Power Electronics and Motor Line KPIs: Benchmark Ranges and How to Improve Them

Track a small set of KPIs on a power electronics or motor line and hold each against a reference band, so a number is either on track or worth a closer look. Use the ranges below as reference points for spotting a gap, not as targets, and compare against your own trend first.

Every figure here shows its basis. A superscript number links to a public source in the Sources list at the end, and the sentence states that source's 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

Power module first-pass yield

Power module first-pass yield is the most sensitive upstream indicator on the line. World-class IGBT and SiC module lines hold 97 to 99 percent first-pass no cited source, and 93 to 96 percent is typical no cited source.

Anything below 90 signals a real excursion or a mis-set test limit no cited source. Read the number with its sample size in mind, and trend it against control limits rather than reacting to one batch.

For example, each unit in a 250-unit batch is 0.4 points of yield, so three extra fails move first-pass yield by 1.2 points.

How to measure

• Count only units that pass with no rework as first-pass.
• Report it per product family, since SiC and silicon lines do not share a baseline.
• Plot each batch against control limits instead of a single target line.

What moves it

• Solder-void control, with a target under 3 percent voiding under the die no cited source.
• Wire-bond or sinter quality.
• Thermal interface material (TIM) application consistency.

Run your own batch numbers with the Power Module First-Pass Yield calculator, and check dispense consistency with the Thermal Interface Material Usage calculator.

Motor line OEE

Overall equipment effectiveness (OEE) is the headline throughput KPI for a motor line. It multiplies availability, performance and quality, so one number hides three different problems.

Across manufacturing, 60 to 70 percent is typical^{1, 2}. Vorne, an OEE monitoring vendor, says most plants sit closer to 60 percent¹, and a 2016 Chalmers study of 23 Swedish companies found a median of 70 percent and an average of 65 percent².

Measured averages can run lower still: a 2018 Chalmers study of Swedish manufacturers found machines averaging 51.5 percent OEE⁴. Below 55 means availability and changeover losses are eating the line no cited source.

World-class discrete assembly runs 80 to 85 percent OEE: the 85 percent end is the world-class mark from Seiichi Nakajima's total productive maintenance (TPM) work in Japanese discrete manufacturing¹, and the 80 percent end has no cited source no cited source.

A 2022 journal study tabulates Nakajima's world-class levels as OEE above 85 percent, with availability above 90, performance above 95 and quality above 99.9 percent³. That split is a useful yardstick for each leg.

For example, a line with 90 percent availability, 85 percent performance and 85 percent quality runs 0.90 x 0.85 x 0.85 = 65 percent OEE.

Decompose the number like that before you act, because it tells you which of the three legs to attack. On motor lines the performance leg usually drags, since winding is machine-paced with manual dress.

How to measure

• Define available hours as scheduled time minus planned maintenance, and freeze that definition.
• Report availability, performance and quality separately, then multiply them.
• Track OEE weekly per line and per product family.

What moves it

• Single-minute changeover on winding fixtures and fewer micro-stops, which typically buy 5 to 10 OEE points faster than chasing availability no cited source.
• Changeover discipline, which drives availability losses.
• First-pass quality at the winding and module steps, which drives the quality leg.

Put a cost on winding time and dress labor with the Motor Winding Labor calculator.

Test stand utilization and burn-in capacity

Test and burn-in capacity is a distinct KPI family because it is where fast build lines quietly stall. Motor test stand utilization should sit at 80 to 90 percent no cited source.

Below that band, setup and queue starvation dominate. Above 90 leaves no buffer for maintenance or re-tests no cited source.

For example, closing a gap from 75 to 85 percent utilization on one stand recovers about 48 hours for every 480 available hours.

Inverter burn-in capacity is the hidden constraint, because its multi-hour cycles do not compress. Measure effective good-unit output per chamber rather than slot count, and add parallel chambers only after confirming the loaded hours are genuine and not re-run churn.

How to measure

• Divide loaded test hours by available hours for each stand.
• Separate first tests from re-tests, so re-runs do not look like real load.
• Count good units out of each burn-in chamber per shift, not occupied slots.

What moves it

• Setup time and queue starvation at the stand.
• Rework volume, which inflates utilization with re-runs.
• Burn-in dwell settings, which fix chamber throughput.

Check your stands with the Motor Test Stand Utilization calculator, and size test cycles with the Drive Test Time calculator.

Good winding capacity

Good winding capacity is gross output discounted by uptime and first-pass yield. It should recover 85 to 90 percent of nameplate no cited source.

A line committing to gross capacity typically lands 10 to 15 percent short of plan no cited source. Plan on good capacity so the commitment matches what the line actually ships.

How to measure

• Start from nameplate output for the scheduled hours.
• Discount it by measured uptime.
• Discount it again by first-pass yield to get good units.

What moves it

• Uptime on the winding machines.
• First-pass yield at winding, including dress quality.
• Planning discipline that commits to good capacity rather than gross.

Model winding throughput and labor with the Motor Winding Labor calculator.

Scrap and rework rates

Scrap rate for power electronics runs 0.5 to 2 percent of build value world-class and 3 to 6 percent typical no cited source. Each point matters more here than in most industries, because scrapped modules carry loaded die value.

Rework rate should stay under 3 percent no cited source. Heavy rework masks a first-pass yield problem and inflates test stand utilization with re-runs that look like real load.

How to measure

• Value scrap at build cost, including the die, not at material cost alone.
• Count rework as units that needed any repair before passing.
• Report both per product family, weekly.

What moves it

• Upstream module assembly quality, which feeds scrap, rework and test load at once.
• Solder voiding and wire-bond quality on high-value modules.
• Fast feedback from test failures to the build step that caused them.

See what scrapped value does to unit cost with the Inverter Assembly Cost calculator.

Warranty field return rate

Warranty return rate is the lagging KPI that validates all the leading ones. World-class power electronics field return runs 0.2 to 0.5 percent per year over the warranty term, and 1 to 2 percent is typical no cited source.

Above 3 percent points to a factory test coverage gap, usually in thermal cycling or high-voltage insulation screening no cited source. If returns climb while first-pass yield holds, your test is passing latent defects.

Warranty cost is a separate measure from units returned. Across U.S. manufacturers filing with the SEC, warranty claims averaged 1.30 percent of product sales in 2025, and accruals averaged 1.43 percent⁵. That is an all-industry cost ratio, not a power electronics return rate.

How to measure

• Count field returns per year as a share of the installed base still under warranty.
• Review it monthly, since returns lag the build by months.
• Tag each return by failure mode so it maps back to a factory test.

What moves it

• Burn-in coverage and dwell time.
• A partial-discharge screen for high-voltage insulation.
• First-pass yield at module assembly, the leading indicator of returns.

Tightening burn-in dwell or adding a screen costs factory hours but pulls returns down and shrinks the warranty reserve. Size that trade with the Drive Test Time calculator.

How to build your own baseline

• Define each denominator once and freeze it, including available hours, first-pass and good capacity.
• Measure fast KPIs weekly and warranty monthly, so trends stay comparable.
• Segment by product family, because blending SiC and silicon lines buries the one that is drifting.
• Use hundreds to thousands of units before trusting a high-yield reading no cited source.
• Pair every KPI with its own control limits rather than a single target line.

Then prioritize by dollars per point, not by which number looks worst. A point of first-pass yield on a high-value module line often beats several points of OEE on a low-cost step, and closing a large test utilization gap can defer a capital purchase entirely.

Sequence the work upstream first. Winding and module assembly quality feed test load, scrap and warranty downstream, so fixing solder voiding or wire-bond quality lifts several KPIs at once.