OEE for Fresh Produce Packing: How to Measure a Packhouse Line

A packhouse does not behave like a bottling hall. The raw material changes with the grower, the field and the weather, and the plan changes with the retailer's order that morning.

That is why a standard OEE setup often gives a packhouse numbers nobody trusts. This page shows how to set it up so the number follows the line, not the fruit.

We sell Fabrico, so test what we say here like any other vendor's claims.

The packhouse line, and where it loses

Most fruit and vegetable packing lines follow the same order. Root crops and apples are washed or brushed, while soft fruit and tomatoes usually go straight from the tip to the sorter.

The pack machine in the middle sets the pace, and the stations before and after it starve it or block it.

The table below is the typical picture, not a measurement. Use it as a first list of stop codes.

Short jams matter more than they look. See our page on micro-stops for how they hide inside a shift.

Grade-out: OEE quality or pack-out?

This choice decides if the packhouse will trust the number. No OEE standard settles it for produce, and plants use two methods.

Method 1, pack-out next to OEE. OEE quality answers one question: of the packs the line made, how many were good?

Fruit that the sorter or the graders remove before the pack is made was never a pack. It goes into pack-out: kilos packed against kilos tipped, by grower lot.

Method 2, grade-out inside quality. The graded-out kilos are turned into packs and counted as quality loss, which gives one number for everything lost between the tipper and the pallet.

We suggest method 1 for a line report. With method 2 the number goes up and down with the delivery, and the line team cannot fix a hail-marked lot.

Method 2 can suit a site report that is meant to show raw material loss. Whichever you pick, write it down and use it on every line, or the lines cannot be compared.

The grade itself comes from the standard and the customer, not from the line. The apple marketing standard in Regulation (EU) No 543/2011, in the copy kept in UK law, allows in Class I a total tolerance of 10 per cent, by number or weight, of apples that miss the class but meet Class II.

Poor fruit still costs line time, and OEE shows that under either method. When a bad lot forces the infeed down, book the lost speed as a performance loss with the reason "raw material".

We would not lower the ideal rate for the lot. The rate stays, and the reason tells you what the lot cost.

Overweight packs are the other loss OEE cannot see, because they pass as good. Track them as their own figure, as our page on overfill and OEE quality explains.

Ideal rate and planned time for short retailer runs

A packhouse may run ten orders in a shift: the same tomatoes in three punnet sizes, under four retailer labels. Two rules keep OEE fair across them.

One ideal rate per pack format

Set the ideal rate at the pack machine, for each pack format: 250 g punnet, 500 g flow wrap, 1 kg net. Use the best rate the line has held on good fruit with a full crew, and write it down.

The grader's kilos per hour is a different number. Keep it for capacity planning, not for OEE.

Changes are stops inside planned time

A label, film or crate change costs output the line could have made. Count it as an availability loss, inside planned time, with its own code.

Give each change type a standard: label only, film and label, full format change. Then track the minutes over the standard, and the slow minutes after restart.

Take breaks out of planned time. Leave wash-downs and water changes that fall inside the shift in, as stops with their own code, so the minutes stay visible.

Then report OEE by order as well as by shift. A shift average hides the two formats that cause most of the stops.

People-paced stations

Many lines still place fruit by hand: into punnets, onto trays or into crates. There the people set the speed, and the machine signal says little.

Set the ideal rate for the station and the pack format with its standard crew, never for a named person. Report by station and shift.

When the lane runs two people short, the count drops. Book that as a staffing reason, so it does not sit in the report as an unexplained slow shift.

Our page on OEE at a manual packing station covers the method in full.

Counting packs with no usable PLC count

A packhouse has many counters, and most of them count the wrong thing. Check each one before you connect it.

The camera earns its place where packs differ in shape and no machine touches them: loose produce into crates, hand-filled punnets, or two formats sharing a belt.

Whatever you use, check it by hand once. Count 200 packs at the outfeed and compare.

Worked example: one shift on a punnet line

These are example numbers and describe no real packhouse. The line packs cherry tomatoes into 250 g punnets, with an ideal rate of 60 packs a minute.

The shift is 450 minutes with 50 minutes of breaks, so planned time is 400 minutes. Stops add up to 80 minutes: 40 on five order and label changes, 15 waiting for fruit, 15 on wrapper jams and film, and 10 blocked at the crate packer.

The line made 15,360 packs. The checkweigher and label check rejected 360, so 15,000 were good.

Availability is 320 of 400 minutes, or 80%. Performance is 15,360 packs against the 19,200 that 320 minutes allow, also 80%.

Quality is 15,000 of 15,360, or 97.7%. Multiply the three: 0.80 x 0.80 x 0.977 = 62.5%.

Check it the other way: 15,000 good packs at 60 a minute is 250 minutes, and 250 of 400 is 62.5%. The two agree.

The full method is on our OEE calculation page.

Now the fruit. Say 4,300 kg was tipped: 3,750 kg went into good packs, 90 kg into rejected packs, 430 kg was graded out and 30 kg was overweight.

Pack-out is 3,750 of 4,300, or 87.2%.

Under method 2, the 430 kg of grade-out would count as 1,720 lost packs. Quality would read 15,000 of 17,080, or 87.8%, against 97.7% under method 1.

Good packs are still 15,000, so OEE stays 62.5% once the 1,720 are added to the total count. Performance then reads 17,080 of 19,200, or 89.0%: the split moves, the total does not.

What to check in OEE software for a packhouse line

Run these in the demo with every vendor, including us. Bring one real shift with its order list.

1. Ideal rate per pack format. Ask to see three formats on one line, each with its own rate, and the rate switching when the order changes.
2. OEE by order. Ask for one shift split by retailer order, with the change between two orders shown as a stop.
3. Change standards. Ask if a label change and a full format change can have different standard times, and where the overrun shows.
4. Counting without a PLC. Name your hand-pack lane and ask how its packs get counted, and how that count was checked.
5. Good count and rejects. Ask where the reject count comes from, and what happens when rejected packs are opened and tipped back.
6. Raw material as a reason. Ask how a slow hour on a poor lot is recorded, and if you can sum it by grower lot.
7. Grade-out and overweight. Ask if they sit in the same report as OEE or stay in your own system, and how the two are matched.
8. Wet floor use. Ask to enter a stop reason with wet gloves on the device you would really use.

For a wider list of products, see our review of OEE software.

Where Fabrico fits

Fabrico is a cloud-based OEE and light MES platform, on the web, iOS and Android.

• Stops: they come from the machine signal, through the PLC or a Fabrico IoT device. The operator adds the reason on a tablet or phone.
• OEE and downtime: availability, performance and quality by line and by shift. Micro-stops are detected from the machine signal.
• Losses in money: not only in minutes.
• Counting: the AI camera is an add-on that counts parts. If a lane has no count, ask us to show it on your packs and crates.
• Your data: REST API, webhooks, and any list exports to Excel or CSV. A stop reason can also be set up to open a maintenance task.
• Not in Fabrico today: production planning and scheduling. Your order plan stays in your planning tool.

Pack-out by grower lot, and how orders and formats reach the line, depend on your setup. Treat them as checks for us, as for any vendor.

Book a demo or Contact us.

What people ask next

What is a good OEE for a fresh produce packing line?

We found no published benchmark we would trust for packhouses, so we give no figure. Compare the line with its own last month, by pack format.

Is grade-out a quality loss in OEE?

It depends on the method the plant picks, as no OEE standard settles it. We suggest reporting fruit removed before the pack is made as pack-out, next to OEE, with only rejected packs in OEE quality.

Some plants count grade-out inside quality instead. Either works if it is written down and used the same way on every line.

Should the ideal rate change with the fruit?

We suggest not. Keep one ideal rate per pack format.

When a poor lot slows the line, record the lost speed with a raw material reason.

How do we count packs on a hand-pack lane?

Count at the outfeed, where finished packs or crates pass one spot. A sensor works for one pack shape, and a camera can count where shapes vary.

Does a label change count as downtime?

Yes, if the line was planned to run. Book it as an availability loss with its own code and a standard time, then work on the minutes over the standard.

Want to run these checks on your own line? Book a demo or Contact us.