#	Pagina
attuale pagina	/open-h2020/projects/218509/results.html

Report

Teaser, summary, work performed and final results

Periodic Reporting for period 1 - PERF-AI (Enhance Aircraft Performance and Optimisation through utilisation of Artificial Intelligence)

Teaser

Summary

For the Operation of Commercial aircraft, airlines use manufacturer data that are common to all aircraft from the same type. However, during its lifecycle, performances (fuel consumption, speed, etc) for each individual aircrat will change as engines are ageing, aircraftt strucrure can me modified, etc. Through PERF-AI, we apply machine learning technologies to flight data to identify the change in performance for each individual aircraft and provide airlines with updated data that in turn will allow to operate more efficiently. Safety Line has been working since several years on the optimization during climb phase using Artificial Intelligence (OptiClimb). Thanks to PERF-AI, we will be able to provide extended services to the airline industry and possibly to manufacturers.

Work performed

The project started in November 2018 and so far we have performed the following tasks :

- decoding and pre-processing of aircraft flight data
- worked on the problem definition and data engineering
- identification of relevant machine learning techniques to issue performance data
- Issue performance tables compatible for the use of Flight Management System

Final results

Due to the amount of data and the number of parameters that influences aircraft performances, it is a challenge to identify statistical models that fit real aircraft performances. New statistical approaches allow to do so and we have already solved a big part of the problem with the help of INRIA. The expected results will allow to optimize flight trajectories according to actual performance and reduce fuel consumption and CO2 emissions generated by the aviation industry.