Does Rework Count as Good in OEE? A Food Plant Answer

Key takeaways

• Good count means good the first time. Rework does not count as good in OEE.
• Reworked units still count in total count, because the line made them.
• Counting rework as good is one of the three common ways OEE gets fudged.
• Track rework on its own: units, labour minutes and cost.
• Your food safety plan decides which rework is allowed at all.

Rework does not count as good in OEE. A pack that had to be relabelled, repacked or re-run was not right the first time, so it stays out of good count. It still counts in total count, and its cost should be tracked on its own.

Fabrico, OEE and light MES software with AI cameras, counts rejects by machine, time and reason, and prices them in money. Which rework is allowed, and where reworked product may go, stays in your food safety plan.

What is good count in OEE?

Quality in OEE is good count divided by total count. OEE itself can be written as good count x ideal cycle time / planned production time. So every unit that leaves good count lowers both Quality and OEE.

Good count means units that were right the first time, with no repair, relabel or second pass. Process defects and reduced yield, two of the six big losses, both land in Quality. Our guide to OEE Quality covers the factor in detail.

Why does reworked product not count as good?

Because it was made twice. The line spent time making a unit that was not right, and then someone spent more time fixing it. If reworked units go back into good count, the Quality rate says the first pass went well when it did not.

Christoph Roser lists counting scrap or rework as good among the three common ways OEE gets fudged. The other two are shrinking the time basis and taking a good day's speed as the ideal. Our article on first pass yield explains the same idea from the quality side.

What kinds of rework happen in food plants?

• Relabelling: a wrong date code or label on otherwise good product.
• Repacking: a damaged case or film around sound product.
• Re-running: packs that failed a weight or seal check, sent back through the line.
• Blending back: product returned into a later batch, where the food safety plan allows it.

In food plants, rework carries its own risks, such as allergen cross-contact. Your food safety plan decides which rework is allowed, how it is recorded and where it may go. OEE only counts the units and their cost.

Rework is also a signal. A pack that needs a new label tells you something went wrong at the changeover or at the coder. Count it, and the cause rises in the Pareto.

Worked example: one shift with rework

Sample line (illustrative numbers, not a customer). A packing line makes 24,000 packs in a shift. A date code error means 600 packs must be relabelled. Another 240 packs are scrapped.

The wrong method hides 600 packs and 2.5 points of Quality. On the right method, the date code error shows up as the biggest quality loss of the shift.

Say the shift ran at 90% Availability and 95% Performance. Counting rework as good would report OEE of 84.6%. The true first-pass OEE is 82.5%.

The rework also has its own cost. Two people spend 90 minutes each relabelling: 180 labour minutes. At an illustrative USD 0.50 a labour minute, that is USD 90 (about EUR 83), plus 600 new labels. None of that appears in OEE, so track it next to it.

How do you track rework without losing sight of it?

1. Give rework its own reject reasons, by cause: date code, label, seal, weight.
2. Count reworked units when they leave the line, not when they are fixed.
3. Track rework labour and materials separately, because they sit outside OEE.
4. If reworked packs are re-run, do not count them twice in good count.
5. Report first-pass Quality and the rework rate side by side each week.

Our guides to scrap vs rework and rework vs repair explain the terms, and quality losses by machine and shift shows how to split them.

Which rework mistakes skew OEE?

• Adding reworked units back to good count, so Quality looks better than the first pass was.
• Not counting rework done off the line, so its cost vanishes.
• One code for rework and scrap, so you cannot tell a fixable loss from a lost unit.
• Rework that breaks the food safety plan. Never, whatever it does to the numbers.

Where does Fabrico fit?

Fabrico is a computer vision enabled Manufacturing Performance Platform. OEE and light MES come first, with rejects counted where they happen.

• Automatic downtime logging, so the stops behind a quality problem are timed.
• The operator confirms the reason in one tap, for stops and for rejects.
• Every loss priced in money, so rework and scrap have a cost.
• OEE live by line and shift, with Quality counted the first time round.

It is the OEE layer of a light MES. You start on one line, with a baseline and a decision date agreed in advance. There is no free plan and there are no tiers: each plant gets a quote.

Frequently asked questions

Does rework count as good in OEE?

No. Good count means units that were right the first time. Reworked units count in total count, because the line made them, but not in good count. Counting rework as good makes Quality and OEE look better than the first pass really was.

What is the difference between rework and scrap in OEE?

Both are Quality losses, because neither unit was good the first time. Scrap is thrown away. Rework is fixed and may still be sold. Give them separate reject codes, so you can see how much of the loss is recoverable and what the fixing costs.

Is first pass yield the same as OEE Quality?

They are close. Both count only units that were good the first time, out of all units made. OEE Quality is measured per line and shift against the line's total count. First pass yield is often reported per process step or per product.

Should reworked product be counted again when it is re-run?

Not as new good count. If reworked packs go back through the line, count the line time they use, but do not add them to good count a second time. Otherwise the same pack appears as good output twice.

How do you reduce rework on a food packaging line?

Start with the reject Pareto by cause and by SKU. Date code and label errors often trace back to changeovers, so check the settings step there. Track the rework rate weekly next to first-pass Quality, and fix the top cause first.

See your rework next to your rejects

Bring one week of reject and rework counts. We will show what each cause costs and where it starts.

Request a demo · Contact us