Aircraft maintenance is shared among many companies, including airlines and tiers of manufacturers, maintenance shops and parts suppliers. This piecemeal maintenance process cannot be efficiently conducted with completely different methods for each airline. Some degree of maintenance program or data format standardization is necessary. Standardizing maintenance programs would permit MROs to handle many airlines without using different inspection intervals, different task cards and differently trained workers. The growth of start-ups, reductions in engineering staff at major carriers and increasing use of leased aircraft is promoting this kind of standardization.
While no one MRO program will fit all carriers, there are efficiencies to be gained by tailoring maintenance programs to unique operations and aircraft configurations. So, there will always be a tradeoff between seeking the gains of standardized methods and the advantages of customized procedures.
Standardizing the format of maintenance and performance data holds much more promise universally. This would ease the flow of information among the many companies involved in maintenance, enabling fleet-wide analyses that could improve reliability and cut costs. There are no long-term tradeoffs here. The hurdles to standardizing data are investment, effort and time.
Apart from these two big areas, standardization also can be sought in smaller but useful ways; for example, in more uniform programs for one carrier's fleet, in more standardized job cards for the same tasks and more similar sign-off requirements.
Legacy carriers once devoted dozens of engineers to developing their own customized MRO programs. “Those resources are no longer available, so there should be a shift to more standard programs,” predicts Jim Ballough, vice president of the Cavok unit at Oliver Wyman.
“Airlines have been reluctant to adopt MPD [maintenance planning document methods from aircraft manufacturers], but that is changing,” Ballough argues.
Ballough expects this standardization to make MRO outsourcing more efficient. “Now the MRO may have five customers with one aircraft type but five different programs. That leads to inefficiencies. Techs have to learn five different programs and that leads to human factors problems. And you must have five different sets of task cards.”
Standardization of MRO work also should help cut costs and delays at lease return, which is becoming more important as short-term leases increase. “Different programs complicate that,” notes Chris Jessup, AAR's senior VP for sales and marketing.
MRO record-keeping already has moved away from the paper documents that did not permit much analysis. But Jessup stresses that there are three types of analysis that records should facilitate: analyzing one fleet type in one airline, multiple types in one airline, and shared data on fleets among many airlines, OEMs and shops.
The digital PDFs now common can be used for analysis within an airline. “But they cannot be shared across organizational boundaries, because systems do not talk to each other,” Jessup says. Even best-of-breed modern MRO systems do not store data in the same format. Carriers must standardize the data formats according to ATA Spec 2000 or ASD S1000D rules to make sharing and multi-organization analysis possible.
One pressure toward standardizing maintenance comes from leases. “If you own an aircraft you can decide how to optimize maintenance on it,” explains Joerg Coelius, manager of maintenance programs in's (LHT's) engineering unit. “But if you lease an aircraft, the leasing company may require you to stick with the OEM's program, the MPD.” For example, LHT offers escalated -400 C check intervals because it has found over time that the MPD 15-month check can be safely extended. LHT uses the 24-month checks on its own aircraft, but not on aircraft taken on short operating leases. “This is mostly for long-term leases,” Coelius explains. “If it is a short-term lease, they would never even think about changing the interval.”
If an airline were to make any change in the program in these circumstances, it would have to pay for a transition check to get the aircraft back on the MPD program, a very expensive proposition.
LHT provides third-party customers a program that is about 90% MPD. The 10% non-MPD part mostly affects cabins because the OEM maintenance planning programs do not tend to require a great deal of work on interiors. “But if you want a nice cabin that people want to fly in, you have to put more man-hours into the cabin,” Coelius says. This may appeal to full-service carriers but, “if you are an LCC, not so much.”
LHT has not seen any increase in maintenance program standardization requests yet, and continues to perform maintenance mostly according to customized airline programs. Customers can bring their own task cards, or LHT can generate the cards based on the optimized programs its engineering unit creates to either cut costs or enhance revenue. Customers can accept or reject the recommendations. Most suggestions for cabins and interiors are accepted, but most for check escalation are not, because of lease requirements.
Coelius would definitely like to see more standardization of data. “Now we get it in Excel or PDF, andand have different formats.” Different formats mean LHT must use different loaders to get data into its database. Smaller MROs may not even have various loaders, making the problem much worse.
“We want it in the industry standard, S1000D,” Coelius says. “We are working closely with operators, MROs and OEMs to get data in the same format.” He expectsdocuments to be mostly, but not entirely, in S1000D, and both 's and Airbus's to be 100% compliant. “We have worked on it for three years. I think it will take three more years.”
AAR has not seen much MRO program standardization across its 25 heavy maintenance customers yet, says Jessup. “They all have customized programs, they standardize as much as they can, but even if you havechecks at the same interval, no one is identical with the other.”
One reason isinspectors, who may require different sign-off procedures or MRO processes. Carriers undergoing consolidation—like Delta and Northwest, and United and Continental—that are trying to standardize within their fleets face additional hurdles from these varying regulatory compliance requirements.
“Of course standardization would help us to reduce cost and time because we could stick to the MPD,” Jessup says. “Customized programs drive costs up.” For example, one carrier may require 80 hr. of training for a task, while another requires only 40 hr. “So we can't shift labor easily between bays, and we have to set up a system to make sure each mechanic is approved for each job by each carrier.”
AAR has set up a program to standardize maintenance for each customer across its five locations. “We want all our facilities to have the same feel and touch to a customer,” Jessup says.
This effort took four years and an investment approaching $2 million to implement over AAR's five facilities. It has improved efficiency for both AAR and airlines, and reactions from customers have been positive. Before standardization, only one or two airlines felt secure enough to use multiple AAR facilities. Now more than 10 use several facilities. This allows AAR to fill its hangars, and airlines to get work done much more flexibly.
Data standardization also matters increasingly at AAR. “We are seeing more and more customer interest in electronic record-keeping, sharing and data mining,” Jessup says. “They want to move from managing and publishing records manually to electronically. Cloud services are kicking it off.”
Chris Reed, managing director of Trax, says carriers that stick entirely with OEM maintenance programs generally have such small fleets that it does not pay to make changes. “It's not worth it for three aircraft, but if it costs $10,000 to make a change that saves 10 minutes each for 120 aircraft, it's worth it.”
Customizing maintenance programs for individual carriers can be done in several ways. Some OEMs offer to customize the program for the airline. “You tell them the changes and they will incorporate these into the standard program and charge you,” Reed explains.
The advantage of customization by the OEM is that revisions will be made to maintenance documents given to airlines. “We can link to that, load it into our system and update MRO requirements,” Reed says. “We can generate job cards from the MRO program and the aircraft maintenance manual, get requirements to carry out tasks, look up subtasks and build that into job cards. It saves a lot of time and resources.”
This compares with airlines that customize their programs, which can be a slow process, but they avoid OEM charges by doing it themselves.
Trax sees data-standardization challenges very directly. It downloads OEM maintenance documents online and the airline merely accepts or rejects them. Integration differs by OEM, however. “You can get Airbus documents in SGML or XML, that is straightforward,” Reed says. “Boeing is messing around. They want to keep control of data.” To exploit online OEM documents, airlines need only document-management systems, sometimes added to and sometimes built into their MRO application.
None of Mxi Technologies' customers are returning to OEM standard MRO programs, says James Elliott, product marketing manager. “Bigger operators see opportunities to improve with customization. Smaller operators see fewer benefits or have no engineering departments, so they may outsource this to third-party providers or OEM fleet management.” Elliott sees outsourcing program design as a step toward letting an MRO or OEM manage all the maintenance on a fleet.
Elliot has heard about increasing standardization of task cards, which reduces the burden of updating these cards. “Maybe the OEMs are getting better at task cards, moving away from updating thousands of task cards.”
On data standardization, Elliott foresees a “step change,” as airlines outsource maintenance systems to Software-as-a-Service (SaaS) providers, which may be MROs, OEMs or simply platforms for many carriers. “This will drive standardization of data,” Elliott predicts. “It will compare apples to apples and increase reliability. Spec 2000 is one option, but S1000D is where we are headed.”
These semi-private clouds will store MRO applications and data, and they may give carriers analyses of data based on fleets much larger than just their own, thus reducing costs and downtime. A carrier needs at least 20 aircraft to analyze reliability, and results improve as hundreds of aircraft are studied. Data analysis might be conducted by OEMs, MROs or community providers. Elliott believes the economic gains will range from “noticeable to very significant.”
But the SaaS model may need more than just MRO solutions to be economic. Elliot says it needs to integrate MRO with other applications such as flight operations and technical documents, as well as integrate with in-house enterprise resources planning (ERP) systems.
has taken data standardization seriously and expects to reap major gains in improved MRO processes. In 2007, Air Canada Maintenance had 84 different IT systems, the majority of which did not talk to each other, explains Alan Butterfield, vice president for maintenance. “Other major carriers had invested in improving their old MRO system, we had not. So there was a huge opportunity to improve.”
Air Canada reviewed leading MRO systems and chose Trax, partly due to its configuration-control functions. The airline had a lot of cross-connectivity parts: line replaceable units, engine parts and other components that were applicable to several models. So it had to work with Trax on mixed modeling routines for the software.
Significantly, the airline decided to put all its MRO data into the ATA Spec 2000 format. “No one had done that for all their systems and data,” Butterfield notes. “Their systems were not as broken as ours had been. We did it for every system, because we wanted all our systems to talk to each other in the same industry standard language. Then we can innovate for the next 20 years.”
The project, begun about 18 months ago, is nearly complete. Data on Embraers, Boeing, Airbus narrowbodies and maintenance tooling has been converted. Remaining are supply chain and data, which Butterfield expects to have done by February 2013.
The carrier already has been reaping rewards. “Now we have hyperlinked data on parts, the applicability of parts, the MRO program, manuals and work plans,” Butterfield emphasizes. “We have all the information you need feeding into the main Trax system.”
One result of data standardization is that Air Canada is now the world's leading user of Aeroxchange, even though it is hardly the world's biggest airline and it has not begun yet to buy exclusively on the electronic exchange, as it plans to do.
“I get work orders and all the data on repairs,” Butterfield says. “If the No Fault Found rate is high, I can look at the fault isolation manual [FIM], talk to mechanics, go back to the shops or change the FIM. There are lots of opportunities to lower MRO costs and do things right the first time.” The carrier can change its basic MRO processes and write the changes into Trax.
Industry-standardized MRO data means Air Canada also can innovate in cooperation with its suppliers, which must be able to communicate in Spec 2000 format. Instead of waiting for his engineering department to drive innovation, Butterfield can now let suppliers, who have all the data, do the hard work of developing improvements. Air Canada engineers merely approve and learn from these changes.
These gains did not come easy. “It was a huge effort,” Butterfield admits. Three-quarters of the cost was commitment of people to the project, cleaning and converting old data. It will require two years for 200 aircraft, even after writing 43 scripts to automate as much conversion as possible.
The business case for adopting Trax and data standardization was partly based on the need for much better configuration control and avoiding the expense of maintaining 84 old IT systems. “We had to resurrect people to program for them,” Butterfield remembers. Payback was estimated at 48 months; “not fast,” but good enough to go ahead. And the best parts are still in the future. “We will be able to innovate much faster,” he predicts.