HVAC Duct Shop KPIs: Yield, Leakage, Productivity and OEE Benchmarks
The KPIs that separate a strong duct shop from a typical one, how to measure each, the levers that move it, and which benchmark figures have a cited public source.
A duct shop turns steel into sellable square feet, and every KPI below measures a piece of that conversion. Use the ranges as reference points for spotting a gap worth investigating, not as targets. Your own trend, measured the same way every month, says more about your shop than any single published figure.
Every figure here shows its basis. A superscript number links to a public source in the Sources list, and the sentence states that source's scope, such as its year and who it applies to. A figure marked no cited source is kept exactly as originally published because no public source stating it has been verified, so treat it as illustrative.
Ranges are as originally published unless a source is cited. Where the original used a word other than typical, such as world class, target or struggling, that word is kept in the cell. The cited OEE and delivery figures come from other industries, as each section explains. 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. The cited OEE, delivery and duct leakage figures come from studies across industries, one customer's supplier scale, and a U.S. Department of Defense specification, and the sentence says so. A figure marked no cited source is kept exactly as originally published because no public source stating it has been verified. It is illustrative, not measured industry data.
Material yield
Material yield is the first KPI because steel is the biggest cost and the most controllable waste. Typical rectangular duct shops run 82 to 88 percent nesting yield no cited source. World-class operations hold 90 to 94 percent through tight nesting software and standardized sheet sizes no cited source.
For example, a shop buying 40,000 lb of steel a month that moves from 85 to 91 percent yield needs roughly 2,600 lb less steel for the same output. At roughly 1.15 to 1.35 dollars per lb, that is 3,000 to 3,500 dollars a month.
How to measure
Divide finished square feet shipped by square feet of steel consumed.
Labor productivity is best tracked as pounds fabricated per labor hour, which normalizes across gauge and size. A typical shop turns 55 to 75 lb per direct labor hour on standard rectangular work no cited source. Strong shops with coil lines and auto-benders reach 90 to 120 no cited source.
Spiral and fitting-heavy work runs lower, 30 to 50 lb per hour no cited source. Segment the metric by product type, or it lies.
How to measure
Divide pounds fabricated by direct labor hours.
Report rectangular, spiral and fitting work separately.
What moves it
Batching like fittings.
Cutting handling moves.
Pushing standard sections to automated lines.
Reserving skilled hands for transitions and custom tie-ins.
Leakage class is the quality KPI that customers and code both watch. SMACNA Seal Class A on rectangular low-pressure duct targets a leakage class of 6 or lower no cited source, and tight shops hold field results at 3 to 4 no cited source.
Project specifications can differ, so check yours. The U.S. Department of Defense's May 2025 specification for HVAC testing and balancing requires Seal Class A on all tested ductwork6.
The same specification caps rectangular duct at leakage class 16 for 1 in. WG pressure class, 8 for 2 and 3 in. WG, and 4 for 4 in. WG and above. Round and oval duct is capped at 8, 4 and 37.
The failure rate tells you what leakage costs in rework. World-class is a first-pass test failure rate under 3 percent no cited source. Typical shops sit at 6 to 10 percent and eat the rework no cited source.
How to measure
Track failures per hundred sections tested.
Pair every failure with a root cause.
Check the leakage class your project specification requires, by pressure class.
What moves it
Longitudinal seam sealing, a usual root cause.
Transverse joint sealing, the other usual root cause.
Better sealant coverage and a torqued lock, which beat re-testing every time.
Throughput measures how much sellable product the shop pushes per shift, and it exposes hidden constraint time. A single-shift rectangular shop commonly runs 900 to 1,400 finished square feet per shift no cited source.
A well-tuned line with a coil line and auto-nesting reaches 1,600 to 2,200 no cited source. Tie output to line rate and staffing to see whether the bottleneck is the coil line, the assembly bench, or the leakage test station.
How to measure
Count finished square feet completed and passed per available shift hour.
Include setup and rework time, so they drag the number honestly.
What moves it
The constraint station: coil line, assembly bench, or leakage test.
A coil line and auto-nesting.
Less setup and rework time at the constraint.
Overall equipment effectiveness
Overall equipment effectiveness (OEE) on the coil line and plasma table catches the time you are paying for but not selling. Typical fabrication OEE lands at 45 to 60 percent1, 2.
That band comes from studies across industries, not duct shops. A 2018 study of Swedish manufacturers measured an average of 51.5 percent, and Vorne reports most manufacturers closer to 60 percent1, 2.
The world-class benchmark usually quoted comes from Seiichi Nakajima's total productive maintenance work: OEE above 85 percent, built from availability above 90, performance above 95 and quality above 99.9 percent3, 2. It applies across industries, not only to metal shops.
For example, a line that is up 90 percent, running at 85 percent of rate, with 96 percent good output scores 0.90 times 0.85 times 0.96, which equals 73 percent.
The biggest lever is usually availability. Single-minute exchange of die (SMED) sets the changeover target at under 10 minutes8.
For example, cutting changeover from 25 minutes to 10 on a line that changes over 12 times a shift saves 15 minutes 12 times, returning 3 hours of run time.
How to measure
Multiply availability, performance, and quality on the coil line and the plasma table.
Log each changeover's length and count per shift.
What moves it
Availability, usually the biggest lever.
Shorter changeovers, which return run time directly.
Scrap and rework
Scrap and rework rate is the KPI that quietly funds or drains margin. Benchmark rectangular scrap at 8 to 12 percent of steel consumed, and spiral and fittings at 12 to 18 percent no cited source. Best-in-class holds rectangular under 7 percent no cited source.
Rework hours as a share of direct labor should sit below 4 percent in a strong shop and often run 8 to 12 percent in a struggling one no cited source.
How to measure
Express scrap as a share of steel consumed, by product type.
Express rework hours as a share of direct labor hours.
Track both weekly.
What moves it
Closing training gaps, a usual cause of a rising rework line.
On-time delivery closes the loop on the whole system. A well-run shop ships 95 percent or more of jobs on the committed date4, 5. Slipping below 90 percent4 usually signals throughput promises that ignore constraint capacity.
Those thresholds line up with published reference points from other industries. Lockheed Martin's April 2021 supplier scorecard rates delivery of 95 to 97.99 percent as Approved, 98 to 100 percent as Excellent, and below 90 percent as Unsatisfactory4.
The IndustryWeek and PricewaterhouseCoopers census of U.S. plants in spring 2000 set the bar for delivery leaders at 98 percent or better, and only about a third of plants reached it5.
Pair delivery with quote-to-actual cost variance. World-class holds within plus or minus 5 percent, while typical drifts 10 to 15 percent no cited source.
How to measure
Count jobs shipped on the committed date as a share of jobs shipped.
Compare each job's actual cost with its quoted cost.
What moves it
Delivery promises based on constraint capacity.
Estimates that use the yield and fitting labor the shop actually achieves.
When variance widens, the culprit is almost always yield or fitting labor that the estimate assumed and the shop never achieved. That points straight back to the yield and productivity KPIs.
Which levers to pull first
Improve these in order, because they stack. Fix yield first since it funds everything, then attack leakage first-pass rate to kill rework, then push throughput and OEE once quality is stable.
For example, take a shop that lifts yield from 85 to 91 percent, cuts test failures from 9 to 3 percent, and raises pounds per hour from 65 to 85. The yield change alone gives 91 / 85, or about 7.1 percent more output from the same steel, and the lower failure rate adds about 6.6 percent more first-pass output. Together that is roughly 10 to 15 percent more sellable output from the same steel and payroll.
Set monthly targets and post them at the coil line. Then review the two or three that are drifting rather than chasing all of them at once.
How to build your own baseline
Pick one KPI, write its definition down, and keep it fixed.
Segment yield, productivity and scrap by product type before comparing.
Collect several months of data before you set a target.
Compare against your own trend first and published ranges second.
Record what changed in the process whenever the KPI moves.
World-Class OEE: Set Targets To Drive Improvement. Vorne Industries (OEE.com). Sections 'What Is World-Class OEE?', 'TPM and World-Class OEE', 'Some Perspective on World-Class OEE'. Accessed September 30, 2026.
Anatomization of the overall equipment effectiveness (OEE) for various machines in a tool and die shop. Journal of Intelligent Manufacturing and Special Equipment, Vol. 3 No. 1, pp. 97-105 (Emerald, CC BY 4.0), DOI 10.1108/JIMSE-01-2022-0004. published 2022. p. 101 (PDF p. 5): sentence "Seiichi Nakajima included the above four “world-class” numbers." and Table 1 "The world-class OEE factor"; p. 103 (PDF p. 7): "worldwide threshold of >85% set by the Japan Institute of Plant Maintenance (JIPM)". Accessed September 30, 2026.
Lockheed Martin's Supplier Scorecard. Lockheed Martin Corporation. published April 2021. 'Delivery Rating Scale' panel. Accessed September 30, 2026.