Blog / Downtime
Downtime · Guide

Minor Stops vs Breakdowns: How to Tell Them Apart

OL By OEE Lab Editorial |Updated 2 November 2026 |6 min read
Short answer

A breakdown is a failure: something broke or stopped working, and the line needs a repair before it can run again. A minor stop is a brief halt with nothing broken, cleared or reset in moments. Sort stops by what the restart needed, with duration only as a clue, because the two have different owners and fixes.

The one question that decides it

Ask this after every stop: what did it take to get running again? That single answer sorts most events.

If a part had to be repaired or replaced, it was a breakdown. If clearing or resetting was enough, it was a minor stop.

The names come from Nakajima's six big losses. Equipment failure is one loss, and idling and minor stoppages is another.

Side by side

BreakdownMinor stop
Restart needsRepair or new partClear, reset, reposition
Usual lengthMinutes to hoursSeconds to a few minutes
How oftenNow and thenMany times a shift
Who restarts itUsually a technicianUsually the operator
Shows up in the logAlmost alwaysOften not at all
Measure withMTBF and MTTRStop count, mean time between stops
Fix byFinding why it failedFinding the repeat pattern
The typical picture. A few real events cross the lines, so use the restart question for those.

Why it matters: same minutes, opposite fixes

Example: two lines each lose 40 minutes in a 480-minute shift. Line A has one 40-minute bearing failure, and line B has 80 stops of 30 seconds each.

Line A: one breakdown, 40 min Line B: 80 stops of 30 s, 40 min
Two 480-minute shifts, drawn to scale for time. Line B's ticks are drawn wider than 30 seconds so you can see them.

Line A needs a failure analysis and perhaps a change to how that bearing is greased. Fix that one thing and the 40 minutes may not return for a year.

Line B needs someone to tally 80 events and find the two or three repeats behind most of them. Sending a technician to "repair" it finds nothing, because nothing is broken.

TPM language calls these sporadic and chronic losses. A sporadic loss is a sudden, visible jump, and a chronic loss is a small one that has become normal.

Where each one lands in OEE

A logged breakdown cuts availability. Whether a minor stop does the same depends on your stop threshold, the length below which a stop is not logged.

Below the threshold, the stop stays inside runtime and pulls down performance. Above it, the stop counts against availability like any other logged stop.

Now run the two lines through the math. Both have an ideal rate of 100 units a minute and make 44,000 units, and the threshold is 2 minutes.

Line ALine B
Logged stops40 min0 min
Runtime440 min480 min
Availability440 / 480 = 91.7%480 / 480 = 100%
Performance(44,000 / 100) / 440 = 100%(44,000 / 100) / 480 = 91.7%
OEE (no scrap)91.7%91.7%
Same OEE, same lost minutes. The factor tells you which of the two problems you have.

Read that last row with care. OEE alone cannot tell these lines apart, so you need the factors and the stop count beside it.

The gray cases

Duration is a poor rule on its own, because real stops do not respect it. Here is how the restart question handles the awkward ones.

  • A jam that takes 12 minutes to clear. Nothing broke, so it is not a breakdown. It is long enough to log, so give it a jam code.
  • A blown fuse, swapped in 6 minutes. A part failed, so call it a breakdown, even though it was quick.
  • A sensor that false-trips 30 times. Each trip is a minor stop.
  • The same sensor, now dead and replaced. That event is a breakdown.
  • A nozzle the operator swaps in 4 minutes. Practice varies here. Pick a rule, write it in your reason codes and keep to it.

Consistency matters more than the exact border. Two crews using two rules make every trend useless.

How to separate them on your line

You can do this with a clipboard for one hour, which is often the moment a team first sees the real count.

1
Stand at the line for one hour
Pick a normal product, on a normal day.
2
Write down every stop
Clock time, seconds lost, and what restarted it.
3
Sort by restart action
Repair goes in one pile, clear or reset in the other.
4
Scale it to a shift
Compare the result with what the log shows for the same line.
A one-hour tally that needs no software.

Example: the hour shows 11 stops. None needed a repair, and they total 6 minutes, while the log for that hour shows nothing.

Scaled to 8 hours, that is about 88 stops and 48 minutes that never reached the report. One hour is a small sample, so repeat it on another shift before you quote the number.

For a lasting view you need a stop signal from the machine, because operator logs cannot catch stops this short and frequent.

Do the short ones really add up?

Often they do, though no neutral source we opened gives a general share. Studies of Swedish plants found speed and short-stop losses to be the largest category (Ljungberg, 1998; Ylipää et al., 2017).

Your own line may differ. A plant with old equipment and long runs may well lose more to failures, so measure before you pick a side.

Common mistakes

  • Coding every stop as "breakdown". Technicians get called for jobs that need no repair, and the failure data turns to noise.
  • Using a time limit as the only rule. A 3-minute fuse swap and a 3-minute jam are different problems.
  • Tracking MTBF on short stops. It swamps the number. Keep MTTR and MTBF for real failures and use mean time between stops for the rest.
  • Ignoring the link between them. A rising count of short stops on one unit can come before its failure, so look at the two lists together.

What to do next

Split last month's stops by the restart question, then rank each pile by minutes. Put them on a downtime Pareto chart, one chart per pile.

For the minor stop pile, go to how to reduce micro-stops. For failures, go to root cause analysis.

See it live in Fabrico

Short stops are the ones nobody can describe afterwards. Where cameras are fitted, Fabrico puts video of each stop, streamed on request next to its machine data, so a jam and a failure are easier to tell apart. See how Fabrico uses cameras for OEE.

What people ask next

How long is a minor stop?

We found no single standard limit, and plants pick their own, often somewhere under 5 minutes. What matters is that one written rule applies to every crew. Stop threshold in OEE shows how to choose it.

Is a micro-stop the same as a minor stop?

Most people use the two terms for the same thing, with micro-stop leaning toward the shortest ones. What are micro-stops covers the idea and why these stops stay hidden.

Do minor stops count as downtime?

It depends on your threshold. Above it, they are logged downtime and hit availability, and below it, they sit in performance. The lost minutes are real either way, as downtime percentage explains.

Which should we fix first?

Rank both piles by total minutes and by how hard the fix is, on the constraint machine first. One failure can outweigh a thousand jams, or the reverse. A Pareto chart settles it with your own numbers.

Should operators fix minor stops themselves?

They normally restart the line, but they should also record the repeats so someone removes the cause. Giving operators basic care tasks is part of autonomous maintenance.

See OEE per line and shift on your own machines

Fabrico is a light MES for OEE. It takes stops, speed and counts from PLCs or IoT sensors, with add-on AI cameras counting output where a machine has no usable signal, and operators tag the cause at the line.

Book a Fabrico demo

Related guides: MTBF for Multiple Machines: Series and Parallel · MTTR Reduction Plan: A 12-Week Template to Use

← More OEE Lab guides · What are micro-stops? →

OEE essentials: OEE calculator · What is OEE · Six big losses · Availability · TEEP · Cost of downtime · OEE benchmark · Free OEE tracker · Free OEE software · How to track downtime · How to implement OEE · Light MES vs full MES