CBL - Campus del Baix Llobregat

Projecte llegit

Títol: Aplicació de Machine Learning per a la predicció de pèrdua de maletes facturades en el SATE

Estudiants que han llegit aquest projecte:

Director/a: FORNÉS MARTÍNEZ, HECTOR

Departament: FIS

Títol: Aplicació de Machine Learning per a la predicció de pèrdua de maletes facturades en el SATE

Data inici oferta: 05-02-2024     Data finalització oferta: 05-10-2024


Estudis d'assignació del projecte:
    DG ENG AERO/TELEMÀT
Tipus: Individual
 
Lloc de realització: EETAC
 
Paraules clau:
Machine Learning, Vols de connexió, Mínim Temps de Connexió, Equipatge facturat, SATE, retard
 
Descripció del contingut i pla d'activitats:
Aquest TFG pretén utilitzar la inteligència artificial coneguda com Machine Learning amb la finalitat de crear una aplicació que permeti preveure la possibilitat de perdre una maleta facturada en el recorregut del SATE.
 
Overview (resum en anglès):
This thesis presents a Machine Learning application for the prediction of the loss of checked luggage in connection flights, using a Random Forest model.
Baggage management is a critical issue for airports and airlines, especially on flights with short connections where the risk of equipment loss is higher. This work has focused on developing a predictive model capable of identifying the flights with the highest risk of luggage loss, based on the analysis of real historical data collected from the Josep Tarradellas Barcelona-El Prat airport.
The model has been designed to detect patterns in the delay of connection flights by means of the difference between the estimated time and the real time of aircraft landing. Through a detailed analysis and a rigorous training and testing process, it has been demonstrated that the Random Forest model can predict with high accuracy the probability of equipment loss depending on the observed operating conditions.
The results obtained show that the model is capable of correctly detecting flights with a high risk of equipment loss, which makes it possible to implement preventive measures in advance, such as prioritising equipment in situations of short connection times or redistributing resources in times of operational delays.

© CBLTIC Campus del Baix Llobregat - UPC