Blockchain in Jordanian construction: an exploratory empirical study of stakeholder priorities and bootstrapped ROI

Highlights

• Pilot-ready blockchain for automating progress payments in public projects
• Evidence-based ROI: positive 5-year NPV under conservative assumptions
• Prioritizes payments and change-orders for high-impact, low-BIM contexts
• Policy actions: regulatory sandboxes, capacity building, modular pilots

1 Introduction

The construction industry is widely characterized by fragmentation, adversarial contracting, and slow digital transformation—challenges that are especially acute in emerging economies (Li et al., 2021; Gartoumi, 2024). In Jordan, public building projects commonly experience cost overruns and schedule delays driven by poor design coordination, delayed payments, change-order disputes, and fragmented supply chains (Matarneh and Hamed, 2017).

These public buildings—such as schools, healthcare facilities, and administrative offices—form key components of the urban fabric in Jordanian cities and are central to delivering sustainable urban services.

Limited uptake of Building Information Modeling (BIM) and weak digital infrastructure further constrain productivity, transparency, and data integration across project participants (Gartoumi, 2024).

Blockchain technology has been proposed as a core enabler of construction digitalization. Its decentralized, immutable ledger and smart-contract capabilities can support secure, automated processes for progress payments, change-order approval, asset tracking, and supply-chain information-sharing (Li et al., 2021; Wang et al., 2020; Dong et al., 2024; Liu et al., 2024a; Kim and Kim, 2024). In this study, the proposed pilots are framed as smart-contract-enabled workflows for progress-payment approvals and change-order processing in public building projects. Recent studies demonstrate promising prototypes and conceptual frameworks for blockchain in areas such as supply-chain traceability, BIM integration, and procurement. At the same time, implementation risks in public projects—particularly legal enforceability of smart contracts, interoperability with legacy systems, privacy/access control, and governance of permissioned networks—remain significant and motivate the phased approach proposed here. However, several important gaps remain, particularly for emerging-market contexts, as summarized below.

Recent work has established blockchain’s potential across multiple construction domains, including supply-chain traceability and information sharing (Wang et al., 2020; Kim et al., 2023; Ding et al., 2024; Heydari and Shojaei, 2025), BIM–blockchain integration for asset tracking and production management (Dong et al., 2024; Liu et al., 2024b), procurement and project-delivery processes (Kim and Kim, 2024), and sustainability and safety applications (Aurangzeb and Yoon, 2025; Nyato et al., 2024). Nevertheless, recent reviews (Li et al., 2021; Gartoumi, 2024; Heydari and Shojaei, 2025) highlight several persistent gaps. First, the evidence base is dominated by conceptual frameworks, technical prototypes, and case studies in high-income, digitally mature markets. There is a clear scarcity of empirical studies from emerging economies with low BIM adoption and weaker digital infrastructure—such as Jordan—particularly for public-sector building projects.

Second, although stakeholder perceptions of blockchain are increasingly discussed, there is limited quantitative modeling of how role, experience, and digital maturity jointly shape the prioritization of specific, operational blockchain use cases (e.g., progress payments, change-order management, safety reporting). Third, despite frequent claims about “efficiency gains” and “improved ROI”, few studies provide transparent, scenario-based NPV/IRR analyses that are explicitly linked to locally grounded adoption priorities and parameterized by stakeholder data.

Jordanian public building projects, already affected by cost overruns, payment delays, adversarial contracting, and low digital maturity (Matarneh and Hamed, 2017; Gartoumi, 2024), offer a pertinent context to address these gaps. Systematic empirical evidence on which blockchain applications stakeholders would prioritize, how organizational and technological characteristics influence these choices, and how financially attractive a focused pilot might be under realistic assumptions is currently lacking.

Beyond blockchain and digitalization, a growing body of work examines how construction waste and carbon reduction can be jointly optimized through advanced modelling and governance frameworks, for example, in city-scale demolition-waste supply chains and regional efficiency studies (Liu et al., 2024a; Wang et al., 2025; Wang et al., 2026). Our study complements this line of research by focusing on blockchain-enabled payment governance in public building projects, rather than on physical waste and carbon-flow optimization.

This paper proposes and empirically illustrates an evidence-informed, three-phase framework for blockchain implementation tailored to stakeholders involved in Jordanian public building projects, with the aim of improving transparency and efficiency in urban public-works payment governance. To our knowledge, it is among the first empirical studies linking stakeholders’ stated blockchain priorities with bootstrapped ROI estimates for a progress-payment pilot in an emerging-economy construction context. Unlike most prior studies that are primarily conceptual, prototype-based, or review-oriented, this study combines stakeholder-based prioritization with an explicit NPV/IRR illustration to support pilot design. The framework is grounded in Innovation Diffusion Theory (IDT) (Rogers, 2003) and the Technology–Organization–Environment (TOE) framework (Tornatzky and Fleischer, 1990). IDT is used to interpret perceived innovation attributes. TOE complements it by organizing adoption conditions into technological, organizational, and environmental dimensions relevant to public procurement and project delivery. A purposive survey of 123 contractors, consultants, and academics was used to (i) identify and rank priority blockchain use cases, (ii) examine how stakeholder characteristics shape these priorities, and (iii) parameterize a realistic pilot-scale cost–benefit model. Analyses include descriptive statistics, chi-square tests with robustness checks, a regularized binary logistic regression with bootstrap inference, and Monte Carlo bootstrapped NPV/IRR simulations.

Against this background, the study makes three main contributions to the blockchain-in-construction literature. Table 1 summarizes how this study differs from prior blockchain-in-construction studies, particularly in linking empirical stakeholder modelling with scenario-based NPV/IRR evaluation in an emerging-economy public-buildings context.

TABLE 1. Prior blockchain-in-construction studies vs. this study.

Context-specific empirical evidence from an emerging economy. The study provides one of the first empirical mappings of blockchain use-case priorities among contractors, consultants, and academics involved in Jordanian public building projects. By focusing on a low-digital-maturity, emerging-economy context, it addresses the geographic and contextual bias noted in recent reviews (Gartoumi, 2024; Heydari and Shojaei, 2025), which are largely dominated by examples from high-income countries.

1.1 Quantitative modeling of adoption priorities using IDT and TOE

Grounded in IDT and the TOE framework, the study employs chi-squared tests and (regularized) binary logistic regression analysis to quantify the relationship between role, years of experience and usage of BIM tools with prioritization of particular use cases of blockchain by stakeholders with a focus on progress payment workflows. In doing so it operationalizes IDT and TOE as much more than conceptual lenses as organizing structures for survey design and empirical analysis.

Integrated perception - ROI perspective to pilot implementation. The empirical patterns of such preferences presented in this study are linked to a Monte Carlo bootstrapped NPV/IRR analysis of a representative progress payment blockchain pilot project of public buildings in Jordan. Unlike most prior work, which is largely conceptual or presents technical pilots with no explicit economic evaluation, this study provides an open illustration of ROI which occurs on a scenario basis and is parameterized by input across the diverse stakeholders, as well as similarly derived cost benefit ranges from the financial literature. While explicitly termed exploratory, and non-representational, such a form of the integrated perception ROIs offers an order of magnitude benchmark for decision makers in settings of low digital maturity.

These contributions together support the development and evaluation of the three-phase blockchain implementation framework presented in the remainder of the paper.

2 Background and context

2.1 Global digital transformation and Blockchain’s potential

Construction is undergoing rapid digital transformation, driven by technologies such as BIM, IoT, cloud platforms, and advanced analytics. However, most project information is still managed through fragmented, centralized systems that create inefficiencies, data silos, and coordination problems (Li et al., 2021; Ding et al., 2024). The contrast between centralized and decentralized architectures commonly used for construction data management is illustrated in Figure 1.

Blockchain’s decentralized and tamper resistant ledger and smart contract capabilities make it a potential alternative architecture for a secure, transparent and auditable set of transactions throughout the project lifecycle.

2.2 The Jordanian construction landscape

Jordan’s construction sector exhibits many of the global industry’s structural weaknesses—fragmentation, adversarial contracting, and siloed data—amplified by low digital maturity. BIM adoption has been estimated at around 15% by 2025 (Gartoumi, 2024), based on recent review evidence on BIM/digital maturity in the region and public projects frequently suffer from cost overruns (≈11.4% on average), schedule delays, and incomplete or inconsistent information (Matarneh and Hamed, 2017). Limited digital skills and low stakeholder awareness further hinder collaboration and the uptake of advanced technologies (Ye et al., 2026). Figure 2 provides an overview of the main characteristics and challenges of the Jordanian construction landscape.

All these characteristics make Jordanian public building projects a relevant test-bed to explore if blockchain can address the payment delays, dispute-prone change management and transparency deficits in an emerging economy context.

2.3 Principles of blockchain and applications in construction

Key blockchain features—decentralization, immutability, transparency, and smart contracts—are directly relevant to long-standing problems in construction. Permissioned ledgers can offer shared, tamper-proof ledgers of contracts, approvals, and asset movement, and smart contracts can be used to automate payments, execute changes of orders, and validate compliance checks (Heydari and Shojaei, 2025; Kim and Kim, 2024). Applied to construction, these capabilities can help improve the integrity of data, open the door to less opportunistic behavior and support more efficient collaboration among clients, contractors, consultants and suppliers.

2.4 Theoretical foundations: IDT and TOE

This study is framed by Innovation Diffusion Theory (IDT) (Rogers, 2003) and the Technology–Organization–Environment (TOE) framework (Tornatzky and Fleischer, 1990). IDT focuses on the impact of perceived relative advantages, compatibility, complexity, trialability, and observability factors. TOE categorizes factors influencing adoption into:

Technological (e.g., perceived advantages, risks, and complexity of blockchain).

Organizational (e.g., resources, skills, existing digital practices such as BIM), and.

External factors (e.g., policy, market and client needs).

IDT and TOE together offer a framework for survey design, responding to stakeholders’ priorities, and the design of the proposed three-phase, blockchain implementation for public building projects in Jordan.

In this study, IDT and TOE are operationalized through observable stakeholder and organizational characteristics. As an exploratory study, these constructs are operationalized using practical proxy variables available in the survey (role, experience, and BIM usage), rather than full multi-item latent scales. Perceived relative advantage and compatibility are reflected in whether respondents prioritize blockchain-enabled progress-payment automation. Organizational and technological contexts are proxied by role (academic, consultant, contractor), years of experience, and BIM usage. These variables are then used in the inferential analyses to test whether organizational role, experience, and digital maturity are associated with prioritizing progress-payment workflows.

These IDT and TOE dimensions are operationalized through the variables listed in Table 2 and examined in the regression analyses described in Section 2.4.

TABLE 2. Mapping of empirical variables to IDT/TOE constructs.

2.5 Conceptual model and hypotheses

Based on the Innovation Diffusion Theory (IDT) and the Technology–Organization–Environment (TOE) framework, this paper presents a conceptual model that connects stakeholder characteristics with digital maturity to the prioritization of blockchain-related progress payment automation. According to the analysis as summarized in Figure 3, we hypothesize that the three variables: stakeholder role (organizational/external context), years of experience (organizational context) and BIM usage (technological context) jointly determine the choice of “progress-payment automation” as top use case of blockchain.

These hypotheses are consistent with prior blockchain-in-construction research that highlights role-based incentives, experience-driven sensitivity to payment frictions, and digital maturity (e.g., BIM use) as adoption-related factors (Li et al., 2021; Kim and Kim, 2024; Gartoumi, 2024).

Based on this model, we test the following hypotheses:

H1 Contractors are more likely than academics to prioritize progress-payment automation.

H2 Consultants are more likely than academics to prioritize progress-payment automation.

H3 More years of experience increase the likelihood of prioritizing progress-payment automation.

H4 Lower BIM use increases the likelihood of prioritizing progress-payment automation.

These hypotheses are then tested with chi-square association tests and with a regularized binary logistic regression model, which are summarized in Section 2.

The model correlates organizational/external characteristics like stakeholder role, organizational experience like years of experience, and technological context like BIM usage to the respondents’ probability of considering blockchain enabled progress-payment automation as their most important use case for blockchain.

3 Methodology and data collection

3.1 Research design and objectives

This exploratory study used a cross-sectional survey to inform and illustrate a three-phase blockchain implementation framework for Jordanian public building projects. Grounded in Innovation Diffusion Theory (IDT) and the Technology–Organization–Environment (TOE) framework, the study addresses three objectives: (i) identify and rank stakeholder priority blockchain use cases; (ii) examine how role, experience, and digital maturity relate to those priorities; and (iii) parameterize a pilot-scale, scenario-based cost–benefit model for a progress-payment application. Analytical methods are descriptive statistics, chi-squared test of associations, regularized binary logistic model (payments vs. others) with bootstrap inference and Monte Carlo bootstrap simulations of NPV/IRR. Full analytic code and extended outputs are provided in the project repository (Supplementary).

3.2 Sample and data collection

Descriptive statistics of the sample (roles, experience, BIM use, and key perceptions) are reported in Table 3. We implemented a purposive, non-probability sampling approach by focusing on information-rich respondents who are experts in the field, i.e., Jordanian public building projects. The final sample was composed of 123 people with professional careers: 48 of them are academics, 43 are consultants, and 32 are contractors. Data were gathered using a semi-structured online survey including organizational profile, years of experience, self-reported performance of the project (typical cost overrun, schedule delay), use of BIM, top priority blockchain use cases, perceived benefits/concerns, and adoption conditions. Participation was voluntary and the answers were anonymous. Because sampling was purposive, results are exploratory and not statistically generalizable to the entire Jordanian construction sector; details on survey instrument, recruitment, and response rates are in the Supplementary Material.

TABLE 3. Respondent profile and descriptive statistics.

This purposive approach may introduce selection bias (e.g., over-representation of respondents more interested in digitalization/blockchain), reinforcing that results support analytical inference within this expert sample rather than population estimates.

This purposive, non-probability strategy was chosen to reach information-rich respondents directly involved in Jordanian public building projects, and our analyses are therefore aimed at analytical inference within this expert sample rather than statistical generalization to the entire Jordanian construction sector.

3.3 Variables and operationalization

Table 3 summarizes respondent counts and key descriptives used below; Table 4 contains the contingency counts for role × top priority.

TABLE 4. Contingency counts and expected counts for role versus top priority (priority_process_1); used for Pearson chi-square and Cramér’s V calculation.

Dependent variables.

Top priority (categorical): respondent’s single top-ranked blockchain use case (see Table 3).

Priority_payments (binary): equals one if the respondent selected “progress payments and approvals” as top priority, 0 otherwise.

Predictors and controls.

Role: Academic (reference), Consultant, Contractor.

Years of experience: continuous (years).

BIM usage: categorical (Never, Rarely, Sometimes, Often).

Additional contextual items (used descriptively and as robustness checks): organization size, typical project type, and self-reported cost/schedule performance.

ROI scenario inputs (baseline):

One-time setup cost: JOD 120,000.

Annual OPEX: JOD 20,000.

Annual benefit: mean JOD 50,000, sd = 20% (JOD 10,000), truncated at zero.

Pilot horizon: 5 years; discount rate: 10%.

These baseline values were selected to be conservative; both the self-reported performance patterns in our sample as well as indicative savings from the literature were represented. The sample size of costs that come in 10%–20% over budget, plus delays in payments that are common in the real world, indicate a significant value loss in current processes (Table 3). Based on empirical and scenario-based analyses of the benefits of digitalization in similar contexts (e.g., Kim and Kim, 2024; Dong et al., 2024; Liu et al., 2024b; Heydari and Shojaei, 2025) it is assumed that the annual benefit (mean JOD 50,000 with 20% standard deviation) is an order of magnitude reduction of administrative effort and payment-related rework for a typical portfolio of public building projects.

Thus, the ROI simulation links stakeholder-reported pain points and literature-based benchmarks to a transparent, scenario-based economic illustration.

These ROI inputs are assumed for exploratory illustration and are anchored to (i) stakeholder-reported payment frictions and (ii) cost/benefit ranges reported in prior digitalization/blockchain literature, rather than calibrated from audited project accounts.

Stakeholder responses on payment delays, dispute frequency, and administrative effort (reported descriptively) were used to select progress payments as the pilot focus and to motivate conservative benefit ranges for the scenario inputs.

Sensitivity scenarios varied annual benefits and costs by ±20% and the discount rate between 8% and 12%; detailed results are reported in the Supplementary Material.

We report three headline cases: pessimistic (benefit −20%, costs +20%, discount 12%), baseline (as specified above), and optimistic (benefit +20%, costs −20%, discount 8%), with full distributions in the Supplementary Material. Table 5 summarizes how the main variables used in the regression analyses maps onto the IDT and TOE constructs.

3.4 Statistical methods—inferential analyses

We performed three concentrated inferential analyses to link the characteristics of the stakeholder, priority use-cases and pilot financial prospects: (A) association testing (B) Regression analysis (C) ROI simulation. Key tables/figures referred to below: Table 3 (descriptives), Table 4 (contingency), Figure 4 (priority by role), Table 5 (binary logit summary), Table 6 and Figure 5 (ROI) and Supplementary Tables S6–S8.

TABLE 5. Logistic regression results.

Percentile bootstrap confidence intervals are provided in Supplementary Tables S7, S8.

TABLE 6. Bootstrapped ROI simulation summary.

3.4.1 Association testing (role vs. top priority)

Pearson chi-square test on the contingency in Table 4; effect size measured with Cramér’s V.

Robustness checks: (i) collapsed-category chi-square retaining categories with ≥10 observations, and (ii) permutation (Monte-Carlo) chi-square with 10,000 replications. (See Figure 4 for role distributions).

3.4.2 Regression analysis (payments vs. others)

Dependent variable: priority_payments = 1 if respondent selected “progress payments and approvals” as top priority, 0 otherwise.

Predictors: years of experience (continuous), role dummies (Consultant, Contractor; Academic ref.), BIM usage categories (Never, Rarely, Sometimes, Often). Additional contextual items used for robustness.

Modeling strategy: we first estimated an exploratory multinomial logistic model (Supplementary Table S6), but due to sparse outcome categories, unstable coefficients, and occasional convergence issues we base inference on a parsimonious L2-regularized binary logistic model (payments vs. others), with the regularization parameter selected by 5-fold cross-validation.

Given the sample size (n = 123) and sparse outcome categories, we use regularization and a parsimonious binary specification to reduce overfitting risk.

Inference: bootstrap resampling (n = 2,000) with percentile confidence intervals and empirical coefficient distributions. Bootstrap confidence intervals and basic model performance (cross-validated accuracy/AUC) are reported in the Supplementary Material. Main odds ratios are summarized in Table 5; full summary of standardized coefficients, CV metadata and bootstrap summaries can be found in Supplementary Tables S7, S8.

3.4.3 ROI simulation for a progress-payment pilot

Monte Carlo simulation (n = 2,000 draws) computed NPV and IRR over a 5-year horizon using baseline inputs: setup = JOD 120,000; annual OPEX = JOD 20,000; annual benefit ∼ Normal (mean = JOD 50,000, sd = JOD 10,000) truncated at zero; discount rate = 10%.

The benefit and cost distributions used in the simulation are assumed for exploratory purposes (with sensitivity checks), rather than empirically fitted to observed pilot cash-flow data.

Outputs reported: empirical mean, median, and 95% percentile interval for NPV and IRR (Table 6; Figure 5). Sensitivity analyses (varying benefits, costs, horizon, discount) are provided in the Supplementary Material.

3.5 Software and reproducibility

All the analyses were implemented with python (pandas, numpy, scipy, scikit - learn, statsmodels, matplotlib). The complete Colab notebook along with anonymized data releases and figure files can be found in the project repository (link in Data Availability).

Monte Carlo bootstrapping (n = 2,000 draws) under the baseline assumptions yields a positive NPV distribution, consistent with the summary statistics reported in Table 6.

3.6 Software, outputs, and reproducibility

All analyses were conducted in Python using the libraries pandas, numpy, scipy, scikit-learn, statsmodels, matplotlib, and seaborn in a Google Colab environment.

The following artefacts are available in the project workspace/repository:

The full Colab notebook implementing the data cleaning, chi-square tests, multinomial and binary logistic regressions, and ROI bootstrapping.

CSV files containing intermediate and final analysis outputs (e.g., contingency tables, regression coefficients, and bootstrapped NPV/IRR draws and summaries).

High-resolution figures (PNG/SVG) corresponding to all figures cited in the manuscript (e.g., role distributions, priority distributions, ROI histograms).

These materials allow independent reproduction of the tables and figures reported in Sections 2, 3.

3.7 Limitations and ethical considerations

The research has several methodological shortcomings. First, it was a purposive sample restricted to the sample participants being contractors, consultants, and academics working in the public building sectors in Jordan; therefore, the results are speculative and cannot be interpreted as estimates of the whole population of the Jordanian construction sector. Second, important variables, including the performance of the project and the use of the BIM, were self-reported; and may be affected by recall or social-desirability bias. Third, the sample size (n = 123) was small, limiting the complexity of the models that could be estimated. An exploratory multinomial model was fitted, but quantitative inference was based on parsimonious, regularized binary logistic regressions to limit overfitting. Fourth, the analysis of ROI is scenario-based: although Monte Carlo Bootstrapping engulfs uncertainty, the results of NPV and IRR are based on presupposed parameters of cost and benefit and discount rate. From an ethical standpoint, the nature of the research conducted was purely among professional stakeholders and non-sensitive topics. Participation was voluntary; respondents were informed of the purpose of the study, assured of anonymity and no person would need to answer a question or could withdraw at any time. No personal identifiers were recorded and all data were stored and analyzed in aggregated and anonymized form.

4 Results

4.1 Sample descriptives

Table 3 reports sample characteristics (n = 123: 48 academics, 43 consultants, 32 contractors). Self-reported typical cost overruns clustered at 10%–20%; BIM usage was generally low (see Table 3; Supplementary Figures S1–S3).

4.2 Priority use cases by role

Counts in Table 4 and the role-by-priority bar chart (Figure 4) show strong cross-role concentration on “progress payments and approvals” and “change-order/variation management” as top priorities.

4.3 Association test (role vs. top priority)

Pearson χ² on the contingency in Table 4: χ² = 6.34, df = 8, p = 0.61; Cramér’s V = 0.16. Robustness checks (collapsed categories and permutation χ², 10,000 reps) produced similar p-values (collapsed χ² = 0.93, df = 2, p = 0.63; permutation p ≈ 0.60). Conclusion: within this purposive sample, role is not strongly associated with selected top priority. The subsequent regression analysis therefore examines role jointly with experience and BIM usage, and its findings should be read as exploratory conditional patterns rather than as contradicting the non-significant chi-square test.

While the chi-square test suggests no strong bivariate association between role and top priority, the subsequent regression considers role jointly with experience and BIM usage, which can reveal conditional patterns within this exploratory sample.

4.4 Binary logistic regression (payments vs. others)

The binary outcome variable was defined as priority_payments = 1 if the respondent selected “progress payments and approvals” as top priority, and 0 otherwise. Predictors included years of experience (continuous), role dummies (Consultant, Contractor; Academic as reference), and BIM usage categories (Never, Rarely, Sometimes, Often). The Supplementary Material reports on additional contextual variables that were investigated during robustness checks.

Given the sample size and sparse outcome categories, we estimated an L2-regularized binary logistic regression model to minimize overfitting. The regularization parameter was selected by 5-fold cross-validation. Non-parametric bootstrap resampling was used to evaluate coefficient uncertainty (n = 2,000 draws). Table 5 summarizes the main coefficient estimates and odds ratios; Supplementary Tables S6–S8 contain complete standardized coefficients, cross-validation metadata, and bootstrap summaries. Percentile bootstrap 95% confidence intervals and approximate p-values for the odds ratios are reported in Supplementary Tables S7–S8.

The exploratory multinomial results and diagnostics are reported in Supplementary Table S6 for completeness.

Key odds-ratio patterns are:

Contractor (vs. Academic): OR = 2.24, showing that contractors are far more likely than academics to prioritize progress payments as the most important blockchain use case.

Years of experience: OR = 1.60, suggesting that more experienced respondents are more likely to prioritize progress payments, However, considering the sample size and design, this estimate should be regarded with caution.

BIM usage–Never (vs. reference category): OR = 1.97, showing that, in comparison to the reference BIM-use group, respondents who report not using BIM are more likely to choose progress payments as their top priority.

Within this exploratory sample, the model suggests that contractors, more experienced respondents, and those reporting lower BIM use are more likely to prefer payment automation, although these patterns should not be interpreted as definitive sector-wide effects. Patterns are exploratory correlational signals, and not definitive causal effects, because of the small sample size and purposive nature of the study.

4.5 ROI simulation (progress-payment pilot)

Monte Carlo simulation (n = 2,000; baseline inputs: set-up cost JOD 120,000; annual OPEX JOD 20,000; annual benefit mean JOD 50,000, sd JOD 10,000; 5-year horizon; discount rate 10%) produced a mean NPV of ≈ JOD 50,143 with a 95% percentile interval of ≈ [JOD 30,717, JOD 69,868], and a mean/median IRR of ≈10% (Table 6; Figure 5).

Sensitivity analyses varied annual benefits and costs by ±20% and the discount rate between 8% and 12%. Scenarios with reduced annual benefits and/or higher costs unsurprisingly lowered NPV and IRR, while scenarios with increased benefits and/or lower costs raised them; changes to the discount rate primarily shifted NPV, with a higher rate compressing NPV and a lower rate expanding it. Across these scenarios, NPV was consistently more sensitive to assumptions about annual benefits and the effective pilot horizon than to moderate variations in costs or discount rates. A simple one-at-a-time sensitivity ranking (tornado-style) confirms that annual benefit is the dominant driver of NPV, followed by pilot horizon, then setup/OPEX, and finally the discount rate (Supplementary). In the pessimistic combinations of lower annual benefits, higher costs, and a higher discount rate, simulated NPVs approach zero or become modestly negative, indicating that under unfavorable conditions the indicative gains may be largely eroded.

The mean IRR being close to the assumed discount rate of 10% is consistent with the conservative benefit assumptions used; the positive NPV reflects the simulated benefit distribution over the 5-year horizon rather than an aggressive performance forecast, and IRR percentile bounds are reported in the Supplementary Material.

Because the IRR is so close to the discount rate, these ROI indications should be viewed as marginal and illustrative rather than as evidence of strong financial outperformance.

4.6 Summary statement

Stakeholders converge on payments and change-order management as leading first blockchain use cases. The binary regression finds contractor role and lesser levels of BIM use to be robust correlates to prioritizing payments; experience has a positive but less precise effect. The scenario-based ROI illustration shows a positive expected NPV under the basic assumptions. Comprehensive tables, diagnostics, and full simulation results are presented in the Supplementary Materials.

5 Discussion: benefits and strategic implications for Jordan

5.1 Interpreting stakeholder priorities through IDT and TOE

Across roles, respondents prioritized progress-payment automation and change-order management as the leading blockchain use cases, with transparency as the main expected benefit and low awareness as the dominant concern. Interpreted through the lenses of Innovation Diffusion Theory (IDT) and the Technology–Organization–Environment (TOE) framework, these patterns are consistent:

From an IDT perspective the relative advantages and compatibility with the existing workflow for automating payments and managing variation will be high, which is part of the reason why they have mass appeal. At the same time, respondents perceive blockchain as complex, and trialability and observability are low. There is limited understanding of distributed ledgers and smart contracts and their legal and cybersecurity implications.

Crossing the TOE lens, projected improvements in transparency and dispute reduction are mainly technology solutions when skill gaps, upfront investment and integration with legacy systems are constituent organizational barriers. Regulatory ambiguity and absence of explicit recognition of smart contracts can be identified as environmental constraints. Together, the views of the IDT and TOE can identify the need to better adopt in Jordan not only technically sound solutions, but interventions that address perceptions of complexity, visibility of benefits, institutional and regulatory uncertainties.

5.2 Quantifiable performance signals and ROI

The statistical analyses and ROI simulations provide initial, order-of-magnitude signals on where blockchain could create value in Jordanian public building projects.

The chi-square results suggest that, in this purposive sample, role is not particularly linked to the first choice of blockchain use case considered, supporting the hypothesis that applications in relation to payment are widely perceived as a natural starting point. The binary logistic regression further suggests that the more experienced respondents and contractors are particularly likely to prioritize progress payments and that lower BIM use is related to a stronger preference for blockchain in this area. These patterns are in line with the idea that the propensity to automate payment workflows increases when stakeholders are more exposed to delays in payments and greater risk of cash-flow problems.

The Monte Carlo bootstrapped NPV/IRR analysis using conservative literature informed assumptions show positive expected NPV for a 5-year progress payment pilot. However, these indicative ROI signals do not remove the need to address implementation risks such as integration complexity, governance of permissioned networks, and the legal treatment of smart-contract-based payment records. Although scenario based and not an exact forecast, this supports the view - also echoed in recent case studies from elsewhere in the world (Dong et al., 2024; Liu et al., 2024a; Kim and Kim, 2024) - that carefully designed blockchain pilots could be financially attractive if they are implemented in high friction processes like payments within the public sector. The sensitivity analyses show that benefits, project duration, and cost assumptions are significant levers, emphasizing the need for rigorous scoping and phased deployment.

The ROI technique is based on problems reported by stakeholders (including cost overruns and payment delay) and efficiency levels found in the literature rather than on ad hoc inputs. The simulation emphasizes progress-payment operations in accordance with the priorities of the survey stakeholders. Assumed benefits informed by inefficiencies identified by respondents and by previous studies.

5.3 Strategic implications for Jordan

The results indicate a number of strategic measures for the Jordanian construction industry and decision makers. Priority should be given to high impact and high friction processes. Progress-payment automation and change-order management are natural pilot candidates. They address common pain points: delayed payments, disputes, and mistrust.

Pilots must be user-friendly. Interfaces need to be intuitive and demand minimum navigation. Responsibility should be assigned clearly so that tasks are traceable. Although handoffs need to be explicit, they should be simple. Workflows need to be written in simple language with step-by-step checklists. Quick-start guides and in-app tooltips should be available so that onboarding time is reduced.

Measurable results need to be monitored. Metrics such as reduction in the payment delay and reduction in the number of disputes should be published on dashboards and in periodical reports. A subset of sites should be managed intensively and the performance data collected and analyzed. Iterations should be planned according to the results observed and defined success criteria should be prescribed in advance. Contingency plans need to be on hand.

Integration with the existing systems should be incremental. Modular architecture and middleware are recommended in order to have partial integration without complete replacement. Blockchain would need to be compatible with existing procurement processes (e.g., DBB, DB, PPP) to minimize legal and institutional friction.

Supportive policy interventions are required. Regulatory sandboxes and model provisions for contracts in the smart contracts should be piloted. Guidance on how far blockchain records can be relied on will need to be issued. Collaboration between government, industry and academia should be promoted in order to align technical, organizational and regulatory measures.

Taken together, these steps will help to increase observability, reduce perceived complexity, and increase speed of diffusion - provided pilots are tightly scoped, measurable and iterated rapidly.

5.4 Limitations and implications for interpretation

The preceding discussion and implications should be interpreted within the context of a number of limitations. The survey employed a non-probability, purposive sampling method, targeting contractors, consultants, and academics involved in Jordanian public building projects, with a limited sample size (n = 123). As a result, the findings support analytical inference about patterns within this expert sample but cannot be treated as statistically generalizable estimates for the entire Jordanian construction sector. Critical variables, including project performance and BIM usage were self-reported, and this could introduce recall and social-desirability biases. The regression models were succinctly specified and regularized to mitigate overfitting, although inconsistencies persist.

In addition, the sample contains a relatively high proportion of academics (48 of 123 respondents), which may over-represent more conceptual perspectives and under-represent some site-level operational constraints compared with a contractor-dominated sample.

The ROI analysis is explicitly scenario-based: while the analysis (i.e., bootstrapping and sensitivity analysis) takes some account of the measurement of uncertainty, outcomes (namely, NPV and IRR) are still dependent on the cost, benefit and discount-rate parameters that we use and the pilot horizon (5 years). As such, the economic results must be taken as simple benchmarks for informing the development of pilot programs and policy debate, rather than as accurate investment projections.

Within the above limitations, the research provides exploratory but actionable evidence/proof of concept about where and how blockchain should be applied for the public building industry in Jordan and offers a holistic perception-ROI value perspective that is not commonly practiced in the overall blockchain-in-construction literature.

6 Three-phase blockchain implementation framework for Jordanian public building projects

This section presents the combined perception–ROI technique in a three-phase approach to introduce blockchain to Jordanian public construction projects. This framework is distinguished from generic technical models by its direct foundation in locally collected stakeholder priorities and its integration of a bootstrapped financial feasibility analysis, providing a roadmap specifically tailored for low-BIM, emerging-economy public procurement contexts. The plan is based on the results of the stakeholder survey and on a Monte Carlo bootstrapped NPV/IRR analysis of typical public building schemes. It is built to fit into existing regulations, to be respectful to low but improving levels of digital competency, and to begin where the perceived value is greatest: change-order management and progress payment automation.

The first phase is a preparatory phase. Existing payment and change order processes for public projects are documented in a straightforward way: who is involved, how long each process takes, where conflicts typically occur, and how information is shared between organizations. This mapping is integrated with a light-touch assessment of digital maturity, including BIM and ERP usage, data quality and key integration gaps.

Short surveys, focused workshops, and document reviews are then used to check and refine stakeholder priorities so that progress-payment and change-order use cases are confirmed rather than assumed. Based on this, clients, contractors and consultants collaborate to draw candidate blockchain workflows that are open, with clear responsibilities, and usable to non-experts.

Meanwhile, a small number of indicators (payment lead times, frequency and duration, administrative effort) are agreed and simple economic parameters are set and compared with the cost–benefit ranges reported in the previous studies. The parameters are then fed into the scenario-based ROI modelling performed in this study.

The second part is dedicated to real projects instead of design. A permissioned blockchain network is implemented on a select few public construction projects to manage progress-payment approvals and change-order processing. Integration is kept modest and stepwise. To maintain legacy systems, middleware links the blockchain layer to existing tendering portals, document management systems, and fundamental BIM. To minimize perceived complexity, project teams are supported by targeted training sessions, written instructions and accessible help desk channels.

Throughout the pilots processes, performance and cost data are collected systematically. Payment processing times, the number and duration of disputes, and administrative effort are tracked alongside implementation and integration costs. These observed cost and benefit streams are then compared with the earlier scenario-based NPV/IRR estimates, and the model assumptions are updated where the evidence shows that expectations have not been met.

Scale-up and institutionalization comprise the third phase. The configurations that show reasonable performance in the prototypes are then further developed and then used in further public projects and–if applicable–in additional asset classes and contract types (DBB, DB, PPP).

The integration with BIM, ERP, and other digital tools is gradually enhanced through the use of API-based middleware and interoperability standards (such as openBIM) to facilitate the progression of change without the need for a significant system replacement. Parallel to this, policy instruments that are supportive are being created, including regulatory sandboxes, model smart-contract clauses for payments and change-order approvals, and explicit guidance on the evidential status of blockchain records. These are developed in collaboration with regulators and procurement authorities. Continuous monitoring and evaluation arrangements are employed to modify both technical solutions and policy settings, as they monitor performance, ROI, and user experience over time.

The three-phase can be used together to form a pathway for applying the perception–ROI technique. Phase 1 involves identifying and prioritizing high impact use cases, with stakeholder information and exploration ROI modelling. Phase 2 includes bounded pilots and provides local economic and operational evidence. In Phase 3, validated arrangements are then scaled-up and integrated into the regulatory and organizational processes of public construction in Jordan.

A schematic overview of the proposed three-phase implementation pathway is shown in Figure 6.

7 Challenges and mitigation strategies

This section summarizes the key barriers to blockchain adoption that were found in the survey and literature, and provides brief, viable mitigation strategies.

7.1 Skills and technological literacy

Challenge: The absence of technical expertise and poor levels of digital literacy among practitioners and project teams make them less prepared to integrate blockchain (BIM/IoT/AI).

Mitigation: Develop blockchain-specific user-centered front-ends with clear visualizations of smart-contract states and transactions and middleware that conceal ledger complexity from end users. Complement this with university-industry workshops focused on smart contract logic and targeted capacity building (short courses, practical pilot training on private key management and gas fee estimation).

7.2 Upfront investment and uncertain ROI

Challenge: uncertain returns on investment and perceived high setup and integration costs.

Mitigation: Time-boxed, co-funded pilots with explicit performance indicators; stepwise deployments emphasized; cost-sharing or public R&D funds; and transparent ROI frameworks based on scenarios to inform choices.

7.3 Privacy, security, and interoperability

Challenge: Cybersecurity risks, granular access control, node governance models, and ambiguous data ownership are critical multi-part data-sharing challenges.

Mitigation: Implement role-based access controls (RBAC) for node permissions, define clear data-ownership protocols in smart contracts, use modern encryption (proxy re-encryption, ZKPs if necessary), channel/ledger partitioning, permissioned blockchains, openBIM data declaration standards, and API-first middleware for connecting legacy systems. Additionally, establish a governance framework for consortium node voting and dispute resolution.

7.4 Regulatory and contractual uncertainty

Challenge: The legal standing of smart contracts, procurement regulations, and how blockchain data are treated as evidence are all unclear.

Specifically, smart contracts face questions on their enforceability as binding contracts, liability for code errors, and conflict with traditional ‘signature’ requirements.

Mitigation: Regulatory sandboxes in public procurement - develop pilot procurement models with legal experts; embed the use of model contract clauses, defensive evidential guidelines; host inception discussions with procurement authorities.

7.5 Organizational inertia and integration complexity

Challenge: Resistance to change, siloed workflows, and costly integration with ERP/BIM/document management impede adoption.

Mitigation: Co-design workflows with end users, prioritize high impact processes (payments/change orders), using modular APIs/middleware for incremental integrations, and demonstrate early wins to help build organizational buy-in. A summary of these challenges, their primary barrier types, and linked mitigation strategies is provided in Table 7.

TABLE 7. Summary of blockchain adoption challenges, their primary barrier types, and corresponding mitigation strategies for Jordanian public building projects.

Concluding note: Improvement in these five areas-together, technically, financially, legally, organizationally, and skills, with a range of coordinated pilot projects, public-private cooperation will enhance the chances of a successful and scalable blockchain adoption in Jordanian public building projects by a significant margin.

8 Conclusion

This paper proposes an evidence-informed, three-phase framework for blockchain implementation in Jordanian public building projects and provides exploratory empirical evidence to inform pilot design. Using a purposive survey (n = 123) of academics, consultants, and contractors, we identify stakeholder priorities (cross-role emphasis on progress-payment automation and change-order management), quantify associations between respondent characteristics and payment prioritization via a regularized binary logistic model, and illustrate pilot economics with Monte Carlo bootstrapped NPV/IRR simulations.

8.1 Key takeaways

Stakeholder priorities: Progress payments and change-order management are the most frequently prioritized blockchain use cases; transparency is the dominant anticipated benefit and low awareness the biggest barrier.

Correlations of prioritization: Contractor role and lower BIM use are robustly associated with selecting payments as top priority; experience shows a positive but less precise association.

Pilot economics - As an exploratory illustration, under a set of conservative, literature-informed assumptions, the 5-year progress-payment pilot shows a positive expected NPV and a favorable IRR distribution in the baseline scenario and under lower discount-rate assumptions. These results should be interpreted as indicative benchmarks rather than guaranteed outcomes.

8.2 Practical implications

To move toward practical implementation, the most promising short-term strategies for Jordan include carefully scoped and user-centered pilot design, high friction process focus (payments, change orders), integration with existing BIM/workflow tools through modular middleware and combination of technical deployment with capacity building and regulatory sandboxes are the most promising short-term strategies for Jordan.

Limitations and final note These findings are from an exploratory study, based on a purposive, non-probability-based sample and scenario-based ROI assumptions, findings are indicative and not generalizable. The contribution here is to combine empirical evidence from stakeholder preference and a transparent illustration of the evolution in terms of ROI to help inform the discussion on piloting activities and form policies on piloting in low-digital-maturity contexts. Future work advisable to enhance the external validity with larger, representative surveys, and with project-level pilot evaluations.

By focusing on transparent, automated payment workflows for public building projects—which our stakeholders prioritized as a key solution to delays and disputes—the proposed framework offers… and contributes to SDG 11 (Sustainable cities and communities) by enhancing the accountability and efficiency of urban infrastructure delivery.

9 Future research directions

Building on this exploratory study, we suggest four priority areas for future research.

9.1 In-depth longitudinal pilot evaluations

Conduct project level, longitudinal assessments of blockchain pilots to measure impacts on payment delays, rework, disputes, time, cost, and quality. Report empirical cost benefit and performance metrics to serve as validation to scenario-based ROI estimates.

9.2 Representative and comparative surveys

Implement larger, probability-based surveys across regions and procurement models (DBB, DB, PPP) to test the generalizability of stakeholder priorities and to compare adoption drivers across contract types and market segments.

9.3 Case studies of technology fusion and integration

Develop detailed deployments and implementations of blockchain and BIM systems with integration into IoT, ERP, and AI systems, through interoperability standards (such as openBIM) and patterns that support incremental adoption.

9.4 Socio-technical and regulatory studies

Discuss institutional, legal validation, cybersecurity governance, interoperability, and cultural factors that can affect adoption (e.g., enforceability of smart contracts, procurement rules, professional norms). Examine regulatory sandboxes and provide examples of clauses in contracts, pair up qualitative interview research and analysis with policy analysis.

For these strands, highlight reproducible approaches (open data/code), mixed approaches (quantitative + qualitative), industry and public client engagement for actionable, generalizable evidence for scaled adoption.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fbloc.2026.1832998/full#supplementary-material

Supplementary material Supplementary material at the source

Footnotes

Edited by: Jingkuang Liu, Guangzhou University, China

Reviewed by: Zhenshuang Wang, Dongbei University of Finance and Economics, China

Preenithi Aksorn, Khon Kaen University, Thailand

Arpit Solanki, Pandit Deendayal Energy University, India

Footnotes

BIM, Building Information Modeling; DB, Design–Build; DBB, Design–Bid–Build; ERP, Enterprise Resource Planning; IDT, Innovation Diffusion Theory; IRR, Internal Rate of Return; IoT, Internet of Things; NPV, Net Present Value; PPP, Public–Private Partnership; ROI, Return on Investment; SDG, Sustainable Development Goal; TOE, Technology–Organization–Environment (framework).