1. Define the production period and counts
Record planned production time, actual run time, total pieces and good pieces for one consistent period. Identify whether reworked pieces are counted as first-pass good output. Do not combine one shift’s run time with another period’s output.
Define planned breaks and changeovers consistently. A model can only reconcile the categories you supply. Moving a loss between categories can change an individual ratio without improving actual good output.
2. Calculate OEE three ways into one result
Availability = run time ÷ planned production time. Performance = ideal cycle time × total output ÷ run time, using the same time unit. Quality = good output ÷ total output. OEE is the product of those three ratios.
Consider 420 planned minutes, 360 run minutes, 400 total pieces, 380 good pieces and a 30-second ideal cycle. Availability is 85.71%, performance is 55.56% and quality is 95%. OEE is 45.24%. A direct check gives 380 × 0.5 ideal minutes ÷ 420 planned minutes = 45.24%.
3. Compare takt with an effective cycle
Takt expresses available production time per unit of customer demand. With 420 available minutes and demand of 300 pieces, takt is 1.4 minutes, or 84 seconds, per demanded piece. It is a demand-paced interval, not a measured machine cycle.
Compare takt with the effective cycle at the actual constraint. A machine’s ideal cycle can look fast enough while downtime, changeovers, product mix, staffing or quality losses prevent the required output.
4. Count scrap and rework once
First-pass yield measures output that passes without rework under the definition you choose. Reworked output may eventually be good, but it still consumes time and resources. Keep first-pass and final yield separate when reviewing losses.
When costing scrap or rework, state whether lost material value, replacement production and rework labor are already included in another category. Adding all of them without a defined boundary can double-count a loss.
5. Use the result to investigate the constraint
Change one assumption at a time: available minutes, effective cycle, changeover frequency or good-output share. Compare the resulting good capacity with compatible demand and with actual observations.
OEE is a descriptive process measure under a chosen basis. It does not identify every cause of a loss, prove that a capacity increase can be sold or certify a process as statistically stable. Inspect the operating context before deciding what to change.
Use the calculators
More manufacturing & operations calculators · Methods, sources and testing