Dosing and Loss-in-Weight Feeding KPIs: Benchmark Ranges and How to Measure Them

Gravimetric dosing lines are judged on accuracy, money and stability: dosing variation, batch pass rate, give-away, feeder utilization, refill frequency, refill error and calibration drift. Use the ranges below as reference points for spotting a gap worth investigating, not as targets, and compare against your own trend first.

Every figure 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 none of the public sources checked for this guide states it, so treat it as illustrative.

At a glance

Dosing coefficient of variation

Dosing coefficient of variation (CV) is the headline accuracy KPI. Typical bulk feeders run a CV of 1.0 to 2.0 percent at rate, and well-tuned loss-in-weight feeders on free-flowing material hit 0.5 to 1.0 percent no cited source.

World-class twin-screw units on cohesive powders reach 0.25 to 0.5 percent at 2 sigma no cited source. Coperion K-Tron's feeder accuracy guide notes that repeatability has traditionally been expressed at two standard deviations over minute-to-minute samples, so state the sigma level with every CV you report¹.

Measure it from at least 30 consecutive catch samples², never from the controller's own display. The same guide's example of a complete repeatability statement is based on 30 consecutive samples of one minute, one kilogram, one belt revolution or thirty screw revolutions, whichever is greater².

How to measure

• Take 30 consecutive timed catch samples at the running rate².
• Make each sample at least one minute or one kilogram, whichever is greater³.
• Report the standard deviation of the sample weights as a percent of their mean, with the sigma level.

What moves it

• Screw geometry and refill frequency, the usual levers if CV sits above 2 percent, rather than loop tuning no cited source.
• Material flow behavior, since cohesive powders are harder to meter than free-flowing ones.
• Feeder type matched to the material, such as twin-screw units for cohesive powders.

Compare your CV against these bands with the Dosing Accuracy calculator.

Batch pass rate

Batch pass rate is the share of batches inside the tolerance window on the first attempt. It should sit at 98 to 99.5 percent for a mature line no cited source.

Below 95 percent you are losing an hour or more per shift to rework no cited source. A single cohesive or micro-heavy formula often drags the plant average down 2 to 3 points no cited source.

How to measure

• Count batches inside the tolerance window on the first attempt, divided by batches made.
• Track it per recipe, not only as a plant average.
• Log which ingredient caused each failed batch.

What moves it

• Feeder resolution matched to the tightest ingredient in the recipe.
• Per-ingredient tolerance bands that combine to a workable batch window.
• Widening the window, but only as a last resort.

See how per-ingredient bands combine with the Batch Tolerance Window calculator.

Give-away

Give-away percentage is the money KPI that plant managers under-watch. World-class fill lines hold average give-away to 0.1 to 0.3 percent above target, while 0.5 to 1.0 percent is common and quietly expensive no cited source.

For example, on a 4,000 t/year line at $2.40/kg, moving from 0.8 percent to 0.25 percent give-away is roughly $53,000 recovered: 4,000,000 kg x $2.40 is $9.6 million of material, and 0.55 percent of that is $52,800.

How to measure

• Take the mean of (actual minus target) / target across a week of batches.
• Weight the mean by batch mass.
• Convert the percentage to dollars at your material cost.

What moves it

• Lower feeder scatter, which lets you reduce the safety bias.
• A safety bias set from measured scatter, never below minimum weight.

Convert give-away into dollars with the Ingredient Variance Cost calculator.

Feeder utilization

Feeder utilization separates a rate problem from a scheduling problem. Continuous process lines target 85 to 92 percent utilization, while batch and high-mix plants realistically run 60 to 75 percent no cited source.

For example, if throughput is short but utilization is already 88 percent, you have a speed limit, not idle time. If utilization is 58 percent, chase changeover, starvation and refill dwell first.

How to measure

• Divide running time by available time, per shift.
• Split available time into running, refill and idle so the gap is visible rather than assumed.

What moves it

• Changeover time between recipes.
• Feeder starvation from upstream supply.
• Refill dwell on each hopper.

Check what rate the feeder can deliver with the Feeder Rate calculator.

Refill frequency

Refill frequency is an underrated stability KPI because every refill drops the feeder into volumetric control. Coperion K-Tron's guide explains why: the rising hopper weight cannot be used for rate control, so a feeder that simply holds its speed during refill acts as a volumetric feeder⁴.

Target fewer than 3 to 4 refills per hour on a primary feeder no cited source. More than 6 means the hopper is undersized for the rate and open-loop time exceeds 5 percent of run time no cited source.

Pair refill frequency with usable-mass ratio, ideally keeping the low trip at 20 to 30 percent so you carry 70 to 80 percent usable inventory no cited source.

How to measure

• Count refills directly from hopper level trips.
• Record the low and high trip points for each feeder.
• Log the time each refill keeps the feeder out of gravimetric control.

What moves it

• Hopper size matched to the feed rate.
• A lower low-trip point, which extends the interval.
• Refill compensation that holds the feed factor learned before the refill.

See how bigger hoppers or lower trips extend the interval with the Hopper Refill Interval calculator.

Gravimetric-to-volumetric delta during refill

The gravimetric-to-volumetric delta during refill is the KPI that bounds accuracy loss. Keep the delta under plus or minus 1 percent by controlling bulk-density swing and keeping refill dwell short no cited source.

A world-class line sees density variation under 2 percent batch to batch and refill dwell under 15 seconds no cited source.

For example, with 4 refills per hour of 15 seconds each, the feeder runs open loop for 60 of every 3,600 seconds, or 1.7 percent of the time. A 2 percent density error during refill then dilutes to about 0.03 percent of the batch, so the diluted batch error stays under 0.1 percent.

How to measure

• Compare delivered mass during refill with the gravimetric rate just before it.
• Track bulk density batch to batch.
• Time each refill from valve open to return to gravimetric control.

What moves it

• Pre-blending for density consistency, if the delta creeps past 2 percent no cited source.
• Faster refill valves.
• Smooth speed changes on return to gravimetric control.

Test refill timing with the Hopper Refill Interval calculator.

Calibration drift

Calibration drift is a maintenance KPI: the feed factor k should not move more than 1 to 2 percent between scheduled checks no cited source. A feeder drifting 3 to 5 percent between checks needs mechanical attention before you retune software no cited source.

The usual mechanical causes are worn screws, bearing drag or a load cell zero problem. Verify weekly on high-volume feeders and monthly on micros using a timed catch test no cited source, and log the drift. Coperion K-Tron notes that experience will dictate the required frequency of calibration checks for any given feeding application, and recommends run charts to trend calibration data over time⁵. Let your drift log set the frequency.

A single-point check can miss a feed factor drop at low speed. Coperion K-Tron's linearity method takes ten consecutive catch samples at each of 5, 25, 50, 75 and 100 percent of full scale⁶.

How to measure

• Run a timed catch test against the controller's feed factor.
• Log the drift at every scheduled check and trend it on a run chart.
• Check linearity across the speed range, not at one point.

What moves it

• Screw wear.
• Bearing drag.
• Load cell zero stability.

Report linearity across the speed range with the Loss-In-Weight Calibration calculator.

A weekly scorecard

Roll these KPIs into a scorecard reviewed weekly, with one owner and one lever per KPI. These are the weekly targets to start from.

Set a stretch target one band tighter than current. The common failure is chasing utilization for output while give-away and CV erode margin unwatched, so pair throughput figures with the accuracy and cost figures every week.

Size throughput with the Feeder Rate calculator and price accuracy with the Ingredient Variance Cost calculator.

How to build your own baseline

• Write each KPI's definition down and keep it fixed.
• Measure accuracy from catch samples, not the controller display.
• Track accuracy, give-away and utilization per feeder and per recipe.
• Compare against your own trend first and published ranges second.
• Record what changed in the process whenever a KPI moves.