We’re excited to share the results of our latest research project in collaboration with the U.S. Navy on attrition risk and aircraft suitability prediction in pilot training using machine learning. In this study, we developed models that accurately predict attrition risk and recommend suitable training aircraft types for student aviators, potentially saving millions of dollars in attrition costs and improving pilot training outcomes.
The cost of training a basic qualified U.S. Navy fighter aircraft pilot is nearly $10M, with the most expensive stage being advanced flight training. Despite screening tests and early-stage attrition, 4.5% of aviators undergo attrition in this stage due to poor flight performance, voluntary withdrawals, and medical reasons. By identifying those with a high risk of attrition early in training, our models could help reduce late-stage attrition and provide financial and operational benefits to the U.S. Navy.
In addition to attrition risk prediction, we trained our models to recommend suitable training aircraft types for student aviators. Our models took into account various factors such as the student’s flight performance, experience, and training goals to make personalized recommendations for aircraft types that would optimize their training outcomes. This could potentially reduce the time and cost to train pilots and improve overall training outcomes.
While this research was funded by the U.S. Navy, we believe it has broader implications for aviation schools and other organizations that train pilots. By leveraging machine learning and predictive analytics, these models could help identify at-risk pilots early in training, optimize training programs, and improve overall pilot training outcomes.
We’re excited about the potential impact of our work and are eager to collaborate with aviation schools and organizations interested in exploring this technology further. If you’re interested in learning more about our research or exploring potential collaborations, please reach out to us through our “Contact Us” page.
Stay tuned for more updates on our work and other projects leveraging machine learning and artificial intelligence to solve complex problems and improve outcomes across industries.