What is OEE and how do you calculate it? Formula, worked example and common mistakes

OEE (Overall Equipment Effectiveness) is a percentage that shows how much of a machine’s or line’s planned production time is truly productive, meaning good parts made at ideal speed. It is the product of three factors: availability, performance and quality. An OEE score of 100% means the equipment ran for the entire planned time, at its ideal cycle time, without producing a single defect.

This guide covers what OEE stands for, the OEE formula, a step-by-step OEE calculation example with numbers that add up, how to build it in Excel, and the calculation mistakes we see most often on the shop floor.

What OEE stands for and why it matters

OEE stands for Overall Equipment Effectiveness. The concept was formalized by Seiichi Nakajima as part of Total Productive Maintenance (TPM) and has become the most widely used KPI for measuring manufacturing productivity at the equipment level.

The value of OEE is not the single number but what it reveals about where time is lost. An OEE of 60% tells you that roughly 40% of planned time is not productive, and that this loss sits in one or more of three buckets. Knowing which bucket is full tells you whether maintenance, scheduling or quality needs to act.

OEE helps answer questions such as:

• How much of its real capacity is this machine or line using?
• Are we losing time to downtime, slow running or scrap?
• Did the improvement we made (SMED, a maintenance plan, a tooling change) actually work?
• Do we need another machine, or are we under-using the ones we have?

The OEE formula: availability, performance and quality

The core OEE formula is:

OEE = Availability × Performance × Quality

Each factor is a ratio between 0 and 1 and measures a different kind of loss.

Factor	Formula	Loss it captures
Availability	Run time / Planned production time	Breakdowns, setups and changeovers, waiting for material
Performance	(Ideal cycle time × Total count) / Run time	Minor stops, idling, running below ideal speed
Quality	Good count / Total count	Scrap, rework, startup rejects

Most OEE errors come from loose definitions, so be precise:

• Planned production time is shift time minus planned breaks and periods with no production scheduled.
• Run time is planned production time minus unplanned stops and setup/changeover time.
• Ideal cycle time is the fastest cycle the equipment can theoretically achieve for that part. It is not the average or “standard” cycle time; use the machine builder’s rating, the tooling or process design value, or the best demonstrated stable cycle.
• Good count is parts that pass quality the first time, without rework.

Multiplying the three factors gives the same answer as this shortcut:

OEE = (Good count × Ideal cycle time) / Planned production time

In other words: if we had made only good parts at ideal speed, what fraction of planned time would that have taken? If the two methods disagree, one of your inputs is wrong.

OEE calculation example, step by step

Take one 8-hour shift on a plastic injection molding machine. The numbers are illustrative, not taken from a real plant.

Input	Value
Shift length	480 min
Planned breaks	30 min
Unplanned downtime (breakdown)	25 min
Mold change and setup	20 min
Ideal cycle time	30 s per part (0.5 min)
Total parts produced	750
Scrap and rework	30

1. Planned production time

480 min shift − 30 min breaks = 450 min

2. Availability

Run time = 450 − 25 − 20 = 405 min

Availability = 405 / 450 = 0.900 → 90.0%

3. Performance

Time needed for 750 parts at ideal speed = 750 × 0.5 = 375 min

Performance = 375 / 405 = 0.926 → 92.6%

So of the 405 minutes the machine was running, 30 minutes were lost to minor stops and slow cycles.

4. Quality

Good count = 750 − 30 = 720

Quality = 720 / 750 = 0.960 → 96.0%

5. OEE

OEE = 0.900 × 0.926 × 0.960 = 0.800 → 80.0%

Check: (720 good parts × 0.5 min) / 450 min = 360 / 450 = 0.800. Both methods agree.

Turning percentages into minutes

Percentages work in a management review; minutes work on the floor. Reading the same example as time lost makes the priority obvious:

Loss	Time	Share of planned time
Availability loss (breakdown + changeover)	45 min	10.0%
Performance loss (minor stops + slow cycles)	30 min	6.7%
Quality loss (30 parts × 0.5 min)	15 min	3.3%
Fully productive time	360 min	80.0%
Total	450 min	100%

The first target is clearly the 45 minutes of downtime. The next question is how that time splits between breakdown reasons and changeover duration.

The six big losses

TPM groups OEE losses into the “six big losses,” each of which lowers one factor:

OEE factor	Big loss	Shop-floor example
Availability	Equipment failure	Hydraulic fault, servo alarm, broken tool
Availability	Setup and adjustment	Mold change, color change, tool change, first-article approval
Performance	Idling and minor stops	Blocked sensor, jammed part, operator waiting
Performance	Reduced speed	Cycle deliberately slowed to work around a quality issue
Quality	Process defects	Flash, short shots, out-of-tolerance parts, rework
Quality	Startup losses	Parts rejected while the process warms up and stabilizes

Minor stops deserve special attention. Stops of a minute or two are almost never written on paper logs, yet they add up over a shift. In manual data they silently disappear into “performance loss”; with automatic machine data each one becomes visible.

World-class OEE: what does 85% mean?

The most widely cited reference is 85% “world-class” OEE, commonly attributed to Nakajima’s TPM work. It is usually quoted with these factor targets:

Factor	Commonly cited reference
Availability	90%
Performance	95%
Quality	99.9%
OEE	≈ 85% (0.90 × 0.95 × 0.999 = 0.854)

Use it with care. It is a benchmark for discrete manufacturing, not an industry average; continuous processes, high-mix lines with frequent changeovers and manual assembly behave differently. And two plants’ OEE scores are only comparable if they share the same definitions of planned time, ideal cycle time and good count. The most useful comparison is a machine against its own history.

How to calculate OEE in Excel

Many plants start measuring OEE in a spreadsheet, and that is a reasonable first step. Create one row per shift and machine:

Column	Content	Example formula
A	Date / shift / machine	—
B	Shift length (min)	manual
C	Planned breaks (min)	manual
D	Downtime (min)	manual
E	Ideal cycle time (min/part)	manual
F	Total count	manual
G	Good count	manual
H	Planned production time	=B2-C2
I	Availability	=(H2-D2)/H2
J	Performance	=(E2*F2)/(H2-D2)
K	Quality	=G2/F2
L	OEE	=I2*J2*K2

Format columns I–L as percentages. Do not cap performance with =MIN(1,…); a value above 100% is a warning sign that the ideal cycle time is wrong.

The spreadsheet math is fine; the inputs are the problem. Downtime is written from memory at the end of the shift, short stops disappear, counts are estimated, and each shift interprets the definitions a little differently. The result is usually an OEE that looks better than reality. For a quick calculation without building a sheet, you can use the free OEE calculator from the Proxus team.

Common OEE calculation mistakes

These are the mistakes we run into most often on OEE projects:

1. Using average cycle time instead of ideal cycle time. This pushes performance toward 100% and hides speed losses.
2. Treating changeovers as planned downtime. Even a scheduled mold or tool change is an availability loss. Excluding it makes SMED improvements invisible.
3. Not recording minor stops. Manual logs rarely capture stops of a few minutes.
4. Counting reworked parts as good. Quality should follow first-pass yield.
5. Averaging machine OEE to get line OEE. On a serial line the bottleneck sets the pace. Calculate line OEE from good parts at the end of the line and the bottleneck’s ideal cycle time.
6. Using OEE to grade people. Once the number is used to blame shifts, data entry stops being honest.
7. Comparing scores built on different definitions. Write the definitions down and apply them everywhere before benchmarking lines or sites.

OEE vs TEEP vs OOE

Metric	Time base	When to use it
OEE	Planned production time	Daily improvement and loss analysis per shift and machine
OOE (Overall Operations Effectiveness)	Operating time	When unscheduled time inside operating hours should count as loss
TEEP (Total Effective Equipment Performance)	All calendar time (24/7)	Capacity planning, investment and extra-shift decisions

TEEP = OEE × loading (planned production time / calendar time). High OEE with low TEEP means the constraint is the schedule, not the machine.

How to improve OEE

There is no single recipe, but the order rarely changes:

1. Measure accurately. Take run state, part count and stop start/end times directly from the machine. Collecting data from the PLC or CNC closes the gaps in manual records.
2. Attach reasons to stops. A short reason list on an operator panel or tablet turns raw downtime into something you can analyze. We cover this in detail in downtime tracking.
3. Pick the biggest loss. Use a Pareto chart and focus on the few reasons that cost the most time.
4. Fix it permanently. Root-cause analysis and maintenance plans for breakdowns, SMED for changeovers, mechanical and sensor fixes for minor stops, parameter tracking for quality.
5. Track the effect with the same definitions. If the definition changes, real improvement and calculation changes become impossible to separate.

When breakdowns dominate availability loss, condition monitoring and predictive maintenance are the natural next step.

Manual vs automatic OEE

Topic	Manual (paper/Excel)	Automatic (machine data)
Downtime	Estimated at end of shift	Second-level resolution from machine signals
Minor stops	Usually missing	Each one visible
Counts	Tallied or estimated	From the PLC/CNC counter
Latency	A day or more	Real time
Consistency	Depends on the person and shift	One definition across all machines
Data entry burden	On operators and supervisors	Only selecting a stop reason

Automatic OEE does not remove the operator. The machine supplies time and count; the person supplies the reason. That split keeps the data accurate and keeps ownership on the floor.

How VOLSOFT approaches OEE projects

VOLSOFT is an industrial software and integration engineering firm. We do not touch machine control logic; we build the data and software layer on top of it. A typical OEE project looks like this:

• Site survey and definitions. We document the machine list, controller brands and network, and agree with the customer on the OEE definitions: what counts as planned downtime, where ideal cycle times come from.
• Machine data collection. Run state, counters and alarms are read from Siemens S7, Mitsubishi, Omron and Allen-Bradley PLCs, Fanuc and Heidenhain CNCs, and OPC UA and Modbus devices, without changing the controller program. See machine data collection.
• OEE and downtime tracking on Proxus. Every delivery runs on Proxus, the industrial data platform our team develops. Data stays on the customer’s own server, OEE and stop reasons are calculated there, dashboards and shift reports are built on top, and data is buffered at the edge if the network drops.
• ERP and reporting integration. Work orders and product data from the ERP let OEE be reported by product and order as well as by machine.

If you want to measure OEE automatically from your machines, see our OEE software page or get in touch to scope a site survey.

Summary

• OEE = Availability × Performance × Quality, or (good count × ideal cycle time) / planned production time.
• The formula is simple; the hard part is defining planned time, ideal cycle time and good count correctly and consistently.
• 85% is a widely cited Nakajima/TPM benchmark; the real comparison is the machine against itself.
• Converting losses into minutes shows where to start.
• Excel is a fine start; seeing minor stops and real speed losses requires data from the machine.