Blockchain-enabled Supplier Risk Scoring for Sustainable Procurement

Thesis Abstract

This study investigates how blockchain-enabled supplier risk scoring can advance sustainable procurement by enhancing transparency, traceability, and data integrity across supply networks facing environmental, social, and governance (ESG) challenges. The problem addressed is the persistent opacity and data fragmentation in supplier risk assessment, which undermines steady access to credible ESG information, increases the likelihood of supplier default or non-compliance, and impedes evidence-based decision-making in sustainable procurement. The aim is to develop and validate a Blockchain-enabled Supplier Risk Scoring (B-SRS) framework that integrates multi-source ESG indicators with immutable transaction records to produce timely, auditable risk scores for strategic supplier selection and ongoing monitoring. Specific objectives are (1) to design a data-architecture model that harmonizes internal procurement data with external ESG datasets and on-chain activity; (2) to construct a risk scoring algorithm combining machine learning and blockchain verification to produce dynamic risk scores; (3) to evaluate the predictive validity of B-SRS against historical supplier performance and ESG incidents; (4) to assess user acceptance and governance implications of deploying blockchain-based risk scoring in procurement decision processes; and (5) to formulate guidelines for scaling the framework across heterogeneous supply networks. Methodologically, the study adopts a mixed-methods, explanatory sequential design. The research uses a cross-sectional survey of 180 procurement professionals from manufacturing and retail sectors to capture perceptions of risk indicators, data quality, and usability of blockchain-based scoring. In parallel, a longitudinal dataset comprising 24 months of supplier transaction records, ESG disclosures, and incident logs from 60 suppliers will be assembled to train and test the scoring model. Data collection instruments include a structured questionnaire validated for construct reliability (Cronbach’s alpha > 0.8) and semi-structured interviews with 20 procurement managers to elicit contextual insights on governance and implementation barriers. The on-chain data will be extracted from a permissioned blockchain deployed using Hyperledger Fabric, supplemented by off-chain ESG metadata stored in secure, auditable distributed storage. The analysis employs (i) descriptive statistics and correlation analysis to characterize data quality and indicator relevance; (ii) feature engineering and supervised learning (logistic regression, random forest, and gradient boosting) to derive a composite risk score, with model selection guided by AUC, F1-score, and calibration metrics; (iii) time-series analysis to assess score stability and predictive lead time; (iv) regression analysis to link risk scores with supplier performance outcomes (delivery reliability, quality incidents, ESG violations) over a 12-month horizon; and (v) thematic analysis of interview data to identify governance, trust, and adoption factors, triangulated with quantitative results. Validity and reliability are ensured through pilot testing, back-translation for survey instruments, k-fold cross-validation, and robustness checks against alternative feature sets. The study also integrates a theoretical lens from the Resource-Based View (RBV) and Stakeholder Theory to explain how blockchain-enabled risk visibility creates competitive capability and enhances stakeholder alignment. Anticipated findings include (a) improved predictive accuracy of supplier risk with B-SRS over traditional risk scoring methods, (b) evidence that immutable, verifiable ESG data reduces information asymmetry and accelerates proactive risk management, and (c) positive user perceptions of transparency, while identifying governance controls required to mitigate data privacy and permissioning concerns. The contribution to knowledge encompasses (i) a novel, implementable B-SRS architecture that fuses on-chain verifiability with off-chain ESG indicators; (ii) an empirically validated risk scoring model with demonstrable predictive validity and operational metrics; and (iii) an integrative governance framework outlining standards for data ownership, privacy, access rights, and auditability in blockchain-assisted procurement. The study concludes that blockchain-enabled supplier risk scoring can meaningfully enhance sustainable procurement by delivering timely, credible risk signals that drive strategic supplier selection and continuous monitoring. Recommendations include developing industry-wide ESG data interfaces, establishing interoperability standards for permissioned blockchains, and creating training programs to cultivate procurement professionals’ competencies in blockchain-enabled decision analytics.