A Blockchain and Machine Learning Integrated Hybrid System for Drug Supply Chain Management for the Smart Pharmaceutical Industry

Abstract

Over the past decade, pharmaceutical businesses have had challenges monitoring their goods across the supply chain, enabling counterfeiters to introduce fraudulent medications into the marketplace. Counterfeit pharmaceuticals are seen as a significant concern for the global pharmaceutical sector. Studies suggest that US pharmaceutical businesses suffer an annual commercial loss of within $200 billion from imitation pharmaceuticals. These medications do not promote patient recovery from the condition but include several hazardous adverse reactions. The World Health Organization (WHO) study data indicates that in underdeveloped nations, one in ten drugs used is fraudulent and of inferior quality. Therefore, a system capable of tracing and tracking medicine distribution at each stage is essential to address the issue of imitation. Blockchain Technology (BCT) can efficiently manage and monitor the Drug Supply Chain (DSC) process. This study proposes and implements an innovative DSC Management System (DSCMS) utilizing BCT and Machine Learning (ML) technologies. The suggested system has two primary components: a BCT-based DSC management engine and an ML-driven drug suggestion system for customers. The initial module implements a DSC monitoring framework utilizing Hyperledger Fabric, designed to monitor and track the drug distribution process inside the innovative pharmaceutical sector. The N-gram and LightGBM algorithms are employed in the ML component to propose the highest-rated medications to pharmaceutical clients. These models have been trained using a widely recognized publicly accessible drug assessment database supplied by an open-source ML library. The ML component is included in this BCT platform via the REST API. The research conducts many assessments to evaluate the efficacy and usefulness of the suggested solution.