Benefits and Challenges of AI-Based Digital Twin Integration in the Saudi Arabian Construction Industry: A Correspondence Analysis (CA) Approach

Abstract

The Fourth Industrial Revolution (4IR) has accelerated the construction industry’s shift toward digital transformation. This progress is mainly driven by the emergence of innovative technologies, including artificial intelligence (AI) and digital twins (DTs). While global research has extensively explored the benefits and challenges of AI-based DTs, the rapid growth of Saudi Arabia’s construction sector—fueled by substantial local investments and international partnerships—underscores the urgent need to examine their specific impact within this context. To address this gap, this study aims to investigate the potential benefits and challenges of integrating AI-driven DTs into Saudi Arabia’s construction industry. To achieve this, a structured literature review and a survey were conducted among architecture, engineering, and construction (AEC) firms, with 106 complete responses analyzed using correspondence analysis (CA). The findings revealed that AI-driven DTs substantially benefit Saudi Arabia’s construction industry. For example, among the 17 identified benefits, the top-ranked ones include AI capabilities to improve analytics, AI’s facilitation of digital twins in modeling complex real-world systems, and the facilitation of strategic decision making. However, several challenges hinder the realization of these benefits, including a lack of standardization of integrated DT and AI in construction projects, a lack of understanding of AI’s capabilities, a lack of logistics and the limited availability of IT infrastructure, and the complexity of AI algorithms. These findings underscore the transformative potential of integrating AI-driven DTs to optimize construction performance, improve decision-making, and address real-world complexities. This study provides actionable insights for stakeholders and recommends future research exploring strategies for overcoming adoption challenges, fostering technological innovation, and capacity building in Saudi Arabia’s construction sector.

1. Introduction

Digital transformation has become essential for the construction industry, especially with the advent of the Fourth Industrial Revolution. This shift, in turn, has brought numerous benefits to the construction field, including increased productivity, enhanced collaboration, and more streamlined processes. In this context, the key technologies driving this transformation encompass building information modeling (BIM), 3D printing, laser scanning, augmented and virtual reality (AR/VR), digital twins (DTs), and the Internet of Things (IoT), all of which play critical roles in advancing and modernizing the construction industry [1]. The successful implementation of digital transformative technologies offers substantial competitive advantages for both organizations and nations; however, some, if not all, continue to encounter significant challenges in adopting these advanced technologies. Therefore, it is possible that one of the core barriers to digital transformation is related to the limited awareness of the full scope of benefits that these digital innovations can bring to the construction industry [2]. Thus, effective digital transformation requires companies to first develop a comprehensive understanding of both the potential advantages and the associated challenges of these technologies. Therefore, using this knowledge, organizations can implement tailored processes and strategies that maximize the value of digital tools while addressing the unique demands and obstacles within the construction sector [3].

Interestingly, the rapid growth of Saudi Arabia’s construction industry has positioned the Kingdom as a key influencer in global construction markets.

The integration of emerging technologies such as artificial intelligence (AI), the Internet of Things (IoT), and digital twins (DTs) has become increasingly vital to the construction sector. These technologies collectively contribute to enhancing project delivery and construction management across various phases of the project lifecycle. Specifically, the adoption of digital twin technologies has demonstrated considerable benefits, such as improved documentation and communication [4] and optimized facilities management and operations [5], as well as the real-time monitoring and control of construction activities [6], improved project efficiency [7], and the risk assessment of different scenarios [8]. Furthermore, the integration of AI significantly augments the capabilities of digital twins by enhancing analytical performance [9], enabling the modeling of complex real-world systems [10] and facilitating more informed and strategic decision making [10].

In alignment with the Saudi Vision 2030, the construction market is projected to witness significant growth in the coming years. According to Mordor Intelligence (2024) [11], the market is expected to reach USD 70.33 billion in 2024, with a compound annual growth rate (CAGR) of 5.37%, ultimately reaching USD 91.36 billion by 2029. Additionally, reports indicate that, since the announcement of Vision 2030 in 2016, the total value of real estate and infrastructure projects launched in the Kingdom has exceeded USD 1.25 trillion [12]. These substantial investments reflect the Kingdom’s strategic commitment to developing its construction sector as part of a broader effort to diversify the economy and achieve the Vision 2030 objectives [11,12].

Despite these advances, the construction sector, similar to its global counterparts, continues to face persistent challenges such as project delays and cost overruns. Nevertheless, Saudi Arabia’s vision for smart city development is evident in projects such as NEOM, which aim to integrate advanced technologies into both construction processes and urban infrastructure. Designed as a technologically driven and sustainable megacity, NEOM incorporates renewable energy sources, AI-enabled governance, autonomous transportation, and a fully digital infrastructure [7]. These initiatives further underscore the Kingdom’s ambition to leverage innovation in pursuit of sustainable development [13].

Recent studies on the adoption of cutting-edge technologies in the construction industry reveal a strong consensus on the key challenges and benefits associated with adopting innovations such as BIM [3,14,15,16,17]; digital twins [1,6,18,19,20,21,22,23,24]; and the Internet of Things (IoT) [25,26,27]. However, according to Xie and Pan [20], little effort has been made to highlight the specific benefits and challenges of implementing AI-based DTs in the construction sector in general; within the context of Saudi Arabia, it remains scarce. As a result, there is limited guidance on implementation strategies and actional recommendations for optimizing these technologies’ integration in the Saudi construction industry. For this reason, this study addresses the research question of how construction professionals in Saudi Arabia perceive the benefits and challenges of integrating AI-based DT into their industry. We also address the solutions needed to mitigate challenges and enhance digital transformation in the construction industry.

Consequently, this research aims to investigate the potential challenges and benefits associated with implementing AI-based digital twins (DTs) in the Saudi construction industry. Given the limited availability of studies within the Saudi context, it is anticipated that the identified challenges may be more closely related to the implementation of DTs themselves rather than the integration of AI technologies. Notably, the existing literature tends to emphasize AI’s technical capabilities and limitations when combined with DTs, with less attention given to the broader implementation context. To address this gap, three main objectives follow: (1) to review the literature to identify the common benefits and challenges related to the implementation of both DTs and AI-integrated DTs in the construction industry; (2) to conduct a survey study to examine the perceived benefits and challenges associated with implementing DTs and AI-driven DTs in the construction sector; (3) to analyze the collected quantitative data using the analytical framework developed based on the literature review (Objective 1), in order to further explore and validate the potential benefits and challenges.

The rest of this study is organized as follows. Section 2 presents the findings of the literature review. Section 3 highlights the adopted research design method and data collection techniques. Section 4 presents the study results. Finally, Section 5 presents a discussion and the conclusions.

2. Literature Review

We conducted a systematic literature review to comprehensively investigate the benefits and challenges associated with the adoption of AI-driven digital twin technologies in the construction sector. To ensure the inclusion of high-quality and up-to-date research, peer-reviewed publications published between 2018 and 2024 were systematically retrieved from reputable academic databases, including Scopus, Web of Science, and IEEE Xplore. The search strategy was guided by the use of carefully selected keywords such as “Digital Twin”, “construction sector”, and “artificial intelligence”, ensuring a focused exploration of literature that specifically addresses the intersection of these emerging technologies within the built environment. This approach enabled the identification of key research trends, critical success factors, and ongoing challenges, forming a robust foundation for the current study.

2.1. Definitions and Application of Digital Twins

Since 2016, the adoption of DT technology has steadily increased within the construction industry. In this context, numerous scholars have proposed varying definitions for DTs within the built environment; despite this, there is a lack of consensus regarding this definition [28]. For example, some scholars have interpreted this concept as a virtual representation of a physically built asset, e.g., Brilakis et al. [5], achieved through digital-twin-enabling technologies, including sensors, communication networks, and 3D models [29]. An additional study conducted by Shahzad et al. [28] compares ten various definitions of the term “Digital Twins” found in the literature. The findings indicate that the DTs approach facilitates real-time updates and enables bidirectional coordination, ensuring that the virtual model accurately reflects a replica of the physical asset.

This is attributed to the transformative abilities of Industry 4.0, which is regarded as the convergence of nine digital technologies, including cybersecurity, big data and analytics, simulation technologies, augmented reality, 3D printing, cloud computing, advanced robotics, the Internet of Things (IoT), and integrated horizontal and vertical systems [30]. Consequently, applying DTs across various industries has demonstrated significant progress and advantages. Conversely, the adoption of DTs in the construction sector remains relatively limited compared to the manufacturing industry [20]. In addition, the applications of DTs can be divided into three purposes: simulation, monitoring, and control [31].

The study employed a structured literature review analyzing recent academic studies to identify benefits/challenges. This combined desk research and stakeholder insights to contextualize findings within Saudi Arabia’s construction sector.

Moreover, several studies focused on exploring the concept of DTs in various contexts. This allowed us to develop a comprehensive understanding of the current state of the digital transformation, including the prevailing challenges, associated solutions, and emerging needs to enhance its physical realism. DT technology is applied across various sectors, including healthcare, meteorology, manufacturing and process technology, education, urban transportation, and the energy industry [4]. Notably, the literature identifies the development of advanced prognostics and health management (PHM) systems as the most prevalent application of DT technology [32].

In the construction industry, DTs are recognized as an up-and-coming advanced digital platform that generates virtual representations of tangible assets [33] This platform operates on the foundation of the IoT [34]. This implies that the successful implementation of DTs depends on the degree of IoT integration employed and the volume of information necessary to create and refine the model [35].

2.2. Benefits and Challenges of Digital Twins

2.2.1. Benefits of Digital Twins

Regarding the potential benefits of DTs in various contexts, the growing application of DTs across multiple fields can be attributed to their significant practical advantages. For example, in 2019, Oracle reported eight key benefits of implementing DTs, which are categorized as follows: better efficiency and safety, scenario evaluation and risk assessment, predictive maintenance and scheduling, improved decision-making systems, real-time remote monitoring and control, the personalization of services and products, strengthened intra- and inter-team collaboration, and better communication and documentation [4]. All of these might contribute to reducing project times and costs during the project life cycle [16].

In terms of DTs’ benefits in the construction industry, Matt Keen [36] identified several key features of DTs technology that are particularly relevant to design and the built environment. These features include BIM, 2D and 3D models, schedules, contracts, construction documents (e.g., submittals, change orders, RFIs), operational data gathered from embedded sensors, and information derived from AI and machine learning technologies. It is essential to note that the relevant studies indicated that the degree of synchronization among these features, whether within a single system or across multiple systems, primarily depends on the effectiveness of the IoT [30]. This synchronization enables operators to access and control various systems, detect faults, and support decision-making processes across different workspaces and their components within the asset.

Along with the factors mentioned above, it could be concluded that the common DT benefits reported in the construction industry include automated progress monitoring, updated as-built drawings and models, resource planning and logistics, safety monitoring, quality assessments, the optimization of equipment usage, worker monitoring and tracking, facility monitoring, facilities management and operations, enhanced decision-making, and the promotion of sustainable development, as shown in

Table 1. Benefits of implementing DTs in the construction industry.

No.	Benefits	References
1	Synchronization benefit	[28,36]
2	Automated monitoring of site progress and reduced operational risk of machines	[5,20]
3	Improved project efficiency	[7,10]
4	Updated as-built drawings and models	[5,22]
5	Resource planning and logistics benefits	[5,10,21]
6	Safety monitoring	[5,10,39]
7	Quality assessment	[5,10,39]
8	Optimization of equipment usage	[5,23]
9	Worker monitoring and tracking	[5,23]
10	Facilities management and operations	[5,23]
11	Promotion of sustainable development	[5,7]
12	Enhanced collaboration among stakeholders	[4,16]
13	Predictive maintenance and scheduling	[4,40,41]
14	Real-time monitoring and control of construction processes	[4,6]
15	Risk assessment of different scenarios	[4,8,39]
16	Support for decision-making processes	[4,5,36]
17	Better documentation and communication	[4,16]

. An example of DT adoption in the construction industry is General Electric’s (GE) implementation of DTs in their wind farms to optimize wind energy production. This initiative led to a notable 20% increase in energy output, generating an additional profit of around USD 100 million [37]. Moreover, research conducted by the McKinsey Global Institute indicates that digital transformation in the construction industry can lead to productivity increases of 14 to 15 percent and cost savings of 4 to 6 percent [38]. Here, it can be inferred that, as Industry 4.0 continues to evolve, further advancements in this domain may result in additional cost savings. Additionally, integrating DTs technology with mobile devices and wearables on construction sites allows for the real-time comparison of a project’s as-built and as-designed conditions. This approach improves the accuracy of project representations and significantly reduces errors and reworkings by enabling the timely transmission of up-to-date information to the field [32].

2.2.2. Challenges of Digital Twins

The rise of Industry 4.0 has dramatically accelerated the adoption of DTs technology across various industries. As a result, this development supports a shift from conventional, labor-intensive project management approaches to more collaborative, technology-focused methods [18]. However, this shift requires considerable time and effort, particularly in modifying existing organizational structures, which often face resistance to change [19]. For instance, Ris and Puvača [42] highlights that digital transformation is not solely about implementing new technologies; it also involves a significant transformation in terms of the organizational culture and workforce dynamics. To ensure the successful implementation process of digital transformation, it is crucial to consider factors related to three main categories: human, workplace, and technology [2,38]. For this reason, several studies have identified these related factors. Furthermore, it is essential to acknowledge the common challenges highlighted in the literature that impede the widespread adoption of DT technology. These include (1) data standardization, (2) data management, (3) data security, (4) the need to update old IT infrastructure, (5) the challenges of connectivity privacy, (6) the security of sensitive data, and (7) the lack of a standardized modeling approach [10,43] and the lack of skilled professionals [4,5,44,45], as shown in

Table 2. Challenges of using DTs.

No.	Challenges	References
1	High implementation costs	[10,28,43,44,45]
2	Lack of data standardization/lack of a standardized modeling approach	[10,40,46]
3	Lack of data management	[10,20,46]
4	Challenges of connectivity privacy	[10,46]
5	Increased demand for power and storage	[10,43]
6	Storage integration difficulties with existing systems	[10,43]
7	The complexity of its architecture	[10,28,43]
8	Lack of IT infrastructure	[10,44,46]
9	Data quality issues, e.g., poor and missing data	[45,46]
10	Sensitive data privacy and security	[10,40,44,46]
11	Lack of trust	[46]
12	Lack of positive and negative expectations of digital twins’ needs	[28,46]
13	Resistance to change to the new system	[19,28,44]
14	Lack of skilled professionals	[4,5,44,45]
15	Issue of interoperability with various sensors	[20,28]
16	Lack of national guidelines/codes of practices	[44,45]

.

In addition to these challenges, the primary barriers expected to restrict the growth of the DT market are the high deployment costs, increased demand for power and storage, integration difficulties with existing systems or proprietary software, and the complexity of its architecture [10,43]. Furthermore, the process of implementing DT technology is costly, requiring significant investment in technological platforms, e.g., sensors and software, infrastructure development, maintenance, data quality assurance, and security measures. As a result, the high initial costs and the complexity of the required infrastructure are anticipated to slow down the widespread adoption of DT technologies [10,43].

2.3. Artificial Intelligence (AI)-Based Digital Twins (DTs): Benefits and Challenges

The three main aspects of DTs are data acquisition, data modeling, and application [24]. To achieve this, DT technologies employ four key technologies to collect and store real-time data, acquire information to generate effective insights and create a digital representation of physical objects. These technologies include the IoT, AI, extended reality (XR), and cloud systems [10].

2.3.1. Artificial Intelligence (AI)-Based Digital Twins (DTs): Benefits

Regarding applications, artificial intelligence aims to mimic the basis of intelligence to create a new intelligence system that can respond similarly to human intelligence [10]. With the assistance of neural networks, machine learning, deep learning, and expert systems [47], AI can support DTs by offering an advanced analytical tool that can automatically analyze the collected data and produce insightful information, forecast results, and offer advice on how to prevent possible issues [24]. For more clarification, with the rapid advancement of digitalization, artificial intelligence (AI) has become a core enabler of digital twin systems. Key AI technologies such as deep learning (DL), machine learning (ML), natural language processing (NLP), and computer vision all enhance the intelligence, adaptability, and automation capabilities of digital twins. To illustrate this, ML supports real-time decision making and predictive modeling [48]. Deep learning, especially convolutional neural networks, aids image-based infrastructure monitoring [49]. NLP enables human–machine interactions by interpreting textual data [50], while computer vision enhances situational awareness in fields such as autonomous driving [51]. Together, these technologies drive the evolution of digital twins across various sectors, including healthcare, energy, and smart cities.

In this context, and according to the literature review, AI-driven DT technology offers three main benefits: (1) AI enhances DTs’ modeling of complex real-world systems, (2) it supports strategic decision-making processes, and (3) AI capabilities improve data analytics and insights. Numerous examples in the literature showcase the current research on smart cities in relation to DTs. Key use cases include the monitoring and comparison of energy consumption, renewable energy integration (e.g., wind turbines), smart grids, intelligent transportation systems, and connected vehicles. For instance, to monitor and compare energy consumption based on environmental factors and human activities, Ruohomäki et al. [52] propose the “mySMARTlife” framework, which leverages IoT advancements across cities to develop a smart city digital twin. Integrating renewable energy sources into the energy grid is a significant area of implementation for digital twins, particularly in developing smart cities where reliable and efficient energy delivery is essential. In this context, Pargmann et al. [53] introduced a cloud-based digital twin monitoring system designed to support farms’ development and real-time supervision, enabling continuous monitoring, analysis, and optimized energy distribution. Similarly, Sivalingam et al. [54] present a case study investigating the deployment of wind farms and energy consumption within smart grid environments. Furthermore, digital twin technologies extend into traffic management. For instance, Kumar and Jasuja [55] present a digital twin framework for vehicles and traffic systems, demonstrating its potential to optimize urban mobility and transportation infrastructure. Several examples of DT applications or platforms have been highlighted in the literature. Notably, GE’s Predix, Eclipse’s Ditto, and IBM’s Watson are identified as prominent contenders for the deployment of digital twin technology [46].

In terms of using AI-enabled DTs to mimic a complex real-world system, this technique leverages data collected by several IoT devices to discover and operate while stimulating real-world manufacturing systems, resulting in the continuous detection of areas for improvement and supporting informed decision making [10]. This technique also helps design systems to improve efficiency and prevent costly redesign during implementation [10]. For example, IBM created the “Watson IoT Platform,” a comprehensive IoT data tool for managing large systems in real time using information gathered from millions of IoT devices [55]. This type of platform has several added features, including cloud-based services, data analytics, edge capabilities, and blockchain services, making it a possible platform for a DTs system [55].

In terms of AI’s role in facilitating strategic decision making, it can be categorized into two distinct functions: the supportive role and the autonomous role [56]. To illustrate this, in its supportive role, AI assists humans by providing accurate and relevant information to enhance decision making while humans retain control over the process. Conversely, the autonomous role involves AI taking over decision-making responsibilities, thereby improving operational efficiency. This is achieved by analyzing collected data and, subsequently, determining and executing the appropriate actions. AI-enhanced optimal decision making involves selecting the most effective solution to achieve specific goals, such as reducing carbon emissions or ensuring cost-efficient investments. Integrating decision-making processes into AI-enhanced DT platforms is expected to yield several benefits, including (1) more robust decision making under uncertain conditions, (2) the proactive inference of human preferences, (3) improved interpretability, and (4) greater transparency in the final decision-making process [57].

In terms of AI capabilities to improve analytics, these advancements have significant benefits in terms of practical applications. For instance, AI can add ‘’what if’’ scenario testing and an AI-based multi-domain analyzers. To illustrate this, AI-aided “what-if” scenario testing significantly expands the functionality of standalone DTs, which are highly effective tools for analyzing various hypothetical scenarios. By incorporating AI, DTs gain advanced capabilities such as (1) addressing operational uncertainties, (2) strengthening system security, and (3) minimizing potential losses during critical events [58]. In addition, Piras et al. (2024) [59] developed a methodology for digitally managing both existing built environments and future constructions. This approach integrates tools such as AI algorithms, DTs, BIM, and the IoT. The study revealed significant benefits, including resource optimization, real-time monitoring, enhanced workplace safety and health, effective material and equipment management, predictive and preventive maintenance, and proactive resource management, as shown in

Table 3. Benefits of integrating AI-based DTs.

No.	Benefits	References
1	AI facilitates digital twins in modeling complex real-world systems	[10]
2	Facilitates strategic decision making	[9,10,60,61]
3	AI capabilities to improve analytics	[9,10]

.

Generally, DTs technology is expected to expand its application across a broader range of sectors and use cases in the future, with solutions increasingly integrating additional technologies. For example, augmented reality (AR) can provide immersive experiences, while artificial intelligence (AI) can enhance analytics, connectivity, and insights [10]. Moreover, the continued advancement of DTs solutions is anticipated to leverage technological innovations, potentially eliminating the need to physically verify real-world objects [10]. In the context of the construction sector, one significant advantage of integrating AI-based DTs technology is its ability to reduce project delays and minimize resource waste, thereby optimizing construction processes and improving overall efficiency [59].

2.3.2. Artificial Intelligence (AI)-Based Digital Twins (DTs): Challenges

Despite the substantial benefits of implementing AI-based DTs, the literature identifies several critical challenges that might hinder their successful implementation, as shown in

Table 4. Challenges of integrating AI-driven DTs.

No.	Challenges	References
1	Complexity of AI algorithms	[20]
2	Data quality issues	[46,56]
3	Lack of understanding of AI capabilities	[59]
4	Ethical concerns	[9,62]
5	Data privacy and security concerns	[9,20,46,59]
6	Lack of standardization of integrated AI and DT in construction projects	[63,64,65]
7	Lack of logistics and availability of IT infrastructure	[66,67,68]

. These challenges include the complexity of AI algorithms, issues related to data quality, limited understanding of AI capabilities, ethical concerns, and data privacy and security risks. Addressing these obstacles is essential for the successful integration of AI-based DTs. To this end, as shown in

Table 5. Resource implementation of AI-based DTs.

No.	Resource Implementation	References
1	Training and education	[5,28,45,59]
2	Access to case studies and best practices	[28,45]
3	Financial incentives or subsidies	[28,52,69]
4	Technical support from vendors	[9,28,45]
5	Resolving data privacy and security concerns	[45,59]

, various studies have proposed solutions such as implementing training and education programs, providing access to case studies and best practices, offering financial incentives or subsidies, enhancing technical support from vendors, and strengthening measures to ensure data privacy and security. These strategies aim to mitigate the identified challenges and promote the broader adoption of AI-based DTs technologies. For example, a review of 88 articles identified significant obstacles hindering the implementation of AI-based DTs. According to Daniel et al. (2024) [45], a key challenge is the lack of demand from clients in both developed and developing countries. To address this issue, the study recommended that access to case studies and best practices could play a pivotal role in raising awareness and increasing demand. Moreover, the high implementation costs represent another significant barrier. In this regard, technical support from vendors was proposed as a potential solution to alleviate financial constraints. Lastly, data privacy and security concerns remain critical issues that must be resolved to foster broader adoption and build trust in AI-based DT technologies. These findings highlight the need for targeted strategies to overcome these challenges and successfully integrate AI-based DTs across industries.

The literature review highlights the growing prominence of DT technology and its integration with AI across various industries. While the transformative potential of AI-based DTs is thoroughly documented, their application in the construction sector remains underdeveloped compared to industries such as manufacturing. AI-based DTs offer substantial advantages, including the ability to model complex systems, enhance decision-making, and improve data analytics, all of which play a critical role in optimizing construction processes by reducing delays, minimizing resource waste, and boosting efficiency. However, several significant obstacles hinder their broader adoption, including the complexity of AI algorithms, challenges with data quality, a limited understanding of AI functionality, ethical considerations, and concerns about data privacy and security. Addressing these barriers requires a combination of targeted solutions, including training initiatives, access to case studies and best practices, financial incentives, and robust technical support. These strategies are crucial for promoting the adoption of AI-powered DT technologies, enabling the construction industry to fully harness their transformative potential.

2.4. Correspondence Analysis (CA)

Correspondence analysis (CA) is a data analysis method that expresses a two-dimensional or higher contingency table in a low-dimensional space to identify the combination of columns and rows [70]. Likert scales, commonly used to measure attitudes, opinions, or perceptions, generate categorical or ordinal data that can be challenging to analyze using traditional statistical methods.

CA is designed specifically for categorical data, which are often encountered in surveys, questionnaires, and factor analysis. It is beneficial when the data involve numerous categories or levels [71]. CA addresses this challenge by transforming contingency tables into a low-dimensional space, enabling researchers to explore the relationships between rows (respondents) and columns (factors) [70]. This method computes the association between factors and highlights their relative importance by identifying the principal components that explain the most variance in the data. For example, in a survey assessing the benefits of technology in a given industry, CA can rank the factors based on their contributions to the overall variance, making it easier to determine the most influential factors in the dataset. This function is particularly valuable when dealing with complex datasets where the relationships between factors and respondents are not immediately apparent [71,72].

Moreover, CA is especially effective as it accounts for the ordinal nature of Likert-scale data by focusing on relative proportions rather than assuming linear relationships. It evaluates the degree to which specific factors contribute to the overall patterns in the data and visualizes these relationships in a two-dimensional space. This visualization aids in understanding clusters of factors or respondents with similar patterns, allowing for the nuanced assessment and ranking of factors [70]. For instance, when evaluating challenges in implementing digital twins in construction, CA can reveal whether specific challenges (such as high costs or a lack of skilled professionals) are closely associated with particular respondent groups. By ranking factors based on their contributions to the principal dimensions of variance, CA provides a robust framework for prioritization and decision making. Its ability to handle categorical data and its interpretive power make it an essential tool for assessing survey-based Likert-scale data in various fields, including education, healthcare, and engineering [72,73].

3. Research Methodology

The methodological process comprises several sequential steps, as shown in Figure 1:

–

Literature review: the process begins with a comprehensive review of the relevant literature to identify the key benefits and challenges associated with adopting DTs and AI-based DT technologies in the construction industry.

–

Survey development and validation: initial survey questions are formulated based on the findings of literature review. These questions are subsequently refined through semi-structured interviews with five field experts, each with over 10 years of experience, to ensure the clarity and relevance of the questions to achieving the study aim. This step is followed by a pilot survey to improve clarity and ultimately strengthen the validity and credibility of the final research findings.

–

Data collection: the validated questionnaire is distributed to the target respondents via a web link to gather the necessary data.

–

Data preparation: the collected survey data are then prepared for analysis. This step ensures that the data are complete, consistent, and properly formatted for processing.

–

Data analysis: finally, the quantitative data are analyzed using correspondence analysis (CA) to identify and prioritize the most significant benefits and challenges of adopting AI-based DT technologies in the construction sector.

3.1. Literature Review: Extracting Related AI-Driven DTs Benefits and Challenges

In this stage, several relevant research studies and review articles were analyzed to identify the benefits and challenges of adopting DTs in the construction industry. Consequently, this review process identified 17 benefits and 16 challenges, as presented in

Table 6. Benefits of implementing DTs in the construction industry.

No.	Benefits	Symbol
1	Synchronization benefits	BDT1
2	Automated monitoring of site progress and reduced operational risk of machines	BDT2
3	Improved project efficiency	BDT3
4	Updated as-built drawings and models	BDT4
5	Resource planning and logistics benefits	BDT5
6	Safety monitoring	BDT6
7	Quality assessment	BDT7
8	Optimization of equipment usage	BDT8
9	Worker monitoring and tracking	BDT9
10	Facilities management and operation	BDT10
11	Promotion of sustainable development	BDT11
12	Enhanced collaboration among stakeholders	BDT12
13	Predictive maintenance and scheduling	BDT13
14	Real-time monitoring and control of construction processes	BDT14
15	Risk assessment of different scenarios	BDT15
16	Support for decision-making processes	BDT16
17	Better documentation and communication	BDT17

and

Table 7. Challenges of using DTs.

No.	Challenges	Symbol
1	High implementation costs	C-DT1
2	Lack of data standardization/lack of a standardized modeling approach	C-DT2
3	Lack of data management	C-DT3
4	Challenges of connectivity privacy	C-DT4
5	Increased demand for power and storage	C-DT5
6	Storage integration difficulties with existing systems	C-DT6
7	The complexity of its architecture	C-DT7
8	Lack of IT infrastructure	C-DT8
9	Quality data issues, e.g., poor and missing data	C-DT9
10	Sensitive data privacy and security	C-DT10
11	Lack of trust	C-DT11
12	Lack of positive and negative expectations of digital twins’ needs	C-DT12
13	Resistance to change to the new system	C-DT13
14	Lack of skilled professionals	C-DT14
15	Issue for Interoperability with various sensors	C-DT15
16	Lack of national guidelines/codes of practices	C-DT16

, respectively.

Given that this study specifically focuses on AI-driven DTs, an additional literature review was conducted to explore the potential benefits and challenges arising from this integration. As a result, three additional benefits and seven further challenges were identified, as shown in

Table 8. Benefits of integrating AI-based DTs.

No.	Benefits	Symbol
1	AI facilitates digital twins in modeling complex real-world systems	B-DT-AI-1
2	Facilitates strategic choice-making	B-DT-AI-2
3	AI capabilities to improve analytics	B-DT-AI-3

and

Table 9. Challenges of integrating AI-based DTs.

No.	Challenges	Symbol
1	Complexity of AI algorithms	C-DT-AI1
2	Data quality issues	C-DT-AI2
3	Lack of understanding of AI capabilities	C-DT-AI3
4	Ethical concerns	C-DT-AI4
5	Data privacy and security concerns	C-DT-AI5
6	Lack of Standardization of Integrated DT and AI in construction projects	C-DT-AI6
7	Lack of logistics and availability of IT infrastructure	C-DT-AI7

, respectively.

This comprehensive review contributed to the development of an analytical framework, which was subsequently used to design the survey questionnaire as shown in Supplementary Materials. The initial list of questions was examined through semi-structured interviews with experts possessing in-depth knowledge of the Saudi construction industry, the application of digital technologies, and broader digital transformation. These interviews included an additional subheading addressing the resources required for the implementation of AI-driven DTs. Furthermore, benchmarking with existing studies, as shown in

Table 10. Shows resource implementation of AI-based DTs.

No.	Resource Implementation	Symbol
1	Training and education	RI-DT-AI1
2	Access to case studies and best practices	RI-DT-AI2
3	Financial incentives or subsidies	RI-DT-AI3
4	Technical support from vendors	RI-DT-AI4
5	Resolving data privacy and security concerns	RI-DT-AI5

, helped to refine and expand the questionnaire content.

This step was followed by a pilot survey to improve the clarity of the questionnaire items and, ultimately, to enhance the validity and credibility of the final research findings. Finally, the validated questionnaire was distributed to the respondents.

3.2. Survey Development and Validation

As outlined above, the survey questions were initially developed through a comprehensive review of relevant literature to achieve the research’s objectives. These questions were subsequently evaluated and refined through semi-structured interviews with five experts from the Saudi construction industry. To ensure the validity and adequacy of the research findings, a pilot survey was conducted to test the effectiveness and clarity of the questionnaire. This two-step process ensured that the survey questions were aligned with the study’s objectives and were clear, relevant, and comprehensive. During this phase, the experts recommended adding and modifying specific components, such as including a “resources” section with several recommendation strategies, as presented in Table 5. The questionnaire was structured into six sections. The first section collected demographic data from participants, while the remaining five sections focused on specific research topics using targeted questions. The questionnaire included closed-ended questions, allowing respondents to score their responses on a five-point Likert scale, ranging from “very low” (1) to “very high” (5). Participants were asked to rate the importance of each factor based on their experiences, reflecting the challenges and benefits associated with digital transformation. No additional responses were received, resulting in 106 complete responses retained for analysis. Given that the population size was unknown, an infinite population was assumed to be used to determine the sample size using Equation (1) [74].

$$sample size={\displaystyle \frac{Z^2\times p\times \left(1-p\right)}{C^2}}$$

(1)

In this equation:

• The sample size was 106.
• Z = 1.96, corresponding to a 95% confidence level.
• p represents the probability of the choice, expressed as a decimal, and was set to 0.5.
• C denotes the confidence interval, which must be less than 0.2.

According to Equation (1), the calculated $C$-value is 0.095, which is below the threshold of 0.2. Therefore, the sample size is deemed acceptable [74,75].

3.3. Data Collection

The questionnaire was developed using the SurveyMonkey platform and distributed primarily through the Saudi Engineering Council, as well as via email and various social media channels. A total of 142 responses were received, of which 106 were fully completed. The survey specifically targeted participants with substantial professional experience to ensure the reliability of the data, reflecting real-world conditions in the field. To qualify for participation, respondents were required to meet the following criteria: hold a specific professional role within their organization, such as architect, engineer, project manager, contractor, or consultant, with an option to specify their role if it was not listed; demonstrate an awareness of digital twin technologies and AI applications; and possess at least five years of experience in the Saudi construction sector. These criteria ensured that the survey captured insights from knowledgeable and experienced professionals, thereby enhancing the validity of the findings.

3.4. Preparing the Survey Data

This section describes the preparation of the survey data for integration into the CA method to assess the benefits and challenges of DTs and the integration of DTs with AI. Data preparation consists of coding and processing missing data. Coding the survey data using a five-point Likert scale is essential for data analysis. This involves assigning numerical values to the participants’ responses to facilitate the quantitative analysis. The coded data were used to convert the linguistic variables (five-point Likert scale) to ordinal variables. The Likert scales of Very low, Low, Medium, High, and Very high were coded as 1, 2,3, 4, and 5, respectively.

The missing data process is essential when dealing with questionnaire data because missing data can significantly impact the integrity and validity of research findings. Missing data can be attributed to the absence of responses or incomplete information in the collected survey. To ensure data completeness, it is crucial to address missing data effectively. Several common methods and techniques address missing questionnaire data. The participants’ responses to the survey were complete, and there were no missing answers; therefore, no type of missing data processing was applied.

3.5. Identification of Significant Benefits and Challenges of Applying the Correspondence Analysis Method

The data were structured into a contingency table where the rows represent the responses and the columns represent the factors. Each cell represents the response using a Likert scale point. The computing of the contribution of factors using a five-point Likert scale through CA involves identifying the relative importance of factors by analyzing the relationships between the rows (respondents) and columns (factors). The process is outlined in the following steps, adapted from [72]. Correspondence analysis (CA) is performed to analyze the association between rows (respondents) and columns (factors). The total of all values in the table is first computed using Equation (2):

$$N={\sum }_{i=1}^n{\sum }_{j=1}^m{a}_{ij}$$

(2)

where n represents the number of rows (respondents), and m represents the columns (factors). Then, the row totals and column totals are calculated using Equations (3) and (4), respectively:

$$r_i={\sum }_{j=1}^m{a}_{ij}$$

(3)

$$c_j={\sum }_{i=1}^n{a}_{ij}$$

(4)

After that, each cell $a_{ij}$ is divided by the grand N as in Equation (5) and $P_{ij} \mathrm{i}\mathrm{s}\mathrm{o}\mathrm{b}\mathrm{t}\mathrm{a}\mathrm{i}\mathrm{n}\mathrm{e}\mathrm{d}$:

$$P_{ij}={\displaystyle \frac{a_{ij}}{N}}$$

(5)

Computing Row and Column Profiles

The row profile ($f_{ij}$) and column profiles ($g_{ij}$) are computed using Equation (6) and Equation (7), respectively.

$$f_{ij}={\displaystyle \frac{a_{ij}}{ri}}$$

(6)

$$g_{ij}={\displaystyle \frac{a_{ij}}{c_j}}$$

(7)

Then, the expected frequencies under independence ($e_{ij}$) can be computed using Equation (8):

$$e_{ij}={\displaystyle \frac{r_i{c}_j}{N}}$$

(8)

The deviations from independence ($d_{ij}$) are shown in Equation (9):

$$d_{ij}={\displaystyle \frac{P_{ij}-e_{ij}}{\sqrt{e_{ij}}}}$$

(9)

The total inertia measures the variance (or dispersion) in the data, which can be explained by the correspondence analysis. It is computed as:

$$I={\sum }_i{\sum }_j{\displaystyle \frac{{\left(a_{ij}-e_{ij}\right)}^2}{e_{ij}}}$$

(10)

Singular value decomposition (SVD) is performed for the D matrix as in Equation (11):

$$D=U\sum V^T$$

(11)

where U is the left singular vectors (row coordinates), Σ is singular values (strength of association), and V is right singular vectors (column coordinates). After that, the principal coordinates for rows (F) and columns (G) (based on the largest singular values) are extracted using Equations (12) and (13), respectively:

$$F=U\sum$$

(12)

$$G=V\sum$$

(13)

The contributions of rows and columns to each dimension are computed using Equation (14) to determine the importance of factors:

$${Contribution}_j={\displaystyle \frac{{\left(v_{ik}{\sigma }_k\right)}^2}{\sum _{j=1}^J{\left(v_{ik}{\sigma }_k\right)}^2}}$$

(14)

where $v_{ik}$ is j-th element of the k-th column of V (column principal coordinates), and ${\sigma }_k$ is k-th singular value. Based on the contribution value of factors, the factors are ranked.

4. Results and Discussion

4.1. Participant Demography

The dataset offers valuable insights into the demographics of participants in the construction industry, focusing on their roles, years of experience, and knowledge of digital twins and AI technologies. A descriptive analysis of each category is presented in Figure 2.

According to the analysis, architects constitute the largest group of participants, comprising 28.3% (30 out of 106), while contractors represent the smallest group, accounting for 6.6% (7 participants). Regarding participants’ years of experience in the construction industry, 36.8% (39 participants) have 11–20 years of experience, and 28.3% (30 participants) have more than 20 years of experience. Together, this indicates that 65% of participants possess over 11 years of experience, reflecting a significant portion of highly experienced professionals. Additionally, approximately 18% of participants reported having between 5 and 10 years of experience. Regarding familiarity with DT technology, 81% of participants reported medium to higher levels of familiarity, while only 5.7% indicated very low familiarity. Similarly, for familiarity with AI-based DT applications, 53% of participants reported medium familiarity, and 21% reported familiarity above medium levels. In contrast, 25.5% indicated low familiarity with this concept. The data highlight that participants are predominantly experienced professionals, with nearly two-thirds having over 11 years of experience in the construction sector. Most respondents demonstrated moderate familiarity with digital twin technology and AI integration. Architects and project managers represent the majority of respondents, ensuring a balanced representation of both strategic and technical perspectives within the industry.

4.2. Adequate Size and Reliability of the Sample

To assess the internal consistency and reliability of the collected data, Cronbach’s alpha coefficient was computed using IBM SPSS software version 26. As shown in

Table 11. Cronbach’s alpha for the five groups.

No.	Group	Cronbach’s Alpha
1	Benefits of using DTs	0.909
2	Benefits of integrating AI-based DTs	0.883
3	Challenges of using DTs	0.883
4	Challenges of integrating AI-based DTs	0.872
5	Resource implementation of AI-based DTs	0.853

, the Cronbach’s alpha values for the groups of benefits of using DTs and advantages of integrating AI-based DTs are 0.909 and 0.883, respectively. In addition, values for the group of challenges of using DTs and the group of challenges of integrating AI-based DTs are 0.883 and 0.872, respectively, as shown in Table 11. The value was 0.853 for the group for the resource implementation of AI-based DTs. These values exceed the acceptable level (0.70). This result indicates that the internal consistency of data for the five groups was satisfied.

Out of the 142 responses received, 106 were completed and analyzed. This sample size is sufficient and reliable [76]. Several attempts were made to expand the sample size and increase participant diversity by sending the survey link to local institutions, including the Saudi Engineering Council, the Contractor Commission, and ACE firms. However, the total number of responses received was 142, of which only 106 were fully completed and analyzed. Figure 2 shows the demographics of the research participants.

The study involves a complex dataset with multiple variables (17 benefits and 16 challenges) related to DT adoption, 3 benefits and 7 challenges for integrating AI-based digital twins (DTs) in the construction industry, and five resources for enablers of the AI-based DT integration.

4.3. Significant Benefits of Using DTs in the Construction Industry

According to Figure 3, three significant BDTs are BDT17 (better documentation and communication), BDT10 (facilities management and operations), and BDT14 (real-time monitoring and control of construction processes). The contribution values of BDT17, BDT10, and BDT14 are 0.062, 0.0614, and 0.0610, respectively, as shown in Figure 3.

Better documentation and communication (BDT17) is a significant benefit of utilizing DT technology in construction. For instance, DTs leverage sensors to provide real-time communication and documentation. This real-time communication and data synchronization across DTs applications enables decision makers to identify potential issues and promptly adapt strategies. Consequently, this approach facilitates real-time project tracking while ensuring adherence to estimated costs, maintaining safety, and meeting upon agreement quality. These findings align with the results of previous studies [4,16].

Facilities management and operation (BDT10) benefits contribute significantly to ensuring project success. The synchronization benefits, real-time monitoring and control of the project process, better communication, and documentation collectively enhance the collaboration process among stakeholders and subsequently result in facilitating the production of accurate and up-to-date as-built drawings and models. A core rationale for employing DTs in facility management is due to their ability to foster interdisciplinary collaboration for space management and streamline the digitization of operations and maintenance processes [28]. The main feature of DTs is real-time monitoring and control via various communication tools to facilitate management and operations for predictive maintenance, operational efficiency, and the elimination of unnecessary adherence to fixed schedules, thus enhancing cost-effectiveness [41]. This is also in agreement with previous studies’ findings, such as [5,22], regarding the critical role of DTs.

The real-time monitoring and control of construction processes (BDT14) represent significant benefits of DT technology in the construction industry. To elaborate, DTs facilitate the real-time monitoring and control of site progress and workforce activities. This capability, in turn, enables the identification of potential risks, resource shortages, and delays that could lead to project progress disruptions and cost overruns. By leveraging these real-time data, DTs allow for accurate predictions of maintenance needs and the development of scheduling strategies that support sustainable maintenance management [23]. This contrasts with traditional preventive maintenance, which is executed according to pre-determined schedules to avoid system failures. Real-time monitoring facilitated by DTs for predictive maintenance improves operational efficiency and eliminates unnecessary adherence to fixed schedules, thus enhancing cost-effectiveness [41]. This also agrees with previous studies’ findings, such as [5,22], as the critical purpose of DT technology is keeping stakeholders informed with updated drawings and models, thereby supporting more effective project management and decision-making. This, in turn, showcases DTs as a more cost-effective system.

It could be concluded that the integration of predictive maintenance within DT frameworks enhances project efficiency and reinforces the role of DTs as transformative tools in facility management sectors.

4.4. Significant Benefits of Integrating AI-Based DTs in the Construction Industry

There are only three benefits to integrating AI-based DTs. They are ranked based on the contribution value, as shown in Figure 4 as B-DT-AI-3 (AI capabilities to improve analytics), B-DT-AI-1 (AI facilitates digital twins in modeling complex real-world systems), and B-DT-AI-2 (facilitates strategic choice-making). The contribution coefficients of B-DT-AI-3, B-DT-AI-1, and B-DT-AI-2 are 0.335, 0.333, and 0.330, respectively, as shown in Figure 4.

The CA results reveal that the integration of AI significantly enhances DTs’ ability to improve analytics (B-DT-AI-3). That is, construction projects are inherently complex, involving a wide range of interrelated elements, including materials, equipment, personnel, weather conditions, and regulatory requirements. Consequently, integrating AI tools, such as machine learning (ML) and deep learning (DL), with DTs technology enables the real-time analysis of large-scale data collected from diverse sources. This integration facilitates project status tracking, provides recommendations for improvements based on environmental conditions, predicts potential system faults, and delivers real-time feedback to project stakeholders [10]. In contrast, traditional modeling techniques often struggle to manage this complexity effectively. Consequently, this highlights the transformative potential of AI-integrated DTs for the construction sector.

Facilitating strategic decision making (B-DT-AI-2) is another significant benefit of integrating AI with DTs. Specifically, AI algorithms further enhance the capabilities of DTs by creating dynamic and evolving scenarios. These scenarios present users with unexpected challenges, thereby providing opportunities to evaluate and improve decision-making processes [57,61,77]. DTs process and model the complex interactions inherent in construction projects with a high degree of precision [28]. This integration, in turn, transforms the decision-making process from a reactive to a proactive and data-driven process [28]. The combined capabilities of DTs and AI offer access to extensive real-time and historical data collected from diverse sources. AI algorithms analyze this data to uncover trends, patterns, and often undetectable correlations through manual analysis [20]. Consequently, this data-driven approach provides a robust foundation for making more informed and strategic decisions, significantly enhancing project outcomes and efficiency.

The research findings align with the literature, confirming that integrating AI-based DTs offers significant benefits. These include enhancing the modeling of complex real-world systems, e.g., Attaran and Celik [10], facilitating strategic decision making, e.g., Rane (2023) [9], Attaran and Celik (2023) [10], and D. J. Wortley (2024) [60], and improving analytical capabilities through AI-driven insights, e.g., Rane (2023) [9], and Attaran and Celik (2023) [10].

4.5. Significant Challenges of Using DT in the Construction Industry

Figure 5 highlights three key challenges hindering the adoption of DTs: C-DT14 (lack of skilled professionals), C-DT16 (lack of national guidelines/codes of practices), and C-DT13 (resistance to changing to the new system). The contribution values of C-DT14, C-DT16, and C-DT13 are 0.068, 0.067, and 0.065, respectively, as shown in Figure 5.

The shortage of skilled professionals who are proficient in utilizing digital twin technology within the construction industry is among the most significant challenges currently facing the sector. This challenge is closely related to human-centered issues and is commonly observed when introducing any new technology or system in different contexts. One of the root causes may be the misalignment between architecture school curricula and the advancements of the Fourth Industrial Revolution, which has led to a noticeable gap between academia and the evolving needs of the industry. Notably, the findings of this study align with previous research, which highlights the impact of insufficient experience and knowledge on the successful adoption of digital twin technology [4,5,44,45].

Regarding the lack of national guidelines/codes of practices (C-DT16), the National Digital Twin Programme (NDTP) emphasizes the importance of developing standards, processes, and tools to ensure that digital twins are safe, secure, trustworthy, and ethical. Without such guidelines, organizations may adopt digital twin technologies in inconsistent ways, leading to fragmented practices and outcomes. The program aims to create a functioning market for digital twins by aligning with broader strategic priorities and ensuring interoperability and scalability [78]. Moreover, numerous IoT applications, including sensors, drones, and laser scanning technologies, are employed to capture real-time project data. Despite their utility, interoperability issues often stem from inadequate IT infrastructure, poor data quality, missing datasets, and a lack of data standardization. These limitations collectively create substantial barriers to the seamless integration and utilization of real-time data within DTs frameworks, subsequently impeding their effective implementation in construction projects. This finding is consistent with [28], as common data standards are essential enablers of DT adoption. Hence, to overcome such challenges, it is essential to invest in robust IT infrastructure, develop standardized data protocols, and enhance data management practices. Such efforts would significantly improve interoperability, facilitating the successful adoption and optimization of DTs technologies in the construction sector.

Resistance to change (C-DT13) remains a significant barrier to the adoption of transformative technology in the construction industry. This resistance persists despite advancements aimed at facilitating the transition. It is predominantly driven by high implementation costs, insufficient skills, and limited stakeholder knowledge. These findings are consistent with the literature highlighting the necessity of addressing organizational, human, and environmental challenges to successfully adopt DTs [28]. To overcome this resistance, it is essential to demonstrate the value of these technologies via training, educational programs, and social media [28].

4.6. Significant Challenges of Integrating AI-Based DTs in the Construction Industry

According to their contribution value, four significant challenges of AI-based DTs in Saudi Arabia’s Construction Industry are C-DT-AI6 (lack of standardization of integrated DT and AI in construction projects), C-DT13 (lack of understanding of AI capabilities), and lack of logistics and availability of IT infrastructure (C-DT7), and the complexity of AI algorithms (C-DT-AI1), as shown in Figure 6. The contribution values of C-DTI6, C-DTI3, C-DTI7, and C-DT-AI1 are 0.1470, 0.1449, and 0.1442, respectively, as shown in Figure 6.

According to the contribution value, one of the most significant challenges hindering the adoption of AI-based digital twins (DTs) in construction projects is the lack of standardization of integrated AI-based DTs in construction projects (C-DT-AI6) that govern their development, implementation, and integration. As highlighted by several researchers, the lack of universal standards results in fragmented deployments [63,64,65], where different platforms and tools adopt varying data structures, modeling techniques, and system architectures, ultimately hindering seamless interoperability across project stakeholders [79,80]. Furthermore, the absence of unified data protocols complicates real-time data exchange, which is essential for the effective functioning of AI-enhanced DTs in construction environments [81,82,83]. Without consistent guidelines to ensure compatibility across platforms, the full potential of AI-based DTs in enhancing predictive analytics, performance monitoring, and decision support remains largely unrealized. Consequently, there is an urgent need to develop industry-wide standards that address these gaps, thereby fostering integration, data consistency, and cross-platform functionality within the digital twin ecosystems in construction [81,84].

A limited understanding of AI capabilities (C-DT-AI3) among stakeholders in the construction industry frequently fosters unrealistic expectations regarding its potential and limitations. Such misconceptions can result in considerable challenges, including inefficient investments in unsuitable technologies and disappointment when AI fails to meet overly ambitious expectations. Bridging this knowledge gap necessitates targeted awareness initiatives and specialized training to ensure stakeholder expectations align with AI’s practical applications and benefits [59].

The complexity of AI algorithms (C-DT-AI1) has been identified as a challenge in the integration of AI DTs within the construction industry. This challenge could arise from the need for specialized expertise and a deep understanding of advanced computational techniques to achieve optimal performance [3]. In this context, the successful implementation of AI-driven DTs solutions necessitates extensive knowledge of machine learning, data analytics, and system integration, underscoring the importance of employing skilled professionals. Equally, the lack of logistics and the limited availability of IT infrastructure (C-DT-AI7) have an equal value in terms of the challenges of using AI-based DTs. Construction projects generate vast amounts of data from diverse sources, including sensors, BIM models, spreadsheets, drones, and laser scanning. However, these captured data can be lost or rendered incomplete due to inadequate IT infrastructure, poor communication systems, or insufficient data redundancy mechanisms. As a result, inconsistencies, missing information, or poorly formatted data hinder the ability to develop a comprehensive and unified view of the project within DT frameworks. These challenges compromise the reliability of DT systems and impede effective decision making and project management [46].

The study’s findings inform policy decisions by advocating for government incentives (e.g., AI-DT adoption subsidies), national data standardization guidelines, and mandatory training programs to address skill gaps, aligning with the Saudi Vision 2030’s digital goals. For industry best practices, the results emphasize prioritizing AI-DT integration in predictive maintenance, real-time monitoring, and stakeholder collaboration, supported by vendor partnerships and phased implementation to mitigate resistance. Future research directions include exploring cultural barriers, ethical AI frameworks, and hybrid methodologies (e.g., multi-case studies with IoT/BIM integration) while expanding diversity in sample demographics and regional representation to generalize findings across the Gulf Cooperation Council (GCC) construction ecosystem.

4.7. Significant Resource Implementation of AI-Based DTs

Figure 7 shows the ranking of resources required to successfully implement AI-based DTs. The three significant resources are training and education (RI-DT-AI1), technical support from vendors (RI-DT-AI4), and resolving data privacy and security concerns (RI-DT-AI5). The contribution values of (RI-DT-AI1), (RI-DT-AI4), and (RI-DT-AI5) are 0.205, 0.199, and 0.198, respectively, as shown in Figure 7.

Comprehensive training and education programs (RI-DT-AI1), technical support from vendors (RI-DT-AI4), and resolving data privacy and security concerns (RI-DT-AI5) are essential resources for facilitating the successful integration of AI with DT technology. Such strategies are instrumental in overcoming core and related challenges, including the complexity of AI algorithms and a lack of professional knowledge and skills, as previously highlighted. By adopting these approaches, organizations can enhance employees’ knowledge and capabilities and increase their willingness to embrace this technology. Furthermore, these strategies help to alleviate associated challenges, thereby expediting the adoption and implementation of DTs technology more efficiently and sustainably. These findings align with previous studies, which emphasize the significance of addressing challenges related to human characteristics and organizational characteristics for the successful implementation of advanced technologies [57,58,61,85].

Notwithstanding the challenges mentioned above, the benefits of AI-based DTs greatly exceed their issues, particularly when leaders and legislators actively put the suggested tactics into practice. The transformative potential of AI-based DTs may be fully realized by eliminating the obstacles with focused strategies such as providing training programs, raising stakeholder awareness, and incorporating supportive regulations. This will accelerate innovation in the construction sector.

The analysis underscores the vast potential of digital twins and AI integration in transforming the Saudi Arabian construction industry. While the benefits include enhanced efficiency, sustainability, and decision making, the challenges—ranging from high costs to data, infrastructure, and trust issues—must be addressed. Resource allocation in the form of training, financial support, and technical assistance is crucial for overcoming these barriers and enabling the successful adoption of AI-based digital twins.

The results of this study’s Correspondent Analysis (CA) offer valuable insights that can guide practical decision-making in the construction industry, especially regarding the adoption of AI-based digital twins (DTs). For instance, the analysis highlighted the primary advantages of integrating AI-based DTs, including automated site progress monitoring, predictive maintenance and scheduling, and improved decision making. Construction companies can leverage these findings by prioritizing investments in technologies that facilitate these benefits, such as IoT sensors and AI-driven analytics platforms for automation and predictive capabilities. Additionally, the study identified significant challenges, such as the complexity of AI algorithms, issues with data quality, and resistance to change. To tackle the complexity of AI algorithms and overcome resistance, construction firms can develop targeted employee training programs. These initiatives should focus on enhancing AI literacy and showcasing the practical advantages of implementing AI-based DTs [86].

This study’s findings largely align with global literature on AI-based digital twins (DTs), emphasizing benefits such as enhanced decision making, predictive maintenance, and complex system modeling, and challenges such as data quality issues, algorithmic complexity, and resistance to change. It advances the discourse by contextualizing these factors within Saudi Arabia’s unique construction landscape, where rapid sectoral expansion under Vision 2030 amplifies the urgency of addressing localized barriers such as interoperability challenges with IoT sensors, cultural resistance to technological adoption, and insufficient IT infrastructure. While prior studies [4,32,33] broadly identify data standardization and workforce skills as hurdles, this research quantifies their significance in the Saudi context through CA, revealing resistance to change as a top-ranked challenge despite practitioners’ awareness of benefits—a nuance that is less strongly emphasized in global literature. Additionally, the prioritization of vendor support and training as critical enablers offers actionable refinements to existing frameworks, addressing gaps in region-specific implementation strategies and aligning with Saudi Arabia’s socio-technical and economic priorities, thereby enriching the global DTs discourse with empirical evidence from an underexplored yet strategically vital market.

5. Limitations and Future Research Directions

This study offers valuable insights into the integration of AI-based digital twins (DTs) within Saudi Arabia’s construction industry; however, it is subject to several limitations that should be addressed in future research. First, the study’s reliance on survey data introduces the potential for respondent bias, as participants’ subjective perceptions and experiences may influence their responses. Second, the sample size of 106 responses, while sufficient for an exploratory analysis, may not fully capture the heterogeneity of the architecture, engineering, and construction (AEC) sector, potentially limiting the generalizability of the findings. Third, the study primarily identifies benefits and challenges without extensively examining the organizational and cultural dynamics that influence the adoption of AI-based DTs.

Future research should adopt a multi-method approach incorporating longitudinal studies, qualitative investigations, and comparative analyses across different construction contexts to advance the understanding of AI-based DT adoption. A multi-case study design would provide a more nuanced understanding of how organizational structures, workforce capabilities, and institutional policies shape the implementation of AI-based DTs. Moreover, integrating theoretical frameworks such as the Diffusion of Innovation theory could offer deeper insights into technology diffusion mechanisms and the factors that facilitate or hinder its widespread adoption. A systematic literature review focusing on AI, DTs, and the construction industry over the past five years could further elucidate research gaps and emerging trends. Additionally, future research should explore the long-term impacts of AI-based DTs, including their role in enhancing decision making, optimizing project performance, and mitigating risks in construction.

Additionally, qualitative research exploring organizational and cultural factors, such as resistance to change, workforce training, and institutional support, would offer valuable context-specific strategies for overcoming adoption barriers. Comparative studies across different regions and construction sectors could also highlight best practices and unique challenges, fostering a more global perspective. Furthermore, research into developing tailored training programs and educational initiatives could address the skill gaps and enhance the understanding of AI capabilities within the construction workforce. Investigating the intersection of AI-based DTs with emerging technologies such as blockchain and advanced analytics could also unlock new dimensions of innovation, further strengthening the digital transformation of the construction sector, as shown in Figure 8.

6. Conclusions

The advancement of digital twin technology has gained significant momentum in recent years, driven by a growing body of academic research and strategic investments from industry leaders aimed at its development and widespread adoption. In this context, this study aimed to explore the benefits and challenges associated with implementing AI-driven digital twins within the Saudi construction industry. This was achieved through a survey study, with the data collected from 106 industry professionals. The findings underscore the transformative potential of digital twins and AI-driven DTs within Saudi Arabia’s construction sector, aligning with global research and practice. The study identified several key benefits of adopting digital twins in the construction industry, including improved documentation and communication, enhanced facilities management and operations, and the real-time monitoring and control of construction processes.

Furthermore, the study revealed that integrating AI-driven digital twins provides significant benefits such as enhanced analytical capabilities, supports the modeling of complex real-world systems, and facilitates strategic decision making. Despite these advantages, several significant challenges persist. These include the absence of standardized frameworks for integrating AI-driven DTs in construction projects, limited awareness of AI capabilities, logistical constraints, inadequate IT infrastructure, and the complexity of AI algorithms.

Furthermore, the correspondence analysis (CA) demonstrated that the research quantifies the prominence of socio-cultural and technical barriers, such as resistance rooted in institutional inertia and skills gaps that hinder adoption despite practitioners recognizing the benefits of DTs. Notably, the research highlights the primary advantages and obstacles associated with DT adoption in Saudi Arabia’s construction projects. Based on these insights, integrating AI-based DTs presents substantial additional benefits for Saudi projects. The study’s conclusions align with the existing literature, demonstrating that AI integration enables intelligent, real-time functionalities, including decision-support mechanisms, “what-if” scenario simulations, and predictive maintenance. Consequently, such capabilities significantly improve cost efficiency and project timelines, ultimately leading to more favorable project outcomes. Simultaneously, the study identifies critical barriers to DT implementation, such as the shortage of skilled professionals, the absence of national guidelines and standardized codes of practice, resistance to transitioning towards new systems, and the high costs associated with implementation. Beyond identifying these challenges, the study extends its scope to explore potential strategies for optimizing the implementation of AI-based DTs. The findings suggest that, while AI applications exhibit immense potential to enhance process efficiency and intelligence, their successful adoption is fundamentally contingent upon three interdependent factors: human, environmental, and technological characteristics. This underscores the necessity of a comprehensive approach that holistically addresses these dimensions to fully unlock AI-based DTs’ transformative benefits. To accelerate the integration of AI-based DTs in Saudi Arabia’s construction sector, stakeholders must prioritize institutional capacity building through targeted training programs, policy standardization to ensure data interoperability and cybersecurity, and the establishment of public–private partnerships. Additionally, cultural change management is essential, necessitating awareness campaigns and leadership advocacy to mitigate resistance and foster acceptance of these emerging technologies. By implementing these strategies, the construction sector can effectively align with Saudi Arabia’s Vision 2030, leveraging AI-based DTs to drive digital transformation.

In terms of the theoretical implications, this study makes a significant contribution to theoretical knowledge by providing empirical evidence from 106 Saudi professionals on their awareness and perceptions of the implementation of AI-driven DTs. It identifies context-specific benefits unique to Saudi Arabia’s digital transformation environment. These findings show the construction sector’s readiness to adopt and leverage AI-driven DT technologies. Additionally, the research highlights key knowledge gaps and barriers that obstruct the complete optimization of AI-driven DTs, thereby enriching academic discourse. By highlighting areas that require further research investment, this study provides a foundation for future investigations to address these obstacles and unlock the full transformative potential of AI-based DTs.

In terms of practical implications, the CA findings identified significant benefits and challenges in adopting such technologies, which led to the formulation of actionable insights and strategies for decision makers, policymakers, and industrial developers. These include:

–

Initiating targeted and structured training programs to enhance the technical and managerial competencies required for the implementation of AI-based DTs;

–

Developing robust data standardization protocols to ensure consistency, interoperability, and reliability across digital systems;

–

Establishing incentive and reward mechanisms to motivate organizational engagement and encourage the adoption of AI-DT technologies;

–

Enhancing IT infrastructure to support the scalability and operational reliability of DT applications across diverse construction environments;

–

Advancing policy development and standardization by introducing data interoperability frameworks, IoT integration policies, and comprehensive cybersecurity measures to enable the safe and scalable deployment of AI-DTs;

–

Fostering public–private partnerships to serve as incubators for pilot projects that validate AI-DT solutions in real-world settings, thereby generating practical evidence of their impact and reducing resistance to technological change.

Integrating these methods into existing construction codes, regulatory frameworks, and national digital transformation strategies will enhance clarity, increase stakeholder confidence, and help institutionalize AI-driven digital twins across project lifecycles. In conclusion, the study offers a dual contribution: it extends theoretical knowledge through context-specific empirical evidence and provides a strategic framework to transform these insights into operational practices and national standards. Doing so equips policymakers, developers, and industry leaders with practical tools to accelerate the integration of AI-based digital twins, thus aligning with the goals of Saudi Arabia’s Vision 2030 and paving the way for a more intelligent, efficient, and sustainable construction industry.

Contrary to expectations, the findings reveal that resistance to change remains a significant barrier for professionals adopting new systems. This resistance is often attributed to a lack of awareness regarding both the benefits and challenges essential for effectively adopting emerging technologies. Although participants demonstrated a reasonable level of understanding concerning AI-based technologies, they still perceive resistance to change as a critical obstacle. Beyond the scope of this study, several underlying factors may contribute to this resistance, including fear of failure, insufficient technical skills and knowledge, dependence on conventional practices, inadequate institutional support, cultural and behavioral influences, and the time and effort required for adaptation. Therefore, addressing these challenges is imperative to facilitating wider acceptance, enhancing implementation efforts, and ultimately unlocking the full transformative potential of AI-driven digital twins in the construction industry.