This project investigates the impact of weather conditions on flight delays, aiming to enhance airline and airport operational efficiency through improved weather prediction and management.
- Examine the effect of various weather conditions on flight delays.
- Develop predictive models to forecast flight delays based on weather data.
- Flight information from the U.S. Department of Transportation. [https://www.transtats.bts.gov]
- Hourly weather data from the open-meteo API. [https://open-meteo.com/]
The project employs multiple modeling techniques, including Linear Regression, XGBoost, and LightGBM, with a focus on ensemble methods for improved accuracy. Model performance is evaluated using RMSE and R² metrics.
Our findings indicate that models like XGBoost and LightGBM, which account for non-linear relationships, significantly outperform linear models in predicting flight delays when incorporating weather data.