Barriers in implementing circular economy principles in public–private partnership smart cities

Rapid urbanization has become a global trend with 68% of the global population now living in cities and this is expected to increase in foreseeable future (UN, 2018). This urbanization trend poses several problems, including traffic congestion, resource exhaustion, environmental pollution and spatial disparities in development (Hasibuan et al., 2025). To address these challenges and improve urban management, many governments have initiated and promoted the concept of “smart city”. A smart city is explained as the effective integration of physical, digital and human systems in infrastructure development to deliver a sustainable, prosperous and inclusive future for urban dwellers (Hasibuan et al., 2025; Sumra et al., 2026). According to ISO PAS (Publicly Available Specification), smart cities involve application of cutting-edge smart technologies and innovative solutions which aim to address many environment, social and economic issues. Previous scholarly works have presented various challenges associated with the construction of smart cities, including budget constraints (Lam and Yang, 2020); lack of public support (Klimek, 2024) and conflicting expectations of various stakeholders (Dincă et al., 2022). Studies and practice documents point to leveraging collaborative solutions through the public-private partnership (PPP) model to develop smart cities. Globally, PPP is a prominent infrastructure delivery mechanism that is keen to facilitate the rapid expansion of technology-savvy urban cities with combined resources from multiple stakeholders. However, PPP smart cities, indifferent to other infrastructure and construction projects, are notable for high carbon emissions (Akomea-Frimpong et al., 2023). For instance, Wu et al. (2024) and Weigert et al. (2022) reported that 23%–35% of global carbon emissions emanate from infrastructure activities (buildings and transport) in the cities. Additionally, PPP smart cities are built on the foundation of linear (make-use-dispose) models, which are not appropriate to attain sustainable and net-zero targets in urban development by 2050 (Ho et al., 2024). Hence, PPP smart city projects need a paradigm shift to more sustainable and low-emissions ways of construction through circular economy (CE).

Circular economy (CE) conserves natural resources and minimizes waste of construction resources (Kwasafo et al., 2024). It has the capability to keep materials in the loop longer and recover these materials for reuse in other projects (Agyekum and Amudjie, 2024). Therefore, integrating CE practices in PPP smart cities has long-term positive environmental and social impacts (Windapo and Moghayedi, 2020).

Despite several attempts to accelerate CE practices, it is still in its infancy in PPP smart infrastructure development globally (Hossain et al., 2020). This was evident as the circularity index has plummeted from 9.1% in 2018 to 7.2% in 2023, showcasing an expanding gap in circularity (UNDP Global Climate Promise, 2024). Circularity index is an indicator of the use and incorporation of recycled materials in a product by evaluating the value of recycled materials to the overall of the products.

The stakeholders of PPP smart cities also lack an understanding of how CE initiatives can be integrated throughout the project lifecycle (Adams et al., 2017). Governments, construction professionals and other stakeholders still require a more coordinated and integrated approach to addressing challenges and driving CE practices (Illankoon and Vithanage, 2023). In practice, there is a lack of shared urgency and intention to integrate circularity in PPP construction projects. This requires a thorough and in-depth understanding of the barriers that hinder the successful conceptualization and implementation of CE practices in PPP smart city projects to drive a sustainable construction sector. Furthermore, it is notable that the barriers to integrating CE principles in the construction sector in general have been extensively studied in empirical and review studies (Kirchherr et al., 2018; Hart et al., 2019; Charef et al., 2021; Munaro and Tavares, 2023; Akomea-Frimpong et al., 2024b). These studies seldom address how such barriers on CE interact within PPP smart city projects and its contextualized solutions. Unlike traditional projects, PPP-based projects involve complex contractual arrangements, multi-actor governance, inter-organizational coordination and shared risk structures, which can reshape how CE principles are adopted or resisted. The current literature also shows a distinctive analysis of each concept separately, leaving a viable opportunity for exploration in this study. Lastly, the very few prior investigations on CE, PPP and smart cities together were undertaken on a country-specific basis, which provides fragmentated and context-dependent insights (Campbell-Johnston et al., 2019; Ezeudu et al., 2021). This approach limits the development of globally coherent strategies for advancing CE adoption in PPP smart cities. Therefore, this study seeks to address this question:

Iossa et al. (2014) has explained PPP projects as “… any contractual arrangement between a public-sector party and a private-sector party for the provision of public services with the following four main characteristics: (1) the bundling of project phases into a single contract; (2) an output specification approach; (3) a high level of risk transfer to the private sector, and (4) a long-term contract duration” (p. 10). While the private sector takes the responsibility of designing, constructing, operating, financing, sharing knowledge, resources and providing technical expertise and capital for the project in PPP arrangements, the public entity provides the permission to the private sector to operate and recoup investments, repay loans and make profit (Ghanbaripour et al., 2023). An investment of US$86.0 billion into urban development, representing 0.2% of the GDP of all low- and middle-income countries, has been recorded through the PPP initiatives in the last decade (World Bank Group, 2025). The PPP model has also been a key backbone in transitioning traditional urban communities into smart cities of modern and technology-savvy projects. The combination of PPP and smart cities results in a terminology in this study as “PPP smart cities” or “PPP smart city projects”. This term can be explained as the support of resources from private and public sectors to build and operate data-driven and technology-savvy infrastructure projects in urban areas. These PPP smart city projects align with Sustainable Development Goal 11 (sustainable communities and cities) with green financing from private and public institutions, together with sustainable management practices (Jha, 2024). The ongoing transformation of major cities into modern communities under the PPP contracts is littered around the world. Examples of PPP smart city projects include Parramatta light rail (Australia), Artificial intelligence (AI) motor traffic directions (Singapore), Shanghai internet-savvy retirement villages (China) and Internet of Things-focused New York city management (USA) (Son and Duong, 2024).

The World Economic Forum (2026) suggests that approximately 90% of construction materials can be reused; however, at the end of a project’s lifecycle, only 20–30% of these resources are recycled. Currently, most construction projects adopt “Take-Make-Dispose” model where they collect the raw materials, use them in construction, and dispose of them at the end of their lifecycle, making it waste (Ghisellini and Ulgiati, 2020; Akomea-Frimpong et al., 2025). This is termed as a linear approach where the end product of construction materials results in an enormous amount of waste thus impacting the environment, climate and natural resources (Ho et al., 2024). The limited nature of construction resources also promotes the development of a unique and innovative circular business model to improve the sustainability of little and relevant resources. The CE in PPP smart cities is then considered as a replacement of the linear economy model in preserving the finiteness of construction materials, which involves the shifting of the processes from “take-make-dispose” to “cradle-to-cradle” model (Ezeudu et al., 2021). Circularized PPP smart city projects also have the potential to mitigate climate change by addressing decarbonization, material management, dematerialization, and systemic and unsustainable practices surrounding construction activities. It is expected that CE could facilitate the reduction of global emissions of 39% by 2030 (Prasad et al., 2023).

The key prevailing challenge of PPP smart projects is largely centred around the general linearity approach of executing the projects (Gil-Garcia et al., 2015). This approach emphasizes the ownership of materials and components until the end of the product lifecycle. This long-established project model is counterproductive and does not promote circularity in urban infrastructure development, and therefore exacerbates waste, amplifies resource inefficiencies and shortens the product lifecycle. To embrace CE in PPP smart city projects, a transformation of the linear model into a sustainable and carbon reduction-friendly approach is vital (Kirchherr et al., 2018). Another critical barrier is restructuring and rethinking of urban project models to fit into the complicated contractual agreements of public and private partnership contracts for CE. The foundation of this challenge emanates from complex organizational culture and unaccommodating systems, which do not advance CE (Hart et al., 2019). Akomea-Frimpong et al. (2024a) argued that many traditional project-based organizations delivering infrastructure projects adopt a culture based on established and unsolicited procurement approaches, which fail to incorporate novel CE strategies. Further, CE demands innovative procurement routes, skilful project professionals who are not hesitant to apply the CE approaches to achieve its goals. However, Salmenperä et al. (2021) presented that most construction professionals in smart city projects lack the skills in CE to deliver sustainable projects. There is also a disjointed effort of the key players in the PPP arrangements to deliver circular and smart city projects. Lăzăroiu et al. (2020) posited that it is because of conflicting interests in the multistakeholder contracts of PPP towards delivering smart city projects. While the private sector favours profit-orientation over environmental mess, public institutions are keen on ensuring ecological and social justice. This put the goal of attaining CE in PPP smart cities at loggerheads.

This study employed a quantitative research approach grounded in a positivist philosophical stance. The research was designed in two stages. In the first stage, a comprehensive literature review of the existing literature was conducted to locate critical barriers to CE implementation in PPP smart cities. The second stage involves gathering views on the critical barriers to CE application in PPP smart cities through an international survey.

A review of past studies was performed to identify critical barriers hindering the adoption of CE in PPP construction projects. Prominent search engines such as Scopus, Web of Science and Google Scholar were utilized to retrieve documents on barriers impeding CE application in PPP smart cities with Boolean keywords search including “circular economy” OR “CE” AND “barriers” OR “obstacle” OR “hinder*” AND “public-private partnership projects” OR “PPP project” OR “construction”. The initial search returned 418 records, which were screened and out of these, 271 of the most relevant articles were selected for further analysis. Through careful content reading and analysis, the barriers were identified, consolidated to form themes, resulting in 35 critical barriers (see Table 1).

The identified critical barriers gathered from the literature for advancing CE in PPP smart cities formed the questionnaire survey. The first part of the survey solicited respondents’ background information. The second part included 35 barriers extracted from the literature. Before administering the survey on a full scale, a pilot test was conducted with nine experts experienced in PPPs construction projects from both academic and industry backgrounds. While the PPP experts did not propose any major changes to the identified barriers, they recommended refining the language of certain barriers for enhanced clarity and comprehension. The respondents were requested to evaluate the 35 barriers in PPP smart cities on a 5-point Likert scale (5 – extremely high, 4 – high, 3 – medium, 2 – low and 1 – extremely low). The survey instrument (that is the questionnaire) was critically assessed and approved together with the entire research by an Ethics Committee from the second author’s institution. Potential participants were identified as they met two inclusion criteria: (1) experience working in PPP construction project settings in urban cities; (2) familiarity with CE applications in PPP and smart city projects. The authors searched and screened potential survey respondents from LinkedIn profiles using their role title, working experiences and working sector before sending them invitations. The authors also tapped into their own professional networks globally and sent invitations. In addition, this study utilized a snowball sampling method where we encouraged those invited to share the invitations with their colleagues. By following this structured approach (Osei-Kyei et al., 2017), a total of 102 survey responses were obtained with six incomplete responses, which were excluded during screening. This process led to a collection of 96 valid survey responses from international PPP professionals.

The participants in the study were from diverse professional backgrounds with the majority being construction facility managers (39%) followed by quantity surveyors (24%), risk consultants (19%), project architects (11%) and project engineers (7%). Approximately 38% of the participants possess more than 10 years of working experience, with almost half (48%) of the participants having a working experience between 6 and 10 years. Most of the participants work in the private sector (52%) and the 48% remaining come from the public sector.

The survey respondents were all over the world, including China, Ghana, Nigeria, Kenya, South Africa, Australia, the United Kingdom, Qatar, Singapore, France, Hong Kong, Norway, Canada, South Korea, New Zealand and the United States (See Figure 1).

The survey data were analyzed using Fuzzy Synthetic Evaluation. First, exploratory factor analysis (EFA) was performed to categorize the barriers in the CE in PPP application in smart cities. Furthermore, a mean score ranking was conducted to rank the identified barriers in order of criticality. To test the internal consistency of the responses and the data reliability, Cronbach’s Alpha coefficient was employed with a value ranging from 0 to 1. The Cronbach’s alpha coefficient for the data collected was 0.847, demonstrating the high reliability and consistency of the data (Nunnally, 1978). The normality of the data set was tested using the Shapiro–Wilk test based on the recommendation. The Shapiro–Wilk test value for the data shows that p-values are 0.00 at 5% significance level for all 35 barriers to CE application in PPPs smart cities. As the results suggest non-normality of the data set ascertained by the Shapiro–Wilk test, a further test of the data set was performed by a non-parametric data analysis technique (Oluleye et al., 2023). The Kruskal–Wallis, a non-parametric test used to check the differences in views between independent groups (more than two groups) of different populations, as in the case of this study, the independent groups being the construction professionals who worked as facility managers, quantity surveyors, risk consultants, project architects and engineers. The results suggest that p-values for all barriers were above 0.05, suggesting no significant difference in opinion of the participants.

EFA was performed to identify the clusters among variables to determine a set of measurement criteria. Varimax rotation was utilized to simplify the interpretation of factors. Second, FSE was undertaken to rank the CE barriers using Likert scales (Wuni and Shen, 2022). FSE utilizes fuzzy logic intelligence techniques to transform human judgement to a quantitative form, which reduces bias and subjectivity in the evaluation process, taking into account the ambiguity and uncertainty that are typically present in the data assessment (Osei-Kyei and Chan, 2018).

To identify the most critical barriers to CE in PPP smart cities, the mean of the collected data was performed. The corresponding standard deviation (SD) was also calculated, showing the relative similarities of agreement among survey participants about the key barriers. The mean value of the CE barriers to integrating CE practices in PPP smart cities ranged from 4.72 to 2.67. This suggests that all the barriers that are impeding CE in PPP smart cities are quite significant. The top three barriers were a lack of leadership commitment towards CE, complex circular design and architecture, and a lack of transparency and information sharing. The results indicate that effective leadership, design and knowledge coordination challenges constrain CE implementation in PPP smart cities. Contrarily, lower-ranked barriers, such as insufficient green procurement policies and a lack of eco-friendly CE tools, while less critical, still indicate systemic technological and policy gaps.

Exploratory factor analysis (EFA) was conducted to evaluate the structure of the factors and determine components within the 35 critical barriers (Lingard and Rowlinson, 2006). The appropriateness of the data samples for the EFA analysis was ascertained by the Kaiser–Meyer–Olkin (KMO) test for sampling Adequacy and Bartlett’s test of sphericity (Ahadzie et al., 2008). The KMO value for this data set was 0.863, which was deemed appropriate for this study (Tabachnick et al., 2013). Furthermore, the corresponding Bartlett’s test of sphericity statistic yielded a significant value of 362.421 at a p-value of<0.05 (Field, 2013). The principal component analysis was employed for factor extraction and grouping into key components. The results produced five groups of barriers to CE principal components in PPP smart city projects with eigenvalues greater than 1.0, explaining 74.775% of the total variance after varimax rotation. Table 2 shows the factor analysis of the data set with five categories with % of variance explained. The categorization of the 35 critical barriers into five-factor components automatically came from the SPSS analysis (Ameyaw and Chan, 2015). The barriers with factor loadings more than 0.7 were kept and reported as shown in Table 2. The retained components represented coherent conceptual dimensions reflecting institutional and systemic inefficiencies, attitudinal and leadership challenges, financial and cost-related constraints, governance and regulatory fragmentation and digital innovation and knowledge limitations. The names of principal components were assigned based on the dominant pattern of characteristics validated through discussions and agreement among all contributing authors of this article.

FSE was performed to ascertain the criticality of the five groupings from the EFA analysis in Section 4.2. The main purpose of the FSE was to determine the critical components (from the most critical component to the least) in Table 2. The FSE commences with the assessment of the weights of each variable within the five groups, which is assessed in Table 2.

In this step, the weightings of each critical barrier (CBR) and the groups (CBRG) were calculated using the mean shown in Table 3.

The following formula was employed to calculate the weightings of all CBR and CBRG.

Wt_i is the weighting function for each CBR/CBRGs to the barrier of CE in PPP smart city projects, and M₀ is the mean score of each group. i denotes the used 5-point Likert scale ranging from 1 to 5, ΣMS_j is the total of all mean scores for all CBRs/CBRGs. The weightings are represented as follows: Wt_i = (Wt₁, Wt₂, Wt₃, Wt₄ … … … … …, Wt_n). For instance, the weighting for CBR19 “inadequate skill set on circular economy” is calculated as follows:

M_j, mean score, for CBR19 is 3.62 and the summation of all mean scores for the corresponding component is 23.14. Therefore, the weighting function (Wt _CBR19) for CBR19 is 0.156. Similarly, the weighting of CBRG₁ is calculated as follows. The same formula was utilized to compute the weightings for the five groups, as shown in Table 4.

The MF of all CBR was obtained by utilizing the survey results achieved from the respondents. The Likert scale was set as 5 – extremely high (EH), 4 – high (H), 3 – medium (M), 2 – low (L) and 1 – extremely low (EL). After carefully examining the survey responses from the respondents, it was found that for CBR16 “insufficient government support”, 1% responded rated this barrier as extremely low, 5.2% rated as low, 14.6% as medium, 20.8% as high and 58.3% as extremely high. Hence, the MF for CBR16 is calculated as follows:

The MF for CBR16 is then expressed as follows: (0.010, 0.052, 0.146, 0.208, 0.583). The MF for all other CBR were computed using the same technique (see Table 4). Consequently, the MF for CBRG was computed drawing from the MF of all CBR to barriers on CE in PPP smart cities and their respective weightings. Therefore, the following formula is used to compute the final evaluation matrix: D_i = W_i X R_i

Here, D_i denotes membership function for CBRG. W_i represents weightings function of CBR and R_i represents the fuzzy evaluation matrix. For instance, the MF for CRBG1 is computed as follows:

Therefore, the membership function for D_CRBG1 = (0.009, 0.024, 0.227, 0.322, 0.418). The same approach was employed to calculate MF for CRBG2, CRBG3, CRBG4 and CRBG5 (Table 4).

Upon completing MF of all CBRG, the criticality indices of all CBRG were calculated. The following formula was utilized to calculate the criticality index:

Where D is the membership function of CBRG and E is on the Likert scale rating (1, 2, 3, 4, 5). The calculation for CBRG1 is shown below and the same formula was employed to calculate the criticality index for all CBRG (see Table 4).

From Table 4, the first and most critical component based on the FSE analysis is governance and regulatory barriers (CBRG4), which scored an index of 4.478. This signifies the major challenge in implementing CE practices in PPP smart cities. It accounted for the 10.379% of the variance explained in the factor analysis with an eigenvalue of 1.447 in the factor analysis which further proves its criticality. This factor component consists of four factors, including complex circular design and architecture, lack of industry standards, lack of national regulatory framework and absence of common international policies. Challenges on circular design and architecture encompass inefficient use of resources so that materials stay in the loop as long as possible, thus increasing carbon emissions and waste. When considering circularity initiatives in PPP smart cities, the design process becomes more complex and has a ripple effect on project costs, schedule and scope (Ceschin and Gaziulusoy, 2016). Adopting a circular design requires extended cross-disciplinary knowledge and involvement of multiple stakeholders early in the project design phase. Akomea-Frimpong et al. (2024a) mentioned that in many jurisdictions, the standards and policies to guide these circular designs for urban areas are absent or improperly implemented. These barriers pose a significant impediment to circular strategies, thus adding complexity to the design. Furthermore, one of the risks that reduces PPP smart city projects performance is complexity and linearity-focused regulatory frameworks. Some policies are outdated and they do not support sustainable materials for smart city developments, which overshadowed CE initiatives in the design (Charef et al., 2022). In addition, contractual policies around multiple stakeholder PPP arrangements make it challenging to promote circularity in the design of PPP smart cities.

The second critical factor component from the FSE analysis is financial constraints and cost pressures, which recorded an index of 4.15 (Table 4). This barrier also accounted for 12.822% of total variance explained with an eigenvalue of 1.842 in the EFA analysis in Table 2. The key sub-group elements of this component include inaccuracies in cost estimation, cost of disposing of wastes, expensive virgin material costs, hiring costs of construction workers and rising insurance costs. The wrong estimation of cost may deter construction professionals from adopting CE as PPP smart cities are very sensitive to delivering within budget and scheduled time (Mousa et al., 2024). Inaccuracy in cost estimates on CE derives from inaccurate estimation of unstable conditions prevailing on ecological matters towards city development. The cost estimations on recycled materials and reuse of resources, waste quantity and distribution may be expensive depending on the economic conditions of where the PPP smart city project is being built. Inaccurate cost data could hamper the proper cost analysis and solutions to finance risks (Feldman et al., 2024). Given the long lifespan of PPP smart city infrastructures, it becomes cumbersome to make an accurate prediction of recycling and material recovery costs if real-time, accurate financial data is not secured. The inaccurate prediction of costs could lead to profligate and unexpected expenses.

The third factor component that hinders CE in PPP smart cities is systematic and institutional barriers. From Table 4, this component was ranked as 3rd with an FSE score of 4.116. It also reported 26.184% of total variance explained and an eigenvalue of 4.182 in the EFA analysis. In PPP smart cities, there is a long-term contractual agreement between the public and private sectors. Unlike traditional contracts where design, build and operate are performed through separate contracts, in PPP the whole execution of projects is granted to one private entity (Son and Duong, 2024). While PPP contract is adopted to gain some financial advantages and transfer risks to the private sector, the PPP suffers from some fundamental problems, including long negotiation times, insufficient flexibility, lack of transparency to the client, inclusion of high risk premium, ineffective delivery, uneven risk allocation and lack of collaboration between the public and private sectors (Wuni and Shen, 2022). These challenges are inherent in PPP contracts, which deter the client from taking on additional risk of applying in CE. Institutional transition costs associated with shifting the linear corporate and project models to CE are another significant impediment to CE adoption (Knoth et al., 2022). A transition to CE requires a significant change in the business model, a shift in design and material selection, extensive upfront planning, changes in procurement strategies, integration of technologies, changes in the supply chain management, upskilling of professionals, outsourcing experts, infrastructure and cost associated with recycling, deconstruction, separation, treatment and storage of CDW.

This is the fourth principal barrier to CE in PPP smart city projects, and it recorded a score of 4.042 in the FSE analysis. Despite the potential of advanced technologies such as blockchain, artificial intelligence (AI) and Building Information Modelling (BIM) in CE and PPP urban development, it comes with many barriers (Charef et al., 2022). This is largely due to the high costs associated with adopting these technologies, which hinder decision-making and stakeholder engagement in CE in PPP smart cities. The stakeholders of city development have been slow to embrace CE and smart technologies, largely due to resistance to change and the lack of specialized skills among professionals for data analysis, programming, and automation (Paiho et al., 2020). Many of these stakeholders are accustomed to traditional construction methods, making it difficult to implement new technologies effectively. Additionally, the initial high investment required for these technologies makes it challenging for contractors, city councils and private financiers to justify the costs, especially without clear, immediate returns.

The least critical component among the five factor groups from the FSE assessment is attitude barriers with an index of 3.989 (Table 4). The variance explained by the factor analysis was 18.347%, which supports its criticality to the discussion. It has important variables such as inadequate leadership, lack of stakeholder collaboration, inadequate competencies, lack of knowledge and awareness and resistance to change. Without appropriate leadership styles, CE cannot be realized in PPP smart city projects as it involves a range of challenges in design and execution. Leadership attributes need to be demonstrated in both private and public entities as a lack of integrated collaboration can fail the application of CE in PPP smart city projects (Oluleye et al., 2023). Also, these projects entail a number of stakeholders from both public and private sectors who have conflicting interests. The intention to apply CE practices is likely to cause confusion because some stakeholders still favour liner model of development to urban projects (Wiesmeth and Starodubets, 2020). Further, in PPP smart city projects, reluctance from either sector can result in fragmented effort, leading to linear approaches. Mere policy implementation is not sufficient for driving CE, if a lack of collaboration persists among stakeholders (Guerra and Leite, 2021). Construction professionals' attitudes towards linear knowledge-focus and immediate attention to monetary benefits hinder the long-term sustainability of PPP smart cities.

This study advances the PPP–CE literature in three important ways. First, by identifying and ranking systemic, financial, governance, attitudinal and digital barriers using the Fuzzy Synthetic Evaluation approach, the study moves beyond descriptive accounts of CE challenges and provides a structured prioritization framework. This contributes to theory by linking CE implementation barriers to institutional and contractual dynamics within PPP smart city projects.

Second, the findings demonstrate that CE integration in PPPs is not merely a technical or environmental issue but is embedded within institutional, financial, and governance arrangements. This reinforces the need to conceptualize CE adoption in PPPs as a multi-level governance and risk-allocation challenge.

Third, the study contributes to emerging debates on circular innovation in infrastructure projects by highlighting how contractual rigidity and lifecycle payment structures can either constrain or enable CE experimentation.

The findings suggest that circular economy integration in PPP smart city projects must be embedded at the procurement and contract design stage. Public authorities should incorporate explicit CE performance indicators and lifecycle-based environmental KPIs into concession agreements, linking circular outcomes to availability payment mechanisms. This alignment ensures that CE objectives are incentivized across the entire asset lifecycle rather than limited to the construction phase.

For private concessionaires and investors, CE should be viewed as a lifecycle risk management and value-creation strategy. Although financial and cost pressures remain significant barriers, circular design approaches, such as durability and resource efficiency, can reduce long-term operational risks. Introducing contractual flexibility clauses can further enable adaptation to emerging circular technologies during extended concession periods.

For financiers and insurers, integrating CE benchmarks into risk assessment and financing models can enhance transparency and improve project bankability. Aligning environmental performance metrics with financial structures can reposition CE from a compliance requirement to a strategically embedded component of PPP value delivery.

The findings indicate that advancing circular innovation in PPP smart city projects requires greater contractual flexibility within concession agreements. Rather than relying on rigid, prescriptive technical specifications, PPP contracts should incorporate performance-based requirements and variation clauses that allow the adoption of emerging circular technologies over time. Embedding adaptive, lifecycle-oriented environmental KPIs can further ensure that circular performance is monitored and progressively improved throughout the concession period. When payment mechanisms are lifecycle-based, CE strategies become financially aligned with long-term asset performance. Approaches such as durability, modularity and resource efficiency can reduce maintenance costs and extend asset usability, thereby strengthening value-for-money outcomes. Integrating environmental performance incentives into availability payments reinforces sustained circular practices and ensures that CE objectives are embedded within the financial logic of PPP delivery rather than treated as peripheral sustainability targets.

This research aims to investigate the critical barriers that impede CE implementation in PPP smart cities. The results devised five unique cluster of barriers, including systematic and institutional barriers, financial constraints and cost pressures, attitudinal barriers, governance and regulatory barriers and digital innovation and knowledge. Complex CE design, paucity of industry standards, regulatory framework and absence of international CE policies are responsible for slow progress towards CE in PPP smart cities. While the response rates cover major continents of the world, including Africa, Asia, Europe, Oceania and North America, covering both developed and developing countries to reflect diverse perspectives, the uneven distribution of expert participants across countries was a key limitation of the study. Future research may attempt to gather a more even distribution of survey respondents, which is larger and geographically balanced. The survey includes participants from both affluent and impoverished countries, where cultural norms, institutional maturity and legal frameworks regarding PPP and CE practices may vary significantly. These contextual factors could influence how respondents interpret and rate certain barriers. Future research might unpack cultural contexts and their association with the CE practices. Most existing studies on CE rely predominantly on quantitative approaches. Future research should explore construction professionals’ insights to capture the nuanced CE paradoxes and project-level constraints that quantitative analyses often overlook.