Predicción de retraso en los vuelos comerciales de Estados Unidos debido a las condiciones climáticas
Loading...
Official URL
Full text at PDC
Publication date
2024
Authors
Advisors (or tutors)
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
Citation
Abstract
En un mundo globalizado, la gestión eficiente de vuelos comerciales es fundamental debido a su impacto en la vida cotidiana. Con el clima volviéndose cada vez más cambiante, anticipar los retrasos en estos vuelos se vuelve esencial para mantener la eficiencia en el transporte aéreo. Este trabajo emplea minería de datos para analizar cómo las condiciones meteorológicas afectan los retrasos de vuelos comerciales en cuatro aeropuertos de Estados Unidos durante 2023. Utilizando técnicas de aprendizaje supervisado, se desarrolla un modelo predictivo que identifica patrones en los datos históricos de vuelos y clima. El objetivo es optimizar la gestión de vuelos y reducir costes, mejorando así la experiencia del pasajero mediante herramientas precisas para anticipar y manejar los efectos del clima en la puntualidad de los vuelos.
Los resultados muestran que el método de XGBoost es el más eficaz para predecir retrasos. Además, se encontró que las precipitaciones y las rachas de viento tienen el mayor impacto en los retrasos de vuelos, superando a otras condiciones climáticas en su efecto sobre la puntualidad.
In a globalized world, efficient management of commercial flights is essential due to its impact on daily life. With increasingly unpredictable weather patterns, anticipating flight delays becomes crucial on maintaining efficiency in air transport. This work uses data mining to analyze how weather conditions affect flight delays at four U.S. airports during 2023. By applying supervised learning techniques, a predictive model is developed to identify patterns in historical flight and weather data. The aim is to optimize flight management and reduce costs, thereby improving passenger experience through accurate tools to anticipate and manage the effects of weather on flight punctuality. The results show that the XGBoost method is the most effective for predicting delays. Additionally, it was found that precipitations and wind gusts have the greatest impacts on flight delays, surpassing other weather conditions in their effect on punctuality.
In a globalized world, efficient management of commercial flights is essential due to its impact on daily life. With increasingly unpredictable weather patterns, anticipating flight delays becomes crucial on maintaining efficiency in air transport. This work uses data mining to analyze how weather conditions affect flight delays at four U.S. airports during 2023. By applying supervised learning techniques, a predictive model is developed to identify patterns in historical flight and weather data. The aim is to optimize flight management and reduce costs, thereby improving passenger experience through accurate tools to anticipate and manage the effects of weather on flight punctuality. The results show that the XGBoost method is the most effective for predicting delays. Additionally, it was found that precipitations and wind gusts have the greatest impacts on flight delays, surpassing other weather conditions in their effect on punctuality.