OEE for Aseptic Filling Lines: Run Length, Stops and Re-Sterilisation

An aseptic line does not behave like a normal filling line. It is sterilised, it runs, and at some point the sterile cycle ends and the line is cleaned and sterilised again.

Everything you make comes out between those two points. That is why a shift OEE number alone tells you little here.

This page shows how to set up OEE and downtime tracking for a UHT plant, aseptic tank, carton or PET aseptic filler, straw or cap applicator and packer. It is about time and counts only.

It covers food and drink lines such as UHT milk, juice and plant-based drinks. Pharmaceutical vial and syringe filling follows other rules and is not covered here.

When a stop means the line must be sterilised again is set by your machine supplier and your own procedures. No OEE page and no OEE software decides that for you.

An aseptic PET line has the same shape with different machines. The preforms or bottles and the caps are sterilised, the filler and capper sit inside the sterile zone, and a labeller and packer follow.

Why an aseptic line needs its own OEE rules

On most filling lines a 20 minute stop costs 20 minutes. On an aseptic line the same stop can end the run.

If the stop goes past the limit your site works to, or the filler loses its sterile state, production cannot simply restart. The line is cleaned first, then sterilised again, and only then fills.

So the real cost of that stop is the stop plus the hours that follow. A stop log that shows "sealing fault, 20 min" hides most of the loss.

Three things follow from this, and the rest of the page works through them.

• Stop length matters as much as stop count. Short stops cost their own minutes. One long stop costs the run.
• Cleaning and sterilising must be split in two. The scheduled ones are planned. The forced ones are a loss.
• Run length is a KPI of its own. It shows whether the line reaches its planned sterile cycle end.

Planned cycle end or forced re-sterilisation: how to book each

The same CIP programme can be planned on Monday and a loss on Tuesday. What matters is why it ran.

Whether planned cleaning sits inside or outside planned production time is a site choice. Our page on CIP time in OEE covers both rules.

The example below keeps planned cleaning outside planned production time, and shows its minutes beside OEE so they never vanish. For the wider split, see planned vs unplanned downtime.

Whichever rule you pick, the forced re-sterilisation must count against availability. If it lands in the planned bucket, the biggest loss on the line disappears.

Where an aseptic line loses availability, speed and product

Availability

On the filler, typical stops are sealing faults, packaging material faults, failed splices and faults in the sterile system itself, such as sterile air or the packaging material steriliser. Each one starts a clock, because a long stop puts the run at risk.

Upstream, the UHT plant fouls over a run and may need an intermediate clean. The aseptic tank is there to cover that gap, and when it runs empty the filler is starved.

Some lines have no aseptic tank, and the UHT plant feeds the filler directly. There a filler stop sends the UHT plant into circulation at once, so log the two machines' stops against each other.

Downstream, a straw or cap applicator or a packer that stops will block the filler within minutes unless there is accumulation between them. A packer jam that nobody clears fast can end a sterile run on a machine that had no fault.

Performance

Many aseptic fillers run at one fixed speed for each pack size. So performance loss is often not slow running but short stops, restarts and ramp-up after a stop.

Set the ideal rate per pack format, from the machine's rated speed for that format. One rate for the whole line makes small packs look good and large packs look bad.

Count short stops from the machine signal, not from the operator's memory. Our page on micro-stops explains why they go missing on paper.

Quality

Aseptic lines throw away packs by design at known moments. These are the usual ones.

• Start of run: the first packs after sterilisation and after each restart.
• End of run: packs lost as the line runs out and empties.
• Packaging material splices: on roll-fed carton fillers, packs that carry a reel or strip splice are typically rejected by the machine.
• Checks: packs the operator opens to check seals and weight.

Count each of these by reason. A run with ten restarts wastes far more start packs than a run with two, and only a split count shows it.

A splice that works costs a few packs, so it is a quality loss. A splice that fails stops the filler, so it is an availability loss and a risk to the run.

On a PET aseptic line there is no reel splice. The matching losses are bottles rejected at the blower, the filler and the capper, plus preform and cap supply stops.

Lab hold is not an OEE loss

Finished pallets often wait for a lab result before they ship. That wait is stock, not line loss, so leave it out of OEE.

If product is rejected later, decide once whether it is booked back to the run that made it. Then check that your software lets you correct a past shift and shows that it was changed.

A worked example: one day on an aseptic filler

These are example numbers and describe no plant. The filler's ideal rate for this pack is 100 packs a minute, so the ideal cycle time is 0.01 minutes a pack.

The day has 1,440 minutes. Planned CIP and sterilisation take 240, which leaves 1,200 minutes of planned production time.

During the day a sealing stop lasts 20 minutes and ends the sterile run. The forced clean and re-sterilisation take another 150 minutes, and other stops add up to 70 minutes.

In real life a forced end also moves the next planned clean. The example leaves that out to keep the sums simple.

Availability: stops total 20 + 150 + 70 = 240 minutes, so run time is 1,200 less 240, or 960 minutes. 960 of 1,200 is 80%.

Performance: at 100 packs a minute, 960 minutes could give 96,000 packs. The filler counted 86,400, which is 90%.

Quality: 2,400 packs were rejected: 900 at starts, 700 at splices, 500 at the end of runs and 300 for other reasons. That leaves 84,000 good packs, or 97.2%.

OEE: 0.80 x 0.90 x 0.972 = 70%.

Check it the other way: 84,000 good packs x 0.01 minutes is 840 minutes, and 840 of 1,200 is 70%. The two agree.

Now look at the sealing stop, which is 20 minutes in the stop log. With the forced re-sterilisation it is 170 minutes, more than twice all the other stops of the day together.

Had it been cleared inside the limit, those 150 minutes would have been run time. Availability would have been 1,110 of 1,200, or 92.5%, not 80%.

For the method behind the three factors, see OEE calculation.

Run length: the KPI to keep beside OEE

Run length is the clock time from the start of production after sterilisation to the end of that sterile cycle. Compare it with the run length you planned.

Here is an example week. Again, example data.

Average achieved run length is 97 hours over 6 runs, about 16.2 hours against 20 planned. Two forced ends cost 23 run hours, before you count the extra cleaning.

Three numbers are worth a line on the weekly report.

• Achieved against planned run length, in hours.
• Share of runs that reached the planned end.
• The reason behind each forced end, with the stop that started it.

Add one more view for the shift leader: stops sorted by length, not by count. The stops that came close to your limit are the near misses, and they are free warnings.

Where to count, and when a camera helps

Take total packs from the filler's own count, so packs the filler rejects itself are included, and good packs after the last reject point. Take the stop signal from the filler, and separate signals from the applicator and the packer if you can.

Sometimes the filler's own count is hard to reach, for example when the machine control is closed to you. Sometimes the saleable unit is a case or a multipack, not a pack.

In those cases a camera that counts parts on the conveyor after the packer is an honest option. It gives a count of what left the line, whatever the pack format.

A camera does not tell you why the filler stopped. The stop still comes from the machine signal, and the reason still comes from the operator.

What to check in OEE software for an aseptic line

Run these in the demo with any vendor, and with us. Bring one real day that had a forced re-sterilisation.

1. Linked losses. Can the forced re-sterilisation be tied to the stop that caused it, so the downtime chart shows 170 minutes on that cause and not 20?
2. Standard times for planned cleaning. Can CIP and sterilisation have a standard length, with the overrun shown apart as unplanned?
3. Stops by length. Can you sort and filter stops by duration, and see the ones that passed a limit you set?
4. Run length. Can it report hours per sterile cycle and how the cycle ended? If not, can you export the data and work it out yourself?
5. Ideal rate per pack format. Does the rate change with the product, or is it one number for the line?
6. Starved and blocked. Can the filler's own stops be told apart from waiting on the UHT plant or the packer?
7. Rejects by reason. Can start, end, splice and check packs be counted apart, from a signal or by operator entry?
8. Late corrections. If product is rejected after lab hold, can a past shift be corrected, and is the edit visible?

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, on the web, iOS and Android. Here is what it does on a line like this.

• 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 in money: downtime can be shown as a cost, not only as minutes.
• Counting: the AI camera is an add-on that counts parts. See camera-based OEE.
• Your data: REST API and webhooks, and any list exports to Excel or CSV, so you can build run length reports your own way.
• Follow-up: a stop reason can be set up to open a maintenance task.

Production planning and scheduling are not part of Fabrico today, so the planned run length comes from your own plan.

All eight checks above apply to us too. Bring your own example day and ask us to show how far Fabrico gets on each one.

Book a demo or Contact us.

What people ask next

Should planned CIP and sterilisation lower OEE on an aseptic line?

That depends on the rule your site picks, and both rules are in use. Keep the planned minutes visible either way, give them a standard time, and book any overrun as unplanned.

What is a good run length for an aseptic filler?

There is no honest general number, because it depends on the machine, the product and your own procedures. Compare achieved run length with your own planned run length, and track how many runs reach the planned end.

Does lab hold count against OEE?

No, product waiting for a lab result is stock, not a line loss. Only product that is finally rejected is a quality loss, booked by the rule you agreed.

Is a packaging material splice a stop or a quality loss?

A splice that works costs a few rejected packs, so it is a quality loss. A splice that fails stops the filler, so it is an availability loss, and a long one can end the run.

Should we look at OEE by shift or by run?

Both. Shift OEE shows which eight hours went badly, and the run view shows why: one stop, one forced re-sterilisation, and the hours lost with it.

Where should we measure OEE on the line?

On the filler, because it sets the pace for the line. Book waiting on the UHT plant or the packer as starved or blocked, so the filler's own losses stay clear.

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