Predictive maintenance is a science still under development. Modern aircraft incorporate a battery of sensors to relay data in real time and post-flight to ground teams, but this information is only as useful as the tools to analyze it are.
Engines account for about 40 per cent of direct maintenance costs, so it’s no surprise that an aircraft’s highest concentration of sensors is found under-wing. CFM’s new LEAP engine produces about 3.5 times the data of the CFM56 powerplant it is designed to replace.
With modern engines such as LEAP and the PW1000G generating about one terabyte of data per flight – roughly the storage capacity of a home computer – engine manufacturers have had to invest in networks of data centers across the globe.
Even more important is software that can sieve these oceans of data for anything out of place or indicative of a trend that might affect future performance.
It is this latter capability that is still being refined.
“We are good at catching problems after they have happened the first time, but we want to use our tools and data to make sure we understand something before it even happens the first time,” says Vijayant Singh, general manager of data analytics at GE Aviation.
When problems occur mid-flight, much is often made of the capabilities of automated health monitoring systems.
These can relay information to ground teams who can, in certain cases, have a fix ready at the gate by the time an aircraft taxis in. However, the vast majority of data is still downloaded once an aircraft is on the ground, primarily due to a lack of infrastructure to handle real-time, in-flight updates.
To find out more about where engine data analytics is heading, plus the different approaches of OEMs and MROs, pick up the forthcoming Engine Yearbook.