Hybrid PoA - DBFT with AI Based Predictive Risk Analytics for Interoperable Healthcare Supply Chains

Abstract

The increased complexity of global healthcare supply chains exacerbates the risks of counterfeit pharmaceuticals, breakdowns in the cold chain, and disruptions to supply in the logistics sector, and there is a need for more intelligent and trustworthy digital infrastructures. Existing blockchain systems struggle with providing performance and predictive risk management for real-time operations with limited scalability and analytical capabilities. This study presents an innovation to the challenges in the following ways: Propose an AI-enabled hybrid Proof-of-Authority and Delegated Byzantine Fault Tolerance (PoA-DBFT) blockchain framework for predictive risk management and end-to-end interoperability across heterogeneous healthcare systems. The methodology is a combination of multimodal IoT datasets, gradient boosted and deep learning-based anomaly detection, FHIR Compliant interoperability services and smart contract-based governance logic. Experimental evaluations with Medical Supply Chain Integrity Dataset illustrates that the proposed model has 95.8% risk detection accuracy, 0.972 AUC-ROC, 1780 TPS throughput and block finality of 1.2 seconds, which is better than five state-of-the-art baselines. The results prove the benefits of combining AI analytics with a hybrid consensus mechanism to substantially boost supply chain resilience, transparency and decision intelligence. The study concludes that this unified architecture offers a strong base for secure, scalable and predictive digital health ecosystems to make way for intelligent next-generation supply chain infrastructures.