AeroSafe: Hybrid Machine Learning-Based Pre-Takeoff Risk Prediction for Private Jets with Natural Language Explanations and Recommendations

DOI

10.2991/978-94-6239-693-7_90How to use a DOI?

Keywords

Aviation safety analytics; Explainable artificial intelligence; Gradient boosting; pre-takeoff risk prediction; Private jet safety; SMOTE; Supervised machine learning; XGBoost

Abstract

The conditions of heterogeneous aircrafts, dynamic environmental factors, and uncertainties related to humans are all involved in the operations of the private jets and pre-takeoff risk assessment is a complicated safety problem. This paper suggests AeroSafe, a hybrid guided machine learning model to predict the risks of pre-takeoff accidents and provide recommendations in the form of explainable advice in natural language. The system combines both the set of operational parameters, the environmental conditions, health indicators of the aircraft, and pilot-specific options into a hierarchical preprocessing pipeline such as the normalization, categorical encoding, and the use of SMOTE-based class balancing. Various classifiers, including Logistic Regression, Decision Tree, Random Forest, XGBoost and Multilayer Perceptron are also evaluated with a comparison on stratified cross-validation. As shown by experimental results, XGBoost results yield the highest accuracy of 92.4% and macro F1-score of 0.91 which creates balanced risk discrimination between low, medium, and high-risk groups. An explainable AI layer produces contextual safety explanation in conjunction with the feature importance scores. The framework suggested offers a deployable, interpretable, and proactive decision-support framework to the purpose of improving the aviation safety of the privatized aviation in the preflight departure phase.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - S. Swetha
AU  - R. Subashini
AU  - E. Murali
AU  - K. Punitha
AU  - Suzan Melissa Wilson
PY  - 2026
DA  - 2026/06/16
TI  - AeroSafe: Hybrid Machine Learning-Based Pre-Takeoff Risk Prediction for Private Jets with Natural Language Explanations and Recommendations
BT  - Proceedings of the International Conference on Intelligent Systems for a Sustainable Future (ISSF 2026)
PB  - Atlantis Press
SP  - 935
EP  - 943
SN  - 2589-4919
UR  - https://doi.org/10.2991/978-94-6239-693-7_90
DO  - 10.2991/978-94-6239-693-7_90
ID  - Swetha2026
ER  -