toolbox
Home / Manufacturing & Production / Tool Changeover Yield Loss Estimator

Tool Changeover Yield Loss Estimator

The Tool Changeover Yield Loss Estimator measures the production loss associated with startup scrap or rework after tool changes. It compares units rejected because of changeover stabilization, first-piece adjustment, setup verification, or related startup conditions with the total units produced in the post-changeover observation windows.

Separating this loss from ordinary steady-state defects helps teams see whether setup quality is consuming material and capacity even after the machine returns to operation. The calculator reports lost units, yield loss percentage, good startup yield, and average loss per changeover so the impact can be reviewed at both batch and setup-event levels.

Post-changeover quality inputs

units
units
changes
Result
—
Changeover yield loss
Lost/rework units—
Good startup yield—
Loss per changeover—

1. Define the observation window
Use a consistent post-changeover window, such as the first fixed number of pieces or a defined stabilization period.

2. Enter total post-changeover units
Count all units produced in those observation windows across the selected changeovers.

3. Enter attributable loss units
Count rejects or rework reasonably linked to setup stabilization or changeover conditions.

4. Enter observed changeovers
Use the number of setup events represented by the totals.

5. Review percentage and per-change loss
Use the percentage for normalized comparison and units per changeover for practical improvement targeting.

Changeover yield loss % = Changeover-related loss units ÷ Post-changeover units × 100
Good startup yield % = (Post-changeover units − Loss units) ÷ Post-changeover units × 100
Loss per changeover = Loss units ÷ Changeovers observed

All unit counts should come from the same defined post-changeover observation window. The method estimates association with changeover conditions; it does not prove root cause for each defect.

What the result means

The main result is the share of units in the defined post-changeover windows that were lost or sent to rework for changeover-related reasons.

Keep the observation window stable; expanding or shrinking it between periods can change the percentage even when setup behavior is unchanged.

Given

  • Post-changeover units = 2,400
  • Changeover-related reject/rework units = 96
  • Changeovers observed = 12

Calculation

Yield loss = 96 ÷ 2,400 × 100 = 4.00%. Good startup yield = 96.00%. Loss per changeover = 96 ÷ 12 = 8.00 units.

Result

Estimated changeover yield loss = 4.00%.

Across the observed setups, an average of eight units per changeover were rejected or reworked within the defined startup windows.

What defects should be classified as changeover-related?

Use defects connected to setup, first-piece adjustment, warm-up, alignment, parameter loading, or stabilization under your plant rules. Avoid reclassifying unrelated steady-state defects just because they occurred after a setup.

Should reworked units count as loss?

Include them if your improvement metric treats rework as a yield loss. If you only track physical scrap, enter scrap units only and label the interpretation accordingly.

How long should the post-changeover window be?

Use a window that captures the stabilization period without drifting into normal steady-state production. A fixed piece count or defined time interval makes comparisons more consistent.

Can a low loss percentage still be important?

Yes. High changeover frequency or expensive material can make a small percentage create meaningful total cost and capacity loss.

How is this different from overall production yield loss?

This metric isolates a defined startup period after changeovers. Overall yield loss includes defects from the full production run and may have many other causes.