Injection Molding KPIs: Benchmarks for OEE, Cycle Time, Scrap and Energy

Use these benchmarks as reference points for spotting a gap worth investigating, not as targets. A press's own trend, measured the same way every week, says more about your molding floor 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 context. A figure marked no cited source is kept as originally published because no public source stating it has been verified, so treat it as illustrative.

At a glance

Overall equipment effectiveness

Overall equipment effectiveness (OEE) is the headline KPI for a molding floor. It multiplies availability, performance and quality, so a loss in any one of them pulls the whole number down.

Typical injection molding shops run 60 to 75 percent OEE no cited source. Published cross-industry averages sit at or below the bottom of that band: 94 usable OEE values from Swedish manufacturers averaged 51.5 percent in a Chalmers University study³, and an OEE vendor observes that most plants are closer to 60 percent⁵.

World-class operations sustain 85 percent or higher^{1, 3}. That line traces to Seiichi Nakajima's total productive maintenance work and applies across discrete manufacturing, not to molding alone.

For example, a press at 75 percent OEE that loses 5 points of availability to changeovers, 5 to slow cycles and the rest to scrap has a clear improvement map.

How to measure

• Calculate it per machine per shift, not as a plant average, because one problem tool can mask a healthy floor.
• Track availability, performance and quality separately, so you know whether the loss is downtime, speed or defects.

What moves it

• Faster changeovers and preventive maintenance, which lift availability.
• Cycles held near the engineered ideal, which lift performance.
• Defects fixed at the tool, which lift quality.

Cycle time index

Cycle time performance is best tracked as an index: actual cycle divided by the engineered ideal cycle. World-class shops hold it at 1.0 to 1.05, running within 5 percent of the theoretical best no cited source. Typical shops drift to 1.15 or worse as processes are dialed back to hide instability no cited source.

For example, if the ideal cycle is 30 seconds and machines average 34, the index is 34 / 30, or 1.13, and the lost capacity is 1 minus 30 / 34, about 12 percent.

The lever is process discipline. A cycle padded to mask a nuisance defect costs capacity on every shot, so fix the defect at the tool instead.

How to measure

• Set the engineered ideal cycle for each tool and material.
• Divide the actual average cycle by that ideal for each job and shift.

What moves it

• Scientific molding.
• Consistent cooling.
• Nuisance defects fixed at the tool rather than hidden by a longer cycle.

Build the engineered ideal from its parts with the Injection Molding Cycle Time calculator.

Scrap rate

Scrap rate is the quality KPI that hits material and machine cost at once. Best-in-class molding runs 1 to 2 percent scrap, and 3 to 5 percent is common no cited source. Anything above 5 signals a process or tooling problem no cited source.

The usual heavy hitters are short shots, flash and cosmetic defects. Track scrap by defect code so a Pareto points at the top two or three causes.

For example, cutting scrap from 5 to 2 percent on a high-volume job recovers roughly 3 percent of both resin spend and press hours, with no capital spent.

How to measure

• Divide rejected shots by total shots.
• Record every reject against a defect code.

What moves it

• A defect-code Pareto that points at the top causes.
• Fixes at the tool or in the process for short shots, flash and cosmetic defects.

Short shots often trace to shot size and cushion, which you can check with the Shot Size calculator and the Cushion Percentage calculator.

Regrind usage

Regrind usage is a cost and quality balance, not a number to maximize or minimize. World-class shops keep regrind blends controlled and documented, commonly 10 to 25 percent on non-cosmetic parts, with the 25 percent ceiling matching the UL 746D allowance². On medical or optical work the share is near zero no cited source.

Under UL 746D, in-house sprue and runner regrind can make up to 25 percent of the hopper when the other 75 percent is the same virgin grade². A resin maker's molding guide for Delrin acetal calls a 3 to 1 ratio of virgin resin to regrind common⁶.

Watch regrind alongside scrap. Rising regrind with rising scrap means you are recycling defects back into defects.

How to measure

• Percent regrind in the blend, against the spec limit for the part.
• Scrap rate on the same jobs, reviewed together with regrind.

What moves it

• Closed-loop control on the blender.
• A firm cap tied to the part spec, so material savings never cost mechanical properties.

See how much runner material each shot generates with the Runner Weight calculator.

Cavity balance and first-pass yield

First-pass yield and cavity balance matter on multi-cavity tools. A well-balanced mold holds cavity-to-cavity weight variation under 3 percent no cited source. Poor balance runs 8 percent or more no cited source and forces you to over-pack good cavities to fill the weak ones.

Track shot-weight standard deviation as a percent of mean, targeting under 1 percent for a stable process no cited source. Yield below 97 percent on an established job usually points to a specific cavity rather than the whole tool no cited source.

How to measure

• Weigh parts from each cavity and compare each with the mean.
• Track shot-weight standard deviation as a percent of mean.
• Report first-pass yield by cavity, not only by tool.

What moves it

• Runner and gate corrections when one cavity consistently short-fills.
• A stable process window, left alone once the tool itself is balanced.

Set the expected weight for each cavity with the Plastic Part Weight calculator.

Uptime and changeover speed

Uptime and changeover speed set how much of a press's available hours actually run. World-class molders exceed 90 percent scheduled uptime, the availability factor in Nakajima's world-class OEE benchmark¹. They turn a mold change in under 20 minutes using SMED no cited source.

Typical shops sit at 75 to 85 percent uptime³ with 45 to 90 minute changeovers no cited source. In the Chalmers study of Swedish manufacturers, machine availability averaged 78.9 percent across industries³.

The Lean Enterprise Institute defines the SMED target as a single-digit changeover, less than 10 minutes⁷. Mold size and quick-mold-change hardware decide how close a molder can get.

For example, a press run 24 hours a day, 5 days a week for 52 weeks has 6,240 available hours a year, and a full calendar year holds 8,760. Every 30 minutes cut from a changeover done twice a shift returns an hour of production per shift on that press.

How to measure

• Scheduled uptime: run time over scheduled time, per press.
• Changeover minutes per event, from the last good part to the first good part.
• Mean time between failures, per press.

What moves it

• SMED: separating internal from external setup and moving internal work outside the stop.
• Preventive maintenance aimed at the longest recurring stops first.

Energy per kilogram

Energy per kilogram of throughput is tracked more and more as electricity dominates conversion cost. Hydraulic presses typically draw 0.8 to 1.3 kilowatt-hours per kilogram of plastic processed⁴, and all-electric machines cut that to 0.4 to 0.7⁴.

MIT measurements published in 2006 fall inside those bands. Across more than 100 energy measurements and calculations, hydraulic machines averaged 3.39 MJ/kg, about 0.94 kWh/kg, and all-electric machines 1.46 MJ/kg, about 0.41 kWh/kg, for the machine alone⁴.

The U.S. Department of Energy credits a switch from hydraulic to all-electric machines with 74 percent machine energy savings, citing Kanungo and Yong (2012)⁸. That is a larger cut than the bands above imply.

Whole-plant figures run far higher, because they include dryers, chillers, compressed air and building loads. DOE's 2017 bandwidth study put current typical on-site energy for U.S. injection molding of polypropylene in 2010 at 4,065 Btu per pound of resin, about 2.63 kWh/kg, against a state-of-the-art 1,659 Btu per pound, about 1.07 kWh/kg⁸.

Right-sizing a job onto a press loaded near 60 to 70 percent of clamp capacity, rather than 30 percent, meaningfully lowers energy per part and frees the big press for work that needs it no cited source.

How to measure

• Meter each press and divide kilowatt-hours by kilograms processed.
• Record whether each figure is machine only or whole plant.
• Compare each job's required clamp force with the press's nameplate.

What moves it

• Moving suitable jobs onto all-electric presses.
• Right-sizing jobs onto smaller presses.
• Auxiliary upgrades that DOE lists, such as barrel heater insulation, low-pressure drying and variable-speed chilled water pumps.

Size the clamp force a job needs, so it runs on a press that fits it. Clamp Tonnage

How to build your own baseline

Improvement is sequential, not simultaneous. Chasing a faster cycle on an unstable tool just raises scrap and cancels the gain.

• Stabilize the process first, so cycle index and scrap are predictable.
• Then attack availability with faster changeovers and preventive maintenance.
• Only then push cycle time toward the engineered ideal.
• Review these KPIs weekly per machine, set a single target per quarter, and tie each to one lever.

For example, a floor at 75 percent OEE gets a written path to 82 and then to world-class, with no guessing about which number to move next.