OEE Explained: Two Lines at 60% With Completely Different Problems

Two packaging lines both report an OEE of 60% for the month. Management asks for one improvement plan for both. That is a mistake: Line A needs a maintenance and changeover project, and Line B needs a quality project. Same score, opposite fixes.

OEE (overall equipment effectiveness) measures how much of the time you planned to produce was turned into good parts at full speed. It multiplies three factors: availability (was the machine running?), performance (was it running at its ideal speed?) and quality (were the parts good?). 100% means only good parts, as fast as possible, with no stops.

How to calculate OEE: a worked shift

One shift, one machine:

• Planned production time: 480 minutes (8 hours, breaks excluded)
• Downtime (a breakdown and a changeover): 60 minutes
• Ideal cycle time: 1 minute per part
• Total parts made: 380
• Good parts: 361

Availability = run time ÷ planned time. Run time is 480 − 60 = 420 minutes. 420 ÷ 480 = 87.5%.

Performance = ideal cycle time × total count ÷ run time. At 1 minute per part, 380 parts should take 380 minutes. The machine ran 420. 380 ÷ 420 = 90.5%. The missing 40 minutes went to slow running and short stops.

Quality = good count ÷ total count. 361 ÷ 380 = 95%.

OEE = 0.875 × 0.905 × 0.95 ≈ 75.2%.

Here is a shortcut worth remembering: OEE also equals good parts × ideal cycle time ÷ planned time. 361 × 1 ÷ 480 = 75.2%. Same answer. In plain words: in 480 minutes, a perfect machine would have made 480 good parts. You made 361.

Two lines at 60%, completely different problems

Now back to the two lines.

Line A: availability 65%, performance 95%, quality 97%. OEE = 0.65 × 0.95 × 0.97 ≈ 60%. The machine runs fast and makes good parts when it runs. It just doesn't run a third of the time. The fix is in breakdowns and changeovers: maintenance routines, faster setups, smarter sequencing. Changeover optimization shows how much sequence alone can save.

Line B: availability 95%, performance 90%, quality 70%. OEE = 0.95 × 0.90 × 0.70 ≈ 60%. It almost never stops, but three parts in ten are bad. More uptime would just make more scrap faster. The fix is process control, materials and inspection, the territory of quality management.

Think of two students with the same 60% average. One skipped a third of the exams and aced the rest. The other sat every exam and got many answers wrong. Nobody would send them to the same tutor.

The six big losses behind OEE

The three factors break down into six typical losses, a classic list from Total Productive Maintenance (TPM):

• Availability: breakdowns; setups and adjustments
• Performance: small stops and idling; reduced speed
• Quality: startup rejects; production rejects

When a line's OEE drops, the first question is not "how do we raise OEE?" but "which of the six grew?" Record losses in these buckets and the improvement plan almost writes itself.

Myth: "Every machine should reach 85% OEE"

An OEE of 85% is often cited as "world-class," usually explained as 90% availability × 95% performance × 99.9% quality. It is a useful reference point from the TPM literature, not a law of physics, and chasing it everywhere can hurt you.

• On a non-bottleneck machine, higher OEE mostly means more parts sitting in front of the next machine. More WIP, not more shipments. Improvement effort belongs at the constraint first; see manufacturing optimization.
• On a high-mix line, frequent changeovers are the price of delivering small orders on time. Cutting them to raise availability can make customers wait.
• Across different machines, comparing OEE is often meaningless. A new high-speed line and a 20-year-old press have different ideal rates and different jobs.

OEE is a diagnostic tool for one machine over time. It gets misleading when it is a trophy. How it should be paired with delivery measures is covered in manufacturing KPIs.

Common OEE calculation mistakes

• Using a "realistic" cycle time instead of the ideal one. Performance looks great and the losses disappear from view. If performance is above 100%, the cycle time is wrong.
• Excluding changeovers from planned time. Changeovers are real availability loss. Hiding them hides the biggest lever on many lines.
• Counting reworked parts as good. Quality counts parts that were good the first time.

FAQ

Is OEE the same as utilization? No. Utilization only asks whether a machine was busy. OEE also asks whether it ran at full speed and made good parts.

Should planned maintenance count as downtime? Most plants exclude it from planned production time, but agree on the rule and keep it the same every month.

How it looks in factory.online

In factory.online, output and scrap confirmations on job orders record the total and good quantities for each run, and the routing holds the time per operation. That gives you the inputs for quality and performance from the same records the planner already uses, and through the Power BI integration you can combine them with your downtime records to split Line A's and Line B's 60% into three factors side by side.

BEEDIGIT collects machine and operator data for OEE, downtime and actual production reporting. These records help the team compare planned capacity with what the equipment actually achieved and investigate recurring losses.

If you have two lines with the same score and one improvement plan, book a demo and let's pull them apart.