Blog

OEE for Breakfast Cereal Lines: Measure Process and Packing Apart

10 Nov 2026 · Updated 14 Nov 2026

12 min.

Measure OEE on a cereal plant in two places: kg per hour on the process, boxes per minute on packing. Worked example and software checks.

In short: OEE for breakfast cereal lines works when you measure two things: the process in kg per hour and the packing line in good boxes per minute. One figure across both hides where the loss sits.

Process (cooking or extrusion, flaking or puffing, toasting, coating): it rarely stops, so its loss shows as slow running and as fines.
Packing (weigher, bagger, cartoner, case packer): its loss shows as jams, short stops and changeovers.
The bin between them links the two. A packing stop fills it, then the process has to slow down.
Giveaway at the weigher is a real loss that OEE quality cannot see, because an overfilled box passes. Track it as its own measure next to OEE.
In the example below, the same shift reads 91.9% on the process and 73.5% on packing.

A cereal plant is two factories joined by a bin. One half cooks, shapes and toasts without a break, and the other half bags, boxes and cases, and stops many times a shift.

Most OEE advice is written for the second half only. This page covers both, and what to ask of software that has to measure them.

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

The cereal line in order, and the two places to measure

The process side runs as a flow: cook or extrude, flake or puff, toast or dry, then coat, dry and cool. Its natural unit is kg per hour.

The packing side makes single items: a multihead weigher drops a portion, a vertical bagger (VFFS) makes the bag, a cartoner puts it in the box, a case packer fills the case. Its natural unit is boxes per minute.

A typical layout, not a drawing of any one plant. Blue is the continuous process, amber is the buffer, green is the packing line. The two white pills are where to count.

Measure 1 sits after cooling, before the bin: a belt scale or a loss-in-weight signal that gives kg per hour of finished cereal.

Measure 2 sits after the cartoner: the checkweigher, which counts boxes that passed and boxes it kicked out.

A case packer count is a useful cross-check, but not the main count. It reacts slowly and hides single rejects.

Why one OEE figure for both halves misleads

The two halves have different ideal rates, different units and different ways of losing. Put them in one number and each hides the other.

The common shortcut is to turn good boxes into kg and divide by what the process could have made.

But packing is normally built faster than the process, so it can stop, catch up and still empty the bin. Measured against the process rate, its stops vanish.

So keep two OEE figures, one per half, each against its own ideal rate. The worked example below shows how far apart they can sit.

The bin: how a packing stop becomes a process slowdown

The process cannot stop cheaply. Product inside a toaster scorches if the belt stops, and an extruder restart costs time and start-up waste.

So when the cartoner jams, the bin takes the product. If packing stays down, the bin reaches its high level and the process turns down, or product is diverted to totes or to waste.

It also runs the other way. When the process trips, the bin empties and the weigher has nothing to drop, so packing is starved, not broken.

Two rules follow for the stop list:

A process slowdown caused by a full bin is booked as blocked by packing, with the bin level as proof. It is a speed loss on the process, and its cause sits on the packing line.
A weigher waiting for product is booked as starved by process. It must not land in the packing line's own breakdown list.

Where a cereal line loses time, speed and product

This is a typical picture, to use as a starting list for stop reasons.

Machine	Typical loss	OEE part
Cooker or extruder	Start-up and shutdown product, die or screw wear, turn-down when the bin is full	Performance, quality
Flaking rolls or puffing	Roll gap drift, broken flakes, thickness out of range	Quality
Toaster or dryer	Moisture or colour out of range, slow running to hold moisture	Quality, performance
Coating drum, dryer and cooler	Clumps, sticky build-up, clean-outs between flavours	Quality, availability
Conveyors, sifter, bin	Breakage at drops and transfers, fines sifted out	Quality
Multihead weigher	Sticky or bridging product, slow combinations on light, bulky cereal, giveaway	Performance (giveaway is tracked apart)
Bagger (VFFS)	Fines in the seal, leaking bags, film splices and tracking	Performance, quality
Cartoner	Carton jams, bag not fully in the box, glue faults, open flaps	Performance, availability
Case packer	Case misfeeds, backs up the cartoner when it stops	Availability

Two points are special to cereal.

Breakage and fines are quality loss on the process. Every drop, elbow and bin fill breaks some flakes, and what the sifter takes out is product you cooked and toasted but cannot pack.

Fines also cause packing stops. Dust in the seal area gives leaking bags, so a breakage problem upstream shows up as bagger rejects and short stops downstream.

Bag seal faults and carton jams are usually cleared in under a minute, so nobody writes them down. They are classic micro-stops, and on a cartoner they often add up to more than the breakdowns.

The packing half has much in common with OEE for snack food lines, which also run a multihead weigher over a vertical bagger.

Setting the ideal rate and the planned time

Ideal rate: set it per product on the process and per box size on packing. A dense granola and a light puffed rice run at very different kg per hour through the same toaster.

Use the best rate the line has held on that product with good quality, not the nameplate and not last year's average. A performance figure over 100% tells you the rate is set too low.

Planned time: take out only what was never meant for production, such as no orders, breaks with the line down and planned deep cleans.

Keep flavour and allergen changeovers inside planned time, as an availability loss. Starved and blocked time stays inside too, even though its cause sits in the other half.

Worked example: one shift, two OEE figures

These are example numbers, round on purpose. They describe no real plant.

Example data. Each bar is drawn to its own scale: minutes for packing, kg for the process. Green is good output at the ideal rate.

Packing line, in boxes

Planned time is 480 minutes. Stops take 80 minutes: 40 for one artwork changeover (same cereal, new film and cartons) and 40 for other stops, so run time is 400 minutes.

The ideal rate for this 400 g box is 60 boxes a minute. The line filled 21,600 boxes, and the checkweigher passed 21,168 of them.

Availability: 400 of 480 minutes = 83.3%
Performance: 21,600 of 24,000 possible boxes = 90%
Quality: 21,168 of 21,600 = 98%
OEE: 83.3% x 90% x 98% = 73.5%

Check it the short way: 21,168 good boxes at 60 a minute is 352.8 minutes of perfect running. 352.8 of 480 is 73.5%, so the two agree.

Process, in kg

The ideal rate for this product is 1,200 kg per hour, so 8 hours could give 9,600 kg. The process never stopped.

But the bin went full, and for 2 hours the process ran at 900 kg per hour, so it made 9,000 kg. The sifter and the quality checks took out 180 kg, leaving 8,820 kg good.

Availability: 8 of 8 hours = 100%
Performance: 9,000 of 9,600 kg = 93.75%
Quality: 8,820 of 9,000 kg = 98%
OEE: 100% x 93.75% x 98% = 91.9%

Check: 8,820 good kg of 9,600 possible is 91.9%. Same answer.

What the two figures tell you

The process lost 600 kg to slow running, and the whole of it came from the full bin. The fix is on the packing line, not on the toaster.

Now the single blended figure: 21,168 good boxes at 400 g is 8,467 kg. Divide by the 9,600 kg the process could make and you get 88.2%.

That is not wrong arithmetic, but it hides a packing line at 73.5%. For the method itself, see our page on OEE calculation.

Flavour and allergen changeovers with dry cleaning

Water in the wrong place is a hygiene risk in a cereal plant, so most changeovers are dry cleans: vacuum, brush, scrape, wipe, then a check before restart.

That makes them slow, manual and very different in length. Plain to honey coated is short, but nuts or chocolate to a product without them is a full clean of the coating drum, the weigher buckets and the bagger forming tube.

For OEE, three things matter:

Time it from last good box to first good box at full speed. The slow restart and the first rejected boxes belong to the changeover too.
Split the reason by type. Use at least: artwork or size change, flavour change, allergen change.
Count the flush product. Product run through to push out the old flavour is quality loss on the process, in kg.

Giveaway at the weigher: the loss OEE quality cannot see

A box filled 4 g over its 400 g target is a good box: it passes the checkweigher and it ships. OEE quality counts it as good, so the metric sees no loss.

But the extra cereal is real money. In the example, 4 g over on 21,600 boxes is about 86 kg given away in one shift.

Do not change the quality figure to make it visible, since the box is sellable. Track giveaway as its own measure next to availability, performance and quality, in grams per box and in kg per shift.

Light, bulky and broken product makes it worse, because the weigher finds fewer exact combinations. Our page on giveaway and overfill covers how to see it and cut it.

What to check in OEE software for a cereal line

Run these checks in a demo with every vendor, including us. Bring one real shift with a packing stop that filled the bin.

Two units on one site. Can the process be measured in kg per hour and packing in boxes per minute, each with its own OEE and its own ideal rate per product?
Slow running without a stop. Ask how a 2 hour run at 75% rate shows up, and how a reason gets attached to it.
Starved and blocked. Ask how a weigher waiting for product is kept apart from a weigher fault, and a process slowed by a full bin from a process fault.
Short stops. Ask where a 20 second carton jam goes: a stop, a speed loss, or nothing. Where is the threshold set?
Rejects and kg loss. Can the checkweigher reject count come in by signal? Can fines and flush product be entered in kg against the process?
Changeover types. Can a changeover carry a type, so you can compare an allergen change against a size change over a month?
Counting without a PLC count. On an older cartoner with no usable count signal, what are the options: a sensor, a camera, a manual entry?

For a wider list of products to run these checks on, see our review of OEE software.

Where Fabrico fits

Fabrico is a cloud-based OEE and light MES platform. Here is what it does, and what it does not.

Stops: they come from the machine signal, through the PLC or a Fabrico IoT device. The operator adds the stop reason on a tablet or phone.
OEE and downtime: availability, performance and quality by line and by shift, on the web, iOS and Android.
Micro-stops: detected from the machine signal, so the cartoner's short jams are not left to memory.
Losses in money: losses can be shown in money, not only in minutes.
Your data: REST API and webhooks, and any list exports to Excel or CSV.
Camera: the AI camera is an add-on that counts parts, worth a look where a packing machine has no usable count signal. You can open the video of a stop, and in the standard setup video is streamed, not stored.
Not in Fabrico today: production planning and scheduling. The order of flavours for the week stays with your planning tool.

A stop reason can also be set up to open a maintenance task, for example on a repeat glue fault.

The seven checks above apply to us too, so ask us to show the process in kg and the packing line in boxes as two lines. Read more about camera-based OEE if counting is your weak spot.

Book a demo or Contact us.

What people ask next

Should a cereal plant report one OEE or two?

Two: one for the process in kg per hour and one for each packing line in boxes per minute. If the board wants one number, report the packing line that feeds the customer and show the process figure next to it.

Is breakage a quality loss or a performance loss?

Quality: fines and broken product that the sifter removes are kg you made but cannot pack. If they are reworked into another product, still count them as loss on this one and log the rework apart.