.png)
Repair Strategy & Technical Solutions
What AOG Means, and Why Component Repairs Take Weeks
AOG means aircraft on ground. Why a component repair is measured in weeks while the operation is measured in hours, and where those weeks actually go.
.png){kind=logo}
AOG means aircraft on ground. Why a component repair is measured in weeks while the operation is measured in hours, and where those weeks actually go.
AOG — aircraft on ground — is the name the industry gives to a scheduling failure that has already happened. It is not really an event of its own.
It is what you see when the clock that governs a component repair, measured in weeks, meets the clock that governs your operation, measured in hours.
Every part of a component MRO business exists to manage that mismatch. Most of what gets written about it addresses the second clock: faster response, quicker turnaround, shorter lead times. Far less is written about why the first one runs as long as it does.
The repair cycle comes apart into four stages, and the weeks do not accumulate evenly across them. The question then splits in two, because a stopped aircraft is almost never one problem.
Sometimes what you are short of is time. Sometimes what you are short of is authorization, and no amount of speed touches that.
AOG stands for aircraft on ground: an aircraft that cannot be released for flight until a defect is cleared. The term says nothing about the cause and nothing about the fix.
It describes a state. The aircraft earns nothing, occupies a stand, and holds a crew and a schedule behind it.
That is why AOG is a status and not a service. On removable components it almost always reduces to a single question: can a serviceable unit be on that aircraft before the schedule breaks?
Notice what the question is not. It is not "how fast can this unit be repaired". Those are different problems, and confusing them is how operators end up paying priority rates for a repair that was never going to finish in time anyway.
A component repair has a minimum duration that does not compress much. The unit has to be received and inspected, and torn down before anyone knows what is actually wrong with it.
Only then do findings get dispositioned, parts sourced, there pair performed, the unit tested and the release issued.
An operation runs on a different scale entirely. A missed rotation is measured in hours, and a stand occupied overnight is already expensive.
These two clocks are not going to meet in the middle. The industry's answer has never been to compress the repair into hours — it has been to decouple the aircraft from its own component, which is an expensive thing to do at scale.
How expensive is measurable. Airlines spent an estimated USD1.4 billion in 2025 on surplus inventory holding alone, according to IATA's assessment of aerospace supply chain bottlenecks published in December 2025.
That is spare parts bought and stored so an aircraft does not stop while a unit sits in a shop. It sits inside a wider bill of more than USD 11 billion for the year.
That is what the gap between the two clocks costs when you bridge it with inventory. And the pressure behind it is not easing: the same assessment puts the average fleet age at 15.1 years.
Meanwhile Oliver Wyman's 2026 fleet and MRO forecast describes a supply chain constrained by engine technical issues and by shortages of raw materials such as composites and titanium.
Older aircraft need more component work, from a system thatis already tight.
Turnaround time is not one number. It is four, and only oneof them happens at the bench. Cycles rarely swell because a technician was slow— they swell in the handoffs, where nobody owns the clock.
The last row deserves more attention than it usually gets. Release is not administrative tidying at the end of a repair.
Under 14 CFR § 145.201, a repair station "may not approve for return to service … any article after a major repair … unless [it] was performed in accordance with applicable approved technical data".
Read practically, that means a unit can be physically finished and still be unable to leave the shop, because the data question behind it has not closed. No expediter fixes that.
Two shops can quote "ten days" and mean numbers that are a week apart.
One is quoting bench time: the days between the work order opening and the release tag being signed. The other is quoting door to door, with transport, customs and paperwork inside the number.
Both are honest. Only one of them is the number you actually live with.
Three questions settle it before a purchase order goes out. Does the clock start on receipt or on work-order release? Does it stop at release or at delivery? And does it pause while a finding waits for an engineering decision, or does it keep running?
There is a related item worth measuring on your own side. A share of units arrive at a shop with no fault anyone can reproduce, and those units still consume induction, bench time and release.
The cheapest week to save is the one never spent on a unit that did not need to come off the aircraft. Better removal data is what saves it.
A stopped aircraft looks like one problem and is usually two. Separating them early decides whether the next call goes to a stores desk or to an engineering desk.
If what you are short of is time, the route is clear and the data exists. You simply cannot wait the weeks.
That is what exchange and rotable availability are for. A serviceable unit goes on the aircraft now, and yours goes through the shop on its own clock.
The trade-offs between repairing your own unit and taking a replacement come down to eligibility, the condition of the unit you hand over, and who carries the inventory.
When the aircraft is already stopped, urgent component support runs on a different footing from a planned removal.
If what you are short of is authorization, speed is not the lever. The unit is repairable in physical terms, but no approved data covers the damage, or the answer coming back is "replace only".
That is a question about approval routes, and what a Designated Engineering Representative can and cannot approve is where it gets settled.
There is a third case, less common and worth naming. Sometimes the schedule broke because the depth of the work was misjudged: are pair was scoped where the unit needed an overhaul, or the reverse.
The difference between the two is not a matter of degree, and getting it wrong once costs a full cycle.
None of the four stages is controlled by a single party, which is why turnaround time resists being fixed by pressure alone.
Induction is mostly yours. Bench time is mostly the shop's. Engineering disposition sits between you, the shop and the approval route. Logistics is shared, and usually under-measured.
The useful move is not demanding a faster number. It is knowing which of the four stages your last three repairs actually lost time in, and asking the next shop what it does about that one.
What is the difference between an AOG and a planned component removal? AOG means the aircraft cannot be released for flight until the defect is cleared. A planned removal happens while the aircraft keeps flying, on a schedule you set. There pair itself can be identical — what differs is the clock. That distinction decides whether availability has to be bought from stock or can simply be scheduled, and it is the single biggest driver of what the event ends up costing.
What is a realistic component turnaround time? It depends on the component, on what the teardown finds, and on whether approved data already exists for that damage. A benchmark number across component types is not meaningful. What is meaningfulis asking for the range on your specific part number, and asking whether the figure quoted is bench time or door to door. Two shops quoting the same number can be a week apart in practice.
Does paying for a faster turnaround actually work? Partly. Priority handling and time-definite transport compress induction and logistics, which are real portions of the cycle. What they cannot compress is an engineering disposition with no approved data behind it, or a release that cannot be signed because the data question is still open. Paying priority rates on a repair that is waiting on approval buys very little.
What should I send with the unit so induction does not cost me a week? A defect description that says what the unit did on the aircraft rather than what the fault code was called. The fault codes and symptom data themselves. The removal record, and the last release document matching the serial number on the unit. Incomplete or contradictory paperwork is one of the most common reasons a unit sits before anyone touches it.
