Aircraft performance models play a key role in airline operations, especially in planning a fuel-efficient flight. In practice, manufacturers provide guidelines calibrated on one single aircraft, with performance modelling for all similar aircrafts (\emph{i.e.} same model) relying solely on that. In particular, it may poorly reflect on the current performance of a given aircraft. However, for each aircraft, flight data are continuously recorded and as such, not used to improve on the existing models. The key contribution of the present article is to foster the use of machine learning to leverage the massive amounts of collected data and update the models to reflect the actual performance of the aircraft. We illustrate our approach by focusing on the estimation of the drag and lift coefficients from recorded flight data. As these coefficients are not directly recorded, we resort to aerodynamics approximations. As a safety check, we provide bounds to assess the accuracy of both the aerodynamics approximation and the statistical performance of our approach. We provide numerical results on a collection of machine learning algorithms. We report excellent accuracy on real-life data and exhibit empirical evidence to support our modelling, in coherence with aerodynamics principles.