OEE for Margarine and Spreads Lines: Remelt, Rework and Fill Weight

Margarine does not wait. Once the emulsion is chilled and worked, it keeps crystallising in the pipe, whether the filler runs or not.

So a stop at packing is never only a stop. Within minutes it becomes product on its way back to the remelt tank.

This page is for people who run tub or block lines and want an OEE number they can trust. We sell Fabrico, so test what we say like any other vendor's claims.

The line, and where the remelt loop sits

Most lines follow the same sequence. Oil and water phases are blended into an emulsion, pasteurised, then chilled in a scraped surface heat exchanger (SSHE) and worked in a pin worker.

Soft spreads go straight to a tub filler. Block and pastry margarines usually pass a resting tube first, so the product is firm enough to form and wrap.

After packing come lidding, date coding, the checkweigher and the case packer. Product that cannot be packed goes back through a remelt unit to the emulsion tanks.

Alfa Laval's published description of its margarine plant systems says that when the packing machine is down, the rework capacity is 100%. That one design fact shapes how you measure the line.

Why the crystalliser's run signal misleads

On many lines the easiest signal to wire is the SSHE or the high-pressure pump. It is also the wrong one.

When the filler stops, the SSHE keeps turning and the product is diverted to remelt. The process side reports "running" while no tub leaves the line.

Take the stop signal and the count from the packing machine. That is where saleable product is made or not made.

Then watch what happens after the stop. Product left in the hopper, the pipe or the resting tube has kept setting, so the restart often gives soft, hard or underweight packs until fresh product arrives.

A short jam at the lidder can cost several times its own length in good output. Your stop log should show both parts: the stop, and the slow or rejected restart.

The remelt loop: lost time or quality loss

The fat is recovered, so remelt feels free. It is not: the line spent time, chilling and sometimes packaging on product it has to make again.

Two cases look alike on the floor and land in different OEE boxes.

• Diverted before the filler. No pack was made. In OEE it shows as lost time: a stop if the filler stood still, slow running if it ran below rate.
• Filled, then rejected. A pack was made and counted, then emptied into remelt. That is a quality loss.

Good count means good the first time. Our page on rework and good count in OEE explains why re-run product is never counted twice.

Keep one more number beside OEE: kilos to remelt per shift. OEE counts packs, and the remelt weight shows how hard the loop is working.

That matters for the product too. How much rework a batch may take is a choice for your own plant and recipe, so set that limit and watch the remelt kilos against it.

Where margarine and spreads lines lose OEE

This is the typical picture, not a measured one. Use it as a first list of stop reasons, then let your own data correct it.

Many of these are jams of a few seconds at the denester, lidder or wrapper. They rarely reach a paper log, which is why micro-stops should come from the machine signal.

Tubs and wrapped blocks need different setups

A plant with both formats should not copy one setup to the other.

Tub lines count tubs. One spread often runs in several tub sizes and lid designs, so format changes are frequent and short, and the soft product forgives a brief stop.

Block lines count blocks at the wrapper. The resting tube holds minutes of product that firms up during a stop, so a stop past a certain length ends in a purge to remelt.

Find that length for your own line and give longer wrapper stops their own reason code. Foil or parchment reel changes are a regular event and deserve a code and a standard too.

Bakery margarine in large cartons runs at a far lower pack rate than retail blocks. It needs its own ideal rate, or its OEE will look bad on a perfect shift.

Setting the ideal rate and planned time

Set the ideal rate per format, in packs a minute. Use the lower of two limits: the rated speed of the filler or wrapper, and what the crystalliser can feed.

Example: an SSHE feeding 3,000 kg an hour gives 50 kg a minute. On 500 g tubs that is 100 tubs a minute, even if the filler could turn faster.

On 250 g tubs the filler is often the limit instead. A single rate for the whole line hides this.

CIP belongs in planned time. Our page on CIP time in OEE sets the rule we use: count the cleaning window as an availability loss with its own code, set a standard per CIP type, and track the minutes over it.

If your plant cleans in a slot with no production planned at all, report it next to OEE so it stays visible.

Recipe changeovers come in kinds, so give each kind a code and a standard. A push from one recipe to a similar one sends the interface to remelt, while a move from a recipe with milk to a dairy-free one needs a full wash.

Fat bands matter too. EU Regulation 1308/2013 defines margarine as at least 80% fat but under 90%, with lower-fat spreads in separate bands, so a move between bands is a real recipe change.

Worked example: one shift on 500 g tubs

These are example numbers and describe no plant.

Planned time is 480 minutes. A recipe changeover with CIP takes 45 minutes and filler-side stops add 35, so the filler runs 400 minutes.

The ideal rate is 100 tubs a minute, or 0.01 minutes a tub. The filler makes 36,000 tubs.

Of those, 720 are rejected for weight, lids or codes and go to remelt, which leaves 35,280 good.

• Availability: 400 / 480 = 83.3%
• Performance: 36,000 / (400 x 100) = 90%
• Quality: 35,280 / 36,000 = 98%
• OEE: 0.833 x 0.90 x 0.98 = 73.5%

The check: 35,280 good tubs x 0.01 minutes is 352.8 minutes, and 352.8 / 480 = 73.5%. Both routes agree.

The filler is 4,000 tubs short of the ideal rate, which is 40 minutes. On this kind of line, look first at restarts after stops and at time spent running slow while product went to remelt.

See OEE calculation for the full method.

Fill weight on tubs and blocks

A tub that is a few grams over is a good unit in OEE. It is also product you gave away.

Fill weight drifts with product temperature and, on whipped spreads, with how much gas is in the product. The restart after a stop is where it tends to move most.

Underweight packs are rejected at the checkweigher and count as a quality loss. Overfill passes, so track giveaway per SKU, the gap between mean fill and target, as its own measure beside OEE.

When a camera count helps

Older wrappers and fillers often have no count you can read from the PLC, or they count machine cycles and not packs that leave. A camera that counts packs on the conveyor after the lidder or wrapper gives a count without touching the machine controls.

It also helps where tubs of different sizes share one conveyor to the case packer. Ask any vendor to show the count on your own packs, with your lighting and your foil.

What to check in OEE software for a margarine line

Run these in the demo with every vendor, and with us.

1. Where does the stop signal come from? Ask for it at the filler or wrapper, and ask how a stop looks while the SSHE keeps running.
2. Can the ideal rate change per SKU and format? Check 250 g tubs, 500 g tubs and a bakery carton on one line.
3. Are short stops caught without the operator? Ask what the shortest logged stop is, and think of a lid jam of a few seconds.
4. Can CIP and each changeover kind have its own code and standard? Look for minutes over standard, not only total minutes.
5. Starved or blocked? Ask how a filler stop caused by the case packer or by an empty emulsion tank is shown.
6. Where do rejects come from? Ask if checkweigher rejects can be read from a signal or must be typed in.
7. Can remelt kilos and giveaway sit beside OEE? If not, ask how the data gets out so you can do it yourself.
8. Can the operator pick a reason with greasy gloves? Test the screen at the line, not in the meeting room.

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 stop 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: they can be shown in money, not only in minutes.
• Your data: REST API and webhooks, and any list exports to Excel or CSV.
• AI camera: an add-on that counts parts. You can open the video of a stop. In the standard setup, video is streamed, not stored.
• Maintenance: a stop reason can be set up to open a maintenance task.
• Not in Fabrico today: production planning and scheduling. Predictive maintenance and suggested stop reasons are on the roadmap.

Remelt kilos, giveaway and checkweigher reject signals are things to ask us about in the demo, the same as any other vendor.

Book a demo or Contact us.

What people ask next

Does remelted margarine count as good output in OEE?

No: good count means good the first time, so a pack that is filled, rejected and remelted is a quality loss. Product diverted before the filler was never a pack, so it shows as lost time.

Where do CIP and recipe changeovers go on a margarine line?

Inside planned time, as availability losses, each with its own code and standard. If cleaning sits in a slot with no production planned, report it next to OEE.

Where should we count on a tub line?

Count total tubs at the filler and good tubs after the lidder, coder and checkweigher. The gap between the two is your reject count.

How do we track fill weight on tubs?

An overfilled tub passes as good, so the OEE quality rate does not show it. Track giveaway, the gap between mean fill and target, as its own measure beside OEE.

What is a good OEE for a margarine line?

We know of no public benchmark for this line type that we would trust. Compare each line with itself, format by format and week by week.

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