Data-Driven Predictive Analytics for Multi-Flight Departure Interdependencies in UAE Airspace

Multiple scheduled flights to the same destination might have variable departure times, which can affect arrival times and cause delays in flights. This study examines how departure time variations affect individual flight performance and the cascade effects on other flight operations by using predictive analytics and utilizing a significant quantity of historical flight data. Advanced data-driven models are used in the research, such as XGBoost, which demonstrated excellent prediction accuracy for on-time departures with an R-squared value of 0.99, RMSE of 9.71 minutes, and MAE of 6.92 minutes. The significance of this study lies in its comprehensive understanding of the interconnected nature of flight operations in a busy airspace like that of the United Arab Emirates(U.A.E) By offering a data-driven foundation for improving operational efficiency and reducing environmental impact, this research contributes valuable strategies for air traffic management, promoting a more sustainable future for the aviation industry.