Despite being recognised for their potential to facilitate the circular economy, current research offers a fragmented exploration into circular business models (CBMs) and Product Service Systems (PSS) within the construction sector. This study addresses this gap by systematically analysing trends and breaking down key knowledge areas.
A mixed-method review process was employed in this study. The quantitative review analysed 389 papers retrieved from Scopus and Web of Science, depicting the temporal development of CBM and PSS research in the construction industry. Screening and exclusion were conducted using the PRISMA framework, resulting in 82 papers for further content analysis.
The quantitative analysis reveals the growing trend in both CBM and PSS publications. The qualitative analysis identifies five interrelated knowledge areas, including the role of data and technology, stakeholder and supply chain networks, organisational learning, material circularity and circular business practices. Moreover, this study discusses the interrelationships between each of these identified knowledge areas.
The findings of this review contribute to the improved understanding of PSS and CBMs in the construction sector. This review also highlights key research gaps and proposes directions for future research.
This paper provides construction industry stakeholders with an empirical basis for understanding the drivers and barriers behind CBMs and PSS. The five knowledge areas identified bridge the practical-knowledge gap, guiding professionals in translating theory to practice.
This review addresses the limited exploration of PSS in relation to CBMs within the construction literature by providing a novel classification of existing research.
1. Introduction
The construction sector, often characterised by its traditional practices and resource-intensive processes, is at a critical point in its development. The construction industry is recognised as being the largest consumer of raw materials (Ghaffar et al., 2020; Zimmann et al., 2016). This growth in raw material consumption is expected to reach unprecedented levels exceeding 190 billion tonnes by 2060 (Elhacham et al., 2020; United Nations Environment Programme, 2020). In addition, estimates show that this industry accounts for 36% of global energy demand and 37% of the total greenhouse gas (GHG) emissions (United Nations Environment Programme, 2021). This is the result of the prevailing linear economic model, characterised by the “take-make-use-dispose” approach, resulting in negative environmental externalities (i.e. negative costs incurred during the production and consumption of goods) (Guerra and Leite, 2021).
In its sustainability transition, the construction industry has witnessed a wider adoption of green technologies, renewable energy and smart building systems (Berawi, 2019). This shift has been spurred by growing environmental concerns related to climate change, resource scarcity and pollution. Actions to reduce toxic material use, decrease GHG emissions and improve waste management practices are becoming increasingly adopted (Larsson, 2018). In line with this, concepts such as Circular Economy (CE) have emerged to facilitate the wider diffusion of sustainability across various industries, including construction.
1.1 Circular Economy
Despite the emergence of CE in the construction industry, the fragmented nature of the supply chain has diminished the potential benefits, resulting in lower returns compared to other sectors (Alashwal and Fong, 2015; Das et al., 2023; Riazi et al., 2020). The CE concept, popularised by the Ellen MacArthur Foundation, provides a restorative approach that aims to maintain products and components at their highest utility and value (Ellen MacArthur Foundation, 2015). In response to increasing sustainability demands, business models in the construction sector have undergone significant transformations, reshaping the entire supply chain (Tokazhanov et al., 2022). In this transition, the adoption of circular business models (CBM) (CBMs) has emerged as a promising pathway to steer the construction sector towards being more environmentally-friendly. These models define the rationale of how an organisation creates and captures value to slow, narrow and close resource loops (Antikainen and Valkokari, 2016; Bocken et al., 2018; Heesbeen and Prieto, 2020).
The impetus for the exploration of CBMs can be attributed to the popularity of CE. This area has received considerable attention in both academia and the industry (Bocken et al., 2014). Geissdoerfer et al. (2017) identified the relevance of business model innovation as the main factor towards the social and technological transformation towards CE. Supply chain networks that advance circularity require business models capable of delivering improved social, economic and environmental outcomes – the so-called “triple-bottom line” (Rashid et al., 2013). However, the inability of companies to rethink their supply chain has been identified as a major challenge, particularly in how they can create and deliver value through CBMs (Lüdeke-Freund et al., 2019). Product and process innovation which focuses on building sustainable products and the production process initially formed the bulk of the research in this area (Geissdoerfer et al., 2017). However, researchers argue this approach is simply inadequate, requiring a more holistic transformation of how a company creates and captures value (Chesbrough and Rosenbloom, 2002; Rashid et al., 2013; Zott and Amit, 2010).
Existing research identifies five types of business models that advance CE. These are resource recovery, circular supply, sharing models, product life extension and product service systems (Lacy et al., 2014; OECD, 2019). Current research on CBM in the built environment reveals a gap in the implementation of CBMs, particularly with product service systems being the least utilised (Guerra et al., 2021).
Despite its significant potential and numerous benefits for advancing CE, Product Service Systems (PSS) models have a relatively low implementation rate within the construction sector. This is largely due to their complexity, compared to other CBM (Guerra et al., 2021). In a PSS model, the manufacturer is the responsible owner of the product/asset during its useful life which is then leased to the customer or building owner. This model shifts the focus from selling only a tangible product to selling a system of products and services to the customer (Andersson and Lessing, 2019). Mont (2002) emphasised that achieving market success requires supporting networks and infrastructure that complement product/service delivery. These systems enable PSS to satisfy customer needs with lower environmental impacts than traditional models. Depending on the level of service intervention, a PSS can be product-oriented, use-oriented or result-oriented (Baines et al., 2007). Furthermore, their level of environmental impact varies depending on the PSS type (Tukker, 2004). Highly servitised PSS models (functional result and pay per use) show a significant potential to reduce environmental impact, compared to product-oriented types (Tukker, 2004). Although they hold great promise to decouple resource use, promote economic growth and deliver customer value (Pieroni et al., 2019), PSS models have received limited application in the construction sector (Guerra et al., 2021).
The current state of knowledge on CBM and product service systems in the construction industry has been examined across several papers. Munaro et al. (2021) offer a categorised analysis for each business model type (circular supply, resource recovery, product as a service, product life extension) across the building lifecycle. The study compares these CBM based on volume of research, with the bulk of the research focussing on circular supply (40%). In contrast, product service systems are minimally represented, comprising only 7% of the reviewed studies (Munaro et al., 2021). Das et al. (2023) tracks the historical development of business model research in construction literature. Simultaneously, this review identifies the focus areas of current business model research to be industrialised house building, strategic organisational transformation, and CBM (Das et al., 2023). Guerra et al. (2021) offers a practical lens to CBM and product service systems, highlighting their level of application across construction companies globally. Business model complexity and the type of business affected the adoption of CBM (CBMs). Product Service System (PSS) was perceived to have a higher level of complexity, limiting its adoption across construction companies globally (Guerra et al., 2021). Despite their unique contributions, however, these studies do not provide a comprehensive picture of the current literature on CBMs and PSS. While they offer useful insights into the exploration of individual CBM types, their levels of application and historical development, they do not fully capture the broader scope of contemporary research. Recent literature increasingly examines multiple dimensions including technological, organisational, and relational factors that shape the implementation of these business models. This fragmented body of knowledge therefore provided the basis for conducting this review, which aims to systematically analyse and categorise the different dimensions of CBM and PSS research.
Thus, the aim of this systematic review is to map how PSS research in construction literature has evolved over time and how it has been applied in connection with CBM. The review will also classify the literature into core knowledge areas to identify the current state of knowledge, gaps, and areas for future research. To achieve this, the research sets out to:
Identify relevant publications in the area of PSS and CBMs through a keyword search.
Highlight key research trends and gaps in relation to PSS and CBM research through bibliometric analysis.
Classify the knowledge contributions and their interrelationships using content analysis.
Propose areas for future research directions based on the findings from this review.
To identify research gaps, this review employs Miles (2017) proposed framework to distinguish between seven types of research gaps. This framework consists of knowledge gap, methodological gap, practical-knowledge conflict gap, evidence gap, empirical gap, theoretical gap and population gap (Miles, 2017). This framework builds upon the gaps identified by Müller-Bloch and Kranz (2015) which include knowledge void, action-knowledge conflict, theory-application void, methodological conflict, evaluation void and contradictory evidence (Müller-Bloch and Kranz, 2015). Miles (2017) provides a comprehensive tool for assessing research gaps when conducting literature reviews (Chigbu et al., 2023; Miles, 2017). Following this framework, this review identifies gaps in theory, knowledge, methodology, practical knowledge, evidence and empirical research, further expanded upon in the discussion section.
2. Methodology
A “mixed-methods” systematic review comprised of a bibliometric analysis followed by an in-depth content analysis was primarily used in this study. A mixed-method systematic review builds on the strength of both quantitative and qualitative approaches, combining them to provide a more comprehensive analysis than using either method independently (Debrah et al., 2023). This approach enabled both a macro-level mapping of research trends and a deeper conceptual development of knowledge areas. The bibliometric analysis aims at identifying research trends, mapping key knowledge areas, and assessing the temporal development in this research domain. Meanwhile, the content analysis uses iterative coding to provide a deeper dive into the enabling conditions for CBMs and PSS, as discussed in the relevant literature. The methodological process followed a systematic approach to identifying and categorising literature, ensuring methodological rigour of the review process, as described below.
2.1 Review protocol- PRISMA analysis
The review process was guided by the PRISMA framework, which stands for Preferred Reporting Items for Systematic Reviews and Meta-Analyses. Using the PRISMA framework, this review systematically identified and analysed relevant literature to fulfil the first objective of the study. This method was employed to reduce the probability of bias and enhance the validity of the scientific process (Moher et al., 2009). In this framework, four key stages, including identification, screening, eligibility and inclusion have been identified (Denyer and Tranfield, 2009), as shown in Figure 1.
The flowchart illustrates the four phases of study selection, arranged vertically from top to bottom: “Identification”, “Screening”, “Eligibility”, and “Included”. Each section contains boxes that include document-shaped images representing the records. In the “Identification” phase at the top, two boxes are positioned side-by-side. Box “1” on the left is labeled “Records identified through database search (W o S)”. It contains two document-shaped images: “1st Search String” with “N equals 68” and “2nd Search String” with “N equals 50”. Box “2” on the right is labeled “Records identified through database search (Scopus)”. It contains two document-shaped images: “1st Search String” with “N equals 228” and “2nd Search String” with “N equals 43”. In the “Screening” phase below, Box “3” is centered and labeled “Records after removal of duplicates”. It contains two document-shaped images: “1st Search String” with “N equals 244” and “2nd Search String” with “N equals 53”. This flows downward into two side-by-side boxes. Box “4” on the left is “Records screened based on abstract” and contains two document-shaped images with “N equals 163” and “N equals 26”. Box “5” on the right is “Records excluded” and contains two document-shaped images with “N equals 81” and “N equals 27”. In the “Eligibility” phase further down, Box “6” on the left is labeled “Full-text article assessment for eligibility” and contains a document-shaped image for “1st and 2nd Search String Combined” with “N equals 189”. Box “7” on the right is labeled “Full-text article assessment for exclusion” and contains a document-shaped image for “1st and 2nd Search String Combined” with “N equals 107”. In the final “Included” phase at the bottom, Box “8” is centered and labeled “Studies included in the qualitative assessment”. It contains a single document-shaped image with “N equals 82”.PRISMA flow diagram. Source: adapted from Moher et al. (2009)
The flowchart illustrates the four phases of study selection, arranged vertically from top to bottom: “Identification”, “Screening”, “Eligibility”, and “Included”. Each section contains boxes that include document-shaped images representing the records. In the “Identification” phase at the top, two boxes are positioned side-by-side. Box “1” on the left is labeled “Records identified through database search (W o S)”. It contains two document-shaped images: “1st Search String” with “N equals 68” and “2nd Search String” with “N equals 50”. Box “2” on the right is labeled “Records identified through database search (Scopus)”. It contains two document-shaped images: “1st Search String” with “N equals 228” and “2nd Search String” with “N equals 43”. In the “Screening” phase below, Box “3” is centered and labeled “Records after removal of duplicates”. It contains two document-shaped images: “1st Search String” with “N equals 244” and “2nd Search String” with “N equals 53”. This flows downward into two side-by-side boxes. Box “4” on the left is “Records screened based on abstract” and contains two document-shaped images with “N equals 163” and “N equals 26”. Box “5” on the right is “Records excluded” and contains two document-shaped images with “N equals 81” and “N equals 27”. In the “Eligibility” phase further down, Box “6” on the left is labeled “Full-text article assessment for eligibility” and contains a document-shaped image for “1st and 2nd Search String Combined” with “N equals 189”. Box “7” on the right is labeled “Full-text article assessment for exclusion” and contains a document-shaped image for “1st and 2nd Search String Combined” with “N equals 107”. In the final “Included” phase at the bottom, Box “8” is centered and labeled “Studies included in the qualitative assessment”. It contains a single document-shaped image with “N equals 82”.PRISMA flow diagram. Source: adapted from Moher et al. (2009)
A database search was initially performed to identify and systematically analyse existing research on this topic. Scopus and Web of Science (WoS) databases were used because of their relative simplicity and accuracy to retrieve relevant articles (Durdyev, 2020). To improve the quality and validity of the analysis, it is important to include all the relevant key terms related to the topic (Debrah et al., 2023). To maximise coverage, two combinations of search strings were used to extract relevant papers on PSS and CBMs within the construction literature (See Table 1). A total of 14 key terms were used for the first search string and 17 key terms in the second search string. This was done in order to compare and assess the temporal development across both research domains. To ensure reproducibility, Boolean operators, truncations and phrase searches were applied across both databases. Table 1 shows the key search terms used in the database search.
Selection of keywords used in database search
| Search string | Keywords |
|---|---|
| 1st search string | “Business Model” OR “Business Model Innovation” OR “Circular Business Model” OR “Circular Business Model Innovation” OR “Business Drivers” OR “Critical Success Factors” OR “Driving Forces” |
| “Circular Economy” OR “Circularity” OR “Closed-Loop System” OR “Performance Economy” | |
| “Construction*” OR “Construction Industry” OR “Built Environment” | |
| 2nd search string | “Product Service System” OR “Servitisation” OR “Servitization” OR “Product Stewardship” OR “Product-as-a-Service” OR “PSS” OR “Product Service Systems” OR “Business Drivers” OR “Critical Success Factors” OR “Driving Forces” |
| “Circular Economy” OR “Circularity” OR “Closed-Loop System” OR “Performance Economy” | |
| “Construction*” OR “Construction Industry” OR “Built Environment” |
The identified key terms were searched in the title, abstract and keywords of each article, to ensure the widest possible coverage of information on this topic across both databases (Lima et al., 2021). The first search string yielded a total of 228 research outputs while the second search string produced 43 results in Scopus, and 68 and 50 results in WoS. In the screening stage, 52 duplicate records from the first search string and 40 from the second search string were excluded. Following the duplicate removal, records were further screened based on title and abstracts. The eligibility criteria for inclusion and exclusion were important to limit the scope of the review. In screening the relevant research papers, the following inclusion criteria was applied:
Journal articles, conference papers and book chapters were considered for this study.
Publications in English language
Studies explicitly focused on CBMs and/or PSS within the construction sector/built environment
On the other hand, publications were excluded for one or more of the following:
Articles referring to topics in medicine, waste, road and transport, automobiles, biotechnology, textiles, food and biowaste
Articles written in any language other than English
Articles lacking full-text availability
3. Findings
3.1 Quantitative review (bibliometric analysis)
The quantitative review uses descriptive time-series analysis to depict emerging trends in this subject area, addressing the second objective set out in this review. The analysis reveals an upward trajectory in the streamlined academic publication of CBM and PSS within the construction literature. However, the proportion of PSS literature compared to wider business model innovation research is still low, as shown in Figure 2. Figure 2 also visualises this temporal development of CBM and PSS research across Scopus and WoS databases.
The line graph illustrates the number of publications across various years. The horizontal axis is labeled “Year” and ranges from “1980” to “2023”. The vertical axis is labeled “Number of Publications” and ranges from “0” to “60” in increments of “10” units. The graph shows four lines. A legend on the left identifies four distinct lines: “C B M publications (Scopus)”, “P S S publications (Scopus)”, “C B M publications (W o S)”, and “P S S publications (W o S)”. The first line, “C B M publications (Scopus)” starts at “1” in “1980”, remains flat until “2014”, then moves in a steep upward direction through the increasing point of “26” in “2019” and peaks at “52” in “2022” before reaching an ending point of “48” in “2023”. The second line, “C B M publications (W o S)”, starts at “0” until “2013”, then moves in an upward direction through the increasing point of “11” in “2020” to reach an ending point of “18” in “2023”. The third line, “P S S publications (Scopus)”, is starting at “0” until “2012”, showing a slight increase to “2” in “2014”, then moves in an upward direction through the increasing point of “7” in “2020” to reach an ending point of “12” in “2023”. The fourth line, “P S S publications (W o S)”, is starting at “0” until “2016”, showing a slight increase to “8” in “2020” and then a dip to “4” in “2021” before reaching an ending point of “14” in “2023”. Note: All data numerical values are approximated.Number of publications on CBM and PSS in Scopus and web of science by year. Source: Authors’ own work using data from Scopus and WoS
The line graph illustrates the number of publications across various years. The horizontal axis is labeled “Year” and ranges from “1980” to “2023”. The vertical axis is labeled “Number of Publications” and ranges from “0” to “60” in increments of “10” units. The graph shows four lines. A legend on the left identifies four distinct lines: “C B M publications (Scopus)”, “P S S publications (Scopus)”, “C B M publications (W o S)”, and “P S S publications (W o S)”. The first line, “C B M publications (Scopus)” starts at “1” in “1980”, remains flat until “2014”, then moves in a steep upward direction through the increasing point of “26” in “2019” and peaks at “52” in “2022” before reaching an ending point of “48” in “2023”. The second line, “C B M publications (W o S)”, starts at “0” until “2013”, then moves in an upward direction through the increasing point of “11” in “2020” to reach an ending point of “18” in “2023”. The third line, “P S S publications (Scopus)”, is starting at “0” until “2012”, showing a slight increase to “2” in “2014”, then moves in an upward direction through the increasing point of “7” in “2020” to reach an ending point of “12” in “2023”. The fourth line, “P S S publications (W o S)”, is starting at “0” until “2016”, showing a slight increase to “8” in “2020” and then a dip to “4” in “2021” before reaching an ending point of “14” in “2023”. Note: All data numerical values are approximated.Number of publications on CBM and PSS in Scopus and web of science by year. Source: Authors’ own work using data from Scopus and WoS
The bibliometric analysis shows a growing interest in CBM research in the construction industry. Particularly, this data shows a percentage increase of 271% (see Table 2) on CBM research between 2018 and 2019. Similar insights can be drawn from the data on the WoS database, despite the relatively slower rate of growth. Based on volume, CBM-related literature comprises 76% of the total across both databases. Table 2 summarises the notable changes in CBM and PSS research across both databases.
Relative and percentage growth of PSS and CBM publications
| Publication | Period | Record at start of period (R0) | Records at end of period (Rf) | Relative growth/decline (Rf-R0) | Percentage Increase/decrease |
|---|---|---|---|---|---|
| CBM (scopus) | 2018–2019 | 7 | 26 | 19 | 271% |
| CBM (WoS) | 2019–2020 | 4 | 11 | 5 | 175% |
| PSS (WoS) | 2020–2021 | 8 | 4 | −4 | −50% |
| PSS (scopus) | 2020–2021 | 7 | 5 | −2 | −29% |
In contrast, PSS-related research does not follow a similar trajectory. Notably, this research area experienced a decrease of 50% between 2020 and 2021 across WoS database (see Table 2). These contrasting differences between CBM and PSS research informed the subsequent analysis, by reaffirming the relevance of this research topic within the academic literature and highlighting key gaps, particularly in relation to PSS-related research. Figure 2 shows the differences in the level of research on CBM and PSS.
3.2 Qualitative review (content analysis)
To address the third objective of this study, a three-level coding process was used to derive meaning from the remaining 82 out of 389 papers. The analytical stage used in this study applies principles to inductively generate theory by directly analysing empirical data (Wolfswinkel et al., 2013; Glaser and Strauss, 1967). This approach was used for two main reasons: (1) to breakdown key themes, patterns and concepts from existing literature using a three-level coding process (open, axial and selective coding) (2) to group the contributions of existing research into knowledge areas. This process involves three levels of coding: open, axial and selective coding (Glaser and Strauss, 1967; Wolfswinkel et al., 2013; Williams and Moser, 2019). The first stage, open coding, involves breaking down the literature into specific themes. This is followed by axial coding, which identifies relationships between open codes and aggregates interrelated concepts into core categories. Finally, selective coding is performed to organise axial codes into overarching knowledge areas. This final stage represents the highest level of abstraction, clustering interrelated axial codes to generate new theories (Williams and Moser, 2019). The coding process allowed multiple codes to be applied to each paper. This ensured an accurate representation of the literature as some papers addressed more than one theme. To improve the methodological rigour of the review process, a two-step quality appraisal process was used. This involved a relevancy assessment, based on the established exclusion and inclusion criteria as well as a methodological evaluation to examine research gaps and identify any potential inconsistencies or conflicting methods (Miles, 2017; Müller-Bloch and Kranz, 2015).
This review followed this coding process using literature as the primary source of data. To minimise bias and strengthen validity, an iterative approach was used across all three coding levels. The iterative coding process, characterised by “code-reflect-refine” cycle, involved multiple rounds of coding to minimise premature classification of thematic areas and ensure all concepts discussed in the relevant literature were thoroughly represented. Information from individual papers were extracted through a full-text analysis, with particular emphasis on their main findings. For instance, studies assessing technologies as enablers of CBM were first broken down into open codes such as “Building Information Modelling (BIM)”, “Artificial Intelligence (AI)”, “Internet of Things (IoT)”, “Digital Twin (DT)” etc. In the second axial coding stage, the codes were structured based on the relatedness of the open codes. Using the previous example, the technological tools were then organised into one axial code “Industry 4.0 tools”. Lastly, the selective coding stage consolidated the axial codes into a single main concept at the highest level of abstraction. Continuing the previous example, “Industry 4.0” was categorised under the broader cluster exploring the role of data and technological innovation. This step-by-step coding process was performed for all identified papers, resulting in the development of five categories. Table 3 shows the breakdown of the coding process for each coding level.
Breakdown of the coding process
| No | Selective codes | Axial codes count | Open codes count |
|---|---|---|---|
| 1 | Data and Technological Innovation | 3 | 21 |
| 2 | Stakeholder and Supply Chain Networks | 2 | 19 |
| 3 | Organisational Learning | 4 | 24 |
| 4 | Material Circularity | 5 | 20 |
| 5 | Circular Business Practices | 4 | 22 |
| Total | 18 | 106 |
During the content analysis, multiple coding per paper was allowed to capture the diverse themes explored in the literature. This analysis resulted in the development of five knowledge areas around CBM innovation in the construction sector. These include the role of data and technological innovation, stakeholder and supply chain networks, organisational learning (OL), material circularity and circular business practices. Table 4 shows the frequency and proportion of these 5 knowledge areas across the 82 relevant papers.
Matrix mapping based on 5 main categories
The conceptual relationships between these categories are further explored in this review using a network visualisation. Gephi software was used to construct a network analysis in which the knowledge areas served as nodes and were linked based on their frequency. This visualisation highlights the interconnectedness of the knowledge areas while simultaneously attributing the authors associated with each study. Figure 3 shows a network visualisation mapping publications across the main knowledge areas.
The network mapping diagram displays five primary teal-colored nodes representing core thematic clusters in academic research. The largest nodes are labeled “Circular Business Practices” at the top, “Material Circularity” in the center-right, and “Stakeholders and Supply Chain Networks” at the bottom. Two slightly smaller nodes are labeled “Organizational Learning” on the left and “Data and Technology” on the lower right. Numerous light blue curved lines radiate from these central nodes to connect with specific academic citations. Under “Circular Business Practices”, citations include “Azcarate-Aguerre et al., 2022a”, “Guerra et al., 2021”, “Ploeger et al., 2019”, “Kristensen and Remmen, 2019”, “Ghafoor et al., 2023”, and “Johansson et al., 2016”. The “Material Circularity” node is linked to “Munaro et al., 2021”, “Meglin et al., 2022”, “Tomatis et al., 2022”, “Anastasiades et al., 2021”, “Das et al., 2023”, “van den Berg et al., 2020”, “Heesbeen and Prieto, 2020”, “Illankoon and Vithanage, 2023”, “van Stijn and Gruis, 2020”, “Bertino et al., 2021”, “Fargnoli et al., 2019”, “Sinclair et al., 2018”, and “Cossu et al., 2012”. The “Stakeholders and Supply Chain Networks” cluster connects to “Giglio et al., 2022”, “Leising et al., 2018”, “Stegehuis et al., 2023”, “Giorgi et al., 2020”, “Xing et al., 2020”, “Nulholz et al., 2019”, “Oluleye et al., 2023”, “Munaro and Tavares, 2023”, “Azcarate-Aguerre et al., 2022b”, “Shooshtarian et al., 2023”, “Domenech et al., 2019”, “Julianelli et al., 2020”, “Wuni, 2023a”, “Hosseini et al., 2014”, “Davari et al., 2023”, “Owojori and Okoro, 2022”, “Talamo et al., 2020”, “Guerra and Leite, 2021”, and “Ollar et al., 2020”. The “Organizational Learning” cluster features links to “Srećković and Šibenik, 2023”, “Hartanto and Chang, 2022”, “Scipioni et al., 2021”, “Ruiter et al., 2022”, “Cruz Rios and Grau, 2020”, “Scipioni and Niccolini, 2021”, and “Lehtimäki et al., 2020”. The “Data and Technology” node is associated with “Argus et al., 2020”, “Cetin et al., 2021”, and “Setaki and van Timmeren, 2022”. Shared citations appearing between clusters, such as “Azcarate-Aguerre et al., 2021”.Network visualisation of authors along the five main categories. Source: Authors’ own work
The network mapping diagram displays five primary teal-colored nodes representing core thematic clusters in academic research. The largest nodes are labeled “Circular Business Practices” at the top, “Material Circularity” in the center-right, and “Stakeholders and Supply Chain Networks” at the bottom. Two slightly smaller nodes are labeled “Organizational Learning” on the left and “Data and Technology” on the lower right. Numerous light blue curved lines radiate from these central nodes to connect with specific academic citations. Under “Circular Business Practices”, citations include “Azcarate-Aguerre et al., 2022a”, “Guerra et al., 2021”, “Ploeger et al., 2019”, “Kristensen and Remmen, 2019”, “Ghafoor et al., 2023”, and “Johansson et al., 2016”. The “Material Circularity” node is linked to “Munaro et al., 2021”, “Meglin et al., 2022”, “Tomatis et al., 2022”, “Anastasiades et al., 2021”, “Das et al., 2023”, “van den Berg et al., 2020”, “Heesbeen and Prieto, 2020”, “Illankoon and Vithanage, 2023”, “van Stijn and Gruis, 2020”, “Bertino et al., 2021”, “Fargnoli et al., 2019”, “Sinclair et al., 2018”, and “Cossu et al., 2012”. The “Stakeholders and Supply Chain Networks” cluster connects to “Giglio et al., 2022”, “Leising et al., 2018”, “Stegehuis et al., 2023”, “Giorgi et al., 2020”, “Xing et al., 2020”, “Nulholz et al., 2019”, “Oluleye et al., 2023”, “Munaro and Tavares, 2023”, “Azcarate-Aguerre et al., 2022b”, “Shooshtarian et al., 2023”, “Domenech et al., 2019”, “Julianelli et al., 2020”, “Wuni, 2023a”, “Hosseini et al., 2014”, “Davari et al., 2023”, “Owojori and Okoro, 2022”, “Talamo et al., 2020”, “Guerra and Leite, 2021”, and “Ollar et al., 2020”. The “Organizational Learning” cluster features links to “Srećković and Šibenik, 2023”, “Hartanto and Chang, 2022”, “Scipioni et al., 2021”, “Ruiter et al., 2022”, “Cruz Rios and Grau, 2020”, “Scipioni and Niccolini, 2021”, and “Lehtimäki et al., 2020”. The “Data and Technology” node is associated with “Argus et al., 2020”, “Cetin et al., 2021”, and “Setaki and van Timmeren, 2022”. Shared citations appearing between clusters, such as “Azcarate-Aguerre et al., 2021”.Network visualisation of authors along the five main categories. Source: Authors’ own work
The following section provides further insights into these categories and expands on the key findings within each knowledge area:
The Role of Data and Technological Innovations: The role of data and technological innovation was discussed across 23% of the analysed literature (See Table 4). Cetin et al. (2021) identified the digital technologies that facilitate a circular built environment. These tools include AI, Blockchain Technology (BCT), Additive/Robot Manufacturing (AMRM), Digital Twins (DT), BIM, Geographic Information Systems, Big Data Analytics, IoT and Material Passports and Databanks. The role of these technologies in minimising construction and demolition waste have been studied across project lifecycle (Illankoon and Vithanage, 2023). A similar study by Setaki and van Timmeren (2022) examined the role of digital technologies across the lifecycle stages. Based on this research, IoT, BIM, AI and Blockchain technologies play a critical role in the design and engineering stages. In the construction phase, IoT, BIM, robotics, 3D printing, and Drones/AR promote the optimal use of material resources. In the demolition phase, BIM, robotics and AI/drones are crucial in recovering by-products and waste (Setaki and van Timmeren, 2022). Oluleye et al. (2023) further examined the specific role of AI in the implementation of circularity across the building lifecycle.
In the commercial viability of CBMs, data quality has emerged as a critical driver for market success. A successful CBM relies on three types of data: (1) Capturing data on raw materials (2) Tracking product data throughout the lifecycle and (3) Verification of raw material and product data (Argus et al., 2020). BIM further enabled by the cloud was discussed as an enabling technology for PSS implementation. For instance, in PSS implementation, component information stored in BIM can be used to facilitate maintenance operations (Fargnoli et al., 2019). Xing et al. (2020) expand on these BIM solutions to include cyber-physical capabilities and cloud-based solutions to help identify, track and exchange components.
In the practical application of PSS, Signify Philips’ ‘lighting as a service’ model incorporates a range of Industry 4.0 technologies, particularly IoT solutions (Cetin et al., 2021; Nobre and Tavares, 2017). Argus et al. (2020) identifies the correlation of data quality and stakeholder responsibility as vital factors in the commercial success of CE business models. Azcárate-Aguerre et al. (2018) also highlights the role of technologies in effectively implementing and minimising the risk of Façade-as-a-Service business models. Emerging AI capabilities can also be leveraged across the building lifecycle stages from design to pre-demolition (Oluleye et al., 2023).
The Role of Stakeholders and Supply Chain Networks: Given the wide range of actors involved, 40% of the identified papers focused on the role of stakeholder management and supply chain networks (see Table 4). The construction industry involves a diverse set of stakeholders including organisations, firms and independent agencies that collectively form a supply chain network (Boons and Baas, 1997; Gordon and McCann, 2000; Leising et al., 2018). Domenech et al. (2019) asserts the role of industrial networks in achieving environmental, social, and economic benefits. This so-called “triple bottom line” (environmental, social, and economic) can also be advanced through networks engaged in remanufacturing processes (Talamo et al., 2020).
In relation to networks, the role of logistics and supply chain has been discussed across the literature as an enabler in the CE transition (Giglio et al., 2022; Julianelli et al., 2020; Lüdeke-Freund et al., 2019). This transition requires companies to rethink their supply chains, and redefine how they create and deliver value (Lüdeke-Freund et al., 2019). These gaps in the industry can be potentially solved through business models that seamlessly integrate actors through mutually beneficial solutions (Berg et al., 2021; Das et al., 2023). The implementation and success of circular strategies is reliant on factors including the synergy, partnerships and collaboration between stakeholders in the supply chain (Geldermans, 2016; Hofmann, 2019; Mokhlesian and Holmén, 2012; Wells and Seitz, 2005). In the built environment, stakeholder collaboration plays a crucial role in the adoption of circular principles along the construction value chain (Zimmann et al., 2016). Guerra and Liete (2021) further assert that multi-stakeholder engagement comprising of government, organisations and academia as essential actors in the exchange of knowledge and information, accelerating the transition towards CE in the built environment. Munaro and Tavares (2023) also highlight the necessity for collaborative and interdisciplinary action between the government and stakeholders to facilitate its sustainability transition. Gue et al. (2020) also define the causal relationships between CE transition supported by governmental policies, investments, and awareness on both the consumer side and the business side. Determining the economic viability of CBM is yet another crucial element in actively engaging stakeholders (Gue et al., 2020). Pareto analysis of stakeholder success factors shows that sustained collaboration, communication and information sharing, early involvement and commitment to be the critical elements (Wuni, 2023a, b). The private sector’s role in the CE transition highlights the need for circular thinking, collaboration in technological innovation and value co-creation across the supply chain (Owojori and Okoro, 2022).
The role of value co-creation based on shared benefits and long-term partnerships was also identified as a critical enabler in the success of Façade-as-a-Service (Azcarate-Aguerre et al., 2021). Ollar et al. (2020) identify spatial and product design, end-user needs and perceptions, regulatory measures and collaborative value creation of circular products as the key opportunities in Sweden. Davari et al. (2023) propose a traceability framework that articulates the various elements defining traceability and the enabling factors to enhance circularity of built assets.
The Role of OL: An organisation’s ability to learn and adapt was discussed across 17% of the identified literature (see Table 4). Learning among actors involved in a network is an essential element in the shift to innovation (Brown et al., 2003; Leising et al., 2018; Quist and Tukker, 2013). Two levels of learning can be identified (Brown et al., 2003; Raven, 2005): First order learning aims to provide new insights on specific problems in relation to the project definitions, norms, beliefs and the vested interest of actors. On the other hand, higher order learning aims to adopt radical and disruptive innovations along with the required change processes (Leising et al., 2018). Across SME’s, precedence of business growth is critical in the implementation of CE (Hartanto and Chang, 2022). Srećković and Šibenik (2023) highlights the role of an organisation’s internal and external environment in the development and implementation of end-of-life business models.
Several OL concepts have been discussed in the literature. For instance, the role of management control tools has been identified as a potential enabler in achieving CE principles (Ruiter et al., 2022). A framework proposed by Scipioni et al. (2021) and Scipioni and Niccolini (2021) highlights the role of interorganisational and intraorganisational learning in relation to CBM (CBMs). Intraorganisational learning happens at the company level, while interorganisational learning occurs at the supply chain level. Learning that occurs within the organisation is critical in the development of CBMs. Knowledge creation, transfer and retention form the building blocks of the OL processes within and across organisations (Scipioni et al., 2021). OL processes are highly influenced by the cultural, regulatory, structural and process factors in an organisation. In another study, three dimensions that impacted OL in the context of CBM’swere identified to be stimulating external environment, collaborative supply chain and resilient organisational features (Scipioni and Niccolini, 2021). Lehtimäki et al. (2020) distinguishes between internal and external factors in CBM innovation of a company. Internal factors that influence a company’s strategic management as well as organisational/technological capabilities are essential in overcoming challenges in the transition towards CE (Lehtimäki et al., 2020). Linked to the role of stakeholders and supply chain, interorganisational tensions arising from conflicting priorities between network actors has been identified as one of the issues hindering OL (Stegehuis et al., 2023). The commercial success of PSS has also been linked to enabling organisational layouts and behavioural change among consumers, companies, government and the broader society (Mont, 2002; Sinclair et al., 2018).
The Role of Material Circularity: The role of material circularity has been extensively discussed across the identified records, comprising 44% of the total publications (see Table 4). In an ethnographic study on demolition projects, van den Berg et al. (2020) highlights the consideration of economic and financial aspects in recovery of materials in the end-of-life phase. Structural and façade elements with higher perceived value were preferred due to their modular nature and their ease-of-disassembly (van den Berg et al., 2020). On the other hand, the structural and non-structural nature of building elements plays a role in the demolition process. Complexity of building, smart material selection and information access are key principles that facilitate the deconstruction process in both structural and non-structural building elements (Bertino et al., 2021). Business model innovations also play a role in enabling the flow of secondary building materials. According to Nuβholz et al. (2019), innovations in resource recovery technologies, partner networks for secondary materials and identifying appropriate customer segment are essential business model drivers enabling secondary resource flow. Nordby (2019) argues the primary driver behind the increase in material circularity to be the development of national targets aimed at reducing greenhouse gas emissions in buildings.
Standarisation and traceability of assets have also been identified as enablers in the circularity of building components in the construction industry. Anastasiades et al. (2021) identify protectionism – a reluctance to modify organisational structure and business routines (Hosseini et al., 2014), on the manufacturer and contractors’ side to be a key problem area preventing standardisation of building components. In addition, supply of secondary raw materials is proposed as a solution in transition to CE through Business Models (Meglin et al., 2022). Refurbishment has also been identified as a potentially sustainable approach in reducing material demand and waste generation for elevator systems (Tomatis et al., 2022).
The Role of Circular Business Practices: Forming 39% of the analysed literature, the role of circular business practices forms one of the key focus areas in the industry’s CE transition. Current trends reveal emerging patterns across CBM research which include repair and maintenance, reuse and redistribution, refurbishment and remanufacturing, recycling, cascading and repurposing and lastly, organic feedstock (Lüdeke-Freund et al., 2019). In the construction context, the development of CBM necessitates the reformation at each stage of the value chain (Das et al., 2023; Tokazhanov et al., 2022; Lüdeke-Freund et al., 2019). Srećković and Šibenik (2023) asserts the lack of business models that focus on the entire lifecycle of a building in the Architecture, Engineering and Construction sector (Srećković and Šibenik, 2023). In terms of building energy retrofits, PSS business model has been identified as a future proofing mechanism for building owners (Azcarate-Aguerre et al., 2022a, b). Successful business cases in construction indicate the importance of strategic partnerships and digital technologies in their implementation (Das et al., 2023). Restructuring of economic incentives between actors is vital to increase investments in energy retrofitting and building facades (Azcarate-Aguerre et al., 2021).
CBMs have emerged as a new pathway for achieving circularity in the built environment. Johansson et al. (2016) present a case for using PSS in urban mining – a novel approach where urban areas are mined instead of the bedrock. PSS also offers opportunities in housing, improving the lifetime performance of building assets and enhancing the overall value for the owner (Ghafoor et al., 2023). However, retention of ownership in PSS presents legal constraints, necessitating a radical change in the traditional ownership model (Ploeger et al., 2019). In the exploration of CBMs, Business Model Canvas (BMC) has emerged as a ubiquitous tool for generating business models (Talukder, 2017; Das et al., 2023). Comprised of nine building blocks, it is a simple yet powerful tool which has been applied across several sectors (Osterwalder and Pigneur, 2010). This tool has served as a benchmark in the development of numerous business models, such as the Sustainable CBM (Antikainen and Valkokari, 2016), CBM Canvas (Lewandowski, 2016), Value Mapping Tool (Bocken et al., 2013), Sustainable Value Proposition Framework (Kristensen and Remmen, 2019), Business Combo Model (Talukder, 2017), The Business Ecosystem Canvas (Humbeck et al., 2020), PSS Business Model (Barquet et al., 2013), Triple Layered BMC (Joyce and Paquin, 2016), Green Business Model Innovation (Henriksen et al., 2012), Environmental Value Proposition Framework (Manninen et al., 2018) and CBM (Garcia-Muiña et al., 2018).
4. Discussion of findings
The findings from this review indicate a growing interest in CBMs, with a considerable increase in research output within the last six years (See Figure 2), consistent with the findings from Norouzi et al. (2021). Despite the upward trend, the exploration of Product Service Systems remains limited, in line with Munaro et al. (2021), reflecting a gap in understanding how PSS can be successfully integrated into the built environment. The key themes emerging from the academic literature include the role of data and technology, stakeholder engagement, OL, material circularity and circular business practices. These qualitative findings were synthesised from relevant papers, addressing the first two objectives of this review.
In relation to prior PSS and CBM concepts, the knowledge areas identified in this study place special emphasis into the challenges of the construction industry. Drawing on construction-focused literature, the findings capture factors that characterise the building industry including material circularity through standardisation and end-of-life recovery and the role of data and technology through Industry 4.0 applications including AI, BIM and IoT. By addressing sector-specific challenges such as fragmented stakeholder networks and slow innovation diffusion, the findings extend existing PSS and CBM understanding across five knowledge areas that are critical for improved adoption in the construction industry. More broadly, this review provides a deeper understanding of the structural factors influencing the understanding of CBMs and PSS, ranging from stakeholder and supply chain networks to OL process. Importantly, these elements are highly interdependent, as shown in Figure 3. For instance, collaborative stakeholder relationships can facilitate the flow of circular materials across the value chain. At an organisational level, learning and capability development play a key role in the implementation of CBMs. Within this whole system, technology acts as a cross-cutting enabler, supporting stakeholder collaboration, material circulation and reducing organisational tensions. Together, these elements play an important role in the success of CBMs and PSS within the construction industry.
Moreover, the findings reveal a strong emphasis on material circularity (44%) and stakeholder and supply chain networks (40%) (See Table 4). However, there is a comparative lack of empirical research exploring OL and decision-making processes (17%). This gap in knowledge stifles the development of circularity strategies at an organisational scale. A methodological gap was identified in the variety of methodological approaches used, making it difficult to develop standardised metrics for assessing the effectiveness of CBMs. A range of methodologies were identified across the 82 papers, with interviews being the most prevalent data collection instrument used, comprising 19.5% of the papers. Subsequently, 17% of studies used case studies, while only 9.8% of studies employed surveys.
The findings also reveal tensions across literature in the current discourse surrounding the application of CBMs and PSS in the construction sector. While digital tools are widely recognised as critical enablers, the success of CBMs equally depends on collaboration, shared value creation and OL. A major barrier to implementing service-based models like PSS lies in the traditional ownership structures that currently dominate the construction industry. While not fully examined in this review due to the limited number of relevant studies (n = 1), a lack of clarity around the legal and contractual factors behind PSS remains a significant factor constraining PSS adoption in the industry, as highlighted by Ploeger et al. (2019). Another tension arises from the theoretical appeal of CBMs and the practical complexities involved in their real-world application. As observed by Guerra et al. (2021), the complexity of a CBM directly impacts its implementability, with simpler models like circular supply and resource recovery experiencing greater uptake compared to more sophisticated models such as PSS. In addition to this, lack of standardisation, traceability and economic viability further hinder the industry adoption of CBMs across the industry.
5. Conclusion and future directions
Drawing on a bibliometric-qualitative analysis, this review highlights key trends and emerging areas of research on CBM (CBMs) and product service systems in the construction industry. The findings contribute to a broader understanding of the drivers and barriers of CE within the context of the built environment. Existing reviews address the macro-level transition to CE (van Stijn and Gruis, 2020) with studies focussing on financial/economic, technological, legal/regulatory and, social/cultural factors (Ababio and Lu, 2023; Caldera et al., 2019; Giorgi et al., 2022; Guerra and Leite, 2021; Hart et al., 2019; Munaro and Tavares, 2023; Srećković and Šibenik, 2023; Wuni, 2023a, b). Moreover, existing reviews on CBMs provide a historical basis (Das et al., 2023), while others provide an overview of the current level of research and application in the construction industry (Munaro et al., 2021; Guerra et al., 2021). In contrast, this review pays special attention to the development of the main knowledge areas associated with PSS and CBMs within the context of the construction industry. By systematically organising and analysing extant literature on PSS and CBMs (see Figure 3), this review discusses the enabling conditions driven by data and technology, stakeholders and supply chain networks, material circularity, OL and circular business practices. It is worth noting that each of these knowledge areas does not exist in a vacuum and is highly interdependent. Collaborative stakeholder relationships, for example, are essential for enabling material circularity strategy like reverse logistics and enhancing traceability across the value chain. Individual stakeholders require learning and capacity development, which are critical to implement CBMs. Within this interconnected system, data and technology act as bridge facilitating communication and collaboration across stakeholders, supporting material flow and minimising organisational tensions.
Material circularity is fundamental to the effective implementation of CBMs, requiring robust frameworks for standardisation and traceability. These practices are essential in closing material loops and designing out waste, which are core principles of CE. The role of shared stakeholder interest and collaborative supply chain networks is vital for the successful implementation of CBMs. Given the fragmented nature of the construction supply chain, collaboration between stakeholders is essential in overcoming barriers to circularity. Reverse logistics, which focuses on returning products from end-user back to the manufacturer, is an approach that can foster stakeholder collaboration. In the successful application of CBMs, the role of data and technology offers numerous opportunities. Quality data combined with appropriate digital technologies hold great potential to improve efficiency and minimise waste. Furthermore, Industry 4.0 technologies are effective in the practical application of CBMs. In the construction literature, business practices informed by circularity have overwhelmingly employed the BMC as a business generation tool. Despite customisations to this business model, this tool hasn’t undergone major changes to address the unique needs of CBMs, presenting an opportunity for refinement. Decision-making processes in organisations is another critical factor in the successful adoption of CBMs. OL enables companies to acquire, develop and implement innovative CBMs, enabling them to adapt to change and achieve circularity goals. This process can either occur internally within a single company or collaboratively across multiple entities.
The primary focus of this study is to improve understanding on the key enablers influencing the commercial application of PSS and CBMs. Drawing on current research trends, five knowledge areas have been identified to inform future research on this topic, addressing the fourth objective of this review. Future opportunities lie in extending these practical concepts to include sector-specific guidelines and policy interventions. These dimensions can further advance theoretical and practical understanding of PSS in the context of CBM, thereby lowering the implementation barriers and improving scalability of these business models. Moreover, legal and regulatory factors influencing the adoption of PSS and CBMs remain largely unexplored, as highlighted in the findings of this review. Although these factors are discussed in broader CE literature, their implications within the specific context of PSS and CBMs are insufficiently examined. These factors are highly context-specific, yielding better insights when examined within certain geographical settings. Future research can build on the five knowledge areas by incorporating legal and regulatory factors, which is an area not fully addressed in this review. Integrating a legal dimension can provide a holistic view of the factors affecting the commercial viability of CBMs, particularly within the construction sector. In terms of methodology, approaches such as longitudinal studies and practical case studies can be better utilised to offer solid empirical evidence on the commercial viability of PSS and CBMs. Future researchers can also expand on generic theoretical frameworks like BMC, tailoring it based on the unique features of PSS and CBMs. While prior studies (discussed in Section 4) have proposed circular-oriented adaptations of the BMC, these frameworks remain only partially comprehensive and often lack empirical validation. The five knowledge areas developed from this study offer deeper insights into PSS and CBMs, helping inform future iterations of the BMC. Incorporating these insights can support the development of a robust framework that better captures the current state of knowledge and factors in the complexities of these business models. Ultimately, this would improve the real-world applicability of frameworks like BMC for PSS and CBMs.
This study is a bibliometric-qualitative review based solely on the analysis of published literature. It does not involve any primary data collection, experiments or the participation of human or animal subjects. Hence, no ethics approval was required to conduct this research. The authors also acknowledge the support provided by the D. E. Napier Scholarship, which contributed to the publication of this journal article.
