AI-Based Prediction of A-CDM Variable Taxi-Time in Low-Data Airports

Thumbnail Image

Files

Durubi_Hala_2025.pdf (1.15 MB)

URL

Journal Title

Journal ISSN

Volume Title

Insinööritieteiden korkeakoulu | Bachelor's thesis
Unless otherwise stated, all rights belong to the author. You may download, display and print this publication for Your own personal use. Commercial use is prohibited.

Authors

Date

2025-09-05

Department

Major/Subject

Computational Engineering

Mcode

ENG3082

Degree programme

Aalto Bachelor’s Programme in Science and Technology

Language

en

Pages

38+11

Series

Abstract

This thesis assesses the performance of machine learning models on the task of variable taxi-time (VTT) prediction as part of airport collaborative decision making (A-CDM) in a low-data airport environment. Based on operational data from the Addis Ababa Bole International Airport in Ethiopia, random forest, histogram-based gradient boosting, and CatBoost were used to predict taxi-in (EXIT) and taxi-out (EXOT) times. While the models obtained reasonable average errors (~2–5 min) for both taxi-in and taxi-out time predictions, their explanatory power was low (R² < 0.1), which reflected the limitation of the use of static metadata only. Feature importance analysis revealed that stand location and aircraft type were the main predictors, with a strong contribution from the hour of operation feature. For airports with limited data, this study sets a performance standard and highlights that future improvements in VTT prediction rely on including more detailed and dynamic data sources.

Description

Supervisor

St-Pierre, Luc

Thesis advisor

Musharraf, Mashrura

Keywords

airport collaborative decision making (A-CDM), airport operations, CatBoost, histogram-based gradient boosting, low-data environment, machine learning

Other note

Citation

Permanent link to this item

https://urn.fi/URN:NBN:fi:aalto-202510218391

Collections

[kand] Insinööritieteiden korkeakoulu / ENG