This study explores the prediction of flight departure delays using data-driven approaches, focusing on the impact of weather conditions. The research addresses the growing demand for air travel and the frequent occurrence of flight delays due to abnormal weather patterns. The study collects datasets from three major international airports—Incheon International Airport (ICN) in South Korea, John F. Kennedy International Airport (JFK) in the United States, and Chicago Midway International Airport (MDW)—spanning over 10 years. The datasets include flight information at six different time intervals (2, 4, 8, 16, 24, and 48 hours) before departure. Various machine learning and deep learning models, such as Decision Tree, Random Forest, Support Vector Machine, K-nearest neighbors, Logistic Regression, Extreme Gradient Boosting, and Long Short-Term Memory (LSTM), are employed to predict flight delays. The models achieve accuracy rates of 0.749 for ICN, 0.852 for JFK, and 0.785 for MDW in 2-hour predictions, and 0.748, 0.846, and 0.772 for 48-hour predictions, respectively. The study also discusses the implications and future research directions, highlighting the importance of long-term delay predictions for international flights. The findings suggest that the proposed models can effectively predict flight delays over extended time periods, contributing to better airport resource management and enhancing the reliability of air travel.This study explores the prediction of flight departure delays using data-driven approaches, focusing on the impact of weather conditions. The research addresses the growing demand for air travel and the frequent occurrence of flight delays due to abnormal weather patterns. The study collects datasets from three major international airports—Incheon International Airport (ICN) in South Korea, John F. Kennedy International Airport (JFK) in the United States, and Chicago Midway International Airport (MDW)—spanning over 10 years. The datasets include flight information at six different time intervals (2, 4, 8, 16, 24, and 48 hours) before departure. Various machine learning and deep learning models, such as Decision Tree, Random Forest, Support Vector Machine, K-nearest neighbors, Logistic Regression, Extreme Gradient Boosting, and Long Short-Term Memory (LSTM), are employed to predict flight delays. The models achieve accuracy rates of 0.749 for ICN, 0.852 for JFK, and 0.785 for MDW in 2-hour predictions, and 0.748, 0.846, and 0.772 for 48-hour predictions, respectively. The study also discusses the implications and future research directions, highlighting the importance of long-term delay predictions for international flights. The findings suggest that the proposed models can effectively predict flight delays over extended time periods, contributing to better airport resource management and enhancing the reliability of air travel.