Intelligent privacy‐preserving data management framework for medicine supply chain system

Abstract In today's era, the pharmaceutical industry has integrated blockchain to secure the sensitive information of medicines, wherein public and private blockchains are used to preserve the security and privacy of the medicine supply chain data. However, conventional blockchains often limit scalability due to limited storage. Moreover, blockchain has loopholes; for example, it is not able to prove the validity of the data prior to being stored in the blockchain, which leads to fake data being added to the blockchain. As a result, it causes an issue of data provenance. Motivated by this, the proposed framework incorporated artificial intelligence (AI) algorithms to enhance the efficiency of the medicine supply chain data. The proposed framework integrated machine learning (ML) and blockchain, where ML classifies the valid and invalid data of the medicine supply chain, whereas blockchain stores only valid data and maintains its security and privacy. This identification helps the blockchain to verify medicine supply chain data before adding it to the blockchain. Additionally, we employed an InterPlanetary file system (IPFS) that saves medicine supply chain data and computes its hash to offer decentralized storage. Further, this hash data is stored on a private Hyperledger Fabric blockchain, which requires minimal storage instead of storing an entire large file. This minimal storage optimizes the process of data storage and retrieval in the Hyperledger Fabric blockchain, which enhances the scalability of the proposed framework. Finally, the result of the proposed framework is assessed in two phases: ML and blockchain, wherein the ML model's performance is measured by statistical measures and the blockchain‐based result is assessed using several performance parameters such as throughput is around (618 transactions per second), latency (0.12 s), response time (11 s) and data rate (282 Mbps).