OEE combines availability, performance, and quality
Overall equipment effectiveness is calculated as Availability multiplied by Performance multiplied by Quality. Availability measures run time against planned production time. Performance compares actual output with the output implied by an ideal cycle time. Quality measures good output against total output. The calculation is only comparable when those inputs use the same definitions and operating context.
| Factor | Formula | Losses exposed | Definition risk |
|---|---|---|---|
| Availability | Run time / planned production time | Breakdowns, setups, waiting, stops | What counts as planned time or downtime |
| Performance | Ideal cycle time x total count / run time | Small stops, reduced speed | Ideal cycle by product, cavity, and process |
| Quality | Good count / total count | Scrap, rejects, startup loss | When rework or later inspection changes good count |
| OEE | Availability x Performance x Quality | Combined equipment-related loss | Equal totals can hide different loss patterns |
Calculate the components before interpreting the total
Assume a resource has 420 minutes of planned production time and runs for 360 minutes. Availability is 360 divided by 420, or 85.7 percent. If the ideal cycle is 0.50 minute and total output is 660 pieces, performance is 0.50 times 660 divided by 360, or 91.7 percent. If 640 pieces are good, quality is 640 divided by 660, or 97.0 percent.
Multiplying the three factors produces an OEE of about 76.2 percent. That figure is not a diagnosis. The component losses show where to investigate: 60 minutes outside run time, production below the ideal rate, and 20 pieces that were not good. Retain the underlying events so a supervisor can separate planned setup, unplanned failure, starvation, blocked flow, minor stops, slow cycles, startup loss, and process rejects.
Do not manage the total alone
Two resources can have the same OEE while one loses time to breakdowns and the other loses output to speed or quality. Improvement work must start with the component and reason evidence.
Agree on the measurement rules before comparing scores
The planned-production-time boundary is the first decision. Determine how breaks, no-demand periods, preventive maintenance, changeovers, warmup, trials, and cleaning are treated. The rule should match the question being asked. A utilization question and an equipment-loss question may need different measures.
Ideal cycle time is equally sensitive. Use a controlled standard tied to product, operation, tooling, cavity count, and relevant process condition. If the standard is padded, stale, or averaged across unlike work, performance can look acceptable while speed loss remains. Define how rework, later inspection, partial cavities, and count corrections affect total and good quantities.
- Store event time and duration, not only summarized percentages.
- Use a reason-code hierarchy that operators can apply consistently.
- Separate state detection from cause assignment when the machine cannot know the cause.
- Keep product and operation context with the cycle standard.
- Assign approval for corrections and definition changes.
Use OEE to investigate equipment-related loss
OEE is most useful when a team can drill from the score into time, rate, and quality evidence and then assign a loss to an owner. A Pareto chart without a stable reason model can direct attention to the easiest code rather than the real cause. Validate automatic states against observation, especially around setup, small stops, blocked flow, and manual cycles.
OEE does not state whether the resource should have been scheduled, whether demand exists, whether the selected job was the right priority, or whether another part of the value stream constrained throughput. Pair it with schedule attainment, delivery, queue, changeover, yield, and constraint measures where those questions matter. Avoid a universal target that ignores equipment role and product mix.
| Question | Use OEE? | Companion evidence |
|---|---|---|
| Where is equipment time or rate being lost? | Yes | Events, reasons, product, operation |
| Did the plant make the required mix on time? | Not alone | Schedule attainment and delivery |
| Is the asset the system constraint? | Not alone | Queue, flow, throughput, constraint analysis |
| Can sites be compared? | Only with common definitions | Rule register and audit samples |
Connect OEE events to orders, quality, and schedule context
A production-monitoring system should associate counts and states with the correct resource, item, operation, order, shift, and reason. Test automatic and manual sources, event buffering, clock alignment, changeovers, corrections, rejected parts, rework, and outages. Confirm that supervisors can move from an aggregate result to the source event.
DELMIAWorks public MES documentation includes production monitoring and OEE functions. The precise calculation rules, machine connections, configuration, and reporting behavior must be verified against the intended release. An integrated record is valuable only if the data definitions and ownership are governed.
Create an OEE data-definition register
Document planned time, downtime threshold, ideal cycle, count sources, good-count timing, reason ownership, corrections, and validation samples before publishing comparisons.
Frequently asked questions
What does OEE stand for?
OEE stands for overall equipment effectiveness. It combines availability, performance, and quality into one equipment-loss measure.
How is OEE calculated?
OEE equals Availability multiplied by Performance multiplied by Quality. Availability is run time divided by planned production time. Performance uses ideal cycle time, total count, and run time. Quality is good count divided by total count.
What is a good OEE score?
There is no universal score that is meaningful without context. Equipment role, product mix, planned-time rules, ideal-cycle standards, data quality, and operating objectives affect interpretation. Use the components and loss evidence to guide action.
Can OEE compare machines or plants?
Only when the compared operations use common definitions, controlled cycle standards, consistent count and reason rules, and enough product and process context to make the comparison valid.
References and method
- OEE.com: Calculating OEE, formula and component definitions
- DELMIAWorks MES data sheet, vendor-documented production monitoring and OEE scope; current configuration requires verification
- Dassault Systèmes DELMIAWorks, current manufacturing ERP and MES architecture
- The tables, diagrams, selection questions, and downloadable register are Morphos 3D guidance. Verify product configuration, interfaces, validation requirements, and commercial packaging for the exact release and operating environment.
Audit the definitions before comparing OEE
Use the register on one representative resource, validate the events against observed work, and resolve definition ownership before expanding the measure.