Flight delays are the airline sector’s most serious concern because they disrupt airlines, passengers, and airports. This study examines the use of big data analytics with machine learning and business intelligence to improve flight delay prediction accuracy. The aim is to analyze and visualize flight delays using big data applied to all arriving and departing domestic flights from January 2019 to August 2023 in the United States by using big data analytics and business intelligence. In addition, it aims to predict flight delays and compare six classification algorithms using big data analytics. This analysis and visualization showed that Aircraft Arriving Late and Air Carrier Delay reasons have the highest percentage of delays compared to other reasons, followed by National Aviation System Delay and weather delay. The security delay has the lowest percentage. The year 2023 has the highest number of delays, and June has the highest number of delays. There are many prediction models for flight delays; however, more development of accurate prediction models is still needed. In this study, the selected models were “Random Forest classifier,” “Logistic Regression classifier,” “Neural Networks classifiers,” “Support Vector Machine (classifier),” “Gradient Boosting classifier,” and “K-Nearest Neighbors” classifier. The comparison used the resulting confusion matrix to summarize prediction results for the following classification problems: F1-score, accuracy, recall, and precision. The model performance evaluation results showed that the Support Vector Machine classifier yielded better results than other models, followed by the Neural Networks classifiers, K-Nearest Neighbors classifier, Random Forest classifier, and the Logistic Regression classifier showed the worst results.