A recent article from Aviation Week titled “Spend Less From Better MRO Planning” focuses on some of the challenges airlines face as they try to control maintenance costs. While the article starts by describing the cost implications of leased aircraft and engines, the real message is the opportunity to lower total aircraft costs through long-term maintenance planning. In a business where the utilization of capital equipment fluctuates according to seasonal demand, oil prices, geo-political circumstances, economic activity, weather and even seismic or geothermal events, it can be especially difficult to implement long-term maintenance planning. Furthermore, the complexity of the equipment (aircraft, engines and components) makes aviation an inherently unpredictable business. With these things in mind, contingencies have to be part of any full strategic maintenance plan.

The author suggests that with enough data about each aircraft, each fleet, each lease contract, each maintenance site, and each route plan, that airlines could optimize their MRO schedules and significantly decrease costs. The article also talks about understanding manpower, skill levels, locations and lead times; these are important. But equally important to controlling costs is the ability to understand the current and desired configuration of each aircraft, and the parts and procedures that will be needed. So while the author speaks at length about the vision, he only briefly touches on the roadmap or requirements to achieve it.

Throughout the article, several MRO software vendors are interviewed and each says basically the same thing, “give us resource and planning data and we can do the calculations.” Guess what? If an airline has the time and talent they can get Microsoft Excel to do the resource allocation calculations. (A point that is made in the article.) Achieving this vision requires the integration of accurate configuration data with flexible execution systems, so that building strategic plans and responding to tactical requirements go hand-in-hand.

In other words, achieving this vision requires a system like Enigma. Here’s why. How does the configuration and maintenance plan for each aircraft and engine get into the planning system? One way is to load it manually from OEM manuals and IPCs. A better way is to use software to deconstruct those manuals against each tail number and automatically load the configuration into the M&E system. How does the as-maintained configuration of each aircraft and engine stay current in the planning system? One way is to manually load the data from completed job cards. A better way is to automatically extract the data from job cards and update the M&E system.

How do revised (new, modified, deleted) parts and procedures get into the planning system? One way is to inspect each revision of every manual and IPC and make the changes manually. A better way is to compare each OEM revision to the current data in the M&E system, note the changes and then apply airline-specific logic to determine which changes are made automatically and which require engineering approval. You get the point, achieving the goal of spending less on maintenance through better MRO planning is a complicated but achievable problem. But it’s more than resource allocation. The key is to ensure that the data that drives maintenance planning and execution remains integrated and accurate so that maintenance reflects the needs of the airline rather than the desires of the OEM or (in lease situations) the owner.

The second part of achieving this vision is to have a system that allows airlines to quickly respond to changing situations. In other words, the ability to handle AOGs (aircraft on ground) and NRs (non-routine events). The article describes how maintenance requirements are needed well ahead of the service date so they can gather long-lead items and prepare job cards. While long-lead items are a valid concern, the unpredictable nature of maintenance means that job card generation should not be considered a long-lead item. Even for heavy maintenance, a significant portion of the work performed is unscheduled. Planning and execution systems need to be integrated and accurate so that non-routine job cards can be dynamically generated and approved. This will minimize delays, while still remaining true to the airline’s current strategic maintenance plan.

Such capabilities are not in the distant future, they exist today at places like Korean Airlines (KAL). KAL recently delivered a case study at the Airline and Aerospace MRO & Operations IT tradeshow in Singapore on this very topic, and we will cover that in this blog in the near future. The key message is, when it comes to MRO planning it’s time to apply a new equation: More Integration (M&E + ERP + Enigma) = Lower Cost + Less Downtime.

A few weeks ago I wrote about the importance of helping mechanics to fix rare problems faster—accelerating the repair of systems that don’t fail very often. Today’s post follows up on that theme as it relates to the aviation MRO industry. Because aircraft systems are generally quite reliable, when something fails unexpectedly it is called a non-routine event (NR). (It should be noted that aircraft systems have multiple redundancies, so an NR is not necessarily a cause for concern. But it still needs to be fixed.)

Non-routine events are a major source of unscheduled aircraft maintenance. (Other industries may refer to this as a break-fix event because something failed outside the normal maintenance schedule.) The question therefore is, “What to do about NRs?” The idea of planning for unscheduled maintenance seems like an oxymoron. After all, how can you plan for something that you can’t predict? And that is really the point; on any complex machine—and an aircraft is really complex—you know something is going to break, you just don’t know what it will be.

To explain the difficulty of the problem let’s look at some real customer data. In a fleet of wide body, long-haul aircraft over a two-year period, an airline found that there were over 3,000 non-routine events that could be traced back to almost 300 different systems. 10 systems caused almost half of the NRs. (In fact, 1 system alone caused 20% of the NRs.) But the majority of the NRs (51%) were caused by failures in one of the other 270+ systems.

When this data is plotted on a graph (above) it is easy to see the trend. The top 10 causes of NRs occured much more frequently than the rest of the systems. (The number 1 cause needs unscheduled maintenance almost once a day and number 10 occurs at least every other week.) The rest of the systems in this study (97%) occurred an average of 4 times per year, forming a long tail on our statistical data. However, since the top 10 problems account for less than half of all non-routine events, if an airline wishes to reduce the overall impact of NRs, then it must address the long tail.

Several conclusions can be drawn from these numbers:
• Most of the systems on an aircraft are quite reliable. (Given the complexity of an airplane and the physical stresses it endures, this is a great testament to the aircraft manufacturers.)
• Mechanics can quickly gain experience fixing the top 10 causes of NRs.
• Mechanics do not gain significant experience fixing the other half of the NRs. (They just don’t see them often enough.)

A typical maintenance approach is to focus attention on the most frequent failures—the top 10. Training to fix the top 10 is pretty straight forward however, the long tail consists of many different problems that happen infrequently. (These are individual systems that rarely break but, when added together, account for the majority of the unscheduled maintenance.) The only way to deal with such a wide-ranging set of problems is to improve the mechanic’s ability to respond to the unknown. Better training is not the solution, automating information systems is.

It should be obvious but when time is tight and a repair is urgent it’s best to bring service information to the mechanic, rather than the other way around. This can be accomplished by integrating maintenance information systems with maintenance planning, inventory and other back-office systems to provide a fully integrated scheduling and execution environment. Then, no matter what type of problem the technician uncovers they can quickly obtain the information they need and begin the process of repair.

Along with our partner Oracle, I’ll be co-presenting an Air Transport World magazine webinar on this topic in a few weeks (December 9, to be specific); if you’d like to attend that webinar, sign up here.