Labeller Drift Stops: Why Label Position Wanders and the Line Stops
Key takeaways
- Drift shows up as rising rejects first and as stops later.
- The usual triggers are a new container lot, a new label roll and speed changes after stops.
- One drift problem can hit Quality, Availability and Performance at once.
- Give "drift adjustment" its own code, apart from web breaks and roll changes.
- A position check after every roll change catches most drift early.
Label drift is a slow change in where the label lands on the container: a little higher, a little lower, skewed or overlapping, until inspection starts rejecting and the labeller stops. It rarely comes from one sudden fault. It usually builds from container changes, a new label roll, glue temperature, dirty sensors or worn parts.
Fabrico, an OEE and light MES platform, logs each labeller stop and shows OEE live by line and shift, rejects included, so a drift that starts in Quality and ends in a stop can be seen as one problem. This guide covers what drift is, what causes it, how it turns into stops, and how to code it.
What is label drift?
A labeller places each label at a set height and position on the container. Drift is when that position slowly moves during a run. Common forms:
- Vertical drift: labels creep higher or lower.
- Skew: labels land at an angle.
- Overlap or flagging: the label ends do not meet cleanly, or a corner lifts.
- Wrinkles: the label does not lie flat.
The difference from a normal labeller fault is time. A web break stops the machine at once. Drift gives a warning, often for an hour or more, in the form of a slowly rising reject count. Our guide to labeller OEE covers the machine's other losses, such as web breaks and splices.
What makes label position drift during a run?
- Container variation. A new bottle lot, or a blow moulder running warmer or cooler, changes the container's shape slightly. See stretch blow moulding OEE for the upstream side.
- A new label roll. Different roll tension, a new label lot or a splice changes how the label feeds.
- Glue or adhesive. On glue labellers, glue temperature and amount affect where and how the label holds.
- Dirty or misaligned sensors. The sensor that detects the label gap or the container can drift with dust, glue or a knock.
- Worn parts. Pads, brushes and grippers wear and place the label less precisely.
- Speed changes. On most lines the labeller speeds up and slows down around short stops, and placement can shift with speed.
- Condensation. Cold filled containers can sweat, and wet surfaces affect how labels sit.
How does drift turn into rejects and stops?
| Stage | What you see | OEE factor | Code it as |
|---|---|---|---|
| 1. Small drift | Labels still in spec, slowly moving | None yet | Nothing, but visible in checks |
| 2. Rejects rise | Inspection rejects containers out of spec | Quality | Label position reject |
| 3. Reject stops | Several rejects in a row stop the labeller | Availability | Reject stop, label position |
| 4. Adjustment stop | Crew stops to reset height, guides or sensor | Availability | Drift adjustment |
| 5. Slow running | Crew lowers speed to keep labels in spec | Performance | Reduced speed, label quality |
Because one cause spreads over three factors, drift is easy to underestimate. Each part looks small in its own report. Code all of it with the same cause so it adds up.
How should drift be coded apart from other labeller stops?
- Web break: the label web tore. A sudden stop.
- Roll change or splice: planned material change, with its own standard time.
- Glue skip: glue missing or uneven on one label.
- Reject stop, label position: the inspection rule stopped the machine.
- Drift adjustment: the crew stopped to reset placement.
- Starved or blocked: charged to the machine that stopped first.
Keep the trigger in the note when you know it, such as "after roll change" or "new bottle lot". Over a few weeks, the notes show which trigger causes most drift.
What does a drifting run look like?
Sample line (illustrative numbers, not a customer): a labeller on a juice line running 300 bottles a minute, with a 3 minute buffer between the filler and the labeller. A new label roll goes on at the start of a six hour run. Rejected bottles are scrapped at a variable cost of $0.25 each, or €0.25 in a euro plant.
| Hour of the run | Label position rejects | Labeller stops | What happened |
|---|---|---|---|
| 1 | 40 | None | Normal |
| 2 | 60 | None | Slight rise |
| 3 | 180 | None | Clear rise, nobody checks |
| 4 | 420 | 3 reject stops of 2 minutes, then a 9 minute adjustment | Label height reset |
| 5 | 50 | None | Back to normal |
| 6 | 45 | None | Normal |
| Total | 795 | 15 minutes |
- Rejects: 795 x $0.25 = about $199 in scrapped bottles. In a normal run of six hours like hours 1, 5 and 6, rejects would have been around 270.
- Downtime: 3 x 2 + 9 = 15 labeller minutes. The reject stops ended inside the 3 minute buffer, so they did not stop the filler.
- The 9 minute adjustment blocked the filler for 9 minus 3 = 6 minutes: 6 x 300 = 1,800 bottles not filled.
- The warning was there two hours before the first stop. A position check after the roll change, or an hourly look at the reject trend, would have caught it.
Which fixes keep labels in place?
- Centerlines per SKU for label height, guide positions and sensor settings, checked at each changeover. Our guide to first off approval after changeover shows a simple check routine.
- A position check after every roll change, not only at the start of the run.
- A container lot check when a new lot of bottles starts.
- Sensor cleaning on a schedule, written into the shift routine.
- Wear parts replaced on a schedule, before placement suffers.
- Fewer speed swings. Fewer short stops around the labeller means fewer speed changes. See reduced speed vs minor stops.
Where Fabrico fits
Fabrico is a computer vision enabled Manufacturing Performance Platform. OEE and light MES come first, with computer vision on stops.
- Automatic downtime logging for every labeller stop, short or long.
- OEE live by line and shift, with rejects next to stops, so drift shows up before it stops the line.
- The operator confirms the reason in one tap, so drift adjustments get their own code.
- A camera clip as evidence for a stop, where a camera watches the labeller outfeed.
- Every loss priced in money and ranked, rejects and downtime together.
You start with a pilot on one line, with a baseline and a decision date agreed in advance. There is no free plan and there are no tiers: each plant gets a quote. See how stops are captured on camera on our computer vision OEE page.
Frequently asked questions
Why do labels drift during a run?
Usually because something changed: a new container lot, a new label roll, glue temperature, a dirty sensor, worn parts or speed changes after stops.
How do I know drift is starting?
Label position rejects rise slowly before any stop. Watch the reject count hour by hour, especially after a roll change.
Does label drift count against OEE?
Yes, in up to three ways: rejects reduce Quality, reject and adjustment stops reduce Availability, and slower running reduces Performance.
Should drift have its own downtime code?
Yes. Code drift adjustments and position reject stops apart from web breaks and roll changes, so the full cost of drift adds up in one place.
Does a labeller stop always stop the filler?
No. It only stops the filler when it lasts longer than the buffer between them. Short reject stops often end inside it.
See your labeller's drift in a demo
Bring one line and a week of labeller stops and rejects. We will show how they connect and what they cost.
Latest from our blog

Real Fabrico tiles, sample data.





