Is Too Much Maintenance Reducing Aircraft Reliability?

An extraordinary amount of data is flowing from aircraft to manufacturers, operators, and data analysis teams, and this informs more efficient maintenance programs and ultimately improved availability. Maintenance support is critical, but fewer visits to the maintenance shop have huge benefits not only for reliability and availability but also for safety.

While it might seem like the term reliability-centered maintenance (RCM) is a modern development, it all started decades ago with a book written in 1978 by two United Airlines maintenance experts: Stanley Nowlan, director of maintenance analysis, and Howard Heap, manager of maintenance program planning. Their work was sponsored by the Department of Defense, which wanted to find better ways to support different types of military equipment.

In an industry in which people are generally taught that aircraft are safe because of time limits for how long certain components can be flown and strict maintenance intervals that must be adhered to, the RCM concept might seem to raise the risk of problems caused by a lack of maintenance. However, this is exactly the opposite.

Essentially, Nowlan and Heap were able to show that more maintenance is not better and, in fact, can contribute to less availability and more failures caused by maintenance actions. Tellingly, the authors concluded that “the failure process is a phenomenon that cannot be avoided by any form of preventive maintenance.” This is an extraordinary statement for an industry that spends so much time and expensive resources on preventive maintenance, under the assumption that doing so will improve reliability.

“At one time, it was believed that all equipment would show wearout characteristics,” they wrote. United Airlines developed conditional-probability curves for components to help ensure that higher overhaul times didn’t reduce reliability. What’s interesting about these curves is that “the presence of a well-defined wearout region is far from universal…Some 89% of the items analyzed had no wearout zone; therefore, their performance could not be improved by the imposition of an age limit…Another 5% had no well-defined wearout zone but did become steadily more likely to fail as age increased. For a few of these items, an age limit might prove useful, provided it was cost-effective. Only 6% of the items studied showed pronounced wearout characteristics.

“Usually, also, the conditional-probability curve shows no marked point of increase with increasing age; the failure probability may increase gradually or remain constant, but there is no age that can be identified as the beginning of a wearout zone. For this reason, unless there is a dominant failure mode, an age limit does little or nothing to improve the overall reliability of a complex item. In fact, in many cases, a scheduled overhaul actually increases the overall failure rate by introducing a high infant mortality rate in an otherwise stable system.”

The authors go on to point out, “It is apparent from our discussion thus far that most statements about the ‘life’ of equipment tell us little about its age reliability characteristics…The definition of reliability is the probability that an item will survive a given operating period, under specified operating conditions, without failure. In discussions of reliability, therefore, it is insufficient to state an operating period alone as the ‘life’ of an item. This statement has no meaning unless a probability of survival is associated with it.”

Nowlan and Heap’s research helped inform the development of the next iteration of the Maintenance Steering Group (MSG) process. MSG-2 was more focused on prescriptive component time limits, but MSG-3 offered OEMs the ability to use data from in-service operations to adjust inspection intervals and develop more intelligent approaches to component replacement requirements.

Maintenance-induced Failures

Although not aimed at the business aviation segment, a study by Daniele Scarpazza and Joseph Hutter uncovered a pertinent and related issue: the prevalence of failures caused by maintenance actions or maintenance-induced failure (MIF). Pilots may be aware of heightened risk when flying an airplane for the first time after maintenance is accomplished, and this study highlights that risk in relation to light airplanes.

In “Quantifying the Risk of Accidents and Serious Incidents Due to Maintenance in General Aviation,” the authors asked: “Does evidence show higher rates of aircraft-caused accidents and serious incidents in GA airplanes just returned to service after inspective maintenance?”

They analyzed general aviation accidents from 2008 through 2024, “comparing the post-maintenance reliability sampled on adverse events caused by aircraft alone, against those caused by human error alone. We find that the answer is yes: the risk is 33.8% higher than baseline in the first hour following an inspection, and it remains higher than baseline for at least the first 31 hours. Heightened pilot and operator caution in the early hours in service after an inspection is therefore justified.”

Research such as the above two examples bolsters efforts in the aviation community to put this knowledge to work. One prominent example is a groundswell among light airplane owners flying their piston engines well beyond the manufacturers’ recommended time between overhauls and calendar limits.

Mike Busch, founder of Savvy Aviation, is a prominent advocate of running piston engines past TBO and maintaining them under an on-condition maintenance program, just like many airlines and business jet operators do with turbine engines. Busch explained that piston aircraft engine components don’t run reliably, then suddenly start failing around TBO time, in an article for Sport Aviation magazine: “But piston aircraft engines don’t exhibit this kind of failure pattern. We know these engines suffer the highest risk of catastrophic failure not when they pass TBO, but rather when they’re fresh out of the factory or field overhaul shop.”

In an examination of NTSB accident reports from 2001 through 2005, Busch showed “that engines fail with disturbing frequency during their first few years and few hundred hours in service after manufacture, rebuild, or overhaul.” Aircraft operated under Part 91 of FAA regulations aren’t required to adhere to manufacturer TBOs, and some owners are taking advantage of that and using Savvy Aviation’s guidance to monitor their engines’ health and successfully and safely fly well past TBO.

Strangely, while the information that Savvy Aviation has gathered is widely available, piston engine OEMs have not changed their TBO recommendations. And one country, South Africa, has even changed its policies and now requires that all piston aircraft engines be overhauled at the specified manufacturer interval, regardless of the type of operation.

RCM in Bizav

A key aspect of RCM is to prevent MIF. The more maintenance that is done, the higher the risk of MIF and the resulting consequences of safety risk and excessive downtime. An incident documented in the NASA Aviation Safety Reporting System archives captured an example of MIF that could have been a factor in an accident.

The crew flying a Gulfstream G200 on a repositioning flight was not only well aware of the need to perform a detailed preflight inspection following a maintenance visit but also faced poor weather on departure, including rain, icing, snow, visibility of 1 mile, and a 300-foot ceiling.

Splitting the preflight duties, the pilot in command (PIC) performed the external inspection while the second in command (SIC) checked over the flight deck. With everything appearing normal, they started the engines, taxied to the active runway, and took off into the weather.

Shortly after takeoff, the pilots were presented with a left and right pitot heat EICAS message. After climbing to a safe altitude, the pilots ran the checklist for that issue and selected OVRD (override) on the pitot heat switch. The pilot flying (PF) checked circuit breakers “but did not notice anything,” according to the report.

After the airspeed indication on the left side deteriorated, the pilots asked ATC for vectors back to the departure airport and for the ILS 11 approach. The right-side instruments appeared normal, so PF transferred control to the SIC. They then selected air data computer reversion and saw that the PIC instrumentation looked normal, then the PIC retook control.

According to the report, after inputting the approach to the FMS and briefing the approach, both the right air data computer and standby instrument’s airspeed decreased to zero, “and crew quickly discussed and decided to disregard all airspeed indications.” They asked for vectors for the ILS approach and “flew using pitch, power, [and] altitude indications with ATC information and vectors to get established on the ILS. Once on the ILS, crew used power settings and ATC as [a] resource to keep the aircraft in [a] stable approach and landed without incident.”

After shutting down, the pilots debriefed the flight and re-checked the flight deck and found that both pitot heat circuit breakers were tripped. While neither pilot’s narrative blamed the maintenance shop for neglecting to ensure that the circuit breakers were reset, it appears that this may have been the case.

Both shared some lessons learned, including adding a circuit breaker check to the pitot heat checklist and performing the preflight inspection together before a post-maintenance flight. “Always use extra caution and diligence when an aircraft is fresh out of maintenance,” one of the pilots wrote.

OEMs and RCM

Digital aircraft records management provider Bluetail has digitized thousands of logbook records and is now using AI to dig up useful information from documents that used to live in dusty boxes or file cabinets in a room in the hangar. This data could prove a boon to OEMs that want to learn more about what breaks and what doesn’t during an aircraft’s life, but it’s still early days in the AI revolution, especially in business aviation.

“There is for sure value to the OEMs and MROs,” said Kent Pickard, Bluetail chief technology officer. “While the customers generally look at things one aircraft at a time, there is value in the anonymized, aggregated data. If I’m going to bring my Challenger 600 in for the five-year inspection, what should I expect? We are actively talking with OEMs in terms of potentially entering agreements to improve their ability to do analysis.”

AIN asked business aircraft OEMs about their approaches to RCM and received responses from Textron Aviation, Gulfstream Aerospace, and Bombardier.

Textron Aviation

“We’ve been at this business for a while,” said Brian Adams, Textron Aviation v-p, aftermarket innovation. Over time, the company’s maintenance programs have evolved, and RCM and MSG-3 are important facets. “We use an MSG-3-based program, but…we look at it as part of a broader system, an intelligent maintenance system from Textron Aviation that allows [operators] to operate our aircraft in a more efficient way. It’s very data-driven.”

That data comes from multiple sources, including field reports and reliability and quality data from Textron Aviation product support team members who work directly with operators, and data from later-model airplanes equipped with onboard diagnostic systems. While many older airplanes’ maintenance records are on paper, making it hard to collect data, last year Textron Aviation implemented a digital maintenance transaction report that helps with more efficient data collection from its 20 company-owned service centers, more than 40 mobile service units, and 300 authorized facilities.

At the same time, Textron Aviation’s safety management system offers technicians a way to give feedback. Technicians often complain that engineers don’t understand how difficult it can be to access a component, and this is one way to provide input on those issues or to highlight risks they discover while working on an airplane. The Textron Aviation parts operation is also a key source of useful information on premature failures and how often parts need to be replaced.

“All that data goes into the development of the MSG-3 program,” Adams said, “and that helps optimize the tasks in that maintenance plan and tailors them specifically for each product. The results we're looking for would be to target any unnecessary inspections and eliminate those, extend the intervals where possible, based on that data, and then the output of that would be improved dispatch reliability, lower overall operating costs for our customers, and increased availability. We’re focused on delivering that value to our customers by doing the right maintenance at the right time, in the most efficient manner.”

Maintenance in the Design

All of the efforts to manage this process begin during the design phase for a new airplane, with an integrated product team that includes members of the quality, reliability, and customer support organizations. “From day one on those advanced design teams, they’re giving input to our engineering team around component sourcing, system architecture, and design for maintainability and reliability,” Adams said. In parallel, a maintenance engineering team develops the initial MSG-3 program, “because the design is going to dictate some of the maintenance requirements. We’ve got to get in early on these programs to have an influence on the cost of operation, the amount of inspection, and the long-term maintenance.”

Because that initial work during the design phase tends to be theoretical, once the airplane enters service, Textron Aviation supplements the MSG-3 program with operational data. “We’re looking,” he said, “as technicians and operators are doing these maintenance tasks, what are they finding? As we get this data from the field, it tells us how well the aircraft and the maintenance program are working.”

Textron Aviation also examines findings from maintenance tracking providers. If a component isn’t meeting its original reliability target, this will surface in the mean time between unscheduled removal metric. Textron Aviation engineers consult with the supplier to improve that component’s performance, all under the auspices of the MSG-3 committee and regulatory requirements.

Ultimately, this process helps Textron Aviation move to more on-condition maintenance, where components are monitored but not necessarily replaced at specific intervals. “That is to help us not put the airplane down and not induce more maintenance requirements than we need,” Adams explained. “It’s a balance. There’s a limit to how far we can push some of these intervals, because there are some required maintenance tasks, or…major inspections. We need to understand what's going on with the aircraft, so we can gather that data and then push the interval out or pull the interval in. It’s a fine balance. We want to limit the amount of downtime and the amount of tasks that we’re inducing during a maintenance event. But also, we want to be able to catch those [problems]. We want to be able to be proactive with our maintenance and eliminate unscheduled events.”

One way to examine whether these programs are working is with a new version of a classic Citation, the recently certified Ascend, which started as the 560 Excel, then XLS, XLS+, and XLS Gen2. The XLS Gen2 has a maintenance interval of 800 hours and 12 months, but the latter is now 18 months on the Ascend. “That’s a significant move,” he said, “and we’re estimating a 33% reduction in the amount of scheduled maintenance events.”

This change was a result not only of extensive data from the 560 fleet but also of some engineering changes on the Ascend. “We learned how to maintain these aircraft more efficiently,” he said. “And so all those things went into the Ascend, and we’re pretty excited about being able to extend those intervals. We’re doing that same process for the Denali and the CJ4 Gen3.”

Putting AI to Work

Textron Aviation isn’t immune to the need to put AI to use. “It wouldn’t be 2026 if we didn’t talk about AI every day,” Adams said. In fact, his team has been working with AI tools since 2024 and developed the Textron Aviation maintenance intelligence system. The idea is to help technicians find what they need quickly in the voluminous maintenance documents that accompany any aircraft.

“It’s used every day in our service centers, not only by our technicians, but also by our engineers who want to look up information about our products,” he said. “It’s taking our maintenance records, maintenance manuals, instructions for continued airworthiness, flight manuals, product catalogs, and databases we have from product support and putting that in the hands of our technicians so they have access to the data they need. It saves them time finding what they're looking for. We’re not simply using what the AI says to do; it is giving us a link to that [information] we can then use during maintenance.”

Another AI use is more predictive maintenance, which will further help reduce unscheduled maintenance and AOG events. The Ascend, for example, has a communications hub that can transmit helpful information via cellular or Wi-Fi connections on the ground. “We’re getting data from all these systems that are smarter, that have more sensors, and different components are reporting their health and their status to us,” he said.

That enabled the recent launch of a condition-monitoring system. “Now customers can select a set of parameters that they might be interested in tracking over time, to see how well their aircraft is performing compared to the fleet, and start to suggest proactive maintenance that could help prevent an AOG. We’re not going to say that you need to put your aircraft down immediately to take care of this, but you may be able to plan it into your next scheduled downtime.

“One of the things we struggle with [in] the MSG-3 program is the lag in the timeliness of the data,” Adams said. “Now we’re going to be able to get more timely information and more specific information around how the aircraft are being operated and how they’re performing in the field at a more granular level.”

Gulfstream

Gulfstream engineers use information from the MyCMP maintenance-tracking service to inform RCM programs. “This enables Gulfstream engineers to continuously monitor and analyze maintenance programs based on fleet data and the in-service experience, and has the corresponding benefit of providing an enhanced and holistic customer maintenance experience,” according to Lor Izzard, senior v-p of customer support.

By evolving to a focus on preventing unscheduled maintenance events rather than fixing something after a discrepancy occurs, he explained, “Gulfstream is leveraging fleet data to drive towards a more predictive maintenance environment, supported by new and integrated tools, such as our Diagnostics Direct mobile application. Diagnostics Direct complements MyCMP and is solely dedicated to diagnostics—with it, operators can also troubleshoot in real time via the app.”

An example of Gulfstream’s application of RCM protocols is the extension of A-Check intervals to other models, matching those of the new G700 and G800. “As reliability improvements have been demonstrated over time, we were recently successful in extending and aligning the A-Check interval to 750 hours on all the other Gulfstream large-cabin models as well as the super-midsize G280, giving operators the full benefit of enhanced maintenance programs and higher aircraft availability,” according to Izzard.

Bombardier

Bombardier said it uses RCM and MSG-3 to optimize inspection intervals, prioritize important tasks, and eliminate unnecessary tasks to enhance efficiency and reliability, both for in-production and in-service aircraft (although not the Challenger 600). “This allows our teams to lengthen inspection intervals (when supported by fleet data), remove redundant tasks, extend or shift component replacements, and work closely with suppliers to improve component reliability. Collectively, these efforts reduce the maintenance burden while maintaining high safety and performance standards.”

To bring useful data back to the maintenance data analysis (MDA) team, Bombardier said it “monitors fleet reliability to quickly identify components that may be failing early.” That team shares information with engineering “to drive product and reliability improvements.” Maintenance intervals can be adjusted based on that information, either up or down, and all changes are reviewed by an industry steering committee as part of the MSG-3 process.

For individual components, Bombardier’s Failure Review Board ranks component reliability according to mean time between unscheduled failure data from the MDA team, combined with warranty information and feedback from operators. “This process helps identify pain points, prioritize critical components, and work with vendors to improve reliability, reducing the maintenance burden and helping maintain aircraft availability even amid supply-chain challenges.”