Managing a supply chain for the aftermarket service environment is fundamentally different from managing manufacturing or distribution supply chains. While the same language is used—forecast, inventory targets, source, deploy, replenish—how these core processes are actually executed is quite dissimilar. The distinction is due to the dramatically different business models and motivators present in many service supply chains.
For example, consider the perspective of an airline or war fighter fleet manager. Here, the need for service usually arises from a costly equipment failure or accident. And therefore, great lengths are taken to avoid service or parts demand. Another critical characteristic of aftermarket service supply chains is the application of used or defective parts. Typically, manufacturers cannot use parts harvested from finished assets or repaired from a defective state. Yet, for service providers—particularly aerospace and defense organizations—the cost of new parts and the length of their manufacturing lead times tip the scales heavily in favor of repaired or used parts. This often results in the reverse-logistics path of the supply chain becoming the primary source of supply.
There also is much greater variability and uncertainty in the base product data for a service provider as compared to a manufacturer. The manufacturer typically works with only the most current revision of a part and can provide a finite assembly schedule and complete unit bill of material (BOM) to the shop floor. The service provider, on the other hand, has neither of those luxuries.
Take, for instance, aircraft of the same fleet type that have unique configurations by tail number. These configuration variations mean that the rules for parts applicability, supersession, or interchangeability can be quite different for two seemingly identical aircraft. Even more challenging, these configurations are represented by product BOMs that often can go deeper than 20 levels. Compounding the configuration confusion is the fact that there is considerable uncertainty around the timing of the maintenance event, the final scope of work to be completed, and the BOMs that will be required to complete it. Unfortunately, there isn’t an easy fix using traditional supply chain management approaches.
When discussing the challenges for assets such as these, the material planning challenge adds yet another layer of complexity. Service supply chains have to deal with assets that are in near continual movement and are comprised of dynamic configuration changes. They also have extremely low-volume demands coupled with high costs and long lead times. And to top it off, they are subjected to heavy regulatory scrutiny.
The traditional approach to maintaining assets such as aircraft has been to take them out of service for very long periods of time. This enables professionals to determine the scope of work required, source parts, and resequence and reprioritize activities. As a result, companies with assets requiring this type of service must plan to maintain higher levels of parts safety stocks in order to cope with the unpredictability of demand.
This approach has two major flaws related to buffering for uncertainty. The first is the elongated periods for which an asset is out of service and the associated technician downtime. The second is being unable to effectively manage the growing stock levels of expensive, low-volume parts.
These challenges today lead us to the emergence of high-velocity maintenance. This is a philosophically different approach, which strives to take assets out of service more frequently, but for much shorter durations. When practiced effectively, the primary goal of increased asset availability is realized, along with improved technician productivity and task turnaround times. Part inventory usage also is enhanced.
The ability to perform maintenance rapidly places much greater emphasis on overall planning—most specifically, the coordination between maintenance and materials. Traditionally, the interaction between these two planning organizations has been somewhat inefficient. In some cases, it can be almost adversarial due to conflicting metrics. Maintenance employees are tasked to perform the work on time, but with no penalty for housing bloated inventory profiles; meanwhile, materials managers are measured by levels of inventory investment.
With high-velocity maintenance, asset availability is the overriding goal, and departmental metrics and motivations are developed to align with it. There are at least three levels of coordination between maintenance and materials that ensure this goal is reached.
1. Macro planning. This is strategic planning that typically occurs on an annual cycle and is updated as required. For instance, for airlines and their original equipment manufacturers, the targeted outputs are driven by several factors, including
Aggregated maintenance plans with monthly projections drive financial plans around the volume of work to be performed in-house versus that which is outsourced. They also create implications for maintenance capacity, including facilities; human resources; and supporting assets, such as test equipment and tooling. The projected levels of work to be performed in turn drive high-level plans and budgets for the materials organizations.
- configuration of the fleet
- aircraft numbers by type
- planned use of the assets to be supported, such as flight
- hours or cycles
- routes flown
- planned maintenance programs
- desired availability goals.
2. Micro planning. Periodic planning cycles look at more detailed schedules by the tail number of the specific aircraft. Included are required maintenance checks, modifications and upgrades, and the removal of life-limited parts (items that must be inspected based upon reaching a limit of number of days on wing, flight hours, cycles, or other activity metric).
These plans are scrutinized to include the work or task cards for each maintenance event to be performed. Each card carries a BOM for the parts and tools known to be required to complete the work. These BOMs are a major component of the development of the dependent-demand plan.
The aircraft activity levels and associated failure or removal rates drive the dependent demand. The supporting application systems infrastructure has evolved to enable plans to be passed in either direction and evaluated, eventually leading to clear results. This has dramatically reduced the time taken to evaluate desired changes to maintenance schedules and enhanced the quality of the material plans for repairing and sourcing parts.
In the aerospace and defense industry, the impact has been huge. More and more carriers and war fighters are looking for ways to use an ever-increasing volume of health data about their aircraft. The data are captured in real time to adjust planned maintenance events and unscheduled part removals caused by component failure. For example, higher-than-expected readings on oil pressure or engine temperatures could predict an oil pump failure or issues inside the engine.
3. Real-time adjustments. Given the cascading uncertainties—from how much the aircraft actually fly, to where they fly, to changes to the maintenance schedules or the list of required parts—there are numerous adjustments that must be made bidirectionally between maintenance and materials. If maintenance wants to pull forward an event, materials must be able to support it with the complete BOM. This presents the greatest challenge both procedurally and from a systems support perspective.
It can even manifest as a cultural challenge, where the individuals in the two planning groups must find their proper counterparts and transfer critical information in complete and accurate transactions. The interaction between people and systems must be in or near real-time—meaning, requests for unplanned, non-stocked parts are expedited immediately, and complete and precise configuration information is required in order to determine the proper part and revision to source.
Uncertainty in materials planning for any maintenance shop, and especially a high-velocity maintenance shop, arises from
Dealing with the dynamics of the schedule requires tight coupling of the respective planning systems. On the materials side, it also necessitates the ability to model different scenarios quickly and to use sensitivity analysis tools to gauge the implications of alternate schedules and work scopes.
- variability around the schedule
- the scope of maintenance tasks
- the total array of parts that will be consumed in the completion of the work.
Determining the complete list of BOMs is a different problem requiring a different solution. A typical maintenance BOM lists only those parts that are completely known to be required to complete an event. In order to see the complete parts list, it’s necessary to determine likely additional part demands not called out from the task work cards for the maintenance event.
These part demands, often called nonroutines, are the greatest contributor to delays and disruptions in shop floor operations.
One approach to predicting this demand involves computing probabilistic BOMs that add parts to the standard maintenance BOM with an associated probability of need. These probabilities are computed by analyzing historic work records for past nonroutines. Note that the aging of assets and engineering or part reliability improvements can create dynamics in the probabilities that require trending of the values over time.
An emerging approach to predicting nonroutine parts demand uses knowledge management tools and expanding databases to test for causality. This may be between what were thought to be independent maintenance events, but actually may be related; symptoms represented in the health data; or a function of time, environmental considerations, or asset use.
Any organization looking to get the most out of its high-velocity maintenance will require a commitment at the highest levels to ensure that the goals and metrics of the maintenance and materials teams are aligned around equipment availability. It’s also essential to foster collaboration within the organization.
The high-velocity maintenance environment must be supported by a system infrastructure that efficiently and effectively transfers information and evaluates and accommodates any implications—and, at the core of this evaluation is the application of tools to manage or minimize uncertainty about the work to be performed and the parts that will enable it.
The benefits of increased asset availability add to top-line revenue for any organization, while also reducing the maintenance and material costs and ultimately increasing profitability.
Ed Wodarski is vice president of solutions for PTC, where he is focused on the commercial aviation and aerospace and defense industries. He has 30 years of experience in the design, development, and deployment of software to support the service life cycle. Wodarski may be contacted at firstname.lastname@example.org.
To comment on this article, send a message to email@example.com.