Integrating Machine Learning and Explainable AI for Decision Support in Supply Chain Management

In modern industries, supply chains are becoming more complicated by introducing significant challenges such as supplier risks, quality control issues, and inventory inefficiencies, and it is necessary to have a strong solution to minimize the risk and optimize the operations so that industries can maintain competitiveness and ensure customer satisfaction in these dynamic and complex business environments. This research presents a comprehensive analysis of supply chain management using a synthetic dataset which contains key metrics of supply chains by applying statistical testing. Furthermore, several machine learning models such as logistic regression, random forest classifier, support vector,achine (SVM), extreme gradient boosting (XGBoost) classifier, and many more models, are employed to predict defect rates to prevent unwanted production loss. Our findings demonstrate that random forest classifier and XGBoost performed effectively with 96% and 96% accuracy, respectively. Explainable AI (XAI) has also been introduced in this article to uncover the contribution of each feature to a model’s prediction and explain individual predictions for better interpretability. This article contributes to the growing field of supply chain management by seamlessly integrating artificial intelligence.