PREdictive MAintenance Analysis


CALL: 2018

DOMAIN: IS - Information and Communication Technologies

FIRST NAME: Jorge Augusto




HOST INSTITUTION: University of Luxembourg

KEYWORDS: Avionics, Predictive Maintenance, aircraft, machine learning, ARINC429

START: 2018-05-01

END: 2018-08-31


Submitted Abstract

Predictive Maintenance (PREMA) is a very hot topic in the aircraft maintenance. Given the detailed maintenance procedures set up by the manufactures, aircraft maintenance is more or less a constant, but very important cost factor for airlines. The potentials to reduce costs are mostly limited to employment cost. That’s why Lufthansa for instance transferred part of their maintenance business to Romania and to the Philippines. The detailed maintenance plans and clearly defined intervals cannot avoid that parts of the aircraft fail in between. Type of failure and location where failure occurs may cause an Aircraft On Ground (AOG) event. This means that the aircraft cannot continue its planned route until the part is replaced. AOG events cause high financial damages for the airline for re-routing, re-booking of passengers and freight, express delivery and exchange of parts etc. Pursuant Eurocontrol costs per minute for long delay of a Boeing 747-400 was calculated with roughly 289 Euros in 2004. Today this means costs of approx. 350 Euros per minute or 21,000 Euros per hour. Existing surveillance systems in the aircraft do not indicate problems of parts and components; they only alert the pilot once the problem occurred. Even though the amount of data available from the aircraft increased from the generation Boeing 747-400 to the following generation 747-8 from 120 sensors and 3,600 sensor signals per minute flight to more than 1,000 sensors and 3,600 sensor signals per minute flight the systematic analysis of this data to predict failure does not sufficiently exist.

This site uses cookies. By continuing to use this site, you agree to the use of cookies for analytics purposes. Find out more in our Privacy Statement