An Intelligent Solution for Improving Procurement Decision Efficiency and Transparency in University Procurement Management Based on Big Data Analytics

Abstract

The expansion of higher education has made university procurement systems increasingly complex, especially in the acquisition of high-value engineering equipment, digital infrastructure, and communication-related laboratory assets. Transparent and efficient p rocurement g overnance i s therefore important for supporting research platforms that rely on advanced sensing, network communication, and electromagnetic engineering equipment. This study develops a big data analytics solution to improve procurement decision efficiency and transparency in university procurement management. The proposed framework includes demand forecasting, intelligent supplier evaluation, anomaly detection and alerting, and end-to-end traceable oversight. A private Hyperledger Fabric blockchain with PBFT consensus was used to strengthen process traceability, while machine-learning models supported demand prediction and supplier risk evaluation. Empirical analysis across 20 representative universities and more than 15,000 procurement transactions shows that average procurement decision cycles were reduced by 50.8%, supplier evaluation accuracy reached 92.6%, the transparency score increased from 53.8 to 88.0, and anomaly detection achieved 90.9%. The results indicate that big-data-driven procurement systems can simultaneously improve efficiency, accountability, and institutional governance quality.