AI-Driven Predictive Analytics for Proactive Supply Chain Risk Management and Resilience in Critical U.S. Industries

The supply chains of industries that are important to the U.S. economy are moving through higher- and higher-stakes environments that feature transportation downturns, unreliable suppliers, uneven inventory levels, traffic jams, weather, port blockages and shifting delivery constraints. Traditional supply chain risk management (SCRM) methods tend to be reactive in nature, looking for operational issues once they have already occurred that are causing problems with the delivery. A promising framework for proactive risk management in supply chains and supply chain resilience is proposed in this research, which is based on the use of machine learning algorithms for the identification and classification of potential logistic risks prior to the emergence of the significant logistic risks. This study employs a dataset of logistics and supply chain information for operations in Southern California including transportation, warehouse, route planning, supplier, environmental, and real-time monitoring information. Key operational parameters like ETA variance, congestion, warehouse inventory, supplier reliability, lead time, weather intensity, port congestion, route risk, driver behavior, driver fatigue monitoring, and cargo conditions are included in the data set. The data preprocessing, exploratory analysis, feature selection, and comparative machine learning modeling are suggested methodologies for the analytical approach. Risk Classification (Low Risk, Moderate Risk, and High Risk) is suggested as the key predictive outcome, and Disruption Likelihood Score, Delay Probability, and Delivery Time Deviation are complementary indicators of the vulnerability of the supply chain and the operational performance. Machine learning models like logistic regression, decision tree, random forest, support vector machine and gradient boosting can be assessed by means of accuracy, precision, recall, F1-score, ROC-AUC and confusion matrix. The framework aims to help identify key risk factors and provide early warning insights into the decision making process for proactive routing, inventory planning, supplier management, transportation scheduling and contingency planning. The dataset's public applications are the following: predictive risk assessment and detection of disruptions, optimization of routes and schedules, analysis of external factors, warehouse and inventories. The goal of this study is to illustrate the role predictive AI can play in enhancing readiness within an organization by transforming the way the supply chain manages disruption from a reactive to a proactive approach, anticipating issues, intervening early, and fostering resilience in decision-making. The results will be used to gain practical insights into logistics operations, and to inform development of a comprehensive plan for increasing supply chain resilience across key U.S. industries.