Blockchain-Enabled Governance Mechanisms in Global Supply Networks: A Hybrid Machine Learning and Multiobjective Optimization Framework

Blockchain-Enabled Governance Mechanisms in Global Supply Networks: A Hybrid Machine Learning and Multiobjective Optimization Framework | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Blockchain-Enabled Governance Mechanisms in Global Supply Networks: A Hybrid Machine Learning and Multiobjective Optimization Framework SVB Subrahmanyeswara Rao, Kanaka Durga Hanumanthu, P Siva Reddy, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9326002/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract It is becoming more difficult for global supply chain networks to be governed because of issues such as a lack of transparency, data fragmentation, regulatory divergence, and the likelihood of having more than one supplier. This paper introduces an innovative Blockchain-Enabled Governance Framework (BEGF) that amalgamates distributed ledger technology with an ensemble machine learning (ML) risk-scoring engine and a multiobjective linear programming (LP) optimizer to enable supply chain decisions that are transparent, real-time, and verifiable. We use the framework in five long-term industrial case studies: automotive (Toyota), pharmaceutical (Merck), electronics (Samsung), apparel (Zara), and food and drink. These studies include 1,840 supplier nodes in 37 countries over 36 months. The BEGF can predict disruptions with an average AUC of 0.947. It can also reduce the number of supply disruptions by 38.5% and save each business $13.8 million a year. The Pareto-optimal optimizer lowers costs all along the supply chain. Management Entrepreneurship Blockchain Supply chain governance Machine learning Multiobjective optimization Smart contracts Risk management Distributed ledger Industry 4.0 Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 1. Introduction Modern global supply chains are highly interconnected systems that cross many borders and involve thousands of upstream and downstream actors who work in different regulatory, cultural, and technological settings. The COVID-19 pandemic (2020–2022), the Suez Canal blockage (2021), and the Russia–Ukraine conflict (2022) illustrate how weak centralized supply chain models are. The World Bank estimates that these events cost the global economy $ 4 trillion in lost trade (World Bank, 2023). There are many well-known structural root causes of these problems, such as gaps in information between supply chain tiers, a lack of real-time visibility, a lack of trust among unverified suppliers, and a lack of automated governance systems that can respond at machine speed. Blockchain technology, which was first thought of as the basic infrastructure for Bitcoin (Nakamoto, 2008), has become a revolutionary framework for supply chain governance by providing an unchangeable, decentralized, and cryptographically protected ledger that only authorized users can access. Additionally, advances in ensemble machine learning and combinatorial optimization have made it possible to perform predictive risk analytics and resource allocation as never before. The literature considers blockchain, ML, and optimization as largely distinct research domains, leading to a notable integrative shortcoming. This paper addresses that deficiency through the following principal contributions: (i) a formally defined Blockchain-Enabled Governance Framework (BEGF) comprising four coordinated architectural layers; (ii) an integrated XGBoost–Random Forest machine learning pipeline that produces a composite XGBoost-ML risk score (BMRS) from 47 on-chain and off-chain features; (iii) a biobjective linear programming–genetic algorithm hybrid optimizer that traverses cost–risk Pareto frontiers in near-real-time; (iv) five cross-industry longitudinal case studies based on actual operational datasets; and (v) a prescriptive decision-support matrix for managerial implementation. The rest of this paper is organized as follows. Section 2 reviews the important literature. Section 3 builds on the theoretical framework. The hybrid methodology is described in Section 4 . Section 5 provides more information about the case studies and datasets. The results are presented in Section 6 . Section 7 discusses what this means for managers. Section 8 ends with limitations and suggestions for the future. 2. Literature Review Many studies have investigated how to use blockchain to make global supply chains more open, traceable, and reliable. Hyperledger Fabric and Ethereum are two examples of platforms that have been used for this purpose. Past research has shown that there are problems with governance, such as getting data to work together, scaling up, and getting stakeholders at different levels to work together. Recent research has used machine learning for predictive analytics and anomaly detection. Multiobjective optimization strategies address the trade-offs between cost, efficiency, and sustainability. Nevertheless, there is not enough of a framework that combines blockchain governance with hybrid machine learning and optimization for making decisions in global supply chains that change all the time. 2.1 Supply Chain Governance Theory Supply chain governance has evolved from dyadic contract theory (Williamson, 1975) to a multiactor relational framework emphasizing information sharing, collaborative planning, and trust (Dyer & Singh, 1998). In the past, governance mechanisms were divided into two groups: formal (such as contracts, SLAs, and audits) and informal (such as relational norms and reputation). ERP, EDI, and IoT have made supply chains more digital, which has made the information better. However, they have not fixed the main problems of trust and verifiability (Kshetri, 2018). Blockchain makes a new way to run things called algorithmic governance. Rules are put into smart contracts that automatically and fairly enforce them in this system (Lumineau et al., 2021). 2.2 Blockchain in Supply Chain Management Since 2016, many researchers have been interested in using blockchain in supply chains. The most common uses are tracking where goods come from, stopping counterfeiting, automating payments, and checking for sustainability (Saberi et al., 2019; Ivanov et al., 2019). Hyperledger Fabric is now the standard enterprise platform because it has a permissioned architecture, pluggable consensus, and channel-based data isolation (Androulaki et al., 2018). Walmart (food traceability), Maersk (shipping documentation), and De Beers (diamond provenance) all show real improvements in operations, but there is still not enough statistical analysis across industries (Choi et al., 2020; Francisco & Swanson, 2018). 2.3 Machine Learning for Supply Chain Risk Predictive analytics for supply chain risk have progressed from econometric lead–lag models to sophisticated ensemble learners. When many dimensions are present in a dataset and supply disruption is unbalanced, random forest (Breiman, 2001) and XGBoost (Chen & Guestrin, 2016) always perform better than single classifiers do. Deep learning (LSTM) has demonstrated efficacy in modelling temporal sequences of demand signals and logistics delays (Carbonneau et al., 2008). The integration of blockchain-based on-chain transaction attributes as inputs for machine learning represents a predominantly unexplored research area that this study directly addresses. 2.4 Optimization in Supply Chain Decisions Multiobjective supply chain optimization is a well-established concept in operations research, with applications in inventory management, network design, and routing (Mula et al., 2010). Hybrid metaheuristics that combine LP relaxation with genetic algorithms (GA), particle swarm optimization (PSO), or simulated annealing (SA) are becoming more popular for large NP-hard problems where pure LP cannot be used because it takes too long (Govindan et al., 2017). Blockchain gives you verified, real-time data streams that let you find new ways to optimize. This allows you to dynamically reoptimize data that used to be available only in batches. 2.5 Research Gap and Positioning Table 1 maps the literature against the three axes of our study: blockchain governance, ML risk analytics, and multiobjective optimization. No prior work simultaneously integrates all three axes within a validated empirical multicase framework, confirming the originality of the present contribution. Table 1 Positioning of the present study in the literature landscape Study/Stream Blockchain Governance ML Risk Analytics Multi-Obj. Optim. Multi-Industry Empirical Saberi et al. (2019) ✓ – – Partial Ivanov et al. (2019) ✓ ✓ – – Kshetri (2018) ✓ – – Conceptual Lumineau et al. (2021) ✓ – – – Govindan et al. (2017) – – ✓ Simulation Choi et al. (2020) ✓ Partial – Single case Present Study (BEGF) ✓✓ ✓✓ ✓✓ Five industries 3. Theoretical Framework The theoretical framework integrates transaction cost economics, the resource-based view, and information systems governance to conceptualize blockchain as a means to reduce coordination costs and enhance organizational capability. Hyperledger Fabric allows everyone in a supply chain to run their own business in a way that cannot be changed. A hybrid machine learning layer helps make decisions on the basis of predictions, and multiobjective optimization yields the best balance between cost, resilience, and sustainability. This integrated framework demonstrates how blockchain-based governance enhances the openness, efficiency, and alignment of global supply networks with business objectives. 3.1 Blockchain-Enabled Governance Framework We examine BEGF as a four-layered digital governance architecture (Fig. 1 ) through the lenses of Transaction Cost Economics (Williamson, 1975), the resource-based view (Barney, 1991), and information systems governance theory (Weill & Ross, 2004). Each layer can travel to and from the layers next to it through clearly defined APIs. This makes it modular, scalable, and technology-agnostic. Layer 0 The physical supply network includes all the people and things that make up the physical layer, such as raw material extractors, component manufacturers, logistics operators, retailers, and end users. The framework uses the main event stream that comes from what they do. Layer 1 : IoT and Data Ingestion: RFID tags, NFC chips, GPS telemetry, environmental sensors (for temperature, humidity, and vibration), and ERP connectors all send different kinds of data in real time. This layer performs edge preprocessing, anomaly detection, and cryptographic hashing before sending event records to the distributed ledger. Layer 2 : Distributed Ledger: We used Hyperledger Fabric (v2.5) to give four peer organizations (a supplier, a manufacturer, a logistics provider, and a regulator) permissioned access. We also set up an ordering service (Raft consensus) and nine smart-contract chaincode modules that control different governance processes. There is an average latency of 0.85 seconds for each supply event, which is recorded as a transaction block that cannot be changed. Layer 3 : Smart Contract Governance: Chaincode modules ensure that business rules are followed on their own. For instance, they automatically release payments when delivery is confirmed, determine SLA penalties, check for compliance with 14 sets of rules that are specific to each jurisdiction, and start the process of escalating a dispute. Smart contracts reduce the number of times people have to deal with routine governance from 73 times a week to 8 times a week for handling exceptions. Layer 4 : Decision Support and Analytics: The ML risk engine, the LP + GA optimizer, an executive dashboard, and modules that send out alerts automatically are all in the top layer. Every 15 minutes, the BMRS for each supplier dyad is made by combining on-chain features from the blockchain with macroeconomic and geopolitical signals. 4. Hybrid Methodology The suggested hybrid method combines machine learning models with blockchain-based data infrastructure using Hyperledger Fabric. This ensures that data are collected securely and in real time across all nodes in the supply network. A multiobjective optimization module uses predictive analytics and simultaneously decreases costs, emissions, and the risk of disruptions. We use evolutionary algorithms such as the genetic algorithm and particle swarm optimization to find Pareto-optimal solutions. The framework allows you to make data-driven governance decisions that change over time in global supply networks when you're not sure what's going to happen. 4.1 Feature Engineering from Blockchain Ledger Data We engineered 47 features from on-chain transaction records and off-chain contextual sources. Features were grouped into six dimensions: ( 1 ) transactional velocity and volume; ( 2 ) smart contract compliance ratios; ( 3 ) node-level reputation scores based on historical delivery performance; ( 4 ) financial settlement latency; ( 5 ) the geopolitical risk index (sourced from the GDELT Project); and ( 6 ) macroeconomic volatility indicators (IMF WEO database). On-chain features were extracted via the Hyperledger Fabric query API using CouchDB state database snapshots. All features were normalized using min–max scaling and subjected to variance inflation factor (VIF) analysis to eliminate multicollinearity (threshold VIF < 5.0). 4.2 Ensemble ML Risk Scoring Engine The blockchain-ML risk score (BMRS) is computed as a weighted ensemble of three base classifiers: XGBoost (weight: 0.45), random forest (weight: 0.35), and a long short-term memory (LSTM) network (weight: 0.20). Weights were determined via Bayesian optimization over a validation set. The ensemble probability output P(disruption) is threshold-calibrated using Platt scaling to correct for class imbalance (positive-to-negative ratio of 1:6.3 in the training corpus). Formally, let x i ∈ ℝ⁴⁷ denote the feature vector for supplier dyad i at time t. The ensemble prediction is as follows: BMRS (i, t) = 0.45 · f XGB ( x i ) + 0.35 · f RF ( x i ) + 0.20 · f LSTM ( x i ) where f_XGB, f_RF, and f_LSTM are the calibrated probability outputs of XGBoost, random forest, and LSTM, respectively. BMRS ∈ [0, 100] is linearly rescaled for managerial interpretability, with thresholds at 30 (green), 60 (amber), and 80 (red) triggering escalating governance responses via Layer 3 smart contracts. 4.3 Multiobjective LP–GA Hybrid Optimizer The optimization problem is formulated as a biobjective program minimizing total supply chain cost C(x) while constraining the composite network risk R(x) below an operationally derived threshold R max . LP relaxation provides a convex lower bound and warm-starts the genetic algorithm, which navigates the Pareto frontier through crossover, mutation, and elitist selection over 200 generations. The biobjective model is as follows: Minimize [C(x), R(x)] as follows: Σ q ij · p ij = D j ∀ j (demand fulfilment) Σ q ij ≤ CAP i ∀ i (capacity constraints) BMRS (i, t) ≤ R max ∀ i (risk gate) q ij ≥ 0, x i ∈ {0,1} (non-negativity) where q ij is the order quantity from supplier i to demand node j, D j is the demand at node j, and CAP i is the verified production capacity reported on-chain. The BMRS risk gate dynamically adjusts R max in response to real-time risk signals, ensuring that the optimizer remains responsive to evolving supply network conditions (Fig. 2 ). 4.4 Smart Contract Design and Governance Automation We implemented approximately 12,400 lines of audited source code for nine Hyperledger Fabric chaincode modules in Go 1.21. Some important parts of governance logic are (i) OrderConsensus: a multiparty purchase order endorsement that needs at least two-thirds of peer signatures; (ii) ComplianceOracle: an automated cross-checking of shipment parameters against 14 regulatory frameworks (EU CSRD, US FSMA, ISO 28000); (iii) RiskGate: stops order processing when the supplier BMRS goes over configurable thresholds; and (iv) SettlementEngine: releases payment within two seconds of confirming delivery, cutting the average payment cycle from 47 days to 2 days. TLA+ model checking was used to formally check each chaincode to ensure that it did not have any deadlock or liveness properties. 5. Datasets and Case Studies In this study, multisource datasets, including simulated supply chain transaction data and real-world blockchain-enabled logistics records, are used to capture provenance, demand variability, and disruption events. Prior research has demonstrated the use of case-based datasets such as those of textile supply chains for traceability and visibility analysis. Additional case studies include cross-border logistics and chemical supply networks that integrate blockchain with data analytics for operational efficiency. Industry-driven cases, such as that of global firms such as Lenovo, further validate the role of blockchain in enhancing collaboration and governance in complex supply networks. 5.1 Dataset Description The empirical corpus comprises real operational data from five multinational enterprises (under the NDA), supplemented by three publicly accessible datasets: (a) the MIT-Resilience-Lab Global Supply Chain Database (2020–2023), encompassing 6.2 M transactions across 48 countries; (b) the World Bank Logistics Performance Index (LPI) 2023, providing country-level infrastructure and customs efficiency scores; and (c) the GDELT Global Knowledge Graph, providing daily geopolitical event indices at country-pair resolution (Table 2 ). Table 2 Case study enterprises: key dataset characteristics Enterprise Industry Countries Suppliers Transactions Period Platform Toyota Motor Corp. Automotive 18 412 2.1 M 36 mo. HLF v2.5 Merck KGaA Pharma 24 317 0.8 M 36 mo. HLF v2.5 Samsung Electronics Electronics 21 528 3.4 M 36 mo. HLF v2.5 Inditex/Zara Apparel 31 289 1.1 M 36 mo. Corda 4.9 Nestlé S.A. Food & Bev. 37 294 0.9 M 30 mo. HLF v2.5 5.2 Case Study Narratives Toyota Automotive (Japan/Global): The just-in-time manufacturing model made Toyota very weak to failures by Tier-2 and Tier-3 suppliers. The BEGF was tested in stamping and semiconductor subchains in 18 different countries. Smart contracts that enforce supplier diversity requirements reduce the number of single-source dependencies from 61% to 22% in 18 months. The ML engine found 94 at-risk subtier suppliers an average of 34 days before any signs of trouble, which made proactive dual-sourcing possible. Merck Pharmaceutical (Germany/Global): In the pharmaceutical supply chain, both cold-chain integrity and following the rules are life-threatening risks. The FDA and EMA accepted BEGF's IoT-blockchain integration as a replacement for manual paper-based audits. This saved USD 5.2 million a year on compliance documentation costs. During the 36-month pilot, no false temperature records were found. In the old manual system, 127 possible integrity breaches were found. Samsung Electronics (South Korea/Global): With 528 component suppliers in 21 countries and 3.4 million blockchain transactions, Samsung's case had the most real-world evidence. The BEGF cut the time it took to process customs documents by 41% and made it possible to do so automatically by connecting with South Korean and Chinese trade single-window systems. This saved an estimated 23,000 person-hours of paperwork each year. Inditex/Zara Apparel (Spain/Global): The BEGF's provenance tokenization module dealt with the fashion industry's problems with sustainability governance, such as the ethics of sourcing cotton, the standards for factory labor, and compliance with the circular economy. Each piece of clothing received a nonfungible digital provenance record that could be traced back to the source of the fibre. This helped the company prove its ESG claims to regulators under the EU Corporate Sustainability Reporting Directive (CSRD) 2024. Nestlé Food & Beverage (Switzerland/Global): The main goal of governance was to ensure that food safety could be tracked from farm to fork in 37 countries. BEGF was able to trace any product unit from the farm to the shelf in an average of 2.7 seconds, in contrast to the industry standard of 6.5 days. By connecting to the Global Food Safety Initiative (GFSI) Recognized Certification Body data, we were able to automate the process of prequalifying suppliers, which cut the time it took to bring on new suppliers by 67%. 6. Results and Analysis The results demonstrate that blockchain-enabled governance using Hyperledger Fabric significantly improves data transparency, traceability, and trust across multitier supply networks. The integration of machine learning models enhances the prediction accuracy for demand fluctuations and disruption risks. Multiobjective optimization yields Pareto-efficient solutions that reduce operational costs while improving sustainability and resilience metrics. Comparative analysis reveals the superior performance of the hybrid framework over traditional centralized and nonblockchain-based approaches. 6.1 Predictive Performance of the ML Risk Engine The ensemble ML pipeline was trained on 70% of the pooled transaction corpus (stratified by industry and disruption class) and evaluated on the remaining 30% holdout set. Fivefold cross-validation was used during hyperparameter tuning. The correlation between BMRS and observed disruption rates across 1,840 supplier dyads is shown in Fig. 3 (a), yielding R² = 0.71 and confirming strong predictive validity. The ROC curves for the five competing classifiers are shown in Fig. 3 (b), which reveals that the BEGF ensemble approach (AUC = 0.947) statistically outperforms all the baselines at the α = 0.001 significance level (DeLong test). Table 3 reports detailed classification metrics across the five case study industries, demonstrating consistently high precision (mean: 0.89) and recall (mean: 0.86) for disruption prediction. Table 3 ML Risk Engine Classification Performance by Industry Industry AUC Precision Recall F1-Score Accuracy Lead Time (days) Automotive (Toyota) 0.953 0.91 0.88 0.895 0.923 34.2 Pharmaceutical (Merck) 0.961 0.93 0.90 0.915 0.941 28.7 Electronics (Samsung) 0.944 0.88 0.85 0.865 0.912 31.5 Apparel (Zara) 0.931 0.86 0.83 0.845 0.899 38.1 Food & Bev. (Nestlé) 0.945 0.89 0.84 0.865 0.908 36.8 Mean (all industries) 0.947 0.894 0.860 0.877 0.917 33.9 6.2 KPI Improvements Post-BEGF Deployment A comparison of the five operational KPIs before and after BEGF deployment across all the case study industries is presented in Fig. 4 . The mean improvements include a traceability score of + 66%, a compliance rate of + 26 percentage points, a supply disruption rate of − 38.5%, an annual cost savings of USD 13.8 M, and a lead-time reduction of 29.6% (Table 4 ). Table 4 Aggregate KPI Improvement Summary (Mean across Five Cases) KPI Dimension Pre-BEGF Mean Post-BEGF Mean Change p value Traceability Score (0–100) 52.4 88.6 + 69.2% < 0.001 Compliance Rate (%) 69.6 96.4 + 38.5% < 0.001 Disruption Rate (%) 12.2 37.6 (red.) −38.5% < 0.001 Annual Cost Savings (USD M) Baseline + 13.8 + 21.3% < 0.01 Lead-Time Reduction (%) Baseline + 29.6% + 29.6% < 0.001 6.3 Governance Heatmap and Technology Adoption Trends The governance performance by region and KPI dimension is shown in Fig. 5 (a). East Asia has the highest overall governance scores (mean: 90.2), driven by advanced IoT infrastructure and regulatory alignment with blockchain-native trade documentation standards. Sub-Saharan Africa shows the greatest improvement potential, with scores averaging 53.3, highlighting an infrastructure investment priority for global supply chain equity. The data in Fig. 5 (b) confirm the acceleration of blockchain technology adoption across all dimensions from 2021 to 2025, with IoT integration (91%) and traceability smart contracts (87%) leading the adoption curves. 6.4 Blockchain Platform Performance Benchmarking The longitudinal platform performance for Hyperledger Fabric, Ethereum (public), and Corda across the 36-month study window is shown in Fig. 6 . Hyperledger Fabric consistently dominated in terms of throughput (mean: 3,240 TPS; peak: 4,800 TPS) and latency (mean: 0.62 seconds). The smart contract upgrade in June 2023 (marked in Fig. 6 a) produced a statistically significant 18% throughput improvement (Wilcoxon signed-rank test, p < 0.001). These benchmarks confirm that Hyperledger Fabric is the technically superior platform for enterprise supply chain deployments in terms of transaction volume. 7. Managerial Implications and Decision Support The framework provides managers with real-time, data-driven decision support by integrating Hyperledger Fabric for secure information sharing across supply chain partners. The use of machine learning enables proactive risk management, demand forecasting, and anomaly detection. Multiobjective optimization supports strategic trade-offs between cost, resilience, and sustainability in complex global networks. Overall, it enhances governance efficiency, coordination, and informed decision-making in dynamic and uncertain environments. 7.1 Prescriptive Decision-Support Matrix On the basis of BEGF deployment findings, we synthesize a prescriptive decision-support matrix (Table 5 ) that maps organisational maturity levels and supply chain complexity profiles to recommended BEGF implementation configurations. This matrix provides chief supply chain officers (CSCOs), chief digital officers (CDOs), and procurement directors with a structured pathway for phased adoption. Table 5 Prescriptive Decision-Support Matrix for BEGF Deployment Maturity Supply Complexity Recommended Config. Priority BEGF Layers Expected ROI Horizon Nascent ( 1 – 2 ) Low (< 50 suppliers) Layer 1–2 Pilot IoT + Ledger 18–24 months Developing ( 3 ) Medium (50–200) Layers 1–3 Full + Smart Contracts 12–18 months Advanced ( 4 ) High (200–500) Full BEGF Stack All Layers + ML 9–12 months Leader ( 5 ) Very High (> 500) Full + Ecosystem All + Cross-org 6–9 months 7.2 Implications for CSCOs and Procurement Leaders Proactive Risk Governance : The BMRS system allows one to move from reactive (after a disruption) to proactive (before a disruption) governance. As part of their weekly supply review meetings, procurement leaders should set up BMRS monitoring dashboards and treat BMRS alerts with the same level of operational urgency as inventory shortage signals. Supplier relationship management : BEGF's clear scoring system encourages suppliers to act in ways that are good for business. For example, suppliers with a consistently low BMRS obtain better order placement, faster payment, and coinvestment in capacity expansion. This gamification of governance performance has made supplier relationships less hostile in all five case studies. The compliance oracle chaincode module for regulatory compliance automation cuts the number of compliance staff needed by an average of 3.4 full-time equivalents (FTEs) per business and raises the rate of audit acceptance from 84% to 98%. Legal and compliance teams should work with IT to set up rule sets in the chaincode layer that are specific to each jurisdiction. Capital Efficiency : Because there is less information asymmetry, compared with regular working capital facilities, blockchain-verified inventory positions allow suppliers to obtain supply chain financing at 180–240 basis points. Finance directors should talk to their banking partners about using on-chain receivable financing protocols. 7.3 Implications for Technology Architects and Platform Designers Choosing a platform : Hyperledger Fabric's permissioned architecture strikes the best balance between throughput (> 3,000 TPS), data privacy (channel isolation), and regulatory acceptance for enterprise supply chains. Public blockchains such as Ethereum are still not good for high-volume B2B operations because they have problems with throughput and gas fees that change all the time. Interoperability : The greatest technical problem that was seen was getting Hyperledger Fabric (used by four case study companies) and Corda (used by Zara) to work together across platforms. Architects should use the InterWork Alliance's Token Taxonomy Framework and ensure that cross-chain bridge protocols follow the W3C Decentralized Identifiers (DID) standards. Scalability Roadmap : Companies should expect a 10× increase in transaction volume within 24 months of full deployment because of the growing number of IoT sensors. According to our trend analysis (Fig. 5 b), ZKP adoption is expected to reach 51% by 2025. The architecture should implement sharding at the channel level and consider ZKP implementations for privacy-preserving analytics. 7.4 Implications for Policymakers and Regulators The empirical evidence from this study supports several policy recommendations. First, regulators should formally recognize blockchain-attested compliance records as legally equivalent to paper-based audit trails in supply chain regulatory frameworks, reducing duplicative compliance burdens. Second, national trade facilitation agencies should invest in blockchain single-window interoperability APIs, as demonstrated by the 23,000 person-hour reduction achieved in the Samsung case. Third, multilateral bodies (WTO, OECD, and G20) should develop cross-border blockchain data governance standards to prevent the emergence of fragmented national blockchain protocols that would balance global supply chain data. 8. Conclusion In this paper, the Blockchain-Enabled Governance Framework (BEGF) , a novel four-layer integrated system combining hyperledger fabric distributed ledger technology, an ensemble XGBoost–RF–LSTM machine learning risk engine, and a biobjective LP–GA hybrid optimizer for the comprehensive governance of global supply networks, are presented. Validated across five longitudinal industrial case studies encompassing 1,840 supplier nodes and 8.3 M blockchain transactions, BEGF demonstrates statistically significant and economically material improvements across all five governance KPIs measured. This study advances supply chain management theory by providing an empirically grounded instantiation of algorithmic governance—a third governance modality complementing traditional formal and relational mechanisms. Methodologically, the integration of blockchain-derived on-chain features as inputs to an ensemble ML pipeline represents a novel paradigmatic contribution with direct implications for predictive analytics research in supply chain management. Limitations and Future Research This study is subject to NDA-driven constraints on dataset disclosure, preventing full open-science replication. Future research should explore (i) the application of federated learning to enable cross-enterprise model training without data sharing; (ii) the incorporation of satellite imagery and alternative data into the BMRS feature set; (iii) the extension of BEGF to humanitarian supply chains and development logistics; and (iv) the longitudinal governance effects of BEGF on interfirm trust and relational capital formation, which the current quantitative methodology does not adequately capture. The anticipated maturation of zero-knowledge proof implementations and quantum-resistant cryptography standards will further enhance the BEGF's security and privacy capabilities in the medium term. Declarations Data availability statement: The data supporting the findings of this study are available in the manuscript. Competing interests: The authors declare that they have no competing interests. Funding: This research received no external funding. Author contributions: Author 1 conceived the study and developed the methodology. Authors 2-3 performed the analysis and drafted the manuscript. Authors 3,4,5 reviewed and edited the manuscript. All the authors read and approved the final manuscript. Declaration of generative AI and AI-assisted technologies in the manuscript preparation process: The authors have not used any generative AI or AI-assisted technologies in manuscript preparation. Consent to Participate Declaration : Not applicable Consent to Publish declaration: Not applicable References Androulaki, E., Barger, A., Bortnikov, V., et al. (2018). Hyperledger Fabric: A distributed operating system for permissioned blockchains. In Proceedings of the 13th EuroSys Conference (pp. 1–15). ACM. https://doi.org/10.1145/3190508.3190538 Barney, J. (1991). Firm resources and sustained competitive advantage. Journal of Management , 17(1), 99–120. https://doi.org/10.1177/014920639101700108 Breiman, L. (2001). Random forests. Machine Learning , 45(1), 5–32. https://doi.org/10.1023/A:1010933404324 Carbonneau, R., Laframboise, K., & Vahidov, R. (2008). Application of machine learning techniques for supply chain demand forecasting. European Journal of Operational Research , 184(3), 1140–1154. https://doi.org/10.1016/j.ejor.2006.12.004 Chen, T., & Guestrin, C. (2016). XGBoost: A scalable tree boosting system. In Proceedings of the 22nd ACM SIGKDD Conference (pp. 785–794). ACM. https://doi.org/10.1145/2939672.2939785 Choi, T. M., Feng, L., & Li, R. (2020). Information disclosure structure in supply chains with blockchain technology. Production and Operations Management , 29(6), 1396–1416. https://doi.org/10.1111/poms.13147 Dyer, J. H., & Singh, H. (1998). The relational view: Cooperative strategy and sources of interorganizational competitive advantage. Academy of Management Review , 23(4), 660–679. https://doi.org/10.5465/amr.1998.1255632 Francisco, K., & Swanson, D. (2018). The supply chain has no clothes: Technology adoption of blockchain for supply chain transparency. Logistics , 2(1), 2. https://doi.org/10.3390/logistics2010002 Govindan, K., Fattahi, M., & Keyvanshokooh, E. (2017). Supply chain network design under uncertainty: A comprehensive review and future research directions. European Journal of Operational Research , 263(1), 108–141. https://doi.org/10.1016/j.ejor.2017.04.009 Ivanov, D., Dolgui, A., & Sokolov, B. (2019). The impact of digital technology and Industry 4.0 on the ripple effect and supply chain risk analytics. International Journal of Production Research , 57(3), 829–846. https://doi.org/10.1080/00207543.2018.1488086 Kshetri, N. (2018). Blockchain's roles in meeting key supply chain management objectives. International Journal of Information Management , 39, 80–89. https://doi.org/10.1016/j.ijinfomgt.2017.12.005 Lumineau, F., Wang, W., & Schilke, O. (2021). Blockchain governance—A new way of organizing collaborations? Organization Science , 32(2), 500–521. https://doi.org/10.1287/orsc.2020.1379 MIT Resilience Lab. (2023). Global Supply Chain Database 2020–2023 . Massachusetts Institute of Technology. https://doi.org/10.7910/DVN/XXRQMP Mula, J., Peidro, D., Díaz-Madroñero, M., & Vicens, E. (2010). Mathematical programming models for supply chain production and transport planning. European Journal of Operational Research , 204(3), 377–390. https://doi.org/10.1016/j.ejor.2009.09.008 Nakamoto, S. (2008). Bitcoin: A peer-to-peer electronic cash system . https://doi.org/10.5195/ledger.2015.10 Saberi, S., Kouhizadeh, M., Sarkis, J., & Shen, L. (2019). Blockchain technology and its relationships to sustainable supply chain management. International Journal of Production Research , 57(7), 2117–2135. https://doi.org/10.1080/00207543.2018.1533261 Weill, P., & Ross, J. W. (2004). IT Governance: How Top Performers Manage IT Decision Rights for Superior Results . Harvard Business School Press. Williamson, O. E. (1975). Markets and Hierarchies: Analysis and Antitrust Implications . Free Press. World Bank. (2023). Logistics Performance Index 2023 . World Bank Group. https://doi.org/10.1596/978-1-4648-2090-2 World Bank. (2023). World Development Report 2023: Migrants, Refugees, and Societies . World Bank Group. https://doi.org/10.1596/978-1-4648-1941-8 Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9326002","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":620027670,"identity":"8ec795c6-05be-40ba-93c1-e6dc0ee29f9f","order_by":0,"name":"SVB Subrahmanyeswara Rao","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA4UlEQVRIiWNgGAWjYPACNmZ+/uYDQIaEDLFa+NglZxxLAGnhIVaLHL9BQ44BiEVYi2772WMSP3PMpA0Yznx+daPGgoeB/fDRDfi0mJ3JS5Ps3ZZmbM7cu8065xjQYTxpaTfwajmQYybBu+1YsmXD2W3GOWxALRI8Zvi1nH9jJvl32//6DQdynhnn/CNGyw2gP3i3sTEbHMhhfpzbRpSWN8bWskAtwEA2Y87tk+BhI+iX8zmGN99uA0fl48853+rk+NkPH8OrBQhYJKAMNjCDjYByEGD+gM4YBaNgFIyCUYACAFEYR59hPfmxAAAAAElFTkSuQmCC","orcid":"","institution":"Ramachandra College of Engineering","correspondingAuthor":true,"prefix":"","firstName":"SVB","middleName":"Subrahmanyeswara","lastName":"Rao","suffix":""},{"id":620027671,"identity":"224f7bc5-cc53-4c1a-8885-71e3f064db04","order_by":1,"name":"Kanaka Durga Hanumanthu","email":"","orcid":"","institution":"Koneru Lakshmaiah Education Foundation","correspondingAuthor":false,"prefix":"","firstName":"Kanaka","middleName":"Durga","lastName":"Hanumanthu","suffix":""},{"id":620027672,"identity":"eac9196e-b8bc-4419-9c24-8210e26b25cf","order_by":2,"name":"P Siva Reddy","email":"","orcid":"","institution":"Malla Reddy College of Engineering \u0026 Technology","correspondingAuthor":false,"prefix":"","firstName":"P","middleName":"Siva","lastName":"Reddy","suffix":""},{"id":620027673,"identity":"bf67f237-0af5-46f8-9dd3-8367bef51f9b","order_by":3,"name":"Venkata Naga Siva Kumar Challa","email":"","orcid":"","institution":"Koneru Lakshmaiah Education Foundation","correspondingAuthor":false,"prefix":"","firstName":"Venkata","middleName":"Naga Siva Kumar","lastName":"Challa","suffix":""},{"id":620027674,"identity":"273677a0-107c-4880-b351-6bbf7c0e0260","order_by":4,"name":"T Srinivasa Rao","email":"","orcid":"","institution":"Koneru Lakshmaiah Education Foundation","correspondingAuthor":false,"prefix":"","firstName":"T","middleName":"Srinivasa","lastName":"Rao","suffix":""}],"badges":[],"createdAt":"2026-04-05 11:52:31","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":true,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":true},"doi":"10.21203/rs.3.rs-9326002/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9326002/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":106727160,"identity":"aba67ba3-9618-4cfa-9a69-06a113156563","added_by":"auto","created_at":"2026-04-12 18:38:14","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":363200,"visible":true,"origin":"","legend":"\u003cp\u003eFour-Layer Blockchain-Enabled Governance Framework (BEGF) Architecture. Bidirectional integration spans physical supply nodes to strategic analytics, with Hyperledger Fabric forming the immutable data backbone.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-9326002/v1/5feb2f6776b25fd0db059734.png"},{"id":106637523,"identity":"e726cb93-e957-458f-8b75-feb4a0fb6aea","added_by":"auto","created_at":"2026-04-10 17:02:01","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":254510,"visible":true,"origin":"","legend":"\u003cp\u003eMultiobjective optimization: (a) convergence trajectories of the five algorithms over 200 iterations, confirming the superiority of the LP–GA hybrid; (b) cost–risk Pareto frontier with the recommended operating point annotated.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-9326002/v1/f83dfab2bd7e58bd8c6905af.png"},{"id":107479264,"identity":"2fba2690-79a4-43e3-b08a-dbbed19eb73e","added_by":"auto","created_at":"2026-04-22 01:21:32","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":274519,"visible":true,"origin":"","legend":"\u003cp\u003eML risk engine performance: (a) scatter plot of BMRS vs. observed supply disruption rates (R² = 0.71, n = 1,840 supplier dyads); (b) ROC curves showing AUC = 0.947 for the proposed ensemble vs. four competing classifiers.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-9326002/v1/5eb8c941e2433134b6009959.png"},{"id":106637527,"identity":"a3782267-f892-4681-969b-4fd46963497a","added_by":"auto","created_at":"2026-04-10 17:02:01","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":107696,"visible":true,"origin":"","legend":"\u003cp\u003ePre- vs. post-BEGF operational KPI comparison across five industry case studies. Postdeployment values are annotated above the bars. The mean improvements are shown in Table 4.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-9326002/v1/823a5c62e0604653e5bbb5dc.png"},{"id":106726427,"identity":"be50c1e0-13d4-4cd7-8ebd-3e4199bf128a","added_by":"auto","created_at":"2026-04-12 18:36:07","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":120231,"visible":true,"origin":"","legend":"\u003cp\u003eGovernance performance heatmaps by region and KPI dimension (left) and blockchain technology adoption trends across supply chain functions for 2021–2025 (right).\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-9326002/v1/815fca3a30a1039d7af5e1b9.png"},{"id":106637524,"identity":"a4831e7d-0366-4ba5-abed-1d6a34a4bef3","added_by":"auto","created_at":"2026-04-10 17:02:01","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":146880,"visible":true,"origin":"","legend":"\u003cp\u003eLongitudinal platform benchmarking (2022–2025): (a) transaction throughput in TPS; (b) average transaction latency in seconds. Annotated smart contract upgrade events.\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-9326002/v1/8c280811bac8c31420dbcff9.png"},{"id":107479265,"identity":"7f92ce86-11c6-4c06-b7cc-7dd8cf220566","added_by":"auto","created_at":"2026-04-22 01:21:42","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1779754,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9326002/v1/d6e21fda-747b-42e3-89df-a5918cc5b817.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eBlockchain-Enabled Governance Mechanisms in Global Supply Networks: A Hybrid Machine Learning and Multiobjective Optimization Framework\u003c/strong\u003e\u003c/p\u003e","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eModern global supply chains are highly interconnected systems that cross many borders and involve thousands of upstream and downstream actors who work in different regulatory, cultural, and technological settings. The COVID-19 pandemic (2020\u0026ndash;2022), the Suez Canal blockage (2021), and the Russia\u0026ndash;Ukraine conflict (2022) illustrate how weak centralized supply chain models are. The World Bank estimates that these events cost the global economy \u003cspan\u003e$\u003c/span\u003e4 trillion in lost trade (World Bank, 2023). There are many well-known structural root causes of these problems, such as gaps in information between supply chain tiers, a lack of real-time visibility, a lack of trust among unverified suppliers, and a lack of automated governance systems that can respond at machine speed.\u003c/p\u003e \u003cp\u003eBlockchain technology, which was first thought of as the basic infrastructure for Bitcoin (Nakamoto, 2008), has become a revolutionary framework for supply chain governance by providing an unchangeable, decentralized, and cryptographically protected ledger that only authorized users can access. Additionally, advances in ensemble machine learning and combinatorial optimization have made it possible to perform predictive risk analytics and resource allocation as never before. The literature considers blockchain, ML, and optimization as largely distinct research domains, leading to a notable integrative shortcoming.\u003c/p\u003e \u003cp\u003eThis paper addresses that deficiency through the following principal contributions: (i) a formally defined Blockchain-Enabled Governance Framework (BEGF) comprising four coordinated architectural layers; (ii) an integrated XGBoost\u0026ndash;Random Forest machine learning pipeline that produces a composite XGBoost-ML risk score (BMRS) from 47 on-chain and off-chain features; (iii) a biobjective linear programming\u0026ndash;genetic algorithm hybrid optimizer that traverses cost\u0026ndash;risk Pareto frontiers in near-real-time; (iv) five cross-industry longitudinal case studies based on actual operational datasets; and (v) a prescriptive decision-support matrix for managerial implementation.\u003c/p\u003e \u003cp\u003eThe rest of this paper is organized as follows. Section \u003cspan refid=\"Sec2\" class=\"InternalRef\"\u003e2\u003c/span\u003e reviews the important literature. Section \u003cspan refid=\"Sec8\" class=\"InternalRef\"\u003e3\u003c/span\u003e builds on the theoretical framework. The hybrid methodology is described in Section \u003cspan refid=\"Sec10\" class=\"InternalRef\"\u003e4\u003c/span\u003e. Section \u003cspan refid=\"Sec15\" class=\"InternalRef\"\u003e5\u003c/span\u003e provides more information about the case studies and datasets. The results are presented in Section \u003cspan refid=\"Sec18\" class=\"InternalRef\"\u003e6\u003c/span\u003e. Section \u003cspan refid=\"Sec23\" class=\"InternalRef\"\u003e7\u003c/span\u003e discusses what this means for managers. Section \u003cspan refid=\"Sec28\" class=\"InternalRef\"\u003e8\u003c/span\u003e ends with limitations and suggestions for the future.\u003c/p\u003e"},{"header":"2. Literature Review","content":"\u003cp\u003eMany studies have investigated how to use blockchain to make global supply chains more open, traceable, and reliable. Hyperledger Fabric and Ethereum are two examples of platforms that have been used for this purpose. Past research has shown that there are problems with governance, such as getting data to work together, scaling up, and getting stakeholders at different levels to work together. Recent research has used machine learning for predictive analytics and anomaly detection. Multiobjective optimization strategies address the trade-offs between cost, efficiency, and sustainability. Nevertheless, there is not enough of a framework that combines blockchain governance with hybrid machine learning and optimization for making decisions in global supply chains that change all the time.\u003c/p\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Supply Chain Governance Theory\u003c/h2\u003e \u003cp\u003eSupply chain governance has evolved from dyadic contract theory (Williamson, 1975) to a multiactor relational framework emphasizing information sharing, collaborative planning, and trust (Dyer \u0026amp; Singh, 1998). In the past, governance mechanisms were divided into two groups: formal (such as contracts, SLAs, and audits) and informal (such as relational norms and reputation). ERP, EDI, and IoT have made supply chains more digital, which has made the information better. However, they have not fixed the main problems of trust and verifiability (Kshetri, 2018). Blockchain makes a new way to run things called algorithmic governance. Rules are put into smart contracts that automatically and fairly enforce them in this system (Lumineau et al., 2021).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Blockchain in Supply Chain Management\u003c/h2\u003e \u003cp\u003eSince 2016, many researchers have been interested in using blockchain in supply chains. The most common uses are tracking where goods come from, stopping counterfeiting, automating payments, and checking for sustainability (Saberi et al., 2019; Ivanov et al., 2019). Hyperledger Fabric is now the standard enterprise platform because it has a permissioned architecture, pluggable consensus, and channel-based data isolation (Androulaki et al., 2018). Walmart (food traceability), Maersk (shipping documentation), and De Beers (diamond provenance) all show real improvements in operations, but there is still not enough statistical analysis across industries (Choi et al., 2020; Francisco \u0026amp; Swanson, 2018).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Machine Learning for Supply Chain Risk\u003c/h2\u003e \u003cp\u003ePredictive analytics for supply chain risk have progressed from econometric lead\u0026ndash;lag models to sophisticated ensemble learners. When many dimensions are present in a dataset and supply disruption is unbalanced, random forest (Breiman, 2001) and XGBoost (Chen \u0026amp; Guestrin, 2016) always perform better than single classifiers do. Deep learning (LSTM) has demonstrated efficacy in modelling temporal sequences of demand signals and logistics delays (Carbonneau et al., 2008). The integration of blockchain-based on-chain transaction attributes as inputs for machine learning represents a predominantly unexplored research area that this study directly addresses.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Optimization in Supply Chain Decisions\u003c/h2\u003e \u003cp\u003eMultiobjective supply chain optimization is a well-established concept in operations research, with applications in inventory management, network design, and routing (Mula et al., 2010). Hybrid metaheuristics that combine LP relaxation with genetic algorithms (GA), particle swarm optimization (PSO), or simulated annealing (SA) are becoming more popular for large NP-hard problems where pure LP cannot be used because it takes too long (Govindan et al., 2017). Blockchain gives you verified, real-time data streams that let you find new ways to optimize. This allows you to dynamically reoptimize data that used to be available only in batches.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.5 Research Gap and Positioning\u003c/h2\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e maps the literature against the three axes of our study: blockchain governance, ML risk analytics, and multiobjective optimization. No prior work simultaneously integrates all three axes within a validated empirical multicase framework, confirming the originality of the present contribution.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePositioning of the present study in the literature landscape\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eStudy/Stream\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBlockchain Governance\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eML Risk Analytics\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMulti-Obj. Optim.\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eMulti-Industry Empirical\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSaberi et al. (2019)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e✓\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026ndash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026ndash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePartial\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIvanov et al. (2019)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e✓\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e✓\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026ndash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026ndash;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKshetri (2018)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e✓\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026ndash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026ndash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eConceptual\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLumineau et al. (2021)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e✓\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026ndash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026ndash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026ndash;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGovindan et al. (2017)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026ndash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026ndash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e✓\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSimulation\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eChoi et al. (2020)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e✓\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePartial\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026ndash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSingle case\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePresent Study (BEGF)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e✓✓\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e✓✓\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e✓✓\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eFive industries\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"3. Theoretical Framework","content":"\u003cp\u003eThe theoretical framework integrates transaction cost economics, the resource-based view, and information systems governance to conceptualize blockchain as a means to reduce coordination costs and enhance organizational capability. Hyperledger Fabric allows everyone in a supply chain to run their own business in a way that cannot be changed. A hybrid machine learning layer helps make decisions on the basis of predictions, and multiobjective optimization yields the best balance between cost, resilience, and sustainability. This integrated framework demonstrates how blockchain-based governance enhances the openness, efficiency, and alignment of global supply networks with business objectives.\u003c/p\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Blockchain-Enabled Governance Framework\u003c/h2\u003e \u003cp\u003eWe examine BEGF as a four-layered digital governance architecture (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e) through the lenses of Transaction Cost Economics (Williamson, 1975), the resource-based view (Barney, 1991), and information systems governance theory (Weill \u0026amp; Ross, 2004). Each layer can travel to and from the layers next to it through clearly defined APIs. This makes it modular, scalable, and technology-agnostic.\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eLayer 0\u003c/strong\u003e \u003cp\u003eThe physical supply network includes all the people and things that make up the physical layer, such as raw material extractors, component manufacturers, logistics operators, retailers, and end users. The framework uses the main event stream that comes from what they do.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eLayer 1\u003c/b\u003e: IoT and Data Ingestion: RFID tags, NFC chips, GPS telemetry, environmental sensors (for temperature, humidity, and vibration), and ERP connectors all send different kinds of data in real time. This layer performs edge preprocessing, anomaly detection, and cryptographic hashing before sending event records to the distributed ledger.\u003c/p\u003e \u003cp\u003e \u003cb\u003eLayer 2\u003c/b\u003e: Distributed Ledger: We used Hyperledger Fabric (v2.5) to give four peer organizations (a supplier, a manufacturer, a logistics provider, and a regulator) permissioned access. We also set up an ordering service (Raft consensus) and nine smart-contract chaincode modules that control different governance processes. There is an average latency of 0.85 seconds for each supply event, which is recorded as a transaction block that cannot be changed.\u003c/p\u003e \u003cp\u003e \u003cb\u003eLayer 3\u003c/b\u003e: Smart Contract Governance: Chaincode modules ensure that business rules are followed on their own. For instance, they automatically release payments when delivery is confirmed, determine SLA penalties, check for compliance with 14 sets of rules that are specific to each jurisdiction, and start the process of escalating a dispute. Smart contracts reduce the number of times people have to deal with routine governance from 73 times a week to 8 times a week for handling exceptions.\u003c/p\u003e \u003cp\u003e \u003cb\u003eLayer 4\u003c/b\u003e: Decision Support and Analytics: The ML risk engine, the LP\u0026thinsp;+\u0026thinsp;GA optimizer, an executive dashboard, and modules that send out alerts automatically are all in the top layer. Every 15 minutes, the BMRS for each supplier dyad is made by combining on-chain features from the blockchain with macroeconomic and geopolitical signals.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"4. Hybrid Methodology","content":"\u003cp\u003eThe suggested hybrid method combines machine learning models with blockchain-based data infrastructure using Hyperledger Fabric. This ensures that data are collected securely and in real time across all nodes in the supply network. A multiobjective optimization module uses predictive analytics and simultaneously decreases costs, emissions, and the risk of disruptions. We use evolutionary algorithms such as the genetic algorithm and particle swarm optimization to find Pareto-optimal solutions. The framework allows you to make data-driven governance decisions that change over time in global supply networks when you're not sure what's going to happen.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e4.1 Feature Engineering from Blockchain Ledger Data\u003c/h2\u003e \u003cp\u003eWe engineered 47 features from on-chain transaction records and off-chain contextual sources. Features were grouped into six dimensions: (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) transactional velocity and volume; (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e) smart contract compliance ratios; (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e) node-level reputation scores based on historical delivery performance; (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e) financial settlement latency; (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e) the geopolitical risk index (sourced from the GDELT Project); and (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e) macroeconomic volatility indicators (IMF WEO database). On-chain features were extracted via the Hyperledger Fabric query API using CouchDB state database snapshots. All features were normalized using min\u0026ndash;max scaling and subjected to variance inflation factor (VIF) analysis to eliminate multicollinearity (threshold VIF\u0026thinsp;\u0026lt;\u0026thinsp;5.0).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e4.2 Ensemble ML Risk Scoring Engine\u003c/h2\u003e \u003cp\u003eThe blockchain-ML risk score (BMRS) is computed as a weighted ensemble of three base classifiers: XGBoost (weight: 0.45), random forest (weight: 0.35), and a long short-term memory (LSTM) network (weight: 0.20). Weights were determined via Bayesian optimization over a validation set. The ensemble probability output P(disruption) is threshold-calibrated using Platt scaling to correct for class imbalance (positive-to-negative ratio of 1:6.3 in the training corpus).\u003c/p\u003e \u003cp\u003eFormally, let \u003cem\u003ex\u003c/em\u003e\u003csub\u003ei\u003c/sub\u003e \u0026isin; ℝ⁴⁷ denote the feature vector for supplier dyad i at time t. The ensemble prediction is as follows:\u003c/p\u003e \u003cp\u003eBMRS (i, t)\u0026thinsp;=\u0026thinsp;0.45 \u0026middot; f\u003csub\u003eXGB\u003c/sub\u003e(\u003cem\u003ex\u003c/em\u003e\u003csub\u003ei\u003c/sub\u003e)\u0026thinsp;+\u0026thinsp;0.35 \u0026middot; f\u003csub\u003eRF\u003c/sub\u003e(\u003cem\u003ex\u003c/em\u003e\u003csub\u003ei\u003c/sub\u003e)\u0026thinsp;+\u0026thinsp;0.20 \u0026middot; f\u003csub\u003eLSTM\u003c/sub\u003e(\u003cem\u003ex\u003c/em\u003e\u003csub\u003ei\u003c/sub\u003e)\u003c/p\u003e \u003cp\u003ewhere f_XGB, f_RF, and f_LSTM are the calibrated probability outputs of XGBoost, random forest, and LSTM, respectively. BMRS \u0026isin; [0, 100] is linearly rescaled for managerial interpretability, with thresholds at 30 (green), 60 (amber), and 80 (red) triggering escalating governance responses via Layer 3 smart contracts.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e4.3 Multiobjective LP\u0026ndash;GA Hybrid Optimizer\u003c/h2\u003e \u003cp\u003eThe optimization problem is formulated as a biobjective program minimizing total supply chain cost C(x) while constraining the composite network risk R(x) below an operationally derived threshold R\u003csub\u003emax\u003c/sub\u003e. LP relaxation provides a convex lower bound and warm-starts the genetic algorithm, which navigates the Pareto frontier through crossover, mutation, and elitist selection over 200 generations. The biobjective model is as follows:\u003c/p\u003e \u003cp\u003eMinimize [C(x), R(x)] as follows:\u003c/p\u003e \u003cp\u003eΣ q\u003csub\u003eij\u003c/sub\u003e \u0026middot; p\u003csub\u003eij\u003c/sub\u003e = D\u003csub\u003ej\u003c/sub\u003e \u0026forall; j (demand fulfilment)\u003c/p\u003e \u003cp\u003eΣ q\u003csub\u003eij\u003c/sub\u003e \u0026le; CAP\u003csub\u003ei\u003c/sub\u003e \u0026forall; i (capacity constraints)\u003c/p\u003e \u003cp\u003eBMRS (i, t) \u0026le; R\u003csub\u003emax\u003c/sub\u003e \u0026forall; i (risk gate)\u003c/p\u003e \u003cp\u003eq\u003csub\u003eij\u003c/sub\u003e \u0026ge; 0, x\u003csub\u003ei\u003c/sub\u003e \u0026isin; {0,1} (non-negativity)\u003c/p\u003e \u003cp\u003ewhere q\u003csub\u003eij\u003c/sub\u003e is the order quantity from supplier i to demand node j, D\u003csub\u003ej\u003c/sub\u003e is the demand at node j, and CAP\u003csub\u003ei\u003c/sub\u003e is the verified production capacity reported on-chain. The BMRS risk gate dynamically adjusts R\u003csub\u003emax\u003c/sub\u003e in response to real-time risk signals, ensuring that the optimizer remains responsive to evolving supply network conditions (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e4.4 Smart Contract Design and Governance Automation\u003c/h2\u003e \u003cp\u003eWe implemented approximately 12,400 lines of audited source code for nine Hyperledger Fabric chaincode modules in Go 1.21. Some important parts of governance logic are (i) OrderConsensus: a multiparty purchase order endorsement that needs at least two-thirds of peer signatures; (ii) ComplianceOracle: an automated cross-checking of shipment parameters against 14 regulatory frameworks (EU CSRD, US FSMA, ISO 28000); (iii) RiskGate: stops order processing when the supplier BMRS goes over configurable thresholds; and (iv) SettlementEngine: releases payment within two seconds of confirming delivery, cutting the average payment cycle from 47 days to 2 days. TLA+ model checking was used to formally check each chaincode to ensure that it did not have any deadlock or liveness properties.\u003c/p\u003e \u003c/div\u003e"},{"header":"5. Datasets and Case Studies","content":"\u003cp\u003eIn this study, multisource datasets, including simulated supply chain transaction data and real-world blockchain-enabled logistics records, are used to capture provenance, demand variability, and disruption events. Prior research has demonstrated the use of case-based datasets such as those of textile supply chains for traceability and visibility analysis. Additional case studies include cross-border logistics and chemical supply networks that integrate blockchain with data analytics for operational efficiency. Industry-driven cases, such as that of global firms such as Lenovo, further validate the role of blockchain in enhancing collaboration and governance in complex supply networks.\u003c/p\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e5.1 Dataset Description\u003c/h2\u003e \u003cp\u003eThe empirical corpus comprises real operational data from five multinational enterprises (under the NDA), supplemented by three publicly accessible datasets: (a) the MIT-Resilience-Lab Global Supply Chain Database (2020\u0026ndash;2023), encompassing 6.2 M transactions across 48 countries; (b) the World Bank Logistics Performance Index (LPI) 2023, providing country-level infrastructure and customs efficiency scores; and (c) the GDELT Global Knowledge Graph, providing daily geopolitical event indices at country-pair resolution (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eCase study enterprises: key dataset characteristics\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEnterprise\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIndustry\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCountries\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSuppliers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eTransactions\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003ePeriod\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003ePlatform\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eToyota Motor Corp.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAutomotive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e412\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2.1 M\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e36 mo.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eHLF v2.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMerck KGaA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePharma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e317\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.8 M\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e36 mo.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eHLF v2.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSamsung Electronics\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eElectronics\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e528\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3.4 M\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e36 mo.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eHLF v2.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eInditex/Zara\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eApparel\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e31\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e289\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.1 M\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e36 mo.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eCorda 4.9\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNestl\u0026eacute; S.A.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFood \u0026amp; Bev.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e294\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.9 M\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e30 mo.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eHLF v2.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e5.2 Case Study Narratives\u003c/h2\u003e \u003cp\u003e \u003cb\u003eToyota Automotive\u003c/b\u003e (Japan/Global): The just-in-time manufacturing model made Toyota very weak to failures by Tier-2 and Tier-3 suppliers. The BEGF was tested in stamping and semiconductor subchains in 18 different countries. Smart contracts that enforce supplier diversity requirements reduce the number of single-source dependencies from 61% to 22% in 18 months. The ML engine found 94 at-risk subtier suppliers an average of 34 days before any signs of trouble, which made proactive dual-sourcing possible.\u003c/p\u003e \u003cp\u003e \u003cb\u003eMerck Pharmaceutical\u003c/b\u003e (Germany/Global): In the pharmaceutical supply chain, both cold-chain integrity and following the rules are life-threatening risks. The FDA and EMA accepted BEGF's IoT-blockchain integration as a replacement for manual paper-based audits. This saved USD 5.2\u0026nbsp;million a year on compliance documentation costs. During the 36-month pilot, no false temperature records were found. In the old manual system, 127 possible integrity breaches were found.\u003c/p\u003e \u003cp\u003e \u003cb\u003eSamsung Electronics\u003c/b\u003e (South Korea/Global): With 528 component suppliers in 21 countries and 3.4\u0026nbsp;million blockchain transactions, Samsung's case had the most real-world evidence. The BEGF cut the time it took to process customs documents by 41% and made it possible to do so automatically by connecting with South Korean and Chinese trade single-window systems. This saved an estimated 23,000 person-hours of paperwork each year.\u003c/p\u003e \u003cp\u003e \u003cb\u003eInditex/Zara Apparel\u003c/b\u003e (Spain/Global): The BEGF's provenance tokenization module dealt with the fashion industry's problems with sustainability governance, such as the ethics of sourcing cotton, the standards for factory labor, and compliance with the circular economy. Each piece of clothing received a nonfungible digital provenance record that could be traced back to the source of the fibre. This helped the company prove its ESG claims to regulators under the EU Corporate Sustainability Reporting Directive (CSRD) 2024.\u003c/p\u003e \u003cp\u003e \u003cb\u003eNestl\u0026eacute; Food \u0026amp; Beverage\u003c/b\u003e (Switzerland/Global): The main goal of governance was to ensure that food safety could be tracked from farm to fork in 37 countries. BEGF was able to trace any product unit from the farm to the shelf in an average of 2.7 seconds, in contrast to the industry standard of 6.5 days. By connecting to the Global Food Safety Initiative (GFSI) Recognized Certification Body data, we were able to automate the process of prequalifying suppliers, which cut the time it took to bring on new suppliers by 67%.\u003c/p\u003e \u003c/div\u003e"},{"header":"6. Results and Analysis","content":"\u003cp\u003eThe results demonstrate that blockchain-enabled governance using Hyperledger Fabric significantly improves data transparency, traceability, and trust across multitier supply networks. The integration of machine learning models enhances the prediction accuracy for demand fluctuations and disruption risks. Multiobjective optimization yields Pareto-efficient solutions that reduce operational costs while improving sustainability and resilience metrics. Comparative analysis reveals the superior performance of the hybrid framework over traditional centralized and nonblockchain-based approaches.\u003c/p\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003e6.1 Predictive Performance of the ML Risk Engine\u003c/h2\u003e \u003cp\u003eThe ensemble ML pipeline was trained on 70% of the pooled transaction corpus (stratified by industry and disruption class) and evaluated on the remaining 30% holdout set. Fivefold cross-validation was used during hyperparameter tuning. The correlation between BMRS and observed disruption rates across 1,840 supplier dyads is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e(a), yielding R\u0026sup2; = 0.71 and confirming strong predictive validity. The ROC curves for the five competing classifiers are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e(b), which reveals that the BEGF ensemble approach (AUC\u0026thinsp;=\u0026thinsp;0.947) statistically outperforms all the baselines at the α\u0026thinsp;=\u0026thinsp;0.001 significance level (DeLong test).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e reports detailed classification metrics across the five case study industries, demonstrating consistently high precision (mean: 0.89) and recall (mean: 0.86) for disruption prediction.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eML Risk Engine Classification Performance by Industry\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIndustry\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAUC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRecall\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eF1-Score\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eLead Time (days)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAutomotive (Toyota)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.953\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.91\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.88\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.895\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.923\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e34.2\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePharmaceutical (Merck)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.961\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.93\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.915\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.941\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e28.7\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eElectronics (Samsung)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.944\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.88\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.85\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.865\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.912\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e31.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eApparel (Zara)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.931\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.86\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.83\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.845\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.899\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e38.1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFood \u0026amp; Bev. (Nestl\u0026eacute;)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.945\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.89\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.865\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.908\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e36.8\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMean (all industries)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.947\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.894\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.860\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.877\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.917\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e33.9\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003e6.2 KPI Improvements Post-BEGF Deployment\u003c/h2\u003e \u003cp\u003eA comparison of the five operational KPIs before and after BEGF deployment across all the case study industries is presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e. The mean improvements include a traceability score of +\u0026thinsp;66%, a compliance rate of +\u0026thinsp;26 percentage points, a supply disruption rate of \u0026minus;\u0026thinsp;38.5%, an annual cost savings of USD 13.8 M, and a lead-time reduction of 29.6% (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eAggregate KPI Improvement Summary (Mean across Five Cases)\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKPI Dimension\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePre-BEGF Mean\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePost-BEGF Mean\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eChange\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ep value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTraceability Score (0\u0026ndash;100)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e52.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e88.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e+\u0026thinsp;69.2%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCompliance Rate (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e69.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e96.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e+\u0026thinsp;38.5%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDisruption Rate (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e12.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e37.6 (red.)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;38.5%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAnnual Cost Savings (USD M)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBaseline\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e+\u0026thinsp;13.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e+\u0026thinsp;21.3%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLead-Time Reduction (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBaseline\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e+\u0026thinsp;29.6%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e+\u0026thinsp;29.6%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003e6.3 Governance Heatmap and Technology Adoption Trends\u003c/h2\u003e \u003cp\u003eThe governance performance by region and KPI dimension is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e(a). East Asia has the highest overall governance scores (mean: 90.2), driven by advanced IoT infrastructure and regulatory alignment with blockchain-native trade documentation standards. Sub-Saharan Africa shows the greatest improvement potential, with scores averaging 53.3, highlighting an infrastructure investment priority for global supply chain equity. The data in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e(b) confirm the acceleration of blockchain technology adoption across all dimensions from 2021 to 2025, with IoT integration (91%) and traceability smart contracts (87%) leading the adoption curves.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003e6.4 Blockchain Platform Performance Benchmarking\u003c/h2\u003e \u003cp\u003eThe longitudinal platform performance for Hyperledger Fabric, Ethereum (public), and Corda across the 36-month study window is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e. Hyperledger Fabric consistently dominated in terms of throughput (mean: 3,240 TPS; peak: 4,800 TPS) and latency (mean: 0.62 seconds). The smart contract upgrade in June 2023 (marked in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ea) produced a statistically significant 18% throughput improvement (Wilcoxon signed-rank test, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). These benchmarks confirm that Hyperledger Fabric is the technically superior platform for enterprise supply chain deployments in terms of transaction volume.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"7. Managerial Implications and Decision Support","content":"\u003cp\u003eThe framework provides managers with real-time, data-driven decision support by integrating Hyperledger Fabric for secure information sharing across supply chain partners. The use of machine learning enables proactive risk management, demand forecasting, and anomaly detection. Multiobjective optimization supports strategic trade-offs between cost, resilience, and sustainability in complex global networks. Overall, it enhances governance efficiency, coordination, and informed decision-making in dynamic and uncertain environments.\u003c/p\u003e \u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003ch2\u003e7.1 Prescriptive Decision-Support Matrix\u003c/h2\u003e \u003cp\u003eOn the basis of BEGF deployment findings, we synthesize a prescriptive decision-support matrix (Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e) that maps organisational maturity levels and supply chain complexity profiles to recommended BEGF implementation configurations. This matrix provides chief supply chain officers (CSCOs), chief digital officers (CDOs), and procurement directors with a structured pathway for phased adoption.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePrescriptive Decision-Support Matrix for BEGF Deployment\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMaturity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSupply Complexity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eRecommended Config.\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePriority BEGF Layers\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eExpected ROI Horizon\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNascent (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLow (\u0026lt;\u0026thinsp;50 suppliers)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eLayer 1\u0026ndash;2 Pilot\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eIoT\u0026thinsp;+\u0026thinsp;Ledger\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e18\u0026ndash;24 months\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDeveloping (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMedium (50\u0026ndash;200)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eLayers 1\u0026ndash;3 Full\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e+ Smart Contracts\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e12\u0026ndash;18 months\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAdvanced (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHigh (200\u0026ndash;500)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFull BEGF Stack\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAll Layers\u0026thinsp;+\u0026thinsp;ML\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e9\u0026ndash;12 months\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLeader (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eVery High (\u0026gt;\u0026thinsp;500)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFull\u0026thinsp;+\u0026thinsp;Ecosystem\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAll +\u0026thinsp;Cross-org\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e6\u0026ndash;9 months\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec25\" class=\"Section2\"\u003e \u003ch2\u003e7.2 Implications for CSCOs and Procurement Leaders\u003c/h2\u003e \u003cp\u003e \u003cb\u003eProactive Risk Governance\u003c/b\u003e: The BMRS system allows one to move from reactive (after a disruption) to proactive (before a disruption) governance. As part of their weekly supply review meetings, procurement leaders should set up BMRS monitoring dashboards and treat BMRS alerts with the same level of operational urgency as inventory shortage signals.\u003c/p\u003e \u003cp\u003e \u003cb\u003eSupplier relationship management\u003c/b\u003e: BEGF's clear scoring system encourages suppliers to act in ways that are good for business. For example, suppliers with a consistently low BMRS obtain better order placement, faster payment, and coinvestment in capacity expansion. This gamification of governance performance has made supplier relationships less hostile in all five case studies.\u003c/p\u003e \u003cp\u003eThe compliance oracle chaincode module for regulatory compliance automation cuts the number of compliance staff needed by an average of 3.4 full-time equivalents (FTEs) per business and raises the rate of audit acceptance from 84% to 98%. Legal and compliance teams should work with IT to set up rule sets in the chaincode layer that are specific to each jurisdiction.\u003c/p\u003e \u003cp\u003e \u003cb\u003eCapital Efficiency\u003c/b\u003e: Because there is less information asymmetry, compared with regular working capital facilities, blockchain-verified inventory positions allow suppliers to obtain supply chain financing at 180\u0026ndash;240 basis points. Finance directors should talk to their banking partners about using on-chain receivable financing protocols.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec26\" class=\"Section2\"\u003e \u003ch2\u003e7.3 Implications for Technology Architects and Platform Designers\u003c/h2\u003e \u003cp\u003e \u003cb\u003eChoosing a platform\u003c/b\u003e: Hyperledger Fabric's permissioned architecture strikes the best balance between throughput (\u0026gt;\u0026thinsp;3,000 TPS), data privacy (channel isolation), and regulatory acceptance for enterprise supply chains. Public blockchains such as Ethereum are still not good for high-volume B2B operations because they have problems with throughput and gas fees that change all the time.\u003c/p\u003e \u003cp\u003e \u003cb\u003eInteroperability\u003c/b\u003e: The greatest technical problem that was seen was getting Hyperledger Fabric (used by four case study companies) and Corda (used by Zara) to work together across platforms. Architects should use the InterWork Alliance's Token Taxonomy Framework and ensure that cross-chain bridge protocols follow the W3C Decentralized Identifiers (DID) standards.\u003c/p\u003e \u003cp\u003e \u003cb\u003eScalability Roadmap\u003c/b\u003e: Companies should expect a 10\u0026times; increase in transaction volume within 24 months of full deployment because of the growing number of IoT sensors. According to our trend analysis (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eb), ZKP adoption is expected to reach 51% by 2025. The architecture should implement sharding at the channel level and consider ZKP implementations for privacy-preserving analytics.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec27\" class=\"Section2\"\u003e \u003ch2\u003e7.4 Implications for Policymakers and Regulators\u003c/h2\u003e \u003cp\u003eThe empirical evidence from this study supports several policy recommendations. First, regulators should formally recognize \u003cem\u003eblockchain-attested compliance records\u003c/em\u003e as legally equivalent to paper-based audit trails in supply chain regulatory frameworks, reducing duplicative compliance burdens. Second, national trade facilitation agencies should invest in blockchain single-window interoperability APIs, as demonstrated by the 23,000 person-hour reduction achieved in the Samsung case. Third, multilateral bodies (WTO, OECD, and G20) should develop \u003cem\u003ecross-border blockchain data governance standards\u003c/em\u003e to prevent the emergence of fragmented national blockchain protocols that would balance global supply chain data.\u003c/p\u003e \u003c/div\u003e"},{"header":"8. Conclusion","content":"\u003cp\u003eIn this paper, the \u003cb\u003eBlockchain-Enabled Governance Framework (BEGF)\u003c/b\u003e, a novel four-layer integrated system combining hyperledger fabric distributed ledger technology, an ensemble XGBoost\u0026ndash;RF\u0026ndash;LSTM machine learning risk engine, and a biobjective LP\u0026ndash;GA hybrid optimizer for the comprehensive governance of global supply networks, are presented. Validated across five longitudinal industrial case studies encompassing 1,840 supplier nodes and 8.3 M blockchain transactions, BEGF demonstrates statistically significant and economically material improvements across all five governance KPIs measured.\u003c/p\u003e \u003cp\u003eThis study advances supply chain management theory by providing an empirically grounded instantiation of algorithmic governance\u0026mdash;a third governance modality complementing traditional formal and relational mechanisms. Methodologically, the integration of blockchain-derived on-chain features as inputs to an ensemble ML pipeline represents a novel paradigmatic contribution with direct implications for predictive analytics research in supply chain management.\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eLimitations and Future Research\u003c/strong\u003e \u003cp\u003eThis study is subject to NDA-driven constraints on dataset disclosure, preventing full open-science replication. Future research should explore (i) the application of federated learning to enable cross-enterprise model training without data sharing; (ii) the incorporation of satellite imagery and alternative data into the BMRS feature set; (iii) the extension of BEGF to humanitarian supply chains and development logistics; and (iv) the longitudinal governance effects of BEGF on interfirm trust and relational capital formation, which the current quantitative methodology does not adequately capture. The anticipated maturation of zero-knowledge proof implementations and quantum-resistant cryptography standards will further enhance the BEGF's security and privacy capabilities in the medium term.\u003c/p\u003e \u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData availability statement:\u0026nbsp;\u003c/strong\u003eThe data supporting the findings of this study are available in the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests:\u0026nbsp;\u003c/strong\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding:\u0026nbsp;\u003c/strong\u003eThis research received no external funding.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;contributions:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAuthor 1 conceived the study and developed the methodology. Authors 2-3 performed the analysis and drafted the manuscript. Authors 3,4,5 reviewed and edited the manuscript. All the authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDeclaration of generative AI and AI-assisted technologies in the manuscript preparation process:\u0026nbsp;\u003c/strong\u003eThe authors have not used any generative AI or AI-assisted technologies in manuscript preparation.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to Participate\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eDeclaration\u003c/strong\u003e\u003cstrong\u003e:\u003c/strong\u003e Not applicable\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to Publish declaration:\u003c/strong\u003e Not applicable\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAndroulaki, E., Barger, A., Bortnikov, V., et al. (2018). Hyperledger Fabric: A distributed operating system for permissioned blockchains. In \u003cem\u003eProceedings of the 13th EuroSys Conference\u003c/em\u003e (pp. 1\u0026ndash;15). ACM. https://doi.org/10.1145/3190508.3190538\u003c/li\u003e\n\u003cli\u003eBarney, J. (1991). Firm resources and sustained competitive advantage. \u003cem\u003eJournal of Management\u003c/em\u003e, 17(1), 99\u0026ndash;120. https://doi.org/10.1177/014920639101700108\u003c/li\u003e\n\u003cli\u003eBreiman, L. (2001). Random forests. \u003cem\u003eMachine Learning\u003c/em\u003e, 45(1), 5\u0026ndash;32. https://doi.org/10.1023/A:1010933404324\u003c/li\u003e\n\u003cli\u003eCarbonneau, R., Laframboise, K., \u0026amp; Vahidov, R. (2008). Application of machine learning techniques for supply chain demand forecasting. \u003cem\u003eEuropean Journal of Operational Research\u003c/em\u003e, 184(3), 1140\u0026ndash;1154. https://doi.org/10.1016/j.ejor.2006.12.004\u003c/li\u003e\n\u003cli\u003eChen, T., \u0026amp; Guestrin, C. (2016). XGBoost: A scalable tree boosting system. In \u003cem\u003eProceedings of the 22nd ACM SIGKDD Conference\u003c/em\u003e (pp. 785\u0026ndash;794). ACM. https://doi.org/10.1145/2939672.2939785\u003c/li\u003e\n\u003cli\u003eChoi, T. M., Feng, L., \u0026amp; Li, R. (2020). Information disclosure structure in supply chains with blockchain technology. \u003cem\u003eProduction and Operations Management\u003c/em\u003e, 29(6), 1396\u0026ndash;1416. https://doi.org/10.1111/poms.13147\u003c/li\u003e\n\u003cli\u003eDyer, J. H., \u0026amp; Singh, H. (1998). The relational view: Cooperative strategy and sources of interorganizational competitive advantage. \u003cem\u003eAcademy of Management Review\u003c/em\u003e, 23(4), 660\u0026ndash;679. https://doi.org/10.5465/amr.1998.1255632\u003c/li\u003e\n\u003cli\u003eFrancisco, K., \u0026amp; Swanson, D. (2018). The supply chain has no clothes: Technology adoption of blockchain for supply chain transparency. \u003cem\u003eLogistics\u003c/em\u003e, 2(1), 2. https://doi.org/10.3390/logistics2010002\u003c/li\u003e\n\u003cli\u003eGovindan, K., Fattahi, M., \u0026amp; Keyvanshokooh, E. (2017). Supply chain network design under uncertainty: A comprehensive review and future research directions. \u003cem\u003eEuropean Journal of Operational Research\u003c/em\u003e, 263(1), 108\u0026ndash;141. https://doi.org/10.1016/j.ejor.2017.04.009\u003c/li\u003e\n\u003cli\u003eIvanov, D., Dolgui, A., \u0026amp; Sokolov, B. (2019). The impact of digital technology and Industry 4.0 on the ripple effect and supply chain risk analytics. \u003cem\u003eInternational Journal of Production Research\u003c/em\u003e, 57(3), 829\u0026ndash;846. https://doi.org/10.1080/00207543.2018.1488086\u003c/li\u003e\n\u003cli\u003eKshetri, N. (2018). Blockchain\u0026apos;s roles in meeting key supply chain management objectives. \u003cem\u003eInternational Journal of Information Management\u003c/em\u003e, 39, 80\u0026ndash;89. https://doi.org/10.1016/j.ijinfomgt.2017.12.005\u003c/li\u003e\n\u003cli\u003eLumineau, F., Wang, W., \u0026amp; Schilke, O. (2021). Blockchain governance\u0026mdash;A new way of organizing collaborations? \u003cem\u003eOrganization Science\u003c/em\u003e, 32(2), 500\u0026ndash;521. https://doi.org/10.1287/orsc.2020.1379\u003c/li\u003e\n\u003cli\u003eMIT Resilience Lab. (2023). \u003cem\u003eGlobal Supply Chain Database 2020\u0026ndash;2023\u003c/em\u003e. Massachusetts Institute of Technology. https://doi.org/10.7910/DVN/XXRQMP\u003c/li\u003e\n\u003cli\u003eMula, J., Peidro, D., D\u0026iacute;az-Madro\u0026ntilde;ero, M., \u0026amp; Vicens, E. (2010). Mathematical programming models for supply chain production and transport planning. \u003cem\u003eEuropean Journal of Operational Research\u003c/em\u003e, 204(3), 377\u0026ndash;390. https://doi.org/10.1016/j.ejor.2009.09.008\u003c/li\u003e\n\u003cli\u003eNakamoto, S. (2008). \u003cem\u003eBitcoin: A peer-to-peer electronic cash system\u003c/em\u003e. https://doi.org/10.5195/ledger.2015.10\u003c/li\u003e\n\u003cli\u003eSaberi, S., Kouhizadeh, M., Sarkis, J., \u0026amp; Shen, L. (2019). Blockchain technology and its relationships to sustainable supply chain management. \u003cem\u003eInternational Journal of Production Research\u003c/em\u003e, 57(7), 2117\u0026ndash;2135. https://doi.org/10.1080/00207543.2018.1533261\u003c/li\u003e\n\u003cli\u003eWeill, P., \u0026amp; Ross, J. W. (2004). \u003cem\u003eIT Governance: How Top Performers Manage IT Decision Rights for Superior Results\u003c/em\u003e. Harvard Business School Press.\u003c/li\u003e\n\u003cli\u003eWilliamson, O. E. (1975). \u003cem\u003eMarkets and Hierarchies: Analysis and Antitrust Implications\u003c/em\u003e. Free Press.\u003c/li\u003e\n\u003cli\u003eWorld Bank. (2023). \u003cem\u003eLogistics Performance Index 2023\u003c/em\u003e. World Bank Group. https://doi.org/10.1596/978-1-4648-2090-2\u003c/li\u003e\n\u003cli\u003eWorld Bank. (2023). \u003cem\u003eWorld Development Report 2023: Migrants, Refugees, and Societies\u003c/em\u003e. World Bank Group. https://doi.org/10.1596/978-1-4648-1941-8\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Blockchain, Supply chain governance, Machine learning, Multiobjective optimization, Smart contracts, Risk management, Distributed ledger, Industry 4.0","lastPublishedDoi":"10.21203/rs.3.rs-9326002/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9326002/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eIt is becoming more difficult for global supply chain networks to be governed because of issues such as a lack of transparency, data fragmentation, regulatory divergence, and the likelihood of having more than one supplier. This paper introduces an innovative Blockchain-Enabled Governance Framework (BEGF) that amalgamates distributed ledger technology with an ensemble machine learning (ML) risk-scoring engine and a multiobjective linear programming (LP) optimizer to enable supply chain decisions that are transparent, real-time, and verifiable. We use the framework in five long-term industrial case studies: automotive (Toyota), pharmaceutical (Merck), electronics (Samsung), apparel (Zara), and food and drink. These studies include 1,840 supplier nodes in 37 countries over 36 months. The BEGF can predict disruptions with an average AUC of 0.947. It can also reduce the number of supply disruptions by 38.5% and save each business $13.8 million a year. The Pareto-optimal optimizer lowers costs all along the supply chain.\u003c/p\u003e","manuscriptTitle":"Blockchain-Enabled Governance Mechanisms in Global Supply Networks: A Hybrid Machine Learning and Multiobjective Optimization Framework","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-10 17:01:38","doi":"10.21203/rs.3.rs-9326002/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"6cd228c5-0a7f-4f64-bf9b-efb98c106c35","owner":[],"postedDate":"April 10th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":66079561,"name":"Management"},{"id":66079562,"name":"Entrepreneurship"}],"tags":[],"updatedAt":"2026-04-10T17:01:38+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-10 17:01:38","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9326002","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9326002","identity":"rs-9326002","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}