Exploring Barriers to the Adoption of Digital Technologies for Circular Economy Practices in the Construction Industry in Developing Countries: A Case of Ghana

Abstract

Despite the potential of digital transformation to enhance resource efficiency and waste reduction, numerous barriers hinder its adoption. This study examines the critical barriers to digital technology adoption for circular economy implementation in the construction industry in developing countries, using Ghana as a case study. A structured quantitative approach was employed, integrating mean score ranking, exploratory factor analysis, and fuzzy synthetic evaluation to assess the severity of identified barriers. Data were collected from construction professionals through structured surveys, and statistical analyses were performed using SPSS, Excel, and RStudio to determine the criticality of the barriers. The fuzzy synthetic evaluation revealed that financial and adoption constraints emerged as the most critical barrier group, followed closely by institutional and knowledge barriers, while technological and market limitations and regulatory and organizational challenges also exhibited significant impediments. In response, this study develops a strategic framework comprising targeted solutions such as financial incentives, capacity building, regulatory reforms, and technological infrastructure development. This framework addresses not only the barriers but also the associated risks, including financial uncertainty, data security threats, and regulatory gaps. This study contributes to the theoretical understanding of digital technology adoption in CE practices and offers practical recommendations for policymakers, industry stakeholders, and academics seeking to foster sustainable construction practices in the construction industry.

1. Introduction

The construction industry is a major global driver of economic growth, contributing approximately 13% to the global GDP and employing 7% of the global workforce [1]. However, it is also one of the most resource-intensive and environmentally damaging industries, responsible for 36% of global energy consumption, 39% of CO₂ emissions, and one-third of total global waste [2,3]. Cement manufacturing for instance, which plays a crucial role in the construction industry, is the second-largest emitter of CO₂, contributing 5–8% of global emissions. Major sources include the calcination of limestone and the high-temperature combustion of fossil fuels for clinker production, which together account for nearly all emissions. Producing one ton of clinker emits about 0.83 tons of CO₂. Cement manufacturing is also highly energy-intensive [4] and highlighted as a big problem for the environment [5]. This environmental footprint, coupled with the growing demand for infrastructure, particularly in developing countries, underscores the urgency for sustainable construction practices. The circular economy (CE) concept has gained global traction as a response to these challenges, advocating for resource efficiency, material recovery, and waste minimization in the built environment [6,7,8]. Despite its potential, the implementation of CE principles remains limited, especially in the construction industries of developing nations, due to various barriers, including technological constraints, lack of regulatory frameworks, financial limitations, and inadequate digitalization [9].

In Africa, waste management has traditionally relied on landfill disposal, a practice that is becoming increasingly unsustainable as land availability diminishes [10,11]. Similarly to global trends, the construction industry of developing economies is a significant contributor to carbon emissions and environmental pollution, releasing hazardous materials that compromise both economic and social sustainability aside its tremendous environmental impact. The persistence of linear economic models in construction not only exacerbates resource depletion but also undermines long-term resilience. Therefore, rethinking conventional business models is imperative for construction firms in developing countries to align with sustainability imperatives.

Circular economy presents a transformative opportunity for the construction industry [11,12], offering pathways to reduce waste, optimize resource efficiency, and lower carbon footprints [13]. By adopting CE principles such as material recovery, reuse, and closed-loop supply chains, the industry can transition toward more sustainable and regenerative practices. Mhlanga et al. [14] emphasize that the construction industry in developing economies is still in its growth phase, making it crucial to integrate circularity at this developmental stage. Neglecting CE principles at this juncture could severely hinder these economies’ ability to contribute meaningfully to the United Nations Sustainable Development Goals (SDGs). Hence, a proactive shift towards circular construction practices is not only beneficial but also essential for ensuring the industry’s long-term sustainability and competitiveness on a global scale [11].

Digital technologies (DTs), including Building Information Modeling (BIM), blockchain technology, digital twins, and the Internet of Things (IoT), have been recognized as critical enablers of CE in the construction industry [15,16,17]. Studies that incorporate DTs, including BIM for tracking materials and the IoT for real-time monitoring of construction materials, highlight their crucial role in promoting CE practices within the construction industry [18]. These technologies facilitate real-time data collection, material tracking, predictive analytics, and decentralized information sharing, thereby optimizing material reuse and reducing waste generation [1]. Despite the documented benefits, the adoption of DTs for CE implementation remains significantly lower in developing economies, where construction practices are often fragmented, informally structured, and reliant on traditional linear models [19]. The case of Ghana is particularly relevant, as the country’s rapid urbanization and infrastructure expansion exacerbate waste management challenges, resource inefficiencies, and sustainability gaps [20]. The limited penetration of digitalization in Ghana’s construction sector further hinders the realization of circularity in material usage, recycling processes, and waste minimization [21].

Existing research has largely focused on digitalization and CE adoption in developed economies, where regulatory policies, technological infrastructure, and financial support mechanisms are more advanced [15,16]. Studies have examined how blockchain enhances construction waste tracking [19], the role of digital twins in material reuse [22], and the leveraging DT for CE implementation [1]. However, there remains a significant research gap in understanding the barriers that hinder the adoption of these technologies in resource-constrained settings. The literature lacks empirical evidence on how institutional, financial, technological, and cultural barriers interplay to obstruct CE-oriented digital transformation in the construction industries of developing countries [9]. This study seeks to fill this gap by critically examining the barriers to adopting DTs for CE implementation in the construction industry of a developing country setting. This study provides a more structured and empirical understanding of the digitalization barriers, setting it apart from previous research that has predominantly relied on qualitative assessments or conceptual frameworks [1]. From an international perspective, this study contributes to the global sustainability discourse in construction by providing insights into the challenges of digital transformation in emerging markets. This study offers comparative insights that could inform policy frameworks and investment strategies aimed at bridging the digital divide in global construction industries [23]. To achieve these objectives, this study answers the research question: what are the barriers to adopting DTs to implement CE practices in the construction industry of a developing country context? By addressing these barriers, this study supports SDG 9 (Industry, Innovation, and Infrastructure), SDG 11 (Sustainable Cities and Communities), and SDG 12 (Responsible Consumption and Production). The focus on digital solutions such as BIM, blockchain, and IoT fosters innovation and infrastructure efficiency, while circular practices promote resource efficiency and waste reduction. These efforts contribute to developing sustainable and resilient infrastructure, essential for urban development in emerging economies. This study addresses the decarbonization priorities emphasized in Conference of the Parties (COP) 26 and COP 27, which call for urgent climate action and sustainable industrial practices. This study’s findings highlight how digital transformation can enhance transparency, efficiency, and sustainability in construction, thus supporting the global climate goals of reducing emissions and fostering green growth pathways, particularly in developing economies.

2. Literature Review

2.1. Circular Economy (An Overview)

The concept of the CE has gained global prominence as an alternative to the traditional linear economic model, which follows a “take-make-dispose” approach. CE emphasizes resource efficiency, waste minimization, and material reuse, aiming to close the loop in production and consumption systems [23,24]. A CE operates on R-strategies, which have evolved beyond the traditional 3R model (Reduce, Reuse, Recycle) into multi-R frameworks, such as the 10R model. This expanded model includes Refuse, Redesign, Reduce, Reuse, Repair, Refurbish, Remanufacture, Repurpose, Recycle, and Recover, aiming to decouple economic growth from resource depletion by optimizing material and product use throughout their life cycle. The primary objective is to preserve product value at its highest possible level, ensuring sustainability and long-term resource efficiency [12,23].

In the construction industry, CE principles are particularly crucial given the sector’s significant contribution to environmental degradation, including excessive resource consumption, carbon emissions, and waste generation [9]. The adoption of CE strategies in construction involves sustainable design, modular construction, material recovery, and DTs to enhance efficiency and sustainability [1].

Despite its potential benefits, the implementation of CE in construction is hindered by various factors, including regulatory gaps, high initial investment costs, and a lack of awareness among industry stakeholders [19]. While developed nations have made considerable progress in integrating CE principles into their construction practices, developing economies, such as Ghana, face significant challenges in transitioning toward circular construction models due to limited policy frameworks and inadequate infrastructure [11,12,22]. Nevertheless, the increasing global focus on sustainability, technological advancements, and evolving regulatory landscapes presents new opportunities for the construction industry to accelerate CE adoption. With emerging strategies such as DTs, urban mining, material passports, and secondary material marketplaces, the construction sector is poised to undergo a transformative shift towards sustainable and resource-efficient practices.

2.2. Digital Transformation in CE

Digital technologies are recognized as critical enablers of CE implementation in the construction industry, facilitating real-time monitoring, efficient resource utilization, and waste minimization [25]. The integration of digital solutions such as BIM, blockchain, IoT, artificial intelligence (AI), and digital twins has revolutionized construction processes by promoting transparency, improving decision-making, and optimizing resource efficiency [1].

BIM, for instance, enables better material tracking, lifecycle assessment, and efficient design practices that align with CE principles [19]. Blockchain technology enhances supply chain traceability and ensures that construction materials are sourced and used in a circular manner [9,26]. Digital twins create virtual replicas of physical structures, allowing for real-time monitoring and predictive maintenance, which reduces material waste and extends the lifespan of buildings [22].

Table 1. Digital Technologies Essential for CE Implementation in Construction.

SN	Digital Technology	Function in CE Implementation	Source(s)
1	Building Information Modeling (BIM)	Enables material tracking, lifecycle assessment, and optimized design for sustainability	[1,9,15,18,19,20,23,25,27]
2	Blockchain Technology	Blockchain fosters supply chain transparency by offering real-time data visibility across manufacturers, suppliers, and recyclers. This enables efficient reverse logistics for returned, reused, or remanufactured products	[1,9,18,19,25,26,27]
3	Internet of Things (IoT)	Facilitates real-time data collection and remote monitoring of construction materials and processes	[1,9,18,23,25,27]
4	Artificial Intelligence (AI)	Supports predictive analytics for waste minimization and optimized resource utilization	[1,9,18,23,25,28]
5	Digital Twins	Provides virtual simulations for predictive maintenance and extended building lifecycle	[9,18,22,23,25,27]
6	Big Data Analytics	Big Data Analytics helps track and analyze materials and products throughout their lifecycle, enabling better recovery, reuse, and recycling	[9,25]
7	Cloud Computing	Supports efficient data storage, access, and collaboration among stakeholders	[9,25]
8	3D Printing	Reduces material waste by enabling precise, on-demand production of components	[9,25]
9	Geographic Information Systems (GIS)	Assists in site selection, resource mapping, and impact assessment for circular construction	[9,25]
10	Robotics and Automation	Enhances efficiency in deconstruction, material recovery, and resource optimization	[9]
11	Material passport	Store detailed material properties, such as composition, toxicity, recyclability, and embodied carbon footprint. This information allows stakeholders to track materials throughout their lifecycle, from procurement to reuse, ensuring a closed-loop system	[9,15,18,20,25,26,27]

Source: Table created by authors.

presents some of the enabling technologies for CE implementation in built environment along with their functions and sources. These technologies play a crucial role in enabling circular construction by enhancing design precision, improving supply chain transparency, and facilitating material reuse.

2.3. Barriers to Digital Technologies Adoption for CE Implementation in the Construction Industry

The increasing focus on utilizing DTs to promote CE practices in the construction industry aligns with the worldwide movement toward Industry 4.0 and sustainability. Nevertheless, there is still a lack of comprehensive insight into how these technologies can be effectively applied to support circular initiatives within the sector [17]. The adoption of DTs for CE implementation in construction is impeded by various challenges, ranging from financial and technological constraints to cultural and regulatory barriers. A significant hindrance is the lack of technical expertise, as the construction industry struggles with a workforce that lacks proficiency in emerging technologies such as BIM, AI, and blockchain [19]. This knowledge gap limits the ability of professionals to fully leverage digital tools for CE adoption. Compounding this challenge is the complexity and interoperability issues of digital tools, where many platforms operate on proprietary systems that do not seamlessly integrate with others, leading to inefficiencies in data exchange and collaboration [22].

Financial constraints remain a crucial obstacle, as the high implementation costs of DTs deter widespread adoption, particularly among small and medium-sized enterprises [1]. Many construction firms are reluctant to invest in new technologies due to economic unpredictability and a lack of immediate returns, despite long-term efficiency gains. Furthermore, limited access to quality data continues to hamper data-driven decision-making in CE initiatives. Without reliable and standardized data, stakeholders face difficulties in tracking material flows and optimizing resource recovery processes [9].

Resistance to change within the industry presents another significant challenge, as many construction firms remain entrenched in traditional workflows, showing reluctance toward adopting new technologies [11]. This is further exacerbated by regulatory and policy gaps, where the absence of comprehensive CE-focused regulations and digital adoption incentives slows progress [25]. Data security and privacy concerns also remain a major issue, as firms are apprehensive about potential cyber threats, unauthorized data access, and the risk of intellectual property breaches [1].

Addressing these barriers requires a multifaceted approach, including upskilling programs for construction professionals, enhanced financial incentives to encourage digital adoption, and robust regulatory frameworks that support CE-aligned digital transformation. By overcoming these hurdles, the construction sector can transition toward a more circular and sustainable future, leveraging digital innovations to enhance efficiency and minimize environmental impact.

Table 2. Barriers to Digital Technology Adoption for Circular Economy Implementation in Construction.

S/N	Barrier	Source(s)	Explanation
B1	Lack of technical expertise	[9,25,29]	Insufficient knowledge and skills to adopt digital tools like BIM, AI, or IoT hinder their effective application in CE practices.
B2	High implementation costs	[9,22,23]	High initial investment in digital tools such as digital twins, blockchain, and RFID discourages stakeholders from adopting them for CE implementation.
B3	Fragmented construction industry	[9,22,25]	The construction industry is highly fragmented, with poor coordination between stakeholders, making the integration of digital tools for CE challenging.
B4	Lack of interoperability of digital tools	[9,23,29]	Many digital technologies (e.g., BIM and material passports) are not fully interoperable across platforms, hindering data exchange and collaboration.
B5	Limited access to quality data	[9,22,23]	The absence of accurate and reliable data on materials, components, and buildings makes it difficult to implement CE strategies effectively.
B6	Resistance to change	[9,25,29]	Cultural resistance within organizations and reluctance among stakeholders to adopt new digital processes slow down CE implementation.
B7	Regulatory and policy gaps	[9,23,25]	Lack of supportive regulations and standards for circular practices and digital technology integration hinders the transition.
B8	Data security and privacy concerns	[9,29]	Concerns about the security and privacy of data collected through digital tools such as blockchain or IoT discourage stakeholders from using these technologies.
B9	High energy consumption of technologies	[9,23]	Digital tools and processes like AI and blockchain often require significant energy resources, contradicting CE’s sustainability goals.
B10	Limited scalability	[9,25,29]	Many digital solutions are developed for specific projects and are not easily scalable across the construction industry.
B11	Lack of stakeholder collaboration	[9,23,29]	Poor communication and collaboration between stakeholders across the value chain limit the adoption of digital tools for CE.
B12	Economic unpredictability	[9,23]	Uncertainty in economic returns from CE initiatives, combined with the high costs of digital technologies, discourages investment.
B13	Limited availability of recycled materials	[9,25,29]	The insufficient supply and demand for recycled materials restrict the development of circular processes, which depend on digital tracking and lifecycle data.
B14	Cybersecurity risks	[9,23]	Increased reliance on interconnected digital systems exposes organizations to potential cybersecurity threats, making stakeholders hesitant to adopt these technologies.
B15	Complexity of digital tools	[9,29]	Many tools are overly complex for practical use, leading to resistance among end-users and a lack of perceived value.
B16	Lack of standardization	[9,23,25]	Absence of unified standards for tools like BIM and material passports leads to inefficiencies in adopting circular construction practices.
B17	Inadequate CE knowledge management	[9,23]	Lack of efficient systems for managing and disseminating CE knowledge limits the ability to implement digital tools effectively.
B18	Lack of trained workforce	[9,23,29]	Insufficient training and expertise in handling advanced digital tools limit their adoption in CE practices.
B19	Lack of CE-specific indicators	[9,23,29]	The absence of clear metrics and indicators for assessing the integration of CE practices using digital tools hinders decision-making.
B20	Poor infrastructure for circular systems	[23,29]	Inadequate infrastructure for recycling, waste separation, and material recovery hampers the effectiveness of digital tools in enabling CE.
B21	Short lifecycle of enabling technologies (RFID)	[9,23]	Technologies like RFID tags often have a shorter lifespan than construction projects, limiting their usefulness in CE practices.
B22	Need for new organizational role and training	[1,25]	The transition to CE and digital transformation requires specialized roles, such as CE strategists and digital technology experts, necessitating extensive training programs.
B23	Lack of financial resources	[19,23]	High costs of acquiring, implementing, and maintaining digital technologies deter adoption, particularly among small and medium enterprises (SMEs) in the construction industry.
B24	Slow uptake of new technologies in the construction industry	[1,11]	The construction industry is traditionally slow to adopt innovation due to entrenched practices, risk aversion, and a lack of regulatory incentives.
B25	Environmental concerns of new technologies	[9,22]	The production, use, and disposal of digital technologies, such as sensors and IoT devices, raise concerns about electronic waste and resource-intensive manufacturing.
B26	Lack of awareness of CE and/or DTs	[23,25]	Many industry stakeholders lack knowledge about CE principles and digital tools, limiting widespread adoption and engagement.
B27	Lack of commitment from stakeholders	[1,11]	The absence of clear incentives, coupled with competing priorities, results in weak commitment from policymakers, investors, and construction firms toward CE adoption.
B28	Lack of built-environment-related data	[19,22]	A lack of standardized and accessible data on building materials, construction processes, and waste streams inhibits informed decision-making and lifecycle analysis.
B29	Unavailability of web-based database for secondary products	[9,25]	The lack of digital platforms to catalog and facilitate the exchange of reused and recycled materials limits circular construction practices and material traceability.

Source: Table created by authors.

presents a summary of the variables identified from the review of pertinent literature related to the subject.

3. Methodology

The research methodology employed in this study is outlined in the following sections.

3.1. Survey Design

This study employed a quantitative research methodology to gather a broad spectrum of respondent perspectives. This approach is crucial for ensuring the generalizability of findings to a larger population and was implemented through structured questionnaire surveys. The survey-based strategy, rooted in the quantitative paradigm, utilized questionnaires to collect standardized data [30]. To minimize researcher bias and maintain objectivity, structured data collection methods and statistical analysis were employed. This methodological rigor enhances the reproducibility of the study and allows for the verification of results, thereby strengthening the reliability and validity of the research findings.

3.2. Questionnaire Design

Following a comprehensive review of relevant literature, 29 potential variables were initially identified as barriers to adopting DTs for implementing CE practices in the construction industry. These variables were then refined and structured for quantitative data collection. The measurement items were identified from previous studies as presented in Table 2, where they had been validated for reliability and content adequacy. For instance, variables such as lack of technical expertise, high implementation costs, and stakeholder resistance have been frequently examined and validated in similar contexts [1,9,11]. These previous studies demonstrated acceptable levels of reliability, as evidenced by high Cronbach’s alpha coefficients and other reliability indicators. By building upon these established constructs, the questionnaire in this study ensures content validity and facilitates comparability with prior research.

The questionnaire used in this study was structured into two key sections, each comprising closed-ended questions designed to collect targeted responses. The first section focused on gathering demographic information, while the second section aimed to capture respondents’ perspectives on barriers to digital DTs for CE implementation within a developing country context. In this section, respondents were asked to rate their agreement levels on a five-point Likert scale ranging from 1 (Strongly Disagree) to 5 (Strongly Agree) regarding the identified barriers. Prior to the data collection, a pilot study was conducted to validate the questionnaire and refine its content. This preliminary assessment aimed to identify ambiguous items or variables that might be unsuitable or lack discriminatory ability among respondents. The pilot study involved seven construction professionals with expertise in circular economy and digitalization, four of whom were academics, while three were industry practitioners actively engaged in circular construction. The pilot phase included face-to-face interviews to assess the clarity and relevance of the research instrument. Based on the feedback obtained, necessary adjustments were made, refining the final set of variables to 27 to enhance the reliability and precision of the study.

3.3. Sampling Method

Given the limited awareness of CE and digitalization among construction professionals in Ghana [31,32], a combination of purposive and snowball sampling techniques was employed to ensure the participation of qualified and knowledgeable professionals. Purposive sampling enabled the identification and selection of respondents who possessed substantial expertise, accessibility, and willingness to contribute meaningfully to the study. The selection of participants was guided by two primary criteria: (1) respondents were required to have extensive professional or research experience in the construction industry, demonstrating a strong understanding of the Ghanaian construction industry and its associated challenges; and (2) they needed to possess in-depth knowledge of CE and digitalization, particularly the barriers hindering their adoption in the Ghanaian context. Initially, 62 construction professionals meeting these criteria were identified through purposive sampling. Subsequently, the snowball sampling technique was employed, wherein additional participants were recruited (making a total of 130) based on recommendations from the initial respondents, ensuring a broader representation of experts with relevant insights on CE and digitalization within the construction industry.

3.4. Data Collection

Following the pilot study, the finalized questionnaires were disseminated to the targeted respondents (quantity surveyors, architects, engineers, project managers, site managers, CE researchers) via email using Google Forms, which was selected due to its ability to ensure respondent anonymity, unlike face-to-face interviews. A total of 150 questionnaires were distributed, yielding 130 responses, resulting in a high response rate of 87%. This response rate was deemed adequate for data analysis, as it significantly surpassed the 20–30% typical response rate observed in construction industry surveys [33].

3.5. Data Analyses

The data analysis process followed a structured three-stage protocol, encompassing mean score ranking, exploratory factor analysis (EFA), and fuzzy synthetic evaluation (FSE). The analyses were conducted using Statistical Packages for Social Scientists (SPSS) 27.0 and RStudio, ensuring a robust and systematic examination of the data. To assess internal consistency and reliability, Cronbach’s alpha was calculated, yielding a coefficient of 0.954, which exceeded the recommended threshold of 0.700 [34]. This high value confirmed the strong reliability and internal validity of the dataset. Subsequently, mean score ranking was employed to determine the relative significance of the identified barriers, based on their probability and severity. Exploratory Factor Analysis (EFA) was then performed to classify the critical barriers. The suitability of the dataset for factor analysis was first confirmed through the Kaiser–Meyer–Olkin (KMO) test (0.801) and Bartlett’s test of sphericity (p < 0.000), reinforcing the adequacy of the data for factor extraction [34,35]. Finally, Fuzzy Synthetic Evaluation (FSE) was applied to assess the overall criticality level of the barriers. This involved computing weightings for each variable, identifying principal factors, determining membership functions, and quantifying their impact using criticality indices [35,36].

The data collected through the structured questionnaire were based on expert judgments and respondent perceptions measured on a five-point Likert scale. Such data are inherently subjective and imprecise, as respondents may interpret and assess the importance of each barrier differently, introducing vagueness and linguistic uncertainty. This fuzziness arises from the qualitative nature of perception-based evaluations and the ambiguity associated with linguistic assessments such as “agree” or “strongly agree” [36]. FSE is particularly suited to handling these challenges, as it can convert subjective judgments into quantitative evaluations through membership functions, allowing a more accurate and objective synthesis of expert opinions [35]. Unlike traditional statistical methods, FSE effectively addresses uncertainty, inconsistency, and lack of precision in multi-criteria decision-making scenarios, which is why it has been widely applied in construction management studies involving similar datasets [35,36,37]. Although there is no universally accepted empirical test for the degree of fuzziness in perception-based data, the nature of Likert scale responses, coupled with the complexity of subjective judgment in evaluating barriers to digital technology adoption, justifies the application of the FSE approach. Previous studies have successfully applied FSE to analyze qualitative survey data in similar contexts without performing separate tests for fuzziness, focusing instead on the subjective and uncertain nature of the input data [35,36].

The multi-tiered analytical approach used in this study ensured a comprehensive and statistically rigorous assessment of the barriers to digital technology adoption for CE implementation in the construction industry. Testing for multicollinearity is essential to ensure that the independent variables in the dataset are not highly correlated, which can distort the results of the factor analysis or other multivariate techniques. Hence, to confirm the absence of multicollinearity among the variables used in the EFA and FSE, a correlation matrix was conducted. No correlations exceeded the threshold of 0.80. Collinearity statistics were further assessed using the Variance Inflation Factor (VIF) and Tolerance values. All VIF values remained below the critical threshold of 5, and all Tolerance values were above 0.10 confirming no multicollinearity concerns were present, ensuring the robustness of the factor structure and subsequent analysis.

4. Results

4.1. Demographic Information of Respondents

The demographic profile of respondents as presented in

Table 3. Demographic Information of Respondents.

Demographic Information	Frequency	Percentage (%)
Profession
Quantity Surveyor	42	32.3
Architect	11	8.5
Engineer	18	13.8
Project Manager	15	11.5
Site/Construction Manager	32	24.6
Researcher/Academic	12	9.3
Total	130	100
Number of years spent in current profession
1–5 years	30	23.1
6–10 years	35	26.9
11–15 years	12	9.2
16–20 years	14	10.8
More than 20 years	39	30.0
Total	130	100
Level of Education
BSc	50	38.5
MBA/MSc./MPhil	34	26.2
PhD	46	35.3
Total	130	100
Category of firm
D1K1	36	27.7
D2K2	36	27.7
D3K3	23	17.7
D4K4	17	13.1
None	18	13.8
TOTAL	130	100

Source: Table created by authors.

reflects a diverse and highly qualified group of professionals in the Ghanaian construction industry, ensuring a comprehensive assessment of the barriers to digital technology adoption for CE implementation. Among the 130 respondents, Quantity Surveyors (32.3%) and Site/Construction Managers (24.6%) formed the majority, followed by Engineers (13.8%), Project Managers (11.5%), Researchers/Academics (9.3%), and Architects (8.5%). This distribution highlights strong representation from cost management, site execution, and technical expertise, all critical for CE adoption.

In terms of industry experience, 30.0% of respondents had more than 20 years of experience, while 26.9% had 6–10 years, and 23.1% had 1–5 years, ensuring insights from both seasoned and early-career professionals. Educational qualifications were also high, with 35.3% holding a PhD, 26.2% a Master’s degree, and 38.5% a Bachelor’s degree, suggesting a strong knowledge base for evaluating CE challenges. Firm categorization varied, with D1K1 and D2K2 firms (27.7% each) representing large construction entities, while D3K3 (17.7%) and D4K4 (13.1%) comprised smaller firms. The presence of professionals across different roles, experience levels, and firm sizes ensures a well-rounded perspective, enhancing the credibility and depth of the study’s findings on digital transformation barriers in CE implementation.

4.2. Mean Score Ranking Analysis

The mean scores were calculated to assess the relative significance of the identified barriers based on their criticality and to determine the average perception of respondents using a five-point Likert scale, where “1 = Strongly Disagree” and “5 = Strongly Agree”. In accordance with the grading system employed in the study, a mean score of 3 or higher signifies that a barrier is considered significant [35]. The results of this analysis are presented in

Table 4. Mean score ranking of barriers to Digital Technology Adoption for Circular Economy Implementation in Construction.

Code	Barriers (Indicator (I))	Mean	SD	Rank
B1	Complexity of digital tools	4.677	0.587	1st
B2	Lack of technical expertise	4.669	0.640	2nd
B3	Lack of commitment from stakeholders	4.654	0.567	3rd
B4	Lack of financial resources	4.639	0.693	4th
B5	Lack of CE-specific indicators	4.615	0.801	5th
B6	Limited scalability	4.615	0.698	6th
B7	Lack of standardization	4.615	0.857	7th
B8	Economic unpredictability	4.608	0.629	8th
B9	High implementation costs	4.585	0.581	9th
B10	Short lifecycle of enabling technologies (RFID)	4.585	0.843	10th
B11	Unavailability of web-based database for secondary products	4.577	0.746	11th
B12	Lack of trained workforce	4.569	0.704	12th
B13	Limited availability of recycled materials	4.561	0.826	13th
B14	Limited access to quality data	4.561	0.826	14th
B15	Poor infrastructure for circular systems	4.554	0.727	15th
B16	Slow uptake of new technologies in the construction industry	4.554	0.636	16th
B17	Lack of awareness of CE and/or DTs	4.554	0.636	17th
B18	Lack of built-environment-related data	4.547	0.769	18th
B19	Need for new organizational role and training	4.546	0.808	19th
B20	Fragmented construction industry	4.546	0.683	20th
B21	Inadequate CE knowledge management	4.546	0.727	21st
B22	Regulatory and policy gaps	4.539	0.881	22nd
B23	Lack of interoperability of digital tools	4.523	0.637	23rd
B24	Environmental concerns of new technologies	4.515	0.770	24th
B25	Data security and privacy concerns	4.485	0.790	25th
B26	Cybersecurity risks	4.477	0.684	26th
B27	Resistance to change	4.477	0.900	27th

Source: Table created by authors.

.

The mean score ranking analysis provides a prioritization of the identified barriers to digital technology adoption for CE implementation in the construction industry. The highest-ranked barrier was the complexity of digital tools (Mean = 4.677, SD = 0.587), indicating that the intricate nature of digital solutions presents a significant challenge for industry stakeholders. This suggests that many professionals lack sufficient familiarity with digital platforms, which complicates their adoption and integration into construction workflows. Similarly, lack of technical expertise (Mean = 4.669, SD = 0.640) was ranked as the second most significant barrier, reinforcing the widespread skills gap in the industry, mostly in developing countries. Without adequate expertise, professionals struggle to effectively utilize and maximize the potential of DTs. Stakeholder commitment (Mean = 4.654, SD = 0.567) was also identified as a key issue, implying that reluctance among decision-makers to invest in CE-aligned digital solutions hinders implementation efforts. Other prominent barriers included financial constraints (Mean = 4.639, SD = 0.693), lack of CE-specific indicators (Mean = 4.615, SD = 0.801), and limited scalability of technologies (Mean = 4.615, SD = 0.698). The financial burden associated with acquiring, deploying, and maintaining digital tools has been a persistent challenge, particularly for small and medium-sized enterprises (SMEs) that may lack the capital to invest in CE-enabling technologies. The absence of standardized CE indicators further exacerbates this issue by preventing industry players from having clear benchmarks for evaluating their digital transformation progress.

Interestingly, data-related challenges such as the unavailability of a web-based database for secondary products (Mean = 4.577, SD = 0.746) and limited access to quality data (Mean = 4.5615, SD = 0.82597) were also highly ranked. This underscores the importance of data accessibility and traceability in ensuring the effective integration of digital tools in CE practices.

4.3. Results of EFA

A factor analysis was conducted using the average rankings of the 27 identified barriers to digital technology adoption for CE implementation. Factor analysis is a robust statistical technique commonly employed to identify underlying structures within a dataset by grouping correlated variables into distinct factors. However, ensuring the appropriateness and validity of factor analysis requires meeting specific methodological standards. One such criterion is the recommended variable-to-sample size ratio of 1:5, which enhances the stability and reliability of the extracted factors [38]. Nevertheless, statistical tests can also be performed to assess the suitability of factor analysis, even when the sample size does not strictly adhere to this recommended ratio. These tests include correlation and anti-image matrix assessments, the Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy, Bartlett’s test of sphericity, and reliability analysis [39]. If these statistical measures produce favorable results, factor analysis can still be carried out with confidence and accuracy, ensuring robust structure detection [35,40].

To validate the suitability of factor analysis in this study, several preliminary tests were conducted. Internal consistency was first examined using Cronbach’s α coefficient, which ranges from 0 to 1, indicating the reliability of the dataset. The analysis produced a high Cronbach’s α value of 0.954, confirming strong internal consistency and instrument reliability. The Kaiser-Meyer-Olkin (KMO) test was then performed to measure sampling adequacy, yielding a statistic of 0.801, which surpasses the 0.6 benchmark used in previous studies [35]. This result confirms that the dataset is well-suited for factor analysis, as a higher KMO value indicates that the variables exhibit sufficient common variance to justify factor extraction. Additionally, Bartlett’s test of sphericity was conducted to assess the interrelationships among variables and determine whether the dataset is appropriate for structure detection. The test yielded a Chi-square value of 5426.557 with a p-value of less than 0.000, confirming that the correlation matrix is not an identity matrix, thereby validating the feasibility of factor analysis.

Given these positive statistical indicators, the EFA was performed using Principal Component Analysis (PCA) for factor extraction. Factors were extracted based on eigenvalues greater than 1, ensuring that only components explaining substantial variance in the dataset were retained. Furthermore, Varimax rotation with Kaiser normalization was applied to simplify the factor structure and enhance interpretability. Varimax rotation was specifically chosen because it minimizes the number of variables with high loadings on a single factor, making it easier to distinguish between factors and improve the clarity of their relationships [41]. To ensure strong factor correlation, variables with absolute values below 0.50 were suppressed, as lower factor loadings may indicate weak associations within the extracted components. The results of the EFA, including the extracted factors and their respective loadings, are presented in

Table 5. Factor extraction and their loading.

Indicator Codes	Factor Groupings	Factor Loading	Eigenvalue	% of Variance Explained	Cumulative % Variance Explained
Criteria 1	Institutional and Knowledge Barriers		15.956	59.097	59.097
B7	Lack of standardization	0.830
B5	Lack of CE-specific indicators	0.770
B15	Poor infrastructure for circular systems	0.714
B21	Inadequate CE knowledge management	0.713
B18	Lack of built-environment-related data	0.712
B11	Unavailability of web-based database for secondary products	0.670
B10	Short lifecycle of enabling technologies (RFID)	0.666
B12	Lack of trained workforce	0.657
B19	Need for new organizational role and training	0.545
Criteria	Financial and Adoption Constraints		1.847	6.842	65.938
B9	High implementation costs	0.815
B17	Lack of awareness of CE and/or DTs	0.803
B16	Slow uptake of new technologies in the construction industry	0.774
B4	Lack of financial resources	0.729
B26	Cybersecurity risks	0.608
B3	Lack of commitment from stakeholders	0.590
B23	Lack of interoperability of digital tools	0.547
B1	Complexity of digital tools	0.525
Criteria	Technological and Market Limitations		1.757	6.509	72.447
B24	Environmental concerns of new technologies	0.819
B25	Data security and privacy concerns	0.816
B13	Limited availability of recycled materials	0.724
B8	Economic unpredictability	0.713
B6	Limited scalability	0.707
Criteria	Regulatory and Organizational Challenges		1.458	5.398	77.846
B2	Lack of technical expertise	0.851
B27		0.707
B14	Resistance to change	0.695
B20	Fragmented construction industry	0.569
B22	Regulatory and policy gaps	0.517

Source: Table created by authors.

.

The EFA provided deeper insights into the underlying dimensions of the barriers by grouping them into four major factors. The cumulative variance explained by these four factors (77.846%) suggests that they effectively capture the primary challenges affecting digital technology adoption for CE implementation.

• Factor 1: Institutional and Knowledge Barriers (59.097% variance explained)

This factor encompasses barriers related to standardization, infrastructure, knowledge management, workforce training, and access to built-environment-related data. The high factor loadings indicate that a lack of institutional support, structured knowledge systems, and digital literacy is one of the most significant obstacles to digital transformation. The presence of poor infrastructure for circular systems (loading = 0.714) further underscores the need for strategic investments in enabling environments, such as material recovery facilities and digital material tracking systems.

• Factor 2: Financial and Adoption Constraints (6.842% variance explained)

The barriers grouped under this factor primarily concern financial challenges, lack of awareness, and the slow uptake of new technologies. High implementation costs (loading = 0.815) and lack of financial resources (loading = 0.729) reflect the capital-intensive nature of digital adoption, which is particularly prohibitive for SMEs. Additionally, lack of commitment from stakeholders (loading = 0.590) reinforces the earlier findings from the mean score ranking that suggest a hesitation among key decision-makers to embrace digital transformation.

• Factor 3: Technological and Market Limitations (6.509% variance explained

This factor includes concerns about data security, environmental impacts, and scalability. The high loading for environmental concerns of new technologies (0.819) suggests that there are uncertainties about the sustainability impact of digital solutions, particularly with regard to e-waste management and energy consumption. Furthermore, limited availability of recycled materials (0.724) indicates a disconnect between digital adoption and material circularity, where technologies designed to support CE are constrained by supply chain inefficiencies.

• Factor 4: Regulatory and Organizational Challenges (5.398% variance explained)

The final factor highlights regulatory, policy, and industry fragmentation issues that impede digital adoption. Lack of technical expertise (loading = 0.851) is the most dominant challenge in this category, reaffirming the critical role of digital skills training in overcoming CE implementation barriers. Resistance to change (loading = 0.695) and fragmented industry structure (loading = 0.569) further illustrate how deeply ingrained traditional practices continue to hinder technological innovation in the sector. The presence of regulatory and policy gaps (loading = 0.517) suggests an urgent need for comprehensive CE-focused policies that provide clear incentives and guidelines for digital transformation.

4.4. Results of FSE

FSE is a robust multi-criteria decision-making (MCDM) tool designed to handle complex systems characterized by uncertainties and subjective judgments [36,37,42]. One of the key advantages of FSE over traditional statistical techniques is its ability to manage the inherent fuzziness associated with expert evaluations, making it highly suitable for studies involving qualitative assessments [37]. In the context of this study, the perceptions of construction professionals regarding barriers to digital technology adoption for CE implementation may be influenced by individual knowledge, experiences, and biases. FSE mitigates this subjectivity by utilizing membership functions to assign quantitative values to qualitative judgments, allowing for a more structured and objective assessment of the barriers [36].

The FSE methodology in this study follows a three-level structured approach: (1) calculating weightings for each barrier and determining critical barriers, (2) computing the membership function for each variable and factor, and (3) quantifying the impact of each principal factor using criticality indices. This systematic process ensures a comprehensive evaluation of the relative importance of different barriers, leading to a more precise prioritization of mitigation strategies. To interpret the results, a five-point Likert scale was adopted, where 1 represents “Strongly disagree” and 5 represents “Strongly agree”. A criticality threshold of 3 was established, meaning that barriers with an index greater than or equal to 3 are considered significant obstacles requiring immediate attention and intervention [36,37,42].

By implementing FSE, this study ensures a data-driven, objective approach to assessing barriers to digital transformation in CE implementation, facilitating evidence-based decision-making for policymakers, industry practitioners, and researchers. The results provide a structured foundation for developing targeted strategies to overcome the most critical barriers hindering the implementation of DTs in the construction sector. The following stages were followed to generate the fuzzy output using SPSS version 27, Excel version 2506 on Microsoft 365, and RStudio 2024.09.0.

Stage 1: The set of indicators (in this case barriers to digital transformation in CE implementation) were developed.

I = {I₁, 1₂, 1₃, 1₄, …, 1_n}; where n represents the number of indicators/variables. In this case, 27 indicators were developed from the review of pertinent literature and the piloting study (see Table 2 or Table 4).

Stage 2: The labels for the set of grade alternatives are established as L = {L₁, L₂, L₃, L₄, L₅}. In this study, the 5—point Likert scale is the set of grade alternatives. Therefore, L₁ = strongly disagree, L₂ = Disagree, L₃ = Neutral, L₄ = Agree and L₅ = Strongly agree.

Stage 3; The weightings of each indicator is established. The weighting (W) was determined from the survey results using Equation (1):

$$\mathrm{Wi}={\displaystyle \frac{Mi}{\sum Mii}},$$

(1)

where Wi is the weightings of the indicators, Mi is the mean score value of the indicator and Mii is the summation of the mean score value of all the indicators within a criterion (the criterion were obtained from the EFA). The result for weightings are shown in

Table 6. Weightings of each indicator and criteria.

Criteria 1: Institutional and Knowledge Barriers	Mean	Weightings of Each Indicator (Wi)	Weightings of Each Grouping (Wc)
B7	4.615	0.112
B5	4.615	0.112
B15	4.554	0.111
B21	4.546	0.110
B18	4.546	0.110
B11	4.577	0.111
B10	4.585	0.111
B12	4.569	0.111
B19	4.546	0.110
	41.154	1.000	0.334
Criteria 2: Financial and Adoption Constraints
B9	4.585	0.125
B17	4.554	0.124
B16	4.554	0.124
B4	4.638	0.127
B26	4.477	0.122
B3	4.654	0.127
B23	4.523	0.123
B1	4.677	0.128
	36.662	1.000	0.297
Criteria 3: Technological and Market Limitations
B24	4.515	0.198
B25	4.485	0.197
B13	4.562	0.200
B8	4.608	0.202
B6	4.615	0.203
	22.785	1.000	0.185
Criteria 4: Regulatory and Organizational Challenges
B2	4.669	0.205
B27	4.477	0.196
B14	4.562	0.200
B20	4.546	0.199
B22	4.538	0.199
	22.792	1.000	0.185
			1

Source: Table created by authors.

. For B7 which falls within criterion 1 (Institutional and Knowledge Barriers), the weighting is calculated as follows:

$${\mathrm{Wi}}_{\mathrm{B}7}={\displaystyle \frac{4.615}{4.615+4.615+4.554+4.546+4.546+4.577+4.585+4.569+4.546}}=0.112$$

Similarly, the weighting of a criterion is calculated by dividing the mean score of that criterion (obtain by summing mean scores of all the indicators under the criterion) by the summation of the mean scores of all the criteria. For instance, the weighting for the “Criterion 1 = Institutional and Knowledge Barriers” is given as

$${\mathrm{Wc}}_{\mathrm{B}7}={\displaystyle \frac{41.154}{41.154+36.662+22.785+22.792}}=0.334$$

Therefore, the weightings of all the other indicators and criteria (shown in Table 6) are calculated using the same approach.

Stage 4: The membership function of each indicator is determined using the Likert scale adopted.

Membership functions (i.e., the degree of an element’s membership in a fuzzy set) normally ranges between 0 and 1. They are derived from Level 2 to level 1 [36]. This implies that the membership functions of the indicators (level 2) are obtained first before calculating the membership functions for each of the criteria (level 1). Membership functions are obtained from the ratings provided by the respondents in the survey with regard to the 5-point Likert scale (i.e., L1 = strongly disagree and L5 = Strongly agree) [35]. For instance, for B7 in the first criterion, 3.1% of the respondents strongly disagreed with regard to the rating, 1.5% of respondents disagreed, 1.5% were neutral, 18.5% agreed and 75.5% strongly agreed to the indicator as a barrier to the adoption of DT for CE implementation in the construction industry based on the rating used. These percentages were converted to ratios (by dividing each percentage by 100) to form the fuzzy matrix (Ri) as shown in Table 6. Following this, the membership function level 2 is formed. Hence, the membership function of B7 in criterion 1 becomes;

$${\mathrm{MF}}_{\mathrm{B}7}={\displaystyle \frac{0.031}{L1}}+{\displaystyle \frac{0.015}{L2}}+{\displaystyle \frac{0.015}{L3}}+{\displaystyle \frac{0.185}{L4}}+{\displaystyle \frac{0.754}{L5}}$$

In FSE, the “+” denotes a notation and not an addition (Ameyaw and Chan, 2016). Therefore, the membership function can also be expressed as (0.031, 0.015, 0.015, 0.185, 0.754). The fuzzy matrix for each indicator was concatenated in Excel to obtain a delimited format in commas. Using the same procedure, the membership functions of the remaining 26 indicators were obtained (shown in Table 6).

Stage 5: The final FSE results for the evaluation (also known as the membership function level 1) are determined through the weighting vector and the fuzzy evaluation matrix expressed as:

D = W_iR_i, where W_i = weightings of all indicators within a particular criterion and R_i is the fuzzy evaluation matrix.

After generating the membership function 2 in excel, the weightings of each indicator was concatenated in a delimited format similar to the fuzzy matrix and the results was exported to RStudio for final computation. RStudio was chosen over Excel to generate the FSE results primarily due to its robust statistical programming environment. While Excel can handle simpler fuzzy matrix calculations, RStudio offers greater precision, reproducibility, and flexibility through script-based execution. This approach ensures that all calculations especially the membership function computations and the final index derivations are transparent and easily verifiable, facilitating data-driven validation and advanced customization of fuzzy logic models. Additionally, RStudio’s extensive libraries and built-in statistical functions streamline complex mathematical operations, making it more suitable for multi-criteria decision-making techniques such as FSE, where iterative refinements, advanced plotting, and code-based record-keeping are critical. The output of the RStudio is presented below and the membership function 1 and the final index also presented in

Table 7. Membership functions and Index.

	Fuzzy Matrix
Criteria 1: Institutional and Knowledge Barriers	Strongly Disagree	Disagree	Neutral	Agree	Strongly Agree	Membership Function Level 2	Membership Function Level 1	Index
B7	0.031	0.015	0.015	0.185	0.754	(0.031, 0.015, 0.015, 0.185, 0.754)	0.010, 0.027, 0.034, 0.235, 0.693	4.573
B5	0.015	0.015	0.062	0.154	0.754	(0.015, 0.015, 0.062, 0.154, 0.754)
B15	0.015	0.015	0.000	0.338	0.631	(0.015, 0.015, 0.000, 0.338, 0.631)
B21	0.000	0.031	0.046	0.269	0.654	(0.000, 0.031, 0.046, 0.269, 0.654)
B18	0.000	0.046	0.031	0.254	0.669	(0.000, 0.046, 0.031, 0.254, 0.669)
B11	0.015	0.015	0.015	0.285	0.669	(0.015, 0.015, 0.015, 0.285, 0.669)
B10	0.015	0.031	0.046	0.169	0.738	(0.015, 0.031, 0.046, 0.169, 0.738)
B12	0.000	0.031	0.031	0.277	0.662	(0.000, 0.031, 0.031, 0.277, 0.662)
B19	0.000	0.046	0.062	0.192	0.700	(0.000, 0.046, 0.062, 0.192, 0.700)
Criteria 2: Financial and Adoption Constraints
B9	0.000	0.000	0.046	0.323	0.631	(0.000, 0.000, 0.046, 0.323, 0.631)	0.000, 0.010, 0.048, 0.292, 0.651	4.584
B17	0.000	0.015	0.031	0.338	0.615	(0.000, 0.015, 0.031, 0.338, 0.615)
B16	0.000	0.015	0.031	0.338	0.615	(0.000, 0.015, 0.031, 0.338, 0.615)
B4	0.000	0.031	0.031	0.208	0.731	(0.000, 0.031, 0.031, 0.208, 0.731)
B26	0.000	0.015	0.062	0.354	0.569	(0.000, 0.015, 0.062, 0.354, 0.569)
B3	0.000	0.000	0.046	0.254	0.700	(0.000, 0.000, 0.046, 0.254, 0.700)
B23	0.000	0.000	0.077	0.323	0.6	(0.000, 0.000, 0.077, 0.323, 0.600)
B1	0.000	0.000	0.062	0.200	0.738	(0.000, 0.000, 0.062, 0.200, 0.738)
Criteria 3: Technological and Market Limitations
B24	0.000	0.046	0.031	0.285	0.638	(0.000, 0.046, 0.031, 0.285, 0.638)	0.006, 0.025, 0.043, 0.258, 0.668	4.557
B25	0.015	0.015	0.046	0.315	0.608	(0.015, 0.015, 0.046, 0.315, 0.608)
B13	0.015	0.031	0.031	0.223	0.700	(0.015, 0.031, 0.031, 0.223, 0.700)
B8	0.000	0.015	0.031	0.285	0.669	(0.000, 0.015, 0.031, 0.285, 0.669)
B6	0.000	0.015	0.077	0.185	0.723	(0.000, 0.015, 0.077, 0.185, 0.723)
Criteria 4: Regulatory and Organizational Challenges
B2	0.015	0.000	0.000	0.269	0.715	(0.015, 0.000, 0.000, 0.269, 0.715)	0.0122, 0.025, 0.043, 0.232, 0.688	4.559
B27	0.031	0.015	0.046	0.262	0.646	(0.030, 0.015, 0.046, 0.262, 0.646)
B14	0.015	0.031	0.031	0.223	0.700	(0.015, 0.031, 0.031, 0.223, 0.700)
B20	0.000	0.015	0.062	0.285	0.638	(0.000, 0.015, 0.062, 0.285, 0.638)
B22	0.000	0.062	0.077	0.123	0.738	(0.000, 0.062, 0.077, 0.123, 0.738)

.

Stage 6: After estimating the membership function at level 1, the index of each criterion is determined by;

A = D_n × L_n = (D₁, D₂, D₃, D₄, D₅) × (L₁, L₂, L₃, L₄, L₅)

where D_n = (D₁, D₂, D₃, D₄, D₅) is the fuzzy evaluation matrix or MF for level 1 and L_n = (L₁, L₂, L₃, L₄, L₅) is the grade alternative. Thus, the index (A) for “criterion 1 = Institutional and Knowledge Barriers” is calculated as follows:

A_{(Institutional and Knowledge Barriers)} = (0.010, 0.027, 0.034, 0.235, 0.693) × (1, 2, 3, 4, 5) = 4.573

Using similar approach, the A for the other three criteria are computed in RStudio and final results presented in Table 7.

R Output

sc <- c(1,2,3,4,5)

c1 <- rbind(c(0.03076923,0.01538462,0.01538462,0.18461538,0.75384615),c(0.01538462,0.01538462,0.06153846,0.15384615,0.75384615)

c(0.01538462,0.01538462,0,0.33846154,0.63076923),c(0,0.03076923,0.04615385,0.26923077,0.65384615),

c(0,0.04615385,0.03076923,0.25384615,0.66923077),c(0.01538462,0.01538462,0.01538462,0.28461538,0.66923077),

c(0.01538462,0.03076923,0.04615385,0.16923077,0.73846154),c(0,0.03076923,0.03076923,0.27692308,0.66153846),

c(0,0.04615385,0.06153846,0.19230769,0.7))

wc1 <- rbind(c(0.11214953271028,0.11214953271028,0.110654205607477,0.110467289719626,0.110467289719626,0.111214953271028,

0.111401869158879,0.111028037383178,0.110467289719626))

print(wc1%*%c1)

## [,1] [,2] [,3] [,4] [,5]

## [1,] 0.01030338 0.0273041 0.03419123 0.2356894 0.6925119

mfc1 <- c(0.01030338,0.0273041,0.03419123,0.2356894,0.6925119)

print(sc%*%mfc1)

## [,1]

## [1,] 4.572802

c2 <- rbind(c(0,0,0.04615385,0.32307692,0.63076923),c(0,0.01538462,0.03076923,0.33846154,0.61538462),

c(0,0.01538462,0.03076923,0.33846154,0.61538462),c(0,0.03076923,0.03076923,0.20769231,0.73076923),

c(0,0.01538462,0.06153846,0.35384615,0.56923077),c(0,0,0.04615385,0.25384615,0.7),c(0,0,0.07692308,0.32307692,0.6),

c(0,0,0.06153846,0.2,0.73846154))

wc2 <- rbind(c(0.125052454888796,0.124213176668065,0.124213176668065,0.126521191775073,0.12211498111624,0.126940830885439,

0.123373898447335,0.127570289550986))

print(wc2%*%c2)

## [,1] [,2] [,3] [,4] [,5]

## [1,] 0 0.009593597 0.04802285 0.2915685 0.6508151

mfc2 <- c(0,0.009593597,0.04802285,0.2915685,0.6508151)

print(sc%*%mfc2)

## [,1]

## [1,] 4.583605

c3 <- rbind(c(0,0.04615385,0.03076923,0.28461538,0.63846154),c(0.01538462,0.01538462,0.04615385,0.31538462,0.60769231),

c(0.01538462,0.03076923,0.03076923,0.22307692,0.7),c(0,0.01538462,0.03076923,0.28461538,0.66923077),

c(0,0.01538462,0.07692308,0.18461538,0.72307692))

wc3 <- rbind(c(0.198176907494936,0.196826468602296,0.200202565833896,0.202228224172856,0.202565833896016))

print(wc3%*%c3)

## [,1] [,2] [,3] [,4] [,5]

## [1,] 0.006108141 0.02456241 0.04314652 0.2580948 0.6680881

mfc3 <- c(0.006108141,0.02456241,0.04314652,0.2580948,0.6680881)

print(mfc3%*%sc)

## [,1]

## [1,] 4.557492

c4 <- rbind(c(0.01538462,0,0,0.26923077,0.71538462),c(0.03076923,0.01538462,0.04615385,0.26153846,0.64615385),

c(0.01538462,0.03076923,0.03076923,0.22307692,0.7),c(0,0.01538462,0.06153846,0.28461538,0.63846154),

c(0,0.06153846,0.07692308,0.12307692,0.73846154))

wc4 <- rbind(c(0.204859939250759,0.196422544718191,0.200134998312521,0.199460006749916,0.199122510968613))

print(wc4%*%c4)

## [,1] [,2] [,3] [,4] [,5]

## [1,] 0.01227446 0.0245022 0.04281523 0.2324489 0.6879592

mfc4 <- c(0.01227446,0.0245022,0.04281523,0.2324489,0.6879592)

print(mfc4%*%sc)

## [,1]

## [1,] 4.559316

5. Discussion of Results

The mean score ranking underscores the significant challenges of digital technology adoption for CE implementation within the construction industry, aligning closely with previous findings reported in similar works [1,25]. The complexity of digital tools, ranking highest, echoes observations made by Thirumal et al. [9], who also noted that insufficient familiarity with digital platforms impedes effective integration in developing economies. The prominence of lack of technical expertise further supports Chi et al.’s [23] conclusions that skill shortages continue to undermine digital transformation efforts, particularly among SMEs.

Additionally, stakeholder commitment emerged as a key barrier, reinforcing earlier studies [19] highlighting that reluctance among decision-makers hinders the adoption of CE-aligned digital solutions. Financial constraints and lack of CE-specific indicators highlight the economic and regulatory gaps often cited by Rodrigo et al. [1], while data-related challenges, including unavailability of a web-based database for secondary products, corroborate Harichandran et al. [25], emphasizing data accessibility and traceability as crucial to integrating digital tools in circular construction. Collectively, these results reiterate the multidimensional hurdles covering technical complexity, skills deficits, financing gaps, and data infrastructure that must be addressed for successful digital technology adoption in pursuit of circular economy objectives.

The findings of the EFA and FSE identified four critical barrier groups that significantly impede the adoption of DTs for CE implementation in the construction industry of Ghana: financial and adoption constraints, institutional and knowledge barriers, regulatory and organizational challenges, and technological and market limitations. These results are consistent with, but also extend, prior studies by providing a more nuanced understanding of how these barriers manifest specifically in the context of a developing economy.

Financial and adoption constraints emerged as the most critical barrier group in this study. This finding is consistent with existing research that emphasizes the financial burden associated with adopting advanced DTs. Thirumal et al. [9] identified high implementation costs and limited access to financing as major challenges in developing countries, highlighting that SMEs often lack the capital required for digital transformation. Similarly, Chi et al. [23] emphasized the economic unpredictability that discourages long-term investment in DTs, particularly when the return on investment is uncertain. The findings of this study corroborate these concerns but also reveal a heightened sensitivity to economic risk among Ghanaian construction firms, where limited access to credit and external funding exacerbates these financial barriers. In the Ghanaian construction industry, Agyekum and Amudjie [43] reinforce these findings by emphasizing that financial limitations significantly hinder the capacity of firms to implement CE principles. Additionally, Kwasafo et al. [44] illustrate that despite the increasing adoption of green construction procurement practices in Ghana, financial constraints remain a limiting factor, especially when firms attempt to integrate digital tools that enhance CE outcomes. These findings collectively underline the need for innovative funding mechanisms and public–private partnerships tailored to the financial realities of Ghanaian firms

Institutional and knowledge barriers were ranked as the second most critical group. Prior studies by Rodrigo et al. [1] and Harichandran et al. [25] also stressed the role of institutional weaknesses and knowledge gaps in hindering CE adoption. Rodrigo et al. [1] highlighted the lack of CE-specific indicators and performance metrics, which prevent firms from accurately measuring and managing their circular initiatives. Harichandran et al. [25] pointed out that a lack of technical expertise and insufficient training opportunities impede the effective use of enabling technologies such as blockchain, IoT, and digital twins. The gap between existing skill sets and needed digital competencies complicates effective technology uptake [19]. This study builds on these findings by demonstrating that in Ghana, these barriers are amplified by limited awareness of CE principles and a fragmented knowledge dissemination system, resulting in low stakeholder engagement and minimal uptake of digital solutions. Amudjie et al. [31] reported a moderate awareness but limited practice of CE principles among built environment professionals in the Ghanaian construction industry. Similarly, Asiedu et al. [45] found that although Ghanaian construction practitioners have a general understanding of reuse and recycling, their comprehension of broader CE concepts such as digital material tracking and lifecycle assessments remains limited. This indicates a pressing need for targeted education and training initiatives to bridge the CE knowledge gap in Ghana’s construction sector.

Regulatory and organizational challenges also ranked highly in this study. This aligns with the work of Thirumal et al. [9], who noted the fragmented nature of regulatory frameworks in developing countries, often leading to inconsistent enforcement and a lack of clear guidelines for CE implementation. Jayarathna et al. [19] further emphasized the role of organizational resistance to change, which is prevalent in the construction industry. The findings of this study confirm these observations, revealing that regulatory gaps and fragmented policy frameworks in Ghana create uncertainty, discouraging firms from investing in long-term digital transformation initiatives. Furthermore, weak institutional leadership and a lack of cohesive industry collaboration further complicate the adoption process. In the Ghanaian context, Agyekum and Amudjie [43] noted that despite existing environmental policies, the lack of clear CE regulations and enforcement mechanisms stifles industry-wide CE adoption. Furthermore, Gyimah et al. [46] argue that the absence of robust circular business models (CBMs) and institutional support hinders the transition to a green economy, which is intrinsically linked to CE practices in Ghana. These findings underscore the necessity for comprehensive policy reforms and institutional leadership to facilitate CE implementation.

Technological and market limitations were also found to be significant barriers. Prior studies by Harichandran et al. [25] and Chi et al. [23] have noted challenges related to data security, interoperability, and the availability of high-quality secondary materials. The findings of this study support these concerns, highlighting that Ghanaian construction firms face additional challenges due to inadequate digital infrastructure and a lack of standardized data protocols. These technological barriers not only limit the efficiency of digital tools but also undermine trust among stakeholders, which is crucial for the successful implementation of CE practices. In Ghana, the limited availability of high-quality recycled materials and the absence of material passports exacerbate these challenges [11]. The findings from Asiedu et al. [45] confirm that the lack of digital proficiency among practitioners further hampers the use of advanced technologies like IoT and BIM for CE implementation. The literature suggests that investments in digital infrastructure and the development of industry-wide standards are crucial to addressing these limitations.

By providing empirical evidence from a developing country context, this study highlights how economic, institutional, and technological limitations are exacerbated in resource-constrained environments. The use of FSE further distinguishes this study by offering a rigorous, quantitative method for prioritizing barriers, providing a more objective basis for decision-making compared to traditional qualitative approaches. The FSE results paint a multifaceted picture of barriers: each criterion, financial, institutional, regulatory, and technological, demands coordinated interventions. While financial and adoption constraints edge out slightly as the most critical barrier group, the close index values among all four suggest that no barrier can be ignored. A combination of targeted funding initiatives, capacity-building programs, regulatory clarity, and technological interoperability improvements is key to accelerating digital adoption for circularity. Over time, systemic reforms including the development of CE-specific standards and data protocols will be essential to embedding digital innovation across the construction industry in developing economies.

6. Framework for Overcoming Barriers to Digital Technologies Adoption for CE Implementation

Based on the study’s aim and empirical findings, a comprehensive framework has been developed to address the barriers to digital technology adoption for CE implementation in the construction industry of a developing country. As illustrated in Figure 1, the framework outlines four key barrier categories, each with targeted strategies to promote effective digital technology adoption within the CE paradigm.

To overcome the financial and adoption constraints identified in this study, a series of targeted interventions are proposed. Governments and policymakers should implement financial incentives, including tax relief schemes, grants, and low-interest loans, specifically aimed at SMEs, which often lack the capital to invest in DTs that facilitate CE practices. Similarly to the recommendations of Thirumal et al. [9], public–private partnerships (PPPs) should be encouraged to co-finance digital transformation projects, distributing risk and reducing the financial burden on individual firms. Additionally, the promotion of open-source and low-cost digital solutions, tailored for the resource-constrained environments typical of developing countries, is essential. Vendors should be incentivized to develop affordable, scalable digital tools, such as simplified BIM applications and lightweight IoT solutions, addressing the affordability gap that Chi et al. [23] identified as a critical barrier in the construction sector.

Addressing technological and market limitations requires both infrastructure improvements and standardization efforts. Investment in digital infrastructure, including reliable internet connectivity and secure data platforms, is fundamental to enabling the real-time data sharing necessary for CE implementation. Echoing Harichandran et al. [25], establishing interoperability standards and common data environments (CDEs) can mitigate issues related to fragmented systems and data silos, ensuring seamless integration of technologies like BIM, blockchain, and digital twins. Furthermore, the implementation of robust cybersecurity and data protection frameworks, aligned with international standards such as ISO 27001 [47], will address concerns about data privacy and security, which remain significant barriers to stakeholder trust and participation.

To tackle institutional and knowledge barriers, the study emphasizes the importance of capacity building and skills development programs. Drawing from the insights of Rodrigo et al. [1], targeted training should be provided for professionals at all levels, focusing on both CE principles and the practical use of enabling DTs. As Pittri et al. [48] highlight, the limited incorporation of sustainability concepts, such as Design for Deconstruction (DfD), within formal education systems has resulted in a significant knowledge gap among construction professionals in Ghana. Addressing this requires the integration of CE and digital transformation content into academic curricula, as well as the establishment of continuous professional development (CPD) programs tailored for industry practitioners [49]. In addition to education and training, the framework emphasizes the importance of awareness campaigns and knowledge dissemination as strategies to foster a culture of innovation and collaboration. Pittri et al. [49] argue that enhancing stakeholder awareness through seminars, workshops, and knowledge-sharing platforms is critical for increasing buy-in and reducing resistance to CE practices. This aligns with the study’s recommendations for promoting cross-sector collaboration and developing knowledge management systems that facilitate the exchange of best practices and successful case studies.

Finally, overcoming regulatory and organizational challenges demands a supportive policy and governance environment. Policymakers must establish clear CE and digitalization policies and guidelines, ensuring that regulatory frameworks mandate the use of digital tools in construction procurement processes. Consistent with Jayarathna et al. [19], strengthening regulatory enforcement and compliance mechanisms is critical to ensuring adherence to CE principles and promoting industry-wide transformation. Encouraging multi-stakeholder collaboration through formal governance structures can foster alignment between government agencies, private sector actors, and academia. Additionally, the adoption of standardized practices and certification schemes, such as CE certification for construction projects or digital proficiency standards for firms, can help drive widespread and sustained adoption of CE-aligned DTs.

The proposed framework provides a holistic and adaptable model that can be used by policymakers, industry stakeholders, and academic researchers to guide the systematic removal of barriers to digital technology adoption in CE. While developed within the context of Ghana, the framework’s structure and strategic directions are applicable to other developing countries facing similar challenges, thereby contributing to international efforts aimed at sustainable development and environmental conservation.

Effectively addressing these barriers requires a comprehensive understanding not only of the barriers themselves but also of the inherent risks they pose, particularly in developing country contexts. Recognizing and mitigating these risks is critical for ensuring the successful and sustainable integration of digital solutions in the construction industry. As Kottmeyer [50] and Ramesohl et al. [51] highlights, the use of DTs presents both significant opportunities and substantial risks, especially where institutional weaknesses, financial constraints, and technological deficiencies prevail.

Financial and adoption constraints pose significant risks such as poor return on investment (ROI), capital loss, and sunk costs, particularly when DTs become obsolete or fail to integrate seamlessly into business processes [23]. These risks are heightened in developing economies, where SMEs often lack the financial resilience to recover from failed technology adoption [50]. To mitigate these challenges, this framework recommends the provision of government-backed financial incentives, the establishment of risk-sharing mechanisms such as guarantee funds, and the implementation of pilot programs that allow incremental adoption and validation of DTs prior to full-scale deployment

Institutional and knowledge barriers carry risks related to resistance to change, fragmented responsibilities, and a lack of organizational readiness, which can lead to project delays, poor implementation, and ultimately, failure of digital transformation initiatives [11,31,43]. In many developing countries, limited technical expertise increases the probability of misusing or underutilizing DTs, thereby undermining the potential benefits of CE initiatives [50]. To address these issues, the framework advocates for comprehensive capacity-building programs, continuous professional development, and the creation of CE knowledge management systems that enhance organizational learning and foster leadership committed to digital innovation.

Technological and market limitations introduce considerable risks such as data breaches, cyberattacks, loss of data integrity, and interoperability failures [25]. These technological risks are compounded in developing countries by inadequate digital infrastructure and a lack of standardized data protocols [50]. As a mitigation strategy, the framework emphasizes the need for investment in robust digital infrastructure, the adoption of internationally recognized cybersecurity standards (e.g., ISO 27001), and the establishment of CDEs to ensure interoperability and secure data sharing across stakeholders.

Regulatory and organizational challenges present risks such as weak regulatory enforcement, legal uncertainties, and organizational inertia. The absence of clear regulations governing CE implementation and DTs may result in inconsistent practices, increasing the likelihood of non-compliance and regulatory sanctions. Moreover, as Kottmeyer [50] points out, the lack of regulatory clarity can lead to power imbalances, where certain actors exploit regulatory gaps to their advantage, undermining fair competition and social equity. To mitigate these risks, the framework recommends the development of clear, enforceable policies and regulations, fostering multi-stakeholder collaboration, and promoting transparent governance structures that ensure accountability and equitable participation.

By explicitly addressing these risk dimensions, the proposed framework offers a holistic approach that not only overcomes barriers but also proactively manages risks associated with DTs adoption for CE implementation.

7. Implications of Findings

The study’s findings on barriers to adopting DTs for CE implementation carry both practical and theoretical significance.

Practically, the identification of high implementation costs, limited technical expertise, and resistance to change highlights a need for shared service platforms, collaborative financing models, and specialized training. Such coordinated strategies can assist construction firms, particularly SMEs in overcoming resource constraints and the complexity of digital tools. Also, the prominence of regulatory and policy gaps suggests that constructing an enabling legal framework is paramount. Introducing fiscal incentives and structured guidelines can provide the clarity required for industry-wide digital transformation aligned with CE objectives. The widespread skills gap in emerging digital solutions indicates an urgent need to incorporate CE-focused digital competencies into curricula. Collaborations with professional bodies and private industry can bolster capacity-building efforts, improving the overall readiness of construction professionals to integrate technology-driven circular practices.

Although the empirical focus of this study is on Ghana, the findings have broader relevance for other developing countries facing similar barriers to digital technology adoption for CE implementation. Many developing economies exhibit comparable characteristics, such as limited financial resources, underdeveloped digital infrastructure, fragmented regulatory frameworks, and insufficient institutional capacity. These shared challenges suggest that the barrier groups identified in this study are not unique to Ghana but are likely prevalent in other developing regions, including Sub-Saharan Africa, Southeast Asia, and parts of Latin America.

The study’s use of FSE offers a replicable methodology that can be applied in different country contexts to prioritize barriers based on local stakeholder inputs. This methodological approach ensures that decision-makers in other nations can adapt the framework to their specific needs, providing a rigorous basis for formulating context-sensitive policies and interventions.

Moreover, the study’s insights on the need for integrated strategies encompassing financial incentives, capacity building, regulatory reforms, and technological upgrades are universally applicable to developing economies striving to enhance their circular practices through digital transformation. The emphasis on multi-stakeholder collaboration and the role of government in facilitating enabling environments further underscores strategies that can be implemented beyond Ghana.

By categorizing barriers into financial, institutional, technological, and regulatory challenges, the proposed framework provides a clear roadmap for developing targeted interventions such as financial incentives, capacity-building programs, and regulatory reforms. Practically, it equips decision-makers with actionable strategies to allocate resources effectively, mitigate adoption risks, and foster collaboration across stakeholders, ultimately enhancing the uptake of digital technologies and promoting sustainable construction practices in developing country contexts.

Theoretically, the findings reinforce the necessity to examine multiple dimensions such as economic, institutional, organizational, and technological within theoretical models of digital adoption. This way, future research can better encapsulate how diverse barriers collectively shape CE-driven transformation. The high prominence of financial, skill-related, and infrastructure obstacles underscores the importance of regional and industry-specific factors. Incorporating these contextual variables into existing technology adoption theories can refine their applicability to developing economies.

8. Study Limitations

While this study provides valuable insights into the barriers to digital technology adoption for CE implementation in the GCI, several limitations must be acknowledged. First, the study relied solely on primary data collected through structured questionnaires, which, although effective for gathering quantifiable responses from a relatively large sample, may not fully capture the complex, contextual, and nuanced nature of the barriers identified. The structured format of the survey restricted respondents’ ability to elaborate on their perspectives, potentially omitting subtle but critical insights. In addition, self-reporting bias and social desirability bias could influence the accuracy of responses. To address this limitation, future research could employ qualitative methods, such as semi-structured interviews, focus groups, or case studies, to obtain richer, more detailed insights into how these barriers manifest in real-world projects.

Second, although the study identified and prioritized 27 barriers based on extensive literature review and empirical analysis, there may be additional barriers not covered by the questionnaire that are relevant to different organizational contexts or emerging technologies. Further research could explore these additional barriers or delve deeper into the identified barrier categories to understand their underlying causes and dynamics.

While this study identified and discussed key risks linked to the major barrier categories, a detailed risk assessment—quantifying their likelihood and impact—was beyond the scope of this research. Future studies could employ comprehensive risk analysis frameworks or models to evaluate these risks systematically, particularly in the context of developing countries where such risks may be amplified by resource constraints and institutional weaknesses.

Lastly, the reliance on data from a single developing country context limits the generalizability of findings. Future studies could expand the geographic scope to include comparative studies across different developing regions to enhance the robustness and applicability of the findings.

9. Conclusions

This study identifies and evaluates the core barriers inhibiting the adoption of DTs for CE implementation in the construction industry of the developing economy of Ghana. Through a multi-layered analytical process comprising mean score ranking, factor analysis, and Fuzzy Synthetic Evaluation, the research reveals a tightly clustered set of obstacles spanning financial, institutional, regulatory, and technological domains. These barriers jointly reflect the complexity of transitioning toward circular practices supported by emerging digital solutions.

Although high implementation costs and limited expertise often stand out, every category of barriers requires coordinated intervention to accelerate digital transformation. A strategic mix of policy initiatives, capacity-building, stakeholder collaboration, and technology frameworks will be vital for facilitating a circular built environment. Future research may employ longitudinal techniques or hybrid decision-making tools to further explore how emerging regulations and evolving digital platforms can mitigate these barriers.