Leveraging Business Analytics to Enhance Supply Chain Resilience and Reduce Disruptions in Critical U.S. Industries

Abstract

The growing complexity of world supply chains, alongside the growing demand instability and uncertainty in operations has intensified the necessity to use data-driven solutions that help to build resilience and minimize disruptions in critical sectors. The study analyses the potential of using business analytics to enhance supply chain resilience in America by basing on the DataCo Smart Supply Chain Dataset as the empirical evidence. The data, which is over 180,000 records at the transaction level, offers detailed information about logistics operation, consumer behavior, the use of payment modes, product categories and delivery performance. This type of multidimensional information provides a one of a kind chance to observe the correlation between delivery delays, operational risks, the change in demand, and frauds. The predictive modeling, clustering methods, and statistical analysis methods have been used to determine the important factors leading to supply chain fragility in the study. In particular, the demand forecasting, risk classification of late-delivery, and fraudulent orders are predicted with the help of machine learning models like Gradient Boosting, Random Forest, Logistic Regression, and LightGBM. These models allow finding the patterns that can be overlooked by traditional means of analysis, such as the correlation between the geographic distribution of the customers, the channel of payment, and the shipping delays. Moreover, clustering algorithms like K-Means and DBSCAN can give more information regarding a behavioral pattern of fraud and the inconsistencies of its operations. This study includes explainability tools, such as LIME and FairML, which makes the behavior of models more transparent such that predictive insights can be interpreted and made operationally feasible. The findings indicate that the key variables of critical concern include shipping mode, type of transaction, demand cycle seasonality and customer location that contribute greatly to resilience and risk. This study is a part of an increasing literature on digital supply chain transformation because it proves how business analytics facilitate proactive decision-making, operational planning optimization, and disruption relief in the U.S. industries. The results have strategic implications on supply chain managers, policymakers, and organizations that strive to develop flexible, information-driven, and resilient supply chain systems.