authors, as well as to create social network maps in order to grasp the developments in intelligent safety and security. Simultaneously, the key literature was reviewed and the field’s knowledge base was analyzed. Furthermore, a keyword time zone diagram was drawn using Citespace software. Burst detection analysis was conducted to precisely identify the keywords at the forefront of research, aiding in the revelation of research dynamics and future development directions in intelligent safety and security.

3. Results

3.1. Trends in Literature Publications

The publication trend of the research literature related to intelligent safety and security encompasses both temporal and global spatial distributions [25]. This trend reflects global interest in intelligent safety and security and the allocation of research efforts. This section delves into the temporal distribution of the literature, the publication of documents in the central active countries, and the regional distribution of the literature and national cooperation networks. These analyses lay the foundation for understanding the current status and knowledge base of the field, as well as identifying research trends, hotspots, and frontiers.

3.1.1. Temporal Distribution Analysis

In this study, the temporal distribution of the literature is thoroughly analyzed to reveal research trends in intelligent safety and security. As shown in Figure 2, the development trajectory of this field can be observed through changes in the volume of the literature, SOTC, and H-index. As shown in the figure, the volume of the literature on intelligent safety and security has been increasing annually, reflecting the growing global attention paid to this field. Based on several indicators, the development of intelligent safety and security can be divided into three stages.

The period from 2013 to 2016 was the preliminary stage, during which h the study scale was small and the impact was limited. However, since 2017, the field rapidly began to grow, entering a phase of rapid development that continued until 2020. During this period, the number of publications and the H-index increased significantly, indicating that the research activity was growing and the influence of the field was expanding. Since 2021, research has entered a mature and stable stage. Although the growth of publications has slowed, the H-index has continued to rise, indicating a steady improvement in research quality and a continued expansion of its influence. Overall, intelligent safety and security has undergone three distinct phases: preliminary, rapid development, and mature and stable, each with its own characteristics.

3.1.2. Country/Region Distribution Analysis

In order to understand the achievements and cooperation of various countries in intelligent safety and security research, VOSviewer was used to conduct an analysis of the literature. The results of the analysis show that a total of 100 countries (regions) have published papers in this field. After screening out fifty-five countries (regions) that have published more than six articles, the top ten active countries and their specific publication data were further extracted, as shown in

Table 2. Top 10 countries in the study of intelligent safety and security.

Rank	Country	Quantity	Percentage/%	SOTC	Total Link Strength
1	China	401	28.64	7362	219
2	USA	201	14.36	6185	196
3	UK	120	8.57	5884	147
4	South Korea	114	8.14	2601	90
5	India	101	7.21	2471	139
6	Saudi Arabia	100	7.14	1473	157
7	Australia	98	7.00	3604	118
8	Italy	94	6.71	2033	75
9	Paskistan	66	4.71	1111	122
10	Spain	55	3.93	1358	66

. China leads in publication volume, with 401 papers, accounting for 28.64% of the total literature. The rapid advancement of China in this field is primarily driven by government policy support, such as the “Smart City” initiative and the “14th Five-Year Plan for Safety Science and Technology Innovation”. Additionally, increased research funding has provided substantial financial support for studies in intelligent safety and security. Meanwhile, China has established a series of high-level research platforms, including National Key Laboratories and National Engineering Research Centers, offering cutting-edge experimental facilities and research environments for studies in intelligent safety and security. Furthermore, industry–academia collaboration has been actively encouraged through initiatives such as research–industry cooperation bases and technology transfer platforms, facilitating the practical application of research findings. Collectively, these measures have provided strong support for the development of intelligent safety and security, solidifying China’s leading position in the field.

Using the visualization tools VOSviewer and SCImago, the countries collaborating in the intelligent safety and security research field are mapped, as shown in Figure 3. This visual representation enables an in-depth analysis of the country-to-country distribution within the field. On the map, different countries are clustered in six colors and the distribution of circular nodes reveals the primary source of the articles. China, the United States, and the United Kingdom are recognized as key contributors to intelligent safety and security research. In particular, China and the United States have been shown to play a leading role among the cooperative countries, demonstrating their muscular strength and extensive influence in intelligent safety and security.

3.2. Academic Cooperation and Major Research Bases

To gain insight into the research trends in intelligent safety and security, a multidimensional analysis of the selected literature will be conducted. First, attention will be focused on author cooperation and institution cooperation networks, with the aim of uncovering essential research institutions and groups and exploring knowledge diffusion paths. Secondly, by analyzing the literature’s subject affiliation and journal distribution, the possible research topics in intelligent safety and security will be preliminarily delineated. These analyses aim to provide a comprehensive understanding of academic cooperation and support subsequent analyses of the current state and the construction of a knowledge system.

3.2.1. Key Institutions and Cooperation Analysis

According to the WoS Core Collection, 2105 institutions have contributed to research in intelligent safety and security. After screening, the top 10 institutions with the most publications were listed (see

Table 3. Top 10 organizations in terms of intelligent safety and security.

Rank	Institution	Country	Quantity	ACI	Total Link Strength
1	King Saud University	Saudi Arabia	21	10.24	17
2	The Hong Kong Polytechnic University	China	19	27.58	16
3	Tongji University	China	18	16.89	8
4	Tsinghua University	China	18	42.17	2
5	Southeast University	China	16	16.69	2
6	Princess Nourah Bint Abdulrahman University	Saudi Arabia	14	4.36	18
7	Queensland University of Technology	Australia	14	47.29	6
8	Delft University of Technology	Netherlands	14	36.86	4
9	Islamabad University of Communications	Pakistan	11	25.73	8
10	King Abdulaziz University	Saudi Arabia	11	23.27	7

for details). Among them, China has four institutions on the list, Saudi Arabia accounts for three, and the rest are from Australia, the Netherlands, and Pakistan. China and Saudi Arabia are recognized as having significant research strengths in this area. In particular, China’s Hong Kong Polytechnic University (19 papers), Tongji University (18 papers), Tsinghua University (18 papers), and Southeast University (16 papers) are among the best in terms of the number of papers published, reflecting China’s strength in intelligent safety and security research. Australia’s Queensland University of Technology has an average citation count of 47.29, demonstrating its global academic influence.

VOSviewer software was used to identify sixty-five research institutions with at least six publications and to map their cooperative relationships (Figure 4). The nodes in the figure represent research institutions, and their sizes and colors reflect the number of articles and the average year of publication, respectively. The connecting lines between the nodes represent the degree of cooperation among the institutions. According to the map analysis, King Saud University in Saudi Arabia exhibits the most frequent collaboration. The Hong Kong Polytechnic University started its research in intelligent safety and security early and has produced a wealth of academic output from 2019 to 2020. From 2020 to 2021, Tsinghua University was found to have led the way with the highest number of publications, and Southeast University was also active. After 2021, institutions such as King Saud University began to gain prominence.

3.2.2. Major Authors and Cooperation Analysis

There are 5430 authors in this field in the WoS Core Collection, demonstrating a wide range of research interests. The top 10 authors with the highest number of publications are shown in

Table 4. The top 10 authors of the number of articles published by intelligent safety and security.

Rank	Author	Country	Institute	Quantities	ACI	Links
1	Li Heng	China	Tongji University	7	32.29	4
2	Yigitcanlar Tan	Australia	Queensland University of Technology	6	58.17	4
3	Huh Jun-Ho	South Korea	Korea Maritime and Ocean University	5	14.2	4
4	Liu Yang	China	Tsinghua University	5	2.2	3
5	Azadeh A	Iran	College of Engineering University of Tehran	5	13.4	0
6	Park Jong Hyuk	South Korea	Seoul National University of Science and Technology	5	14	0
7	Almongren Ahmad	Saudi Arabia	King Saud University	4	13	8
8	Shah Munam Ali	Pakistan	Comsats University Islamabad	4	18	7
9	Yu. Yamtao	China	The Hong Kong Polytechnic University	4	26.75	4
10	Buller David B.	America	The Hong Kong Polytechnic University	4	12.75	6

. Among these ten authors, Li Heng topped the list with seven published papers; his research mainly focused on the field of intelligent buildings, with an average citation of 32.29, demonstrating the high impact of their research. He was followed by Yigitcanlar Tan, who published six related papers with an ACI index of 58.17, focusing on intelligent technology and AI research and making significant contributions to technological progress and the development of innovative security. In addition, Yu Yamtao from the Hong Kong Polytechnic University has also garnered attention, and his high citation count fully reflects his academic influence.

The collaborative network of the authors reveals the core dynamics and collaborative relationships in scientific research. In intelligent safety and security, an in-depth analysis of authors and their networks was conducted to identify a core group of authors and their partnerships. The VOSviewer tool was used to conduct a partnership analysis and generate a network diagram of the cooperative relationships, as shown in Figure 5. In this figure, each node represents the author and its size reflects the number of publications. At the same time, the lines represent the cooperative relationships between the authors and the thickness of the lines indicates the closeness of the cooperation. The results revealed 56 cooperative clusters in this field, but most of the clusters had a small number of authors and were not closely related. This indicates that in intelligent safety and security, communication and cooperation between research teams need to be improved, and a more robust cooperation system must be developed.

3.2.3. Journal Distribution Analysis

Academic journals serve as crucial platforms for scholars to disseminate research findings, and the distribution of journals in intelligent safety and security was analyzed. In this study, journals with more than four articles were selected for cluster analysis, and fifty-four nodes were obtained, yielding fifty-four nodes and forming a total of fourteen clusters.

Table 5. Top 10 journals in the field of intelligent safety and security.

Rank	Journal Title	Quantity	ACI	Impact Factor (2022)
1	Sustainability	551	11.76	3.9
2	International Journal of Environmental Research and Public Health	66	12.36	4.614
3	Energies	37	37.19	3.2
4	Applied Sciences	34	17.41	2.7
5	IEEE Transactions on Intelligent Transportation Systems	29	34.66	9.55
6	International Journal of Human–Computer Interaction	25	16.12	4.7
7	Systems	22	3.23	1.9
8	Sustainable Cities and Society	19	63	11.7
9	Renewable and Sustainable Energy Reviews	19	93.47	15.9
10	Energy Research and Social Science	18	30.89	8.514

lists the top 10 journals by publication volume in this field, including Quantity, ACI, and impact Factor.

Renewable and Sustainable Energy Reviews and Sustainable Cities and Society rank among the top in average citations per article, underscoring their significant contributions to the knowledge base of intelligent safety and security. These journals focus on sustainable development and urban safety, reflecting the growing integration of intelligent safety and security measures with renewable energy and smart grid systems. As research in intelligent safety and security continues to intersect with sustainability, these journals are likely to become key platforms for examining the safety of green energy technologies and the resilience of urban infrastructure. In terms of impact factor, Renewable and Sustainable Energy Reviews and Sustainable Cities and Society continue to exhibit strong performance, reinforcing their authority and broad readership. Articles published in these journals may serve as foundational studies that guide subsequent research. IEEE Transactions on Intelligent Transportation Systems ranks third, focusing on technological innovations and applications in intelligent transportation systems. Its high impact factor indicates the significant academic influence of its research findings in this field. Research areas covered by this journal, such as intelligent transportation system security, have also driven technological innovations and applications in intelligent safety and security, providing critical insights and guidance for future studies on transportation security. Although Sustainability has published a large number of papers in this field, its relatively low average citation count suggests a focus on early-stage research, policy discussions, or case studies rather than high-impact theoretical contributions.

Considering various metrics, Renewable and Sustainable Energy Reviews, Sustainable Cities and Society, and IEEE Transactions on Intelligent Transportation Systems are among the most authoritative journals in the field of intelligent safety and security. High-impact journals are likely to drive advancements in the following key areas: (1) enhancing the integration of AI-driven security models with post-quantum cryptographic techniques in intelligent safety and security applications to strengthen system resilience against emerging threats; (2) strengthening the synergy between intelligent safety and security technologies, climate change mitigation, and sustainable development goals to foster a greener and more secure development model; and (3) facilitating the cross-disciplinary integration of cybersecurity, energy security, and urban safety to develop a more comprehensive and resilient intelligent safety and security framework.

Using VOSviewer, the distribution of significant journals in intelligent safety and security research was analyzed, revealing that 1400 articles were published in 233 different journals. To more accurately assess the influence and partnerships of journals, it was found that only 47 of the 54 selected nodes had cooperative relationships. Subsequently, the co-citation knowledge map of publications in intelligent safety and security research was generated, as shown in Figure 6. In this figure, the top three clusters are represented by light blue, dark blue, and purple. Journals within these clusters play a pivotal role in intelligent safety and security research. Notably, these journals exhibit a high volume of publications and close partnerships with other journals.

3.3. Current Situation and Basic Knowledge of Research Field

Co-citation analysis, introduced by Henry Small in 1973 [26], occurs when two articles are simultaneously cited by a third article [27], indicating their connection and influence. Today, this analysis has been extended to authors and journals, aiding in the understanding of domain knowledge evolution, the impact of key journals and literature, and research trends. By revealing the flow of knowledge within the discipline and providing comprehensive domain insights, this approach enables researchers to advance their scholarship.

3.3.1. Highly Cited Journals Analysis

Journal co-citation analysis is a method for deeply exploring the internal correlations among journals through external citations. If two articles are cited simultaneously by other articles, it implies that they are related or similar. This method effectively reveals the complex relationships among journals and identifies core journals in each field [28]. It not only enhances the understanding of academic connections among journals, but also provides strong support for scholars in selecting references, planning research directions, and evaluating journal impact.

In this section, VOSviewer software was used to conduct an in-depth analysis of the co-citation network of journals in intelligent safety and security. Special attention was paid to journals with more than 100 citations, resulting in the selection of 101 key nodes. A network graph containing five clusters was then constructed, as shown in Figure 7. In this figure, the node size represents the number of citations, while the node spacing reveals the closeness of the relationship between journals. As shown in Figure 7, five clusters—red, green, blue, purple, and yellow—represent different research cores and directions in the field of intelligent safety and security. The red cluster is primarily related to analytical chemistry, including IEEE Access. The green cluster focuses on accident analysis and prevention and intelligent transportation systems, including Accident Analysis and Prevention. The blue cluster centers on the environment, ecology, and sustainable development, including Sustainability. The yellow cluster focuses on engineering technologies and are represented by Automation in Construction. And the purple clusters are closely related to energy topics, such as Renewable and Sustainable Energy Reviews. This analysis highlights the research hotspots and emerging trends in intelligent safety and security.

According to

Table 6. Top 10 journals with high citations for intelligence and security.

Rank	Source	SOTC	Total Link Strength
1	Sustainability	1415	34,768
2	IEEE Access	1287	43,358
3	Accident Analysis and Prevention	924	19,580
4	Sensors	808	25,714
5	Automation in Construction	780	22,576
6	Journal of Cleaner Production	775	26,953
7	Renewable and Sustainable Energy Reviews	760	30,870
8	IEEE Transactions on Intelligent Transportation Systems	712	16,894
9	Energy Policy	638	19,461
10	Safety Science	578	16,750

, Sustainability, IEEE Access, and Accident Analysis and Prevention rank as the top three in total citations, with 1415, 1287, and 924 citations, respectively. An in-depth analysis of these clustering results and highly cited journals reveals that Sustainability, IEEE Access, and Accident Analysis and Prevention have emerged as core journals in different research directions within intelligent safety and security research, owing to their high citation frequency and influence.

3.3.2. Core Literature Analysis

A total of 1400 high-impact literature records in intelligent safety and security were screened, and the top 10 most cited core articles were identified through WoS searches. It was found that seven of these involved inter-institutional cooperation, while five involved cross-border collaboration. This indicates that research on intelligent safety and security is both extensive and in-depth, and the cooperation model offers more possibilities and opportunities. Most of the highly cited articles originate from collaborative research, particularly cross-border collaboration, highlighting the significance of cooperation in advancing research in this field. Such collaboration facilitates resource sharing and the leveraging of complementary advantages and may facilitate the anticipated progress of global intelligent safety and security technologies.

As shown in

Table 7. The top 10 most cited core literature.

NO.	Title	Journal	Author	Year	IN	CN	Cations
1	Internet of Things (IoT): a vision, architectural elements, and future directions	Future Generation Computer Systems—The International Journal of Escience	Gubbi, J et al. [29]	2013	1	1	33
2	Climate-smart agriculture for food security	Chemical Engineering Journal	lipper L et al. [32]	2014	18	11	32
3	Blockchains and smart contracts for the internet of things	IEEE Access	Christidis, K et al. [33]	2016	1	1	27
4	Smart cities: definitions, dimensions, performance, and initiatives	Journal of Urban Technology	Albino, V et al. [30]	2015	3	2	23
5	Preparing a nation for autonomous vehicles: opportunities, barriers and policy recommendations	Transportation Research Part A: Policy and Practice	Fagnant, DJ et al. [34]	2015	2	1	23
6	Current trends in smart city initiatives: some stylised facts	Cities	Neirotti, P et al. [31]	2014	1	1	22
7	Iot security: review, blockchain solutions, and open challenges	Future Generation Computer Systems	Khan, MA et al. [35]	2018	2	2	22
8	Social barriers to the adoption of smart homes	Energy Policy	Balta-Ozkan, N et al. [36]	2013	9	2	21
9	Blockchain based decentralized management of demand response programs in smart energy grids	Sensors	Pop, C et al. [37]	2018	2	2	18
10	A systematic literature review of blockchain-based applications: current status, classification and open issues	Telematics and Informatics	Casino, F et al. [38]	2019	6	1	18

, the most cited article is “Internet of Things (IoT): A vision, architectural elements, and future directions” by Gubbi et al. [29]. The paper, cited 33 times, focuses on the central role of wireless sensor networks in the construction of the IoT and delves into the key technologies and application areas of the IoT. Notably, the article also discusses the close connection between the IoT and blockchain. The IoT and blockchain have become prominent research hotspots among the other highly cited literature. The third most cited article, “Blockchains and Smart Contracts for the Internet of Things”, explores in detail the potential and prospects of blockchain applications in the IoT. However, the article also points out that before applying blockchain technology to the IoT, it is necessary to fully consider issues such as transaction privacy and the value of digital assets. In the field of smart cities, the team of Albino, V. [30] elaborated on the meaning of “smart city” through a literature review and proposed performance evaluation methods and implementation measures. Neirotti, P. [31] conducted an analysis of smart city practices across various fields and studied the factors influencing smart city planning.

In addition, these highly cited articles underscore the significance of collaborative research, with many arising from multi-author, cross-institutional, and even cross-border cooperation. Such collaboration promotes resource sharing, academic exchanges, and innovation—key drivers of scientific and technological advancement. Therefore, it is essential to prioritize collaborative research and encourage active participation in such endeavors.

3.3.3. Knowledge Base Analysis

The analysis of highly cited literature serves as a crucial method for identifying research hotspots and emerging trends within a discipline, with the relationships between publications highlighted through co-citation analysis. In this study, VOSviewer was employed to examine the highly cited literature in the field of intelligent safety and security. Publications with a citation frequency of at least 15 were selected, resulting in the identification of 43 nodes that form four clusters, as shown in Figure 8. The node size indicates the citation frequency, while the distance between nodes reflects the strength of their interrelationships. This analysis offers valuable insights into the field’s knowledge base and provides a key reference for future research.

Research on blockchain technology (red cluster): In this cluster study, several scholars have explored the application and potential of blockchain technology from different perspectives. Aitzhan et al. [39] validated the decentralized energy trading system concept by combining blockchain, multi-signature, and anonymous cryptography. Andoni et al. [40] focused on the energy sector, revealing the latest technological advances of blockchain in the energy industry through a literature review and business case analysis. Bhutta et al. [41] conducted a comprehensive survey of the evolution, architecture, development framework, and security issues of blockchain technology. In addition, Xie [42] conducted a comprehensive survey on blockchain applications in smart cities, covering various aspects such as citizen services, healthcare systems, and supply chain management. Hu [43] proposed a privacy-enhanced blockchain-based access control framework for the IoT, utilizing zero-knowledge proofs. By designing account-hidden transactions and verification rules, the framework, which incorporates account-hidden transactions and verification rules, ensures both capability-based access control and requester anonymity while maintaining low overhead performance. These studies demonstrate blockchain technology’s broad application prospects and importance.

Research on the IoT (Blue Cluster): The IoT is a global network of connected entities. Roman et al. [44] conducted a study of the core issues and challenges of the IoT regarding security and privacy. They provided valuable suggestions for the security construction of the IoT. Da et al. [45] provided a comprehensive overview of the applications of IoT in various industries and their key technologies. Sicari [46] focused on the main challenges of IoT security and proposed solutions. Based on the reanalysis of internet technology and telecom management networks, Li et al. [47] studied the overall architecture of the trusted security system based on the IoT and the specific implementation of the trusted security system based on the IoT. Zanella [48] discussed the technologies, protocols, and architectures of the urban IoT and provided practical guidance in conjunction with the Padua Smart City project. Zheng [49] paid attention to the privacy of linkable data in the intelligent IoT, a less explored area in previous studies. Hu [50] introduced the JCRA framework, which combines access control and wireless resource management. By incorporating smart contracts and task offloading mechanisms, it aims to reduce computational overhead and enhance both the access efficiency and reliability of IoT systems.

Smart city research (green cluster): Allam et al. [51] conducted a comprehensive review of the potential of AI in urban construction and proposed a new framework to closely integrate AI with urban development, balancing elements such as culture, metabolism, and governance. At the same time, Braun [52], Zhang [53], and Silva [54] focused on security and privacy in smart cities. Braun proposed solutions to five smart city challenges, including privacy protection, cybersecurity, data sharing, AI utilization, and mitigating the cascading effect of failures. Zhang discussed the application and architecture of smart cities in depth and proposed research countermeasures to the security and privacy challenges in innovative healthcare, transportation, and energy. Silva provided an overview of smart cities’ features, architecture, and real-world applications, providing a clear framework for understanding intelligent cities. In addition, Abbas et al. [55] proposed a decentralized data management system for smart and secure transportation, using blockchain and IoT to address data vulnerabilities in a sustainable smart city environment.

Research in the field of smart home (yellow cluster): A smart home refers to the residential home that integrates intelligent technology to provide users with personalized services. Marikyan [56] provided an analysis of the definition and characteristics of smart homes, discussed their advantages and challenges in implementation, and provided some suggestions for further research. Solaimani [57] systematically presented the latest advances and future challenges in smart home research, employing a business model framework to provide researchers with a broader perspective. Stojkoska [58] constructed a comprehensive framework for integrating smart home devices into cloud computing IoT solutions based on a literature review and explored practical design challenges such as data processing and communication protocol interoperability.

3.4. Research Hotspots and Frontier

3.4.1. Research Hotspot Analysis

Through a study of the distribution of literature topics and their evolution trends over time, insights into hot research areas, changes in analytical perspectives, and innovations of research methods across different periods can be intuitively gained. Keywords constitute the essence of the research topic of an academic paper, as they accurately convey the core content of the article. In this article, a detailed keyword co-occurrence analysis of the literature published between 2013 and 2023 was conducted with the help of Citespace software, revealing the research hotspots and development context in intelligent safety and security [59].

Before conducting the visualization analysis, the Citespace software parameters were set in detail. The time slice was set to 1 year to capture year-over-year changes. The Pathfinder algorithm was used to optimize the map, and keywords with the top 4% co-occurrence frequency were selected. As a result, a map containing 117 nodes and 629 connections was drawn. In addition, the top 10 high-frequency keywords were extracted, as shown in

Table 8. The top 10 keywords with the highest frequency in the field of intelligent safety and security.

Rank	Keywords	Occurrences	Centrality
1	Safety	208	0.74
2	System	208	0.20
3	Internet	191	0.73
4	Smart city	140	0.22
5	Management	137	0.13
6	Model	136	0.37
7	Technology	128	0.32
8	Challenges	99	0.16
9	Framework	88	0.07
10	Big data	75	0.23

. These keywords not only indicate the research hotspots but also reveal future trends.

Figure 9 is the result of keyword co-occurrence. In this figure, the size of each node intuitively reflects the frequency of keyword occurrence, and the larger the node, the more frequently the keyword appears in the literature, thus highlighting its importance in the field of intelligent safety and security. The node’s color shows the co-occurrence time of the keyword, and the darker the color, the more frequently the keyword appears in a later period, which reveals the dynamic evolution of the research topic in the field of intelligent safety and security. Additionally, the centrality index in Table 8 is a crucial indicator for measuring the hub role of a node in the research field [60]. When the centrality is greater than 0.1, the node can be considered to have strong centrality. Combined with the data analysis presented in Table 8, it can be observed that the keywords “Safety”, “Internet”, “Model”, “Technology”, and other keywords not only appear frequently but also possess a significant centrality. These keywords are undoubtedly the research hotspots in intelligent safety and security and are of great significance in revealing the research trend in this field.

Based on the co-occurrence relationship of keywords, corresponding clusters are formed, with the subject words of each cluster representing different research directions in the field of intelligent safety and security. As shown in Figure 10, a total of four clusters are generated. The Q-value of the cluster graph is 0.7586, and the S-value is 0.9481. According to these Q and S values, the cluster structure is significant and provides adequate clustering information [61]. The four clusters that are larger in all clusters are the #0 cluster (IoT), #1 cluster (intelligent transportation systems), #2 cluster (perceived trust), and #3 cluster (smart grid).

Cluster #0 (IoT): The keyword of the largest node in this cluster is “Safety”. The larger nodes included are “Technology”, “Model”, and “Risk”. This cluster focuses on the security and pattern recognition of the IoT and delves into its various potential risks. Qi et al. [62] discussed the security of IoT as a whole, compared the security issues of IoT with those of traditional networks, and explored the open security issues of IoT. Granjal et al. [63] specifically discussed the security problems of protocols customized for the IoT and revealed the hidden security risks of these protocols. Roman et al. [64] analyzed the applicability of public key encryption and pre-shared keys to it in-depth and discussed the practicability of a link-layer key management system in detail. It is worth mentioning that Hoffmann [65] proposed an innovative key exchange protocol, which significantly enhanced communication security by generating shared keys through public exchange information. Faisal [66] summarized the most advanced key management schemes and technologies for different scenarios, such as mobile ad hoc networking, wireless sensor networks, IoT environments, and also discussed the different issues related to IoT environments, identifying the causes and impacts related to IoT security vulnerabilities.

Cluster #1 (intelligent transportation systems): The keywords of this cluster are mainly “impact”, “intelligent transportation systems”, and “road safety”. The papers in this cluster mainly discuss the intelligent transportation system, traffic safety, and their far-reaching influence. Feng Shuo [67], from Tsinghua University, explored intensive reinforcement learning methods and found that D2RL-trained intelligent abilities significantly improved assessment speed. Lian [68] highlighted the importance of high-resolution real-time big data, VRUs security, and multi-source information fusion to improve collision detection or prediction. Kaffash [69] systematically reviewed big data algorithms in intelligent transportation systems (ITS) and pointed out the room for improvement in comprehensive analysis in this field. Zhu et al. [70] thoroughly studied the implementation framework of extensive data analysis in intelligent transportation systems and comprehensively demonstrated the wide application and far-reaching impact of extensive data analysis in this field.

The central nodes of Cluster # 2 (perceived trust) include “behavior”, “autonomous vehicles”, and “user acceptance”. Sun, X [71] investigated cybersecurity issues in the autonomous vehicle environment and described the latest connected vehicle protection strategies in detail. Cui [72] focused on the safety challenges and solutions of autonomous vehicles, providing guidance for future development. Khan [73] thoroughly analyzed the advantages and disadvantages of autonomous vehicles, identifying potential attacks and providing a comprehensive understanding of safety. Berrada [74] explored the user acceptance of autonomous vehicles, including adoption time, attitudes, and perceptions of shared travel. Wang et al. [75] summarized the safety-related issues of autonomous vehicles through the theoretical analysis of autonomous driving systems and the statistical investigation of disengagement and accident reports from road tests. Tu et al. [76] studied the probability and severity of self-driving vehicle accidents based on five potential categories and proposed a safety risk assessment framework for self-driving vehicle road testing.

The keyword of the largest node in Cluster # 3 (smart grid) is “management”. The larger nodes, “renewable energy” and “energy trading”, reflect the current research hotspot: smart grids and the utilization of renewable energy. Gungor et al. [77] comprehensively investigated the potential applications and communication needs of smart grids, including power system management, energy trading, EV charging, and distributed energy management. Hossain et al. [78] discussed the vital role of the smart grids in renewable energy. Aloul et al. [79] reviewed the latest research progress on intelligent grid security and revealed its unique security vulnerabilities. Metke [80] researched the critical security technologies of the smart grids, particularly the application of public key infrastructure and trusted computing, which are essential to securing the smart grid. Wang [81] studied the network security issues of smart grids in detail, focusing on security requirements and network vulnerabilities, and proposed solutions and future research directions. Aggarwal et al. [82] proposed a four-tier energy trading architecture for smart grids.

3.4.2. Research Frontiers Exploration

The time zone chart of Citespace software is a powerful tool that clearly displays the development of research hotspots over time, providing a solid foundation for exploring research frontiers. By studying keywords, an intuitive keyword time zone map is generated, as shown in Figure 11. In this figure, the horizontal axis represents the time dimension, while the vertical axis highlights the diversity of research. Nodes represent keywords, with their size reflecting the frequency of occurrence and the color depth indicating their emergence. These nodes and connections uncover research frontiers and intrinsic connections, demonstrating the diversity of research and the importance and abruptness of keywords. In summary, Citespace’s time zone map serves as an essential tool for staying at the forefront of research and gaining insight into the dynamic evolution of the field. The top three keywords are “Safety”, “Internet”, and “Model”. From 2013 to 2015, research primarily focused on IoT technology and its security and limitations, as reflected in keywords like “Safety”, “Technology”, and “Model”. From 2015 to 2020, the research direction shifted to keywords like “Smart grid”, “Privacy”, “Management”, and “Network”. The research hotspots from 2020 to 2023 include “User acceptance” and “Algorithm”.

In bibliometric research, frontier analysis relies on changes in the number of occurrences of words or phrases. This helps pinpoint research dynamics. Keyword burst detection can reveal trends and shifts in research, capturing emerging or changing directions. From the perspective of burst keyword ranking (see

Table 9. Top 10 keywords of burst intensity in intelligent safety and security research.

Rank	Keywords	Begin	End	Strength	Year
1	Smart grid	2016	2018	7.77	2016
2	Climate change	2016	2019	6.46	2016
3	Renewable energy	2017	2019	6.27	2017
4	Risk	2014	2017	5.89	2014
5	Road safety	2014	2020	5.25	2014
6	Networks	2017	2018	4.87	2017
7	Intelligent transportation systems	2015	2020	4.39	2015
8	Communication	2019	2020	4.28	2019
9	Management	2017	2018	4.1	2017
10	Systerm	2016	2017	3.32	2016

), “Smart grid” (7.77) ranks first and is the core research object of intelligent safety and security. This is followed by “Climate change” (6.46), which reflects the study’s primary objective. “Renewable energy” (6.27) and “Risk” (5.89) are also popular. In particular, “Road safety” has the longest burst duration as it directly relates to life and property safety. It is coupled with intelligent transportation systems to promote road safety research innovation, making it a complex and in-depth research field.

4. Discussion

4.1. Challenges for Future Research

IoT security and privacy concerns: The large-scale deployment and distributed nature of IoT devices introduce significant security risks, such as data breaches and malicious attacks. A key challenge today is how to leverage blockchain technology to build a secure and reliable identity authentication and access control system for IoT devices. First, a blockchain-based distributed identity authentication system can provide tamper-proof access control, enhancing data integrity and privacy protection. Next, an AI-driven intrusion detection system can be established, using machine learning to analyze network traffic, detect anomalies in real-time, and prevent attacks. Furthermore, strengthening lightweight encryption technologies, such as elliptic curve cryptography, can provide effective security protection for resource-constrained IoT devices.

Application challenges of blockchain technology: While blockchain technology holds the potential to enhance smart city security, realizing its full potential requires a deeper understanding of its characteristics and challenges. In terms of performance, blockchain technology currently faces issues such as slow transaction speeds and low throughput, which makes it difficult to meet the demands of large-scale applications. Sharding technology can be improved by dividing the blockchain network into multiple shards, with each shard handling a portion of the transactions, thereby increasing transaction speed and throughput. Additionally, some transactions can be offloaded for off-chain processing, such as through state channels or sidechains, thus reducing the load on the main chain. Furthermore, developing new consensus mechanisms, such as proof of stake or delegated proof of stake, can improve transaction efficiency and reduce energy consumption. Blockchain technology has limited scalability, which makes it difficult to meet the growing demands of data volume. Therefore, in addition to improving sharding technology, cross-chain technology needs to be enhanced to enable interoperability between different blockchain networks, facilitating cross-chain data and asset transfers and improving blockchain network interoperability. Blockchain technology, by nature, is a public and transparent distributed ledger, making it difficult to protect user privacy. For instance, all transaction records on the blockchain are publicly visible, which could lead to user privacy breaches. This can be addressed by allowing users to prove the validity of a statement without revealing any private information, such as using zero-knowledge proofs for identity verification or data authentication, which ensures privacy protection.

Smart grid protection: In the context of smart grids, the integration and security of renewable energy are key issues. How can a secure and efficient energy trading platform be built? Based on existing research, an AI-driven predictive maintenance model can be developed, utilizing big data analysis to predict equipment failures and improve grid operational efficiency. Strengthening the application of smart contracts in energy trading, blockchain technology can be used to automatically execute electricity transactions, enhancing transparency and security. Countermeasures against the Distributed Denial of service attacks can be implemented, and machine learning algorithms can be used to detect abnormal traffic and prevent malicious attacks.

4.2. Theoretical and Practical Contributions

From a theoretical perspective, this study establishes a comprehensive knowledge framework for intelligent safety. By integrating multiple visualization techniques (VOSviewer, CiteSpace) for the first time, it systematically constructs a comprehensive knowledge system for intelligent safety and security research, providing a solid theoretical basis for future research. Furthermore, this study promotes interdisciplinary collaboration, as intelligent safety and security spans multiple disciplines, including computer science, engineering, and safety management. It identifies key research areas, such as IoT security, intelligent transportation systems, and smart grids, offering valuable theoretical support for advancing interdisciplinary efforts. Finally, the study explores the security implications of emerging technologies, identifying blockchain, artificial intelligence, and big data as key research hotspots in intelligent safety and security. It also highlights potential future directions, such as the integration of AI and blockchain for enhanced security applications. At the practical level, this study provides enhanced policy guidance. Findings indicate that China, the United States, the United Kingdom, and South Korea are at the forefront of global intelligent safety and security research. Governments can bolster policy support by promoting standardization efforts and establishing dedicated research funds to foster deeper collaboration between academia and industry. Moreover, this study aids in the practical implementation of technologies. Current research primarily focuses on IoT security, intelligent transportation, and smart grids; however, a significant challenge persists in translating these research findings into practical applications. Future research should prioritize application scenario analysis to enhance the practical relevance of findings. Finally, this study emphasizes the need for greater global collaboration. Current research in intelligent safety and security is largely dominated by individual institutions, with limited cross-institutional and international cooperation. It is recommended that the academic community strengthens international collaborations, promotes data-sharing initiatives, and facilitates the exchange of technological expertise to advance the global development of intelligent safety and security.

4.3. Future Prospects

At present, the industrial transformation driven by the IoT, intelligent manufacturing, and intelligent control will profoundly affect the global industrial layout in the future. Informatization, intelligence, and sustainability have rapidly penetrated various disciplines, leading to various new fields of activity and forms of cooperation. In recent years, intelligent safety and security has been added to the direction of graduate students in the second-level discipline. The integrated development of new technologies driven by informatization, intelligence, and greening will significantly enhance the original innovation power of the safety discipline, which constitutes a critical challenge for the future development of the discipline. Considering the present research status and challenges, the following analysis of the future development of research in this field is presented in this paper:

Information security faces unprecedented challenges with the proliferation of IoT devices and the deepening of the digital society. The widespread deployment of IoT devices introduces security threats such as data breaches, hackers exploiting devices, and more. Therefore, the development of new security models and architectures to protect IoT data and user privacy is one of the focal points of future research. Blockchain technology, owing to its distributed ledger nature, is considered a powerful tool for enhancing data reliability and transparency in intelligent environments. However, blockchain technology faces security challenges, including DoS, eclipse, and double-spending attacks. To tackle these issues, artificial intelligence (AI) technologies, including machine learning, deep learning, and federated learning, are required to enable high-precision anomaly detection and mitigation. Consequently, the integration of blockchain and AI will be a significant direction for future research. In the future, smart homes can enhance the security of smart home devices by adopting measures such as choosing strong passwords, implementing robust authentication mechanisms, and employing HTTPS protocols and VPNs to prevent threats like eavesdropping.

In 2022, the Academic Degrees Committee of the State Council and the Ministry of Education of China released The Catalogue of Disciplines and Majors for Graduate Education (2022), which clarifies that from 2023, safety science and engineering can be granted as a first-level discipline for the degree of engineering and management. At the same time, The Professional Introduction of Safety Science and Engineering Disciplines and Their Basic Requirements for Degrees further refined the catalog of secondary disciplines, mainly including the direction of intelligent safety and security. This adjustment signifies that intelligent safety and security is officially recognized as a second-level discipline under the safety science and engineering framework. This will prompt future researchers and professionals to conduct more in-depth and extensive research and practice in this field. Such changes will undoubtedly bring new development opportunities for intelligent safety and security, outline a more straightforward path for the evolution of China’s safety science and engineering disciplines, and expand the scope. By attracting more elite talents, we can collectively promote the long-term development of China’s intelligent safety and security field.

5. Conclusions

This article presents a bibliometric analysis of the literature on intelligent safety and security from the SCI and SSCI databases, covering the period from 2013 to 2023. The study examines various aspects, including publication trends, academic collaboration, research status, knowledge base, and emerging research hotspots, with the aim of providing a comprehensive understanding of the current state and future directions in the field.

(1)

Research in intelligent safety and security can be categorized into three phases: steady development from 2013 to 2016, stable growth from 2017 to 2020, and rapid expansion from 2021 to 2023. Geographically, significant collaboration has been established between China, the United States, the United Kingdom, and South Korea, with China emerging as the core of this collaboration due to its significant status and influence. Looking ahead, Chinese institutions and scholars are expected to play a pivotal role in further advancing research in this field, owing to their substantial contributions.

(2)

China has shown substantial research strength, with four Chinese institutions ranking among the top 10 in terms of the number of published papers. Notably, King Saud University, The Hong Kong Polytechnic University, and Tongji University have been particularly active, with extensive academic collaborations established by scholars from these institutions. Research in this field is marked by multidisciplinary integration, primarily focusing on environmental protection and sustainable development. The most frequently cited journals include Renewable and Sustainable Energy Reviews, Sustainable Cities and Society, and Energies. Co-cited journals can be broadly categorized into four main areas: intelligent transportation systems, Environmental Ecology, and Chemical and Engineering Technology, with Renewable and Sustainable Energy Reviews, Sustainable Cities and Society, and IEEE Transactions on Intelligent Transportation Systems being the core journals.

(3)

The research hotspots in intelligent safety and security can be summarized into four core categories. The first is IoT security, focusing on ensuring the stable operation of IoT devices and systems, preventing security threats and attacks, and safeguarding data integrity and confidentiality. The second is intelligent transportation systems, which aim to improve traffic efficiency and safety through intelligent technologies, with real-time data processing and intelligent decision-making being crucial for enhancing traffic conditions. Research also emphasizes traffic safety, addressing not only accident prevention and response but also the long-term social, economic, and environmental implications. Additionally, smart grids and renewable energy are gaining attention for their role in the intelligent monitoring and optimization of power systems through advanced information and communication technologies, with renewable energy serving as a key component of sustainable development.