The core idea of this page: OEE is a single number that answers one question — of all the time you scheduled a line to run, how much of it produced good product at design speed? It is a loss-finding tool, not a scoreboard. Used properly it tells you where the next hour of capacity is hiding; used badly it becomes a league table that operators learn to game.
Why packaging lines need a loss metric, not a speed metric
A packaging line can hit its rated speed in a demonstration and still lose half its scheduled time to changeovers, minor stops and rejects. Measuring output per hour hides this, because a fast machine that stops constantly and a slower machine that never stops can post the same figure. OEE separates the two by multiplying three ratios that each isolate a different failure family.
Availability
Availability is run time divided by planned production time. Planned production time excludes breaks, planned maintenance and shifts with no demand — but it must include changeovers if changeovers are part of running that line. The most common measurement error is quietly moving changeover minutes out of planned time, which inflates availability and makes the number useless for improvement.
Performance
Performance compares actual output to the theoretical output at rated speed for the time the line actually ran. Ideal cycle time comes from the machine rating, not from the best hour anyone has ever seen. If you use an observed best as the ideal, performance approaches one hundred percent and the small-stop losses disappear from view.
Quality
Quality is good units divided by total units started. Decide in advance whether rework counts as good (it usually should not for packaging, where a reprocessed pack may carry a second seal) and whether giveaway — overfilling — counts as a quality loss. For filling lines, giveaway is a real cost loss even when the pack passes inspection.
The six big losses behind the three ratios
OEE is only actionable when decomposed into the losses that feed it. The classic set maps neatly onto packaging equipment.
Availability losses
Breakdowns (mechanical and electrical failures, including the failure to restart after a jam) and setup and adjustment losses (changeover, format change, cleaning between products). On packaging lines, setup losses often dominate, which is why SMED and OEE improvement tend to travel together.
Performance losses
Small stops and idling (jams cleared in under a minute, starved infeeds, blocked outfeeds, sensor nuisance trips) and reduced speed (running below rated speed to protect fragile product or avoid jams). Small stops are the hardest to see because nobody logs a twenty-second jam — which is exactly why they account for so much hidden capacity.
Quality losses
Process defects and rework (seal failures, misfilled packs, label misplacement, rejects at checkweighing) and startup and yield losses (product lost bringing the line up to spec after a changeover or a stop). Startup scrap is the reason two lines with identical steady-state performance can have very different OEE.
What a realistic benchmark looks like
World-class OEE is often quoted at eighty-five percent, which implies roughly ninety percent availability, ninety-five percent performance and ninety-nine percent quality. Very few packaging lines reach that without years of focused work. A more useful frame for most plants: measure for a month without changing anything, then set a target that removes one named loss rather than a target that raises the composite number.
Beware comparisons between categories. A high-speed form-fill-seal line running one product for weeks should beat a filling line that changes format twice a shift. Comparing them punishes the harder machine and teaches nobody anything.
How to measure OEE without gaming it
Instrument the line, not the operator
Use PLC counters where they exist: machine run signal, cycle count, reject count. Manual logbooks drift toward round numbers. If you must log manually, log the loss reasons from a fixed list so the data can be grouped later.
Define the boundaries once
Write down what counts as planned time, what counts as a stop, and what counts as good product. Post the definition at the line. When the definition changes, restate history or start a new baseline — do not splice two definitions into one trend line.
Review losses weekly, not the score
The weekly meeting should look at the top three losses by minutes and by units, pick one, and assign a countermeasure with an owner and a date. The composite number is the headline; the loss Pareto is the work.
Common failure modes
The three patterns that kill OEE programmes are: measuring only when convenient (reporting the best shift), attributing every stop to "operator error" so no engineering fix is ever made, and treating OEE as a performance review for the line crew. None of these survive contact with a well-run line for long, because the crews stop reporting honestly and the number improves while output does not.
Verification checklist
Before trusting an OEE figure, confirm: planned time is documented and includes changeovers; ideal cycle time comes from the machine rating; reject counting matches the checkweigher and the case packer; startup scrap is counted; and the loss reasons reconcile to the total lost time. If the three ratios multiply to the reported OEE and the loss minutes add up to the gap, the measurement is sound.
Source: Overall equipment effectiveness — Wikipedia
"Overall equipment effectiveness (OEE) is a measure of how well a manufacturing equipment is utilized compared to its full potential, during the periods when it is scheduled to run."
Frequently asked questions
Is OEE the same as line efficiency?
No. Line efficiency usually compares output to a nominal target; OEE decomposes the gap into availability, performance and quality so each loss can be attacked separately. Efficiency alone cannot tell you whether to fix a changeover or a jam.
Should changeover time count against availability?
Yes, if the line is scheduled to be in production and the changeover is part of that schedule. Excluding it is the most common way OEE figures become fiction.
What is a good OEE for a packaging line?
Most lines start between forty-five and sixty percent. Treat eighty-five percent as an aspiration, and set improvement targets against named losses rather than against the composite number.
Related reading: OEE and downtime analysis, changeover and SMED, factory acceptance testing, and the end-of-line packaging category.
Related guides
- End-of-Line Machines: Top 10 Compared (2026) — Ten end-of-line machines compared across case packing, palletizing, sh…
- Case Packing vs Palletizing: Which to Automate First — case packing and palletizing are two different automation decisions — …
- Conventional vs Robotic Palletizing — conventional (layer) palletizers build a whole layer at once and win o…
- Shrink Wrapping vs Stretch Hooding — shrink wrapping and stretch hooding both stabilize a pallet load with …
- Packaging Line Design and Integration Guide: Plan a Line That Runs — a packaging line is designed from the product out, not from the machin…
- FAT and SAT for Packaging Machinery: Acceptance Testing Explained — FAT and SAT are the two acceptance gates that transfer risk from the s…
- Packaging Line OEE and Downtime Analysis: Measure the Losses That Matter — OEE turns a vague “the line is down too much” into three m…
- How a Case Packer Works — a case packer takes individual packs and builds them into a shipping c…
