Efficiency

The fragmentation gap

When a line looks slow, the instinct is to blame the machine. In our study, the sites that stopped most often averaged 12.54 points lower OEE performance than the steadiest. With interruption time taken out, they were not running slower. We call it the fragmentation gap.

12.54 pts
Performance gap, frequent-stop vs steady sites
81.21%
Average performance, steady sites
68.67%
Average performance, frequent-stop sites
Published
Read4 min
TopicEfficiency
The short version
  • The steadiest sites averaged 81.21% OEE performance; the sites with the most frequent interruptions averaged 68.67%.
  • In OEE, interruption time sits inside performance, so a line that stops often looks slow.
  • With interruption time taken out, the frequent-stop sites were not running slower.
  • The gap is time lost to interruptions, so a faster machine does not close it.

Measure this on your own line. One sensor per line, live within 48 hours of the kit arriving.

A slow line, or a line that stops?

A line isn't hitting target, and the instinct is to push the machine harder or budget for a faster one. Before that, check how often it stops. In our study, that is where the difference lay.

What the sites showed

In our 2025 whitepaper we grouped sites by how often they were interrupted (interruptions are what standard OEE calls minor stops or micro-stops). The steadiest sites averaged an OEE performance of 81.21%. The sites with the most frequent interruptions averaged 68.67%.

The fragmentation gap

That 12.54-percentage-point gap is the fragmentation gap. It comes from time lost to interruptions, not from machines running slower between them.

Why it is not a speed tax

In standard OEE, interruptions shorter than the stop threshold are not taken out of availability, so their time sits inside performance. A line that stops often therefore looks slow, even if it runs at its rated speed every second it moves.

We checked this against the whitepaper's own figures. With interruption time taken out, the frequent-stop sites were not running slower than the steady ones. We first called the gap a speed tax. That name suggested the machines slow down, which the data does not show, so we no longer use it.

Augos reports interruptions apart from speed. Speed is measured only while the line moves, so each loss shows on its own and neither hides the other.

What it means for a line

Because the time is lost while the line is stopped, a faster machine does not close the gap. The lever is the interruptions themselves: the most frequent stop causes, and the time between them.

How the figure was made

The comparison uses three-month site averages for August to October 2025: four sites with the most frequent interruptions against three steady ones. Sites with performance above 100% were left out, because a rated speed set too low makes performance unreliable. It is an association across a small sample, not proof of cause.

The fragmentation gap comes from the Interruption Whitepaper. To see how often a given line is interrupted, start with its time between interruptions and its operational-health band.

Common questions

What is the fragmentation gap?

The difference in OEE performance between sites with frequent interruptions and steady sites. In our 2025 whitepaper it measured 12.54 percentage points, on three-month site averages for August to October 2025.

Is it a speed tax?

No. We used that name at first, but it suggests the machines slow down between stops. With interruption time taken out, the frequent-stop sites were not slower. The gap is time lost to the interruptions themselves.

Why can't faster machines close it?

Because the time is lost while the line is stopped, not while it runs. A faster machine still stops as often.

How was it measured?

From three-month site averages: four sites with the most frequent interruptions against three steady ones. Sites with performance above 100% were left out, because a rated speed set too low makes their performance unreliable.

Book a demo

See it on your own line.

A 30-minute call with a manufacturing specialist. We look at one of your lines together and show you what the first week of data would say about it.

  • Live within 48 hours of your kit arriving
  • One sensor, one terminal, no PLC and nothing to integrate
  • R4,950 per line per month, ex VAT. 90 days, then a month's notice
What happens nextA manufacturing specialist replies within two business hours to set a 30-minute call. We start with your objectives, so the demo answers your exact need and we can say honestly whether we can meet your expectations. Then we show you the parts of the platform that bear on those objectives, on demo data, so you can judge the fit for yourself. If it fits, a full evaluation of your line follows.
Start with your details

Step 1 of 3. Next: your line and a time that suits you. No lead is created until you confirm on the last step.

Step 1 of 3 · Your details