Menu
Book a demo
OEE for Wine Bottling Lines: Changeovers, Dry Goods and Downtime

OEE for Wine Bottling Lines: Changeovers, Dry Goods and Downtime

How to measure OEE on a winery bottling line: planned time on part-week running, changeovers by type, dry goods waits, rework, and 8 software checks.
OEE for Wine Bottling Lines: Changeovers, Dry Goods and Downtime

In short: OEE for wine bottling lines is mostly a changeover and dry goods story. Runs are short, the line bottles only some days, and the filler waits for glass, corks and labels more often than it breaks.

  • Count OEE only on the hours you planned to bottle. Use TEEP for the days the line stands.
  • Time each changeover from the last good bottle to the first good bottle, and name its type: wine, bottle shape, closure or label set.
  • Log waits for glass, closures, capsules and labels as starved stops with the material named, not as machine faults.
  • Bottles pulled for a skewed label or a creased capsule are quality loss, even when you rework them by hand.

A winery bottling line is not a soft drink line. It may run two days a week, fill four wines in a day, and change bottle, closure and label between them.

So the usual OEE advice, built for one product running all week, misses where your time goes.

This page walks the line machine by machine, sets the planned time and the ideal rate, and works one example day. We sell Fabrico, so test what we say like any other vendor's claims.

The line, machine by machine, and where each one loses

Most winery lines have the same seven steps. Small lines combine the rinser, filler and corker in one monobloc, but the losses stay the same.

Wine bottling line: where time goes Typical losses, not measured data Depalletiser Waits for glass Rinser Fallen bottles Filler Sets the line rate Corker or capper Closure feed jams Capsule applicator Creased capsules Labeller Skew, reel changes Case packer Blocks line if down Blue: most changeover work Count total at filler, good at packer
A typical winery bottling line in sequence. The losses shown are typical, not measured data.

The filler sets the pace, so read line OEE there. Everything upstream can starve it and everything downstream can block it.

That is why the first machine to stop matters more than the machine that is standing. A filler that waits because the labeller ran out of reel has no fault of its own.

Planned time: the line bottles only some days

Many winery lines bottle two or three days a week, and not at all during harvest. Those idle days are not an OEE loss.

OEE asks one thing: of the time you planned to bottle, how much became good bottles at the ideal rate? Planned time is the bottling shift, minus breaks if the line stops for them.

TEEP asks a different thing: of all calendar time, how much did the line use? On a line that runs two shifts a week, TEEP will be a single-digit number, and that is normal.

Keep the two apart. OEE tells you how well a bottling day went.

TEEP tells you if you need a faster line or only more bottling days.

Three planned time rules save arguments later:

  • Start-up and end-of-day cleaning: sterilising the filler and the final rinse are needed work. Decide once if they sit inside planned time, write it down, and never change it quietly.
  • Wine not ready: a tank still waiting for filtration or lab release stops the line during planned time. Log it as a starved stop, cause "waiting for wine".
  • Crew pulled to the cellar: if the shift was planned and the people left, the line lost that time. It stays in OEE, with its own reason.

Changeovers: the biggest loss on short runs

With many SKUs and small lots, a line can spend more time changing than any single fault costs it. Measure the changeover as its own state, not as a stop with a vague reason.

Time it from the last good bottle of the old wine to the first good bottle of the new one. Stopping the clock when the mechanic leaves hides the test bottles and the restart.

Not all changeovers are the same job, so name the type. A label swap and a move from cork to screw cap should never share one average.

Change typeWhat gets touchedWhere the loss shows up
Wine onlyFiller drain and rinse, sometimes a filter change, hoses to a new tankAvailability, plus the first bottles checked for fill height
Label setFront, back and neck reels, sensor teach, label height and positionAvailability, plus test bottles scrapped or relabelled
Bottle shape or sizeStar wheels, guides, filler and corker height, labeller stations, case packer formatAvailability, then slow running until the line settles
Closure typeCork to screw cap means another head or another machine, and often no capsuleAvailability, usually the longest change on the line

Keep the changeover inside planned time. It is an availability loss, and on this kind of line it is the first one to work on.

Some plants give each change type a standard time and count only the overrun as loss. That is a fair method too, if the standard is written down and the full minutes stay visible.

For how to shorten them, see our guide to changeover time. The order of the day's wines matters as much as the wrench work.

Dry goods: when the filler waits for glass, corks or labels

Winery lines often stop for things that are not machines. The glass pallet is the wrong mould, the labels came late from the printer, or the cork lot is still in the store.

If these stops go into the log as "filler stopped", the filler gets blamed for a purchasing or warehouse problem. Give them their own reason group: waiting for materials.

Then name the material. Glass, closures, capsules, labels and cartons each have a different owner and a different fix.

  • Glass: wrong bottle on the pallet, broken layers, wet or collapsed layer pads at the depalletiser.
  • Corks and caps: hopper run empty, feed chute bridged, a new lot that sits tight in the jaws.
  • Capsules: stuck together in the stack, wrong diameter for the neck.
  • Labels: reel wound the wrong way, wrong vintage printed, a splice that breaks the web.
  • Cartons: case packer out of blanks, which blocks everything upstream.

A Pareto of these waits, by material, is often the cheapest improvement on the line. It needs a phone call, not a spare part.

Speed: set the ideal rate per bottle and per SKU

One ideal rate for the whole line is wrong on a winery line. A 750 ml Bordeaux bottle, a 375 ml half bottle and a magnum do not run at the same speed.

Set the ideal rate per format, from the best steady speed the line has shown on that format. Do not copy the filler nameplate if the labeller with three labels cannot keep up.

Performance loss then has two parts. One is running below that rate on purpose, often in the first minutes after a changeover.

The other is short stops: a fallen bottle at the rinser, a capsule that did not drop, a label sensor that misses.

These short stops rarely get written down. If your system takes stops from the machine signal, you can see micro-stops as a count and a total, and find which machine makes them.

Quality: label skew, capsule faults and hand rework

Wine is rarely thrown away at the line. The bottle is pulled, the label is stripped, and someone relabels it by hand at a table.

That habit hides quality loss. The bottle ends up in a case, so the day's count looks clean, but it did not come out right the first time.

Count a bottle as good only if it went through the line once and passed. Pulled bottles are quality loss, and the hand rework is extra labour on top.

The usual causes on this line:

  • Label skew, wrinkles and bubbles: often worse on cold wine, when the glass sweats before the labeller.
  • Capsule creases and tears: from spinner or heat settings, or a capsule lot that does not suit the neck.
  • Fill height and cork depth: most common in the first bottles after a changeover.
  • Start-up bottles: test bottles used to set label position. Count them, since they are real glass, wine and labels.

A worked example: one bottling day

These are example numbers and describe no real winery. One shift of 480 planned minutes, three wines in the same 750 ml bottle, so two changeovers.

The ideal rate is 50 bottles a minute, or 3,000 an hour.

One bottling day, 480 planned min Example data, 50 bottles a minute ideal Planned time: 480 min Changeovers, 2 of them: 70 min Waiting for dry goods: 20 min Faults and jams: 30 min Slow running, short stops: 54 min Rejects and rework: 6 min Good bottles: 300 min OEE: 300 of 480 min = 62.5%
Example data. Each bar is cut from the right end of the planned time, on one scale. What is left in green is good bottles at the ideal rate.
Step (example data)NumbersResult
Planned timeOne bottling shift480 min
Stop time70 changeover + 20 dry goods + 30 faults120 min
Availability360 run min of 48075.0%
Performance15,300 filled of 18,000 possible (360 x 50)85.0%
Quality15,000 good of 15,300 filled98.0%
OEE0.75 x 0.85 x 0.980462.5%

Check it the short way. 15,000 good bottles at 50 a minute is 300 minutes of perfect running, and 300 of 480 is 62.5%, so the two routes agree.

Of the 300 pulled bottles, say 200 were relabelled by hand and sold. They still count as quality loss here, because they did not pass the first time.

Now look at what the day says. Changeovers took 70 minutes, more than faults and dry goods together.

A plan with two wines instead of three would have given back 35 minutes before anyone touched a machine.

Slow running and short stops come second at 54 minutes. Nobody wrote those down, which is why they need a machine signal.

And TEEP? If the line runs two days like this a week, that is 960 planned minutes of 10,080 in the week.

62.5% of that share is about 6%. It is not a bad number, it only says the line has room.

The full method is in our OEE calculation guide.

Mobile and contract bottling

If a mobile bottling truck comes to you, the line is not yours, but the day is. You still pay for the hours it waits for your wine, your glass or your labels.

The same split works: changeovers by type, waits by material, pulled bottles counted. Agree with the contractor before the day who records the stops and where the bottle count comes from.

You may not be allowed to wire into the contractor's controls. Ask every vendor, including us, how they would get a count and a run signal on a line you do not own.

What to check in OEE software for this line

Run these in every demo, with any vendor. Bring one real bottling day with at least two changeovers as the test case.

CheckHow to test it
Is a changeover its own state, with a type?Ask to see a label change and a closure change as separate lines, each timed from last good bottle to first good bottle.
Is the ideal rate set per SKU or format?Run a 750 ml and a magnum in the same day. Performance should use two rates, not one.
Do idle days stay out of OEE?Enter a week with two bottling days. OEE should cover those two days, with TEEP or utilisation shown apart.
Can a stop be a wait for a named material?Log one wait for glass and one for labels. The downtime Pareto should show them apart from machine faults.
Does it show which machine stopped first?Take a labeller stop that backed up to the filler. Ask where those minutes land.
Where do the counts come from?Ask where total and good bottles are counted, in bottles or cases, and what happens on a line with no usable counter.
Where do reworked bottles go?Enter 200 bottles relabelled by hand. They should lower quality, not vanish into the good count.
How do short runs add up?Three runs in one day should roll up by minutes, not as an average of three percentages.

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

Where a camera that counts bottles helps

Many winery lines are older monoblocs, or a mix of machines from different makers and decades. There may be no PLC count you can reach, or none you trust.

A camera that counts parts can give you a bottle count at the filler exit or a case count at the end of the line, without touching the controls.

A fixed photo-eye often needs moving at each bottle change. Ask every camera vendor, including us, to show the count holding on your darkest glass and on two bottle shapes.

See how camera-based OEE works before you decide.

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: so a wait for labels has a price, not only minutes.
  • Camera: the AI camera is an add-on that counts parts. You can open the video of a stop. In the standard setup the video is streamed, not stored.
  • Your data: REST API and webhooks, and any list exports to Excel or CSV.
  • Maintenance: a stop reason can be set up to open a maintenance task.
  • Not in Fabrico today: production planning and scheduling. The order of the day's wines stays in your planning tool.

The eight checks above apply to us too. Bring a real bottling day and ask us where each of its lost minutes would land.

Book a demo or Contact us.

What people ask next

What is a good OEE for a wine bottling line?

There is no honest single number. A line with four changeovers a day will sit below a line that runs one wine all shift, with the same crew and machines.

Compare the line with itself: same format, same run length, month on month.

Should changeovers count against OEE?

Yes, if they happen in planned time. They are an availability loss, and on short runs usually the largest one.

Hiding them makes the score look better and the bottling day no shorter.

OEE or TEEP for a line that runs two days a week?

Both, for different questions. OEE judges the bottling days.

TEEP shows how much of the calendar the line uses, which helps when you decide between more bottling days and a bigger line.

Can we measure OEE on a mobile bottling line?

Yes, for the hours it is at your site. You need a run signal, a bottle count and stop reasons.

Agree with the contractor who records them before the day starts.

How do we count bottles if the line has no usable counter?

Add a sensor at the filler exit, or use a camera that counts parts. Then check the count against cases packed for a few runs before you trust it.

Do we need a full MES for one bottling line?

Usually not for OEE and downtime alone. A light MES covers stops, counts, reasons and OEE by line and shift.

Check what you need beyond that before you buy more.

Want to see your own bottling day in these terms? Book a demo or Contact us.

Track OEE and downtime in real time
See where lines lose time
Book a live demo.
See where lines lose time
See where lines lose time

Real Fabrico tiles, sample data.

By clicking the Accept button, you are giving your consent to the use of cookies when accessing this website and utilizing our services. To learn more about how cookies are used and managed, please refer to our Privacy Policy and Cookies Declaration