Deep Learning and Blockchain-Driven Security for Pharmaceutical Supply Chains using Genetic Algorithm and SVM

The pharmaceutical cold chain is a specialised logistics subset characterised by distinct storage and transportation protocols. To enhance product credibility and safeguard public health, pharmaceutical items require reliable traceability throughout their entire life cycle. Cold chain products incur higher storage and shipping costs compared to standard supply chain products, thereby rendering the pharmaceutical cold chain costly. Accurate forecasting of cold chain demand is essential to maintain competitiveness and reduce transportation and warehousing costs. Data preparation encompasses scalability and transformation. Normalisation techniques such as min-max and z-score standardise data. The objective of feature selection is to reduce inaccuracy while choosing a limited number of features. The SVM serves as the fitness function for the GA, selecting the best representative variables for classification in our GA-SVM hybrid model. The testing results indicate that the SVGA algorithm achieves an accuracy of 98.78% in classification. The approach exceeds existing alternatives and is marginally more secure, as indicated by the research. This study highlights optimised data processing and accurate demand forecasting in the pharmaceutical cold chain supply chain. The hybrid GA-SVM model enhances the reliability and efficiency of the pharmaceutical cold chain by augmenting product traceability, reducing costs, and refining classification accuracy.