Explainable Predictive Maintenance for Aircraft Systems

This project focuses on helping engineers understand and trust the results produced by artificial intelligence systems used to monitor aircraft health. Modern aircraft generate large amounts of data, and AI models can detect unusual behavior or potential faults, but their outputs are often difficult for non-AI specialists to interpret. In this project, the interns will develop methods to (1) simulate realistic fault scenarios when real fault data are limited, (2) structure diagnostic reasoning using engineering fault logic, and (3) translate AI model outputs into clear, plain- English explanations that follow how aircraft systems work. The expected benefit to the partner organization is improved confidence in AI-based maintenance tools, better alignment between AI results and engineering judgment, and stronger decision support for ma

Faculty Supervisor:

Soumaya Yacout

Student:

Partner:

Bombardier Inc

Discipline:

Engineering

Sector:

Manufacturing; Transportation and warehousing

University:

Polytechnique Montréal

Program:

Accelerate

Current openings

Find the perfect opportunity to put your academic skills and knowledge into practice!

Find Projects