Few in the aviation supply chain dispute the potential of data to transform the industry. Predictive maintenance, the most obvious benefit spawned by data analytics, has already given a strong indication of the high levels of optimisation that can be expected if data is leveraged efficiently.
“When you talk about digital transformation, at the core, you're solving a business problem,” contends Karine Lavoie-Tremblay, Director of Commercial Engines Digital Transformation at Pratt & Whitney.
“It's coming now to a level of availability and maturity that really empowers our employees with the tools that they need to make the right decisions at the right time.”
Sajedah Rustom, the CEO of AJW Technique, praises digitalisation for its ability to allow for faster and wiser decisions.
“When we talk about digitalisation in aviation, I think first and foremost, we’re talking about business transformation. Without a strong foundation, there's no technology that's going to work for your environment,” she concedes.
But there is a potential roadblock: data is viewed as a valuable commodity and airlines are hesitant to share their data with manufacturers, and vice versa.
And that’s a major problem heading onwards.
Airlines, original equipment manufacturers (OEMs), aftermarket players and the maintenance, repair and overhaul (MRO) sector have access to vast amounts of data.
This can be anything from flight routes and passenger information to the operational history of nearly every imaginable individual part and component on any given aircraft.
Modern aircraft models are outfitted with thousands of sensors and generate massive amounts of data – thousands of terabytes every single day.
While ‘big data’ has been something of a buzzword in the industry for several years now, it is beginning to deliver on its promise, according to Matthew Eli Eli, Head of Planning at Satair.
“The treasure trove of data within this industry means we now know more than ever before,” he says.
“We know the aircraft. We know where they're flying. We know the configuration of the aircraft. We know the checks they’re going into, and we know the tasks that are going to be performed.”
However, simply collecting massive amounts of data isn’t enough, warns Lavoie-Tremblay. You need to know what to do with it.
“Data for data is interesting, but that doesn't necessarily drive value,” she warns.
“You always need to think about: Where am I going to use this data? What am I using it for? And the more we know, the better suited we will be to respond to our customers’ needs.”
Predictive maintenance, which enables airlines to better predict and manage maintenance efforts, is emerging as a highly effective use of data.
For example, reveals Lavoie-Tremblay, Pratt & Whitney can monitor up to four million engine data points on every single flight.
The data collected through its EngineWise system allows P&W’s customers to predict when their engines will need repair and optimise the scheduling of maintenance to minimise ground time.
Predictive maintenance allows for what Satair’s Eli Eli describes as a holistic “bottom up” overview of what’s needed to keep an aircraft in service.
Sensor data and maintenance logs alert airlines to maintenance needs well in advance, helping to plan better and avoid costly groundings.
As computational power and data storage capabilities continue to improve, airlines, manufacturers and MROs get access to more and more information that allows them to further optimise their processes.
“We're looking at the aircraft going through checks. We're looking at the tasks being performed and we're looking at the material requirements to perform those tasks,” adds Eli Eli.
Predictive maintenance is now “the industry standard”, concurs Joost Groenenboom, Aviation Principal at the consultancy ICF.
“Aircraft like the 787 and the A350 produce a significant amount of information on their systems and the health of their systems – how valves are opening and closing, how systems are behaving – and based on that information, you can make certain assumptions,” he explains.
“You can see that certain systems are not operating in the way that they should be, and hence there might be an issue with it.”
Developments in data collection, analysis and implantation through machine learning and artificial intelligence are enabling robotics to play an increasingly large role in MRO.
The robotics are getting much better at dealing with variable tasks thanks to mass data enabling machines to learn how to recognise patterns – essentially turning the variable into something predictable.
Furthermore, complete with added sensors and wireless communications, these intelligent machines have the potential to bring about human-robot collaboration by enabling robots to respond to vocal or image-based commands.
According to Goh Poh Loh, Executive Vice President at Singapore Technologies Aerospace Ltd, these advances may enable MRO providers to maintain their competitive positions in an industry where they’re increasingly being squeezed out.
The full realisation of predictive maintenance depends on readily available big data being shared throughout the industry.
A major roadblock, however, is that data is still viewed as a commodity and is, for the most part, tightly protected by the MROs, OEMs, manufacturers, airlines and leasing companies.
All of them have a sense of ownership; the data helps them to remain competitive in the marketplace.
“When an airline’s aircraft produces this data, it’s obviously their data and they can use it. But if OEMs can gather data from all of the world’s airlines, they're able to very, very accurately predict certain things,” enthuses ICF’s Groenenboom.
Jan Stoevesand, Lufthansa Technik Group’s Senior Director of Analytics and Data Solutions, concurs that the aircraft operators are the owners of the data.
“At Lufthansa, we developed a simple model for this, and we call it the three Cs: Control, Choice and Competition. It means the operator of the aircraft controls the data and gets it encrypted, after which they can choose what to do with the data and whom to share it with,” he says
“This will lead to healthy competition amongst the companies that are able to turn data into insights. This competition is good – it will help us and the industry as a whole to develop better prognostic solutions.”
Stoevesand contends that competition is healthy when it comes to data usage, but urges collaboration concerning its acquisition
“As an industry, we need to separate data acquisition from data usage. We have to build a Great Wall of China between these two things. We should collaborate on data acquisition and then compete on data usage – it just makes so much more sense for all of us,” he urges.
Why the democratisation of data is necessary
Many experts – including Micheál Armstrong, CEO of Armac Systems – agree that the industry must be more collaborative about sharing the data.
“Data needs to be democratised – we should be competing over the analytics, not the data. We need to have almost a taxonomy around this data so that we can all agree and share and benefit from this data. That’s almost a bigger challenge than the algorithms,” warns Armstrong.
“We’ve seen how monopolies are created when one person gets the data. You get Google, Facebook, Instagram – and everybody else gets squeezed out. We need to use blockchain technology to democratise the data and protect people’s confidentiality, but let the ledger information into the public domain, so that we’re not competing on the data. If we see the data as proprietary, then big data is gone.”
Sajedah Rustom, the CEO of AJW Technique, said she’d like to see even more data sharing across the industry.
“Big data needs to be open. We need to be able to make decisions on the aggregate. We need to work together in collaboration as opposed to competition and really open up some of the possibilities around digital transformation,” she says.
“For me, the foundation of next-generation technologies being successful is really the industry coming together.”
Wouter Kalfsbeek from KLM also believes data collaboration is the way forward, but he also raises an important issue that would arise alongside a potential democratisation of data.
“We all have data and are able to generate value for ourselves. If we put it together, we can generate even more value, but how do you divide that value so everyone benefits more or less equally? That’s the question we haven’t really figured out yet,” he explains.
The data itself doesn’t give any one player a leg-up. All the data in the world is useless if it can’t be effectively analysed and acted upon.
All aircraft and engine manufacturers today offer data tools as service products, and while the technology still has a way to go to fully realise its potential, the competitive direction is clear – the focus should be shifted on developing tools to best analyse data, not hoard it.
According to Rhonda Walthall, Technical Fellow in Prognostics & Health Management at UTC Aerospace, there is already a shift in the industry underway from data ownership and control to collaboration.
“Stakeholders are coming together looking for opportunities to partner together and to share data yet still protect their intellectual property,” she says.
Players like Airbus, with their Skywise open platform, are already starting to sow the seeds of this shift toward collaboration. Other popular aviation data platforms include Lufthansa Technik’s Aviatar, Honeywell Forge and Enspan.
Skywise, launched in June of 2017, is aimed at combining data from Airbus in-service aircraft with airline and OEM data, in order to conduct in-depth analysis aimed at anticipating and optimising processes such as maintenance.
To bolster the data available for their analysis, Airbus offers free anonymised data to airlines that submit their own.
The number of aircraft covered by Skywise increased threefold from 3,500 to 10,000 in 2019 – approximately a third of the world’s passenger planes – before the pandemic stalled matters.
As of today, around 10,000 aircraft are covered, of which 5,700 are Airbus, which amounts to nearly half its global fleet.
The popularisation of digital services and the move toward data sharing is likely to have a trickle-down effect on the rest of the supply chain, including logistics providers and distributors such as Satair.
It’s not hard to imagine the effect these services could have on the aftermarket business, as it will provide distributors and logistics providers with valuable insight into what parts the customers need and when they need them.
Moving forward, these instances of collaboration and sharing, in conjunction with more intelligent products that can help us get better data from components, according to Walthall, will help the industry build what she calls the “digital thread for a component throughout its lifecycle”.
And the impact of predictive maintenance isn’t restricted to suppliers and airlines, as cost savings on maintenance very often result in customers being offered less expensive tickets.
Perhaps more importantly, predictive maintenance improves reliability. When airlines have an overview of maintenance requirements, delays and cancellations are less likely to plague customers.
And predictive maintenance increases passenger safety. When components are removed before failure, the aviation industry dramatically reduces the risk of safety incidents.