Pultrusion KPIs: Uptime, Scrap, Fiber Volume and Throughput Benchmarks

Use these benchmarks as reference points for spotting a gap worth investigating, not as targets. Profiles differ widely in section, resin and reinforcement, so your own trend per profile says more than any single published figure.

Every figure shows its basis. A superscript number links to a public source in the Sources list, and the sentence gives that source's scope. A figure marked no cited source is kept exactly as originally published because it has not been matched to a public source that states it, so treat it as illustrative.

At a glance

Line uptime

Line uptime is the headline KPI. Typical pultrusion lines run 65 to 78 percent uptime no cited source, while world-class lines sustain 85 to 92 percent no cited source. The gap is mostly changeover and roving breaks, not catastrophic failure.

The dominant lever is changeover reduction. Moving die swaps from 3 hours to under 90 minutes, using preheated spare dies and staged creels, can add 8 to 12 points of uptime no cited source. That is worth roughly 10 to 15 percent more good feet per year with zero capital on the line itself no cited source.

For example, saving about 1.5 hours per die swap adds 8 to 12 points of uptime when a swap comes roughly every 12 to 19 scheduled hours.

How to measure

• Divide run-hours by scheduled hours per shift.
• Log every stop to the minute.
• Code each stop so changeovers and roving breaks stay separate from failures.

What moves it

• Preheated spare dies for faster die swaps.
• Creels staged before the changeover starts.
• Fewer roving breaks during the run.

See what recovered uptime is worth in output with the Line Throughput calculator.

Scrap rate

Scrap rate is where good lines separate from average. Typical runs carry 5 to 9 percent scrap by length no cited source, while best-in-class lines hold 2 to 3.5 percent no cited source.

Startup purge dominates, at 15 to 40 ft per changeover no cited source. The highest-leverage fix is a longer average run, because the fixed purge is spread over more good footage.

For example, going from a 300 to a 900 ft average run cuts startup-dominated scrap by roughly two-thirds. With a 30 ft purge, purge scrap falls from 30 / 330, or 9.1 percent, to 30 / 930, or 3.2 percent.

How to measure

• Divide scrapped feet by total feet pulled.
• Segment scrap by cause code: startup, exotherm upset, dimensional and surface.

What moves it

• Longer average run length.
• A shorter startup purge at each changeover.
• Stable cure that avoids exotherm upsets.

Convert scrapped feet into pounds with the Profile Weight Per Foot calculator.

Fiber volume consistency

Fiber volume consistency governs both strength and cost. Target fiber volume fraction (Vf) for structural profiles is 0.50 to 0.65 by volume, inside the range published data report for thermoset pultruded parts¹.

World-class lines hold the shift-to-shift standard deviation under 1.5 percentage points, while typical is 2.5 to 4 no cited source. Tight Vf control means you can safely design closer to spec and cut 3 to 6 percent of glass out no cited source.

How to measure

• Measure Vf by burn-off or acid digestion on a cut coupon each shift.
• Do not rely on the theoretical end count alone.
• Track the shift-to-shift standard deviation, not just the mean.

What moves it

• Creel tension uniformity.
• Bath viscosity control within plus or minus 100 cP no cited source.
• A confirmed end count after every roving change.

Check the fraction a given roving count and section should give with the Fiber Volume Fraction calculator.

Throughput in good pounds per hour

Throughput yield in good pounds per hour is the KPI that pays the bills. The gap between world-class and typical lines is entirely uptime and scrap compounding, so report actual good pounds per hour against theoretical to expose the losses.

For example, a single die on a 0.6 lb/ft profile at 2.4 ft/min nets 0.6 x 2.4 x 60 = 86.4 lb/hr theoretical. Applying this guide's uptime and scrap ranges, a world-class line realizes about 71 to 78 lb/hr and a typical line about 51 to 64.

The primary lever after uptime is cure optimization. Raising die zone temperatures 10 to 20 C where the resin system allows can lift safe pull speed 10 to 20 percent without demold defects no cited source. For scale, the European Pultrusion Technology Association (EPTA) notes that the impregnated reinforcement is heated up to 160 °C inside the die, depending on the resin system².

How to measure

• Compute theoretical rate as weight per foot times pull speed times 60.
• Report actual good pounds per hour against that theoretical rate.
• Split the gap into uptime loss and scrap loss.

What moves it

• Uptime, first.
• Cure optimization through die zone temperatures, where the resin system allows.

Run your own profile through the Line Throughput calculator.

Pull speed utilization

Pull speed utilization measures how close you run to the cure-limited ceiling. Many lines operate at 70 to 80 percent of achievable speed out of caution, while disciplined lines run 90 to 95 percent no cited source.

The ceiling depends on the profile. EPTA puts thermoset pultrusion speeds anywhere from 0.02 to 3.0 m/min, about 0.07 to 10 ft/min, depending on wall thickness, reinforcement and the complexity of the cross-section³.

Establish the ceiling empirically by pushing speed until surface or gel defects appear, then back off 8 to 10 percent no cited source.

At constant uptime and yield, every 5 percent of speed recovered is 5 percent more throughput on the same fixed labor and overhead.

How to measure

• Trend actual encoder speed against the validated ceiling for each profile.
• Validate the ceiling again when the resin system or die changes.

What moves it

• An empirically validated speed ceiling for each profile.
• Operators who can see actual speed against that ceiling, so output is not silently left on the table.

Check the cure-limited speed with the Cure Die Dwell calculator and set line speed with the Pull Speed calculator.

First-pass dimensional yield and surface quality

First-pass dimensional yield and surface quality round out the quality KPIs. Aim for over 97 percent of pulled length within tolerance on wall thickness and straightness; typical lines sit at 92 to 96 percent no cited source.

The recurring defects are resin-rich corners, dry fiber on the surface, and warpage from uneven cure. Balance die heat zones so peak exotherm sits at 60 to 70 percent of die length no cited source. For reference, EPTA gives pultrusion die lengths of approximately 0.75 to 1.5 m, about 2.5 to 5 ft⁴.

How to measure

• Divide pulled length within tolerance on wall thickness and straightness by total pulled length.
• Log surface defects by type: resin-rich corners, dry fiber and warpage.

What moves it

• Balanced die heat zones that place peak exotherm correctly.
• Controlled puller-to-cutoff tension.
• A resin bath held within a 200 cP viscosity window across the run no cited source.

Track bath usage across the run with the Resin Bath Consumption calculator.

Die-heater energy intensity

Energy intensity is the efficiency KPI worth tracking annually. Well-run glass pultrusion lines consume 0.05 to 0.10 kWh per pound of profile at the die heaters no cited source, while poorly insulated or oversized-die setups hit 0.15 to 0.20 no cited source.

This KPI is minor per foot but compounds. For example, trimming from 0.15 to 0.08 kWh/lb on a 350,000 lb/yr line at $0.12/kWh saves 0.07 x 350,000 x $0.12 = $2,940, about $2,900 a year.

How to measure

• Divide die zone kWh by good pounds.
• Meter the die heaters separately from the rest of the line.

What moves it

• Insulated die faces.
• Heater wattage sized to the section.
• Avoiding long idle-hot periods between runs.

Convert footage into pounds for the denominator with the Profile Weight Per Foot calculator.

How to build your own baseline

• Write down each KPI's definition and keep it fixed.
• Report every KPI per profile, since profiles differ widely.
• Collect several months of data before you set a target.
• Compare against your own trend first and published ranges second.