If safety is paramount in any industry, it’s for aviation firms like IAG. Any incident involving an aircraft makes headlines—and rightly so. The consequences can be devastating, both in terms of human lives and the broader impact on airlines and the industry. That means that maintenance is a high priority for airlines, as they need […]
If safety is paramount in any industry, it’s for aviation firms like IAG. Any incident involving an aircraft makes headlines—and rightly so. The consequences can be devastating, both in terms of human lives and the broader impact on airlines and the industry.
That means that maintenance is a high priority for airlines, as they need to know their vehicles are in tip-top condition, but it can also be timely and expensive, and often very complex.
Every commercial jet engine must meet strict regulatory intervals while also fitting around flight schedules, parts inventory and shop capacity. Planners juggle thousands of variables, yet one late part or an unexpected route change can unwind months of work.
A plane can be deposited with engineers believing there is one problem with an engine, only for another to be discovered. And often the causes of faults aren’t easy to determine. So, scheduling maintenance – whether routine or emergency – is a big challenge. Yet downtime is costly, too.
Maintenance is a costly part of running an airline. The industry is set to spend over $100 billion annually on maintenance, repair and overhaul (MRO) by 2030 according to Strategic Market Research’s Aircraft MRO Market Size & Forecast report.
To manage this, International Airlines Group (IAG) has leveraged its own internal development team to build an AI system that it claims can optimise engine maintenance.
Designed in-house at IAG’s Barcelona AI Lab, the new Engine Optimisation System (EOS) provides engineering and maintenance teams across IAG airlines with suggestions for the most efficient maintenance schedules.
AI taking off
International Airlines Group is the parent company of British Airways, Iberia, Aer Lingus, Vueling, and Level, with over 700 planes across its portfolio.
The AI technology uses algorithms to assess millions of different scenarios and identify the optimal maintenance schedule for our fleets, reducing potential delays which can happen when parts are not available in the necessary location.
Engineering and maintenance teams benefit from an intuitive, colour-coded Gantt Chart that suggests shop visits, available spare engines and parts, as well as timing for installations on aircraft.
The new system is already being used by the airline giant through its Aer Lingus brand. The Irish carrier is using the EOS to schedule maintenance on its fleet of CFM56 after a six-month testing period.
Why pick Aer Lingus first? IAG’s chief AI scientist Ben Dias explains that it was due to the state of the airline’s datasets, which made it much easier to build the model for the EOS.
“One of the reasons we went with Aer Lingus first was because they had really good data,” he explains to TI at an event held in London last week.
“They were already advanced in their data transformation journey and had implemented a data platform, which meant we could access that more easily.
“For our other airlines, they are interested in adopting the technology, but we need to get their data in a good place before we can plug it in. That’s the carrot.”
Data was also one of the reasons that the project involved multiple iterations – EOS is currently on its 11th version, according to Dias.
It was developed internally by Dias’s team. IAG has dedicated Innovation Labs in Barcelona and London, where leading AI experts and cutting-edge technologies are deployed to address business opportunities in aircraft maintenance, customer experience, as well as operational efficiency.
The team of around 70 experts develop projects for both internal use, and to potentially be white labelled. Dias confirmed the airline firm had also worked with a number of external technology partners to build EOS, though IAG declined to disclose who these companies are or share what AI model the system operates on.
“We work with our own Center of Excellence,” Dias adds. “I have a small internal team. But we had experience in this space, and we knew this was a difficult problem. That’s why you don’t have an easy off-the-shelf solution.”
Owning the intellectual property is vital, as it means IAG can refine the model as its network, fleet mix and shop capacity change. Also, the group avoids vendor lock-in.
Dias admits the “first few tries didn’t quite work”, with one of the big complexities coming from the inclusion of spare engines, which also have their own maintenance cycle.
“If you just ignore spare engines and look at just the engines on the aircraft, it’s a linear problem. It’s a simple one. As soon as you have a pool that you can swap in and out, it becomes more complicated.”
Runway success
Another key benefit the EOS system offers IAG’s to keep a plane in the air for longer for business reasons, it has a significant impact on the engineers on the ground.
Given the lifecycle of engines and the maintenance timeline, this could cause significant disruption. But EOS allows teams to simulate these potential pivots to create an optimised schedule.
“It is a strategic tool,” adds Dias. “If the business needs to make a change to its strategy, we can run different scenarios to optimise the schedule, rather than it constantly shifting.”
This also helps finance teams to sync more easily with the engineers. If the maintenance team need to tweak a schedule, the finance team can see the impact on cashflow through a separate interface created by Dias and his team.
The model does not simply predict when an engine might need service, it chooses the slot that minimises ground time across a 700-aircraft portfolio.
According to IAG representatives speaking at the event, Iberia will be the next Group airline to use the EOS program, with all other IAG carriers to follow by the end of 2025.
“Our new Engine Optimisation System shows how IAG is using AI to tackle real-world challenges and improve maintenance across the group,” concludes Dias.
“Transformation is at the heart of our overall business strategy and is powered by innovation including AI.
“The system gives our people the data and tools they need for smarter planning and better teamwork.”
