Automated Data-Driven Flight Scoring

Traditionally, lessons learned from prior accidents have driven advancements in aviation safety. With recent technological advancements and the resulting availability of flight data, the approach to analyzing trends has shifted from reactive/proactive to predictive — identifying potential risks before they become hazards. This research employs machine learning to assess how effectively student pilots execute assigned tasks, analyzing flight parameters to provide semi-real-time feedback on pilot performance, flight safety, and quality.

This digital twin approach to flight assessment provides objective, comprehensive evaluation of flights, giving students insight into current progress and a plan of action for improvement. It increases standardization in debrief feedback without adding to instructor workload, and improves the pilot certification process by ensuring pilots meet required standards for safe operations, reducing accidents.

← Back to Research