Predictive aircraft maintenance: established practice or future focus?

Gorodenkoff/ Shutterstock

One of the most prevalent challenges in the airline industry is the need to reduce costs and delays, while maintaining and improving aircraft operational reliability. Currently, airlines and MROs are trying to leverage data and technological progress to better predict and manage aircraft maintenance efforts through predictive maintenance (PdM).

What is predictive maintenance and why do airlines need it?

Aircraft maintenance is an integral part of ensuring an aircraft is safe for operations. Poor maintenance planning can lead to devastating financial results for air carriers and keep aircraft grounded, passengers waiting and can even lead to flight cancellations. Additionally, an inaccurate overview of maintenance causes overstocking of surplus aircraft parts, resulting in air carriers losing vast sums of money. 

To increase operational reliability and cost saving measures, aircraft operators follow aircraft maintenance programs. There are three well-known types of maintenance: reactive, preventive and predictive. 

Reactive maintenance refers to a timeline in which a particular part of an aircraft is used to its limits and repairs are only performed after a failure. This method is usually costly and dangerous for operational safety. Therefore, many aircraft operators use preventive aircraft maintenance (PM), also known as planned maintenance, which refers to a determined timeline of checks on certain airplane components. 

The United States Federal Aviation Administration (FAA) defines preventive maintenance as “simple or minor preservation operations and the replacement of small standard parts not involving complex assembly operations.” Anything that requires taking apart a structural component does not come under the definition of ‘maintenance’, according to the FAA.

The biggest challenge in preventive aircraft maintenance is that it’s not always easy to pinpoint a suitable time to carry out checks. For example, by scheduling aircraft maintenance very early, operators waste a certain component life cycle that might still be usable.  On the other hand, scheduling aircraft maintenance too late can lead to a complete component failure, which is not only an issue financially but a huge risk to operational safety. 

The term predictive maintenance (PdM) is contrary to the reactive or preventive approach that are still widely in use. PdM requires knowledge of the correct information at the suitable moment to perform the necessary aircraft maintenance. PdM uses data analysis and techniques to detect possible defects in the aircraft, which helps airlines carry out better maintenance planning.

“It is a predictive tool versus a reactive one. By not only predicting that the part may fail, but providing various solutions for that pending failure, allows an MRO to waste no time deciding which solution is best,” Martin McConnel, director and client relationship manager at Willis Towers Watson said at a WTW Global Aerospace MRO event. 

During the last decade, technological developments have laid the groundwork for predictive maintenance readiness. The widely adopted use of sensors on aircraft and data routers has allowed tools, such as machine learning or artificial intelligence (AI), to have applications in maintenance planning.

The primary data sources used for predictive maintenance include flight crew messages and maintenance messages, parametric data, including sensor, flight data recorder (FDR) data, pilot write-ups, maintenance write-ups, component removals and component shop findings.

“The idea is to decrease cases of AOG [aircraft on ground is a term in aviation maintenance indicating that a problem is serious enough to prevent an aircraft from flying], delays and flight cancellations,” Rodolphe Parisot, vice president digital and innovations at Air France Industries KLM Engineering & Maintenance (AFI KLM E&M) was quoted in a statement. 

Established practice or future focus? 

The use of analytical data to predict various anomalies and avoid unscheduled aircraft maintenance is not a new practice in the aviation sector. However, the innovation of analytical tools used in the quest to predict when maintenance is needed in real-time is always in constant development. 

There are many MRO companies, such as Lufthansa Technik, Air France Industries KLM Engineering & Maintenance and Collins Aerospace, that provide predictive aircraft maintenance solutions for air carriers across the globe. However, it is worth noting that it’s not only MROs providing such maintenance solutions, but also Original Equipment Manufacturers (OEMs). In fact, OEMs were initially at the forefront of introducing such digital solutions. 

According to John Maggiore, Managing Director of Maintenance and Leasing Solutions at Boeing Digital Aviation and Analytics, Boeing has a long history of using analytics in providing services to improve the operation and maintenance of commercial aviation. 

“In the mid-1990s, Boeing, in collaboration with its 777 customers, began sharing data via the In-Service Data Program. Sharing operational data allowed for additional analysis that helped further boost the dispatch reliability of this successful family of airplanes. This programme has since expanded to cover most Boeing models,” Maggiore said at the Predictive Aircraft Maintenance Conference in 2017. 

However, despite the wide use of analytical data in the aviation sector, sophisticated predictive maintenance tools were introduced not that long ago. In 2017, the best-known OEMs, Boeing and Airbus, launched predictive maintenance tools. Boeing introduced Boeing Analytx, while Airbus presented Skywise. 

As for MRO companies, AFI KLM E&M has a predictive maintenance tool called Prognos, which was developed in 2015. According to the company, Prognos retrieves data from an aircraft, both inflight and during turnarounds. 

“Now, with the quantity of data available, the automatic transfer of data, the decreasing cost to store data and to perform the corresponding analytics, AFI KLM E&M has been able to develop its own solution, PROGNOS, which predicts failures which are not seen by legacy solutions,” James Kornberg, Director Innovations at Air France Industries KLM Engineering & Maintenance (AFI KLM E&M) was quoted in Predictive Aircraft Maintenance Conference event. 

AFI KLM E&M’s predictive maintenance tool can be implemented to provide data on aircraft engines, auxiliary power units (APUs), inventory and more. According to the MRO giant, the Prognos is already implemented on Airbus A320 Family aircraft, Airbus A330, Boeing 737, Boeing 747, Boeing 777. It can also be installed on new generation aircraft such as Boeing 787 Dreamliner, Airbus A380, and Airbus A350 aircraft. 

American MRO company, Collins Aerospace, has also established its own predictive maintenance tool called Collins Ascentia. The Collins Ascentia, according to the company, shows a 30% decrease in potential delays and cancellations related to components and systems monitored on the Boeing 787 fleet. Additionally, it also shows a 20% decrease in unscheduled removals for various components on the Boeing 777 fleet.

Lufthansa Technik, one of the biggest MRO solution providers, has also developed its own predictive maintenance instrument called AVIATAR. Airlines, including Wizz Air or Sichuan Airlines, are already utilizing this technology as a predictive maintenance tool to optimize fleets and decrease manual and paper-based processes. 

“We will support our partner Sichuan Airlines in optimizing their technical operations processes and help to eliminate unscheduled maintenance related events,” Johannes Bussmann, CEO of Lufthansa Technik was quoted in the company’s statement in January 2021. 

Popularity and difficulties in implementation

The supply from MROs and OEMs for predictive aircraft maintenance solutions may suggest that a strong demand in the market already exists. However, that isn’t always the case as PdM implementation does come with some negatives. While predictive maintenance ultimately seems like a key measure to maintain aircraft efficiently and eliminate unscheduled maintenance, it also has difficulties in successful execution.

According to findings from an Oliver Wyman annual MRO survey in 2016, the global fleet could generate upwards of 98 million terabytes of data annually by 2026, creating new opportunities for better predictive maintenance implementation. 

Now, that is a huge amount of data to be processed and interpreted. In airlines, as in other industries, the gap between aspirations for AI and the actual impact from the technology can be wide. One of the biggest hurdles preventing operators from investing fully in predictive maintenance initiatives is the data, specifically its completeness, and the ability to sync data from different sources, departments and formats. 

Chris Spafford, Oliver Wyman partner and report co-author said: “The obvious challenge for carriers is a focused execution, which produces tangible and demonstrable improvements in cost and reliability. For OEMs accelerating adoption and profitably monetizing investments in predictive maintenance will be a significant challenge.”

Another primary concern is data security. Due to the enormous amount of data that needs to be processed, it is critical to guarantee that equipment performance data cannot be accessed by outside parties, and that outside parties are not able to control predictive maintenance systems. At a more baseline level, it also remains important to protect information such as customer data.

Another difficulty with the wide implementation of predictive maintenance softwares focuses on old generation aircraft. New generation aircraft such as the Boeing 787 Dreamliner or the Airbus A350XWB, have more capabilities to implement such tools. So, capabilities for previous-generation aircraft are not available. According to the Oliver Wyman report, this creates “new storage, organization, and application challenges”. 

The report continued: “As a result, many operators report modest big data programs, reflecting limited readiness for these new challenges.” 

Predictive maintenance implementation can also be an expensive investment. Its implementation requires manpower, monitoring equipment, skill and experience to accurately interpret data in order to demonstrate the reliable utilization of the technology. Combined, this means that, initially, aircraft condition monitoring software may come at a higher cost than preventive maintenance. 

In an Oliver Wyman MRO Survey, published in 2016, 19% of respondents said they used predictive maintenance systems for all aircraft in its fleet, while 25% revealed that predictive systems were only used for select aircraft. But the vast majority, 56%, did not use PdM at all. Predictive aircraft maintenance systems are mostly being applied to engine maintenance, component maintenance and airframe maintenance, respectively. 

According to the latest data, the market remains uncertain regarding predictive aircraft maintenance with 44% indicating increasing or new investment, but nearly 25% indicating a decreasing, or no planned, investment during the COVID-19 pandemic, as per Oliver Wyman MRO Survey 2020 report. 

Despite the obstacles in PdM implementation, there is a long-standing interest from industry leaders who are in search of more cost-efficient methods to improve aircraft maintenance performance. But visions of a fully AI-enabled future in the aviation MRO sector may have come to a standstill following the COVID-19 pandemic, which has forced airlines, MROs and OEMs to consider greater cost-cutting measures. 

 

Related Posts

Subscribe

Stay updated on aviation and aerospace - subscribe to our newsletter!