Food supply chains (FSC) in the United States face growing disruptions from production delays, transportation bottlenecks, regulatory changes, labor shortages, and market uncertainty. These challenges increase the need for reliable methods to assess and strengthen supply chain resilience. This study aims to predict FSC resilience and identify the factors that most strongly influence resilient performance. To address this objective, six key predictors are examined, including supplier reliability, production flexibility, digitalization index, financial stability, transportation risk, and geopolitical risk. Seven machine learning (ML) models, including Logistic Regression (LR), K Nearest Neighbors (KNN), Support Vector Machine (SVM), Random Forest (RF), AdaBoost, XGBoost, and Artificial Neural Networks (ANN), are trained and compared using hyperparameter tuning and cross-validation. Results show that ensemble models perform best, with XGBoost achieving the strongest overall predictive performance. The findings support contingency planning, resource allocation, resilience improvement, and food security decision-making across the supply chain.