Aerospace and Defense Manufacturing KPIs: Buy-to-Fly, Yield, Escapes and Delivery
The KPIs that matter in aerospace and defense manufacturing, what prime supplier scorecards actually require, how to measure each KPI, and the levers that move it.
Aerospace and defense primes score their suppliers on a short list of numbers. Use the ranges below as a reference point for finding your biggest gap, not as targets, and trust your own trend over any single industry figure.
Every figure here shows its basis. A superscript number links to a public source in the Sources list at the end, including published prime scorecards and supplier requirements, and the sentence states that source's year and scope. A figure marked no cited source is kept exactly as originally published because no public source stating it has been found, so treat it as illustrative.
At a glance
Aerospace and defense manufacturing KPI benchmarks at a glance
Ranges are as originally published unless a source is cited. Labels such as world-class in a cell are the original's own words. PPM means defective parts per million shipped, and Not stated means the original gave no figure for that column.
Method: How the figures in this guide are sourced
A figure with a superscript number is stated by the cited source, listed under Sources with its publisher, date and access date. Scorecard thresholds belong to the named prime and may differ at yours. A figure marked no cited source is kept exactly as originally published because no public source stating it has been found. It is illustrative, not measured industry data.
Buy-to-fly ratio
Buy-to-fly ratio, the weight of material bought divided by the weight of the finished part, is the material efficiency KPI that separates strong shops from wasteful ones. Typical 5-axis machined aluminum aerostructures run 3:1 to 8:1 no cited source, while titanium forgings and hogouts drift into 10:1 to 20:11, 2, 3.
Published titanium and aero engine buy-to-fly figures
World-class programs pull the ratio down by moving from plate hogouts toward near-net forgings, castings, or additive preforms, which can cut a 12:1 machined ratio to 3:1 or better3. For titanium, Oak Ridge reported in 2013 that near-net powder and additive routes let aerospace makers decrease buy-to-fly from over 10 to 1 to less than 1.5 to 1 on some lower yield parts3.
Track it per part number and flag anything above 10:1 as a candidate for near-net redesign no cited source. Each point of ratio reduction directly returns raw metal spend, often the largest cost on the part.
How to measure
Divide the weight of stock bought by the weight of the finished part, per part number.
Use the stock actually ordered, not the nominal part envelope.
Recompute the ratio whenever the stock form or the supplier changes.
What moves it
Near-net forgings or castings in place of plate hogouts.
Additive preforms for high-removal titanium parts.
Redesign of the part numbers flagged as high-ratio candidates.
First pass yield and its cousin, rework rate, define your conformance floor. Typical aerospace machined-detail first pass yield sits around 90 to 94 percent no cited source; world-class flight-hardware lines hold 97 to 99 percent no cited source. Space hardware programs chase yields above 99 percent because a single escape can be a mission no cited source. For rework rate, world-class is under 2 percent versus a typical 4 to 7 percent no cited source.
The main lever is process capability on key characteristics. RTX's 2025 aerospace supplier quality requirements (ASQR-01) set a minimum Cpk/Ppk of 1.33 on key characteristics4, and AM General's 2023 supplier quality manual requires a Cpk of 1.33 or better on major characteristics5.
Pair capability with tool-life monitoring and first-part probing. Measure yield at the operation level, not just final, so you can see where units actually fall out.
How to measure
Divide units that pass an operation without rework or scrap by units started at that operation.
Roll operation yields up to a part-level figure, and keep both views.
Report rework rate as reworked units over units produced in the same period.
What moves it
Process capability on key characteristics, targeted at a Cpk of 1.33 or better4.
Inspection burden, inspection hours as a share of labor hours, is a KPI you want low without starving quality. Typical aerospace shops run 15 to 25 percent inspection burden no cited source. Efficient shops using in-process probing, statistical sampling per AS9138, and reduced-inspection agreements push toward 8 to 12 percent while holding escape rate flat no cited source.
The improvement lever is not cutting inspection blindly; it is earning inspection reduction through demonstrated capability and moving checks upstream into the machine. A shop stuck at 30 percent burden usually has an unstable process it is inspecting its way around, which is far more expensive than fixing the process no cited source.
How to measure
Divide inspection hours by total labor hours for the same period.
Include operator in-process checks as well as final inspection.
Track escape rate alongside, so a lower burden is not bought with escapes.
What moves it
In-process probing that moves checks into the machine.
Statistical sampling under AS9138 where capability is demonstrated.
Reduced-inspection agreements earned with capability data.
Supplier escape rate is the KPI primes watch most closely, expressed in defective parts per million (PPM) shipped. RTX's Supplier Rating System, applied across RTX businesses over a rolling 12 months, rates a high-volume supplier Performing below 100 PPM, Progressing at 250 PPM or less, and Underperforming above 250 PPM6.
Suppliers shipping fewer than 100,000 pieces in 12 months are rated on escapes counted instead: one or fewer rates Performing, six or fewer Progressing, and more than six Underperforming6. A typical developing supplier runs 2,000 to 10,000 PPM no cited source.
Because one escape to a prime can cost 5,000 to 50,000 dollars in containment and corrective action, this KPI ties directly to whether you keep the contract no cited source. Track escape trends monthly rather than reacting to single events.
How to measure
Divide defective parts found by the customer by parts shipped, then multiply by one million.
Use a rolling 12 month window, as the RTX rating does6.
Below 100,000 pieces in 12 months, count escapes rather than PPM6.
On-time delivery (OTD) and nonconformance report (NCR) cycle time govern your standing on the scorecard. Prime supplier scorecards typically demand 95 percent OTD6, 7, and world-class suppliers sustain 98 to 99 percent7.
Anything under 90 percent rates Underperforming at RTX and Unsatisfactory at Lockheed Martin6, 7. RTX counts a lot as on time from three days late to 30 days early (business days) against the statistical date, so compare OTD figures only when the window is stated6.
Delivery rating bands on two prime supplier scorecards
RTX rates on a rolling 12 months6. The Lockheed Martin bands are from its scorecard revised in April 20217.
Nonconformance disposition cycle time is the days from NCR open to material review board (MRB) close. It should target under 5 working days world-class versus a typical 10 to 20 days no cited source, because parked material both blocks delivery and ties up cash.
How to measure
Measure OTD against the customer's commit date, with the on-time window written down.
Measure NCR cycle time in working days from open to MRB close.
Age the open NCR queue so old dispositions stand out.
What moves it
Disposition staffing sized to the engineer hours the NCR queue needs.
A standing MRB cadence rather than meetings called when the queue is already full.
First article and audit performance are lagging indicators of program health. World-class first article inspection (FAI) approval is first-time-right above 90 percent, meaning fewer than one in ten AS9102 packages bounces back no cited source. Typical shops see 20 to 35 percent first article inspection report (FAIR) rejections driven by ballooning errors and missing form data no cited source.
On the audit side, aim to close AS9100 surveillance findings in under 30 days with zero majors per cycle no cited source; multiple majors signals a system, not a part, problem. For comparison, one certification body, TUV Rheinland of North America, gives a client up to 60 days from the audit date to submit root cause and proposed corrective action for each nonconformity8. That is a deadline, not a target.
How to measure
Divide FAIR packages accepted on first submission by packages submitted.
Log each rejection by cause, such as ballooning errors or missing form data.
Measure audit finding closure in days from the audit date.
What moves it
A ballooning check before a FAIR package leaves the shop.
Complete AS9102 form data on every package.
System-level corrective action when major findings repeat.
Plan the hours with the AS9100 Audit Load calculator, so audits and first articles do not become fire drills before a customer visit.
Rework capacity and material certificate lead time
Rework capacity utilization and material certificate lead time are the operational KPIs that quietly wreck schedules. Keep planned rework hours under 10 to 15 percent of skilled-technician capacity so a defect spike has headroom no cited source. A shop already at 80 percent rework utilization has no shock absorber no cited source.
Material certificate lead time is the days from order to certified stock in hand. World-class is under 10 days for standard alloys versus 6 to 12 weeks for specialty titanium or melt compliant with the Defense Federal Acquisition Regulation Supplement (DFARS) no cited source.
Track it per alloy and hold safety stock on long-lead material. Certification delay, not machining, is the hidden pacing item on most aerospace jobs.
How to measure
Divide planned rework hours by skilled-technician hours available each week.
Measure certificate lead time per alloy, from purchase order to certified stock on the shelf.
Aerospace Supplier Quality Requirements (ASQR-01), Revision 15. RTX Corporation (Pratt & Whitney, Collins Aerospace). published May 22, 2025. PPAP process capability section, Table IV and following paragraph (PDF p. 18). Accessed September 30, 2026.
Supplier Quality Manual. AM General LLC. published February 2023. Section 3.3.11.4, Requirements for process capability indices. Accessed September 30, 2026.
Supplier Rating System User Guide. RTX Corporation (Raytheon Supplier Rating System). published 2026. p. 8, 'RTX performance criteria' table; p. 7 quality rating; p. 16 on-time window. Accessed September 30, 2026.
Lockheed Martin's Supplier Scorecard. Lockheed Martin Corporation. published April 2021. 'Delivery Rating Scale' panel. Accessed September 30, 2026.