Machine learning and statistical analysis in fuel consumption

Investigate how to use machine learning to predict fuel consumption in aircraft operation. Every year more than millions of Euros are lost through inaccurate planning of the aircraft fuel loaded in the aircraft.


We consider several different sources like route, aircraft and engine type, pilot and weather characteristics and find n algorithm to a simulate the fuel consumption measured in liters for your route. Using our system, that you can also use in EFB, you can evaluate using machine learning methods  different routes, required fuel and understand the risks based how data collection frequency affects the prediction and which features are most influential for fuel consumption.

Find that a lower collection frequency of 10 minutes is preferable to a higher collection frequency of 1 minute. I also find that the evaluated models are comparable in their performance and that the most important features for fuel consumption are related to the road slope, vehicle speed and vehicle weight.

This product is currently under revision and a new version will be deployed in GIGA Hub March 31st 2020.