For most of the past four years, MRO parts planners operated in unusual conditions. Reduced utilization created a buffer that absorbed planning errors. Parts that were ordered too early sat in inventory rather than causing excess against demand. Removals that were missed in the forecast were often covered by safety stock built from the prior period's lower flying rates.
That buffer has eroded. Utilization has returned to levels that require the planning function to be accurate, not just approximately right. And the supply side has not recovered at the same pace as demand. The combination creates a planning environment where the cost of a missed forecast has gone up substantially.
Utilization Recovery Is Not Uniform
The recovery in flight operations from 2022 onward has not been uniform across aircraft types or route categories. Narrow-body regional and short-haul markets recovered faster than widebody international operations. Some fleet types flew more cycles in 2023 and 2024 than in comparable pre-2020 periods because the mix of aircraft available shifted as older widebodies were parked.
For MRO planners, this means historical demand data from 2020 and 2021 is almost unusable as a forecast baseline. Data from 2018 and 2019 is more relevant for calibrating removal rates, but the current fleet composition, age distribution, and route structure may differ enough from that period that direct reuse of those rates is still risky.
The practical implication is that planners who built their reorder parameters from two- or three-year historical averages are working with data that includes a significant segment of low-utilization years. Those averages understate current demand. The stockouts that result are a predictable consequence of using the wrong baseline, not bad luck.
Lead Times Have Not Recovered at the Same Rate as Demand
Aerospace component supply chains went through significant disruptions from 2020 onward, and lead time normalization has been slow. For many common LRU categories, vendor quoted lead times that were 6 to 10 weeks pre-2020 have settled in the 12 to 20 week range as of mid-2025, with significant variance by commodity type and component family.
Repair station TATs for rotable components have been similarly stretched. Repair shops that cut workforce during the low-utilization period brought workers back more slowly than flying hours recovered, creating a backlog that has been working through the system for two years. Some commodity-specific repair shops are still running 50 to 90 days behind their historical TAT norms.
Extended lead times mean that the penalty for a late reorder decision is larger than it used to be. If the window to reorder before reaching a stockout was 8 weeks when lead time was 6 weeks, that window shrinks to near zero when lead time is 18 weeks. Planners who used to have some recovery room when they missed a demand signal now have none.
Why Spreadsheet-Based Planning Does Not Scale to This Environment
The majority of MRO parts planning at smaller operators, including most regional carriers with fleets under 30 aircraft, still runs on spreadsheet models or basic planning modules in legacy MRO systems. These tools do a reasonable job of tracking inventory balances and generating reorder alerts against fixed min/max parameters. They do a poor job of adjusting demand forecasts dynamically as utilization rates change, fleet composition shifts, or removal patterns evolve.
A spreadsheet min/max parameter set in early 2023 against 2022 utilization data does not automatically update when that aircraft type's flying hours increase 35% over the following two years. The planner has to notice that the baseline has shifted, update the parameters manually, and catch up before the stockout occurs. In practice, parameter reviews happen quarterly at best and are often deferred when the planning team is busy.
This is not a criticism of planners. It is a description of what spreadsheet-based planning requires: regular manual recalibration of parameters against current conditions. When conditions change slowly, infrequent recalibration is tolerable. When conditions change as rapidly as they did from 2022 to 2025, the gap between the model and reality grows faster than manual review cycles can close it.
What a Demand Forecasting Tool Changes in Practice
We should be specific about what we mean by a forecasting tool, because the term is used loosely. We are not describing a reporting layer on top of existing MRO data that shows historical removal trends. We are describing a system that takes current utilization data, tail-number removal histories, rotable pool status, and lead time inputs, and produces a forward-looking demand probability for specific components over a 30, 60, and 90-day horizon.
The practical change in a planner's workflow is that instead of deciding whether to reorder a component based on current inventory versus a static min/max threshold, the planner is looking at a projected demand curve that shows when inventory is expected to reach zero and whether it will cross that point before the next planned replenishment arrives. The decision becomes: does this forecast signal warrant adjusting my reorder date, and how confident am I in the signal?
This does not eliminate judgment. An experienced planner knows which component categories have reliable removal patterns and which are highly variable. A good forecasting tool provides confidence intervals alongside point estimates, so the planner can evaluate the signal quality before acting on it.
The Planners Who Benefit Most
We have spent the past several months talking with parts planners at growing regional carriers and independent MRO shops. The consistent pattern we hear is that the planning workload has grown faster than the team size. Planners who were managing 1,200 active part numbers two years ago are managing 1,800 or 2,000 today, with the same tools and the same number of hours. The cognitive load of tracking removal trends, adjusting for current utilization, and monitoring supplier lead times across that many SKUs is simply beyond what manual review can sustain.
A forecasting tool does not replace the planner. It reduces the number of part numbers that require active daily attention by surfacing the ones with elevated demand probability in the near term and allowing the planner to focus on decisions with the highest urgency. The parts that are not at risk in the next 60 days do not need daily monitoring. The parts that are approaching a likely stockout in the next three weeks do.
The operating environment in 2025 rewards precision parts planning in a way it has not since before 2020. The planners who build that precision into their process now will be better positioned for the capacity cycles ahead, not just the current one.
See Aerotrax on your own fleet data
30-day pilot on your removal history. CSV upload, no integration project required.
Request a pilot