Log defects across categories, shifts, and lines easily. See DPU and DPMO with trends instantly. Export reports, prioritize fixes, and raise quality every day.
| Date | Shift | Line | Units | Defects | OPU | DPU | DPMO |
|---|---|---|---|---|---|---|---|
| 2026-02-17 | A | Line 1 | 1,250 | 18 | 6 | 0.0144 | 2,400.00 |
| 2026-02-18 | B | Line 2 | 980 | 21 | 5 | 0.0214 | 4,285.71 |
| 2026-02-19 | Night | Line 1 | 1,100 | 12 | 6 | 0.0109 | 1,818.18 |
Higher inspection volume improves confidence in the defect picture. If you inspect 1,000 units and find 20 defects, DPU equals 0.020 and the defect rate is 2.0%. Doubling inspection to 2,000 units with the same 20 defects halves DPU to 0.010, showing a meaningful shift in stability. Record shift, line, and lot so teams can validate whether changes are real or just sampling noise.
Raw defect counts mislead when production volumes differ. PPM scales defects to one million units, so 18 defects in 1,250 units equals 14,400 PPM, while 21 defects in 980 units equals 21,429 PPM. Use these normalized rates to compare shifts, lines, or suppliers without arguing about sample size. Track results over time and flag sudden jumps that may need containment and escalation.
Complex products have multiple chances for failure per unit. Opportunities per unit (OPU) captures that complexity, letting DPMO benchmark processes fairly. For example, 12 defects across 1,100 units with 6 opportunities gives DPMO near 1,818, while the same defects with 3 opportunities doubles DPMO, signaling higher opportunity-level risk. Define OPU from critical checks and keep it consistent across audits to preserve comparability.
DPMO connects to expected performance levels. The calculator estimates yield with e−DPU, so DPU 0.0109 gives yield about 98.91%. It also approximates sigma as Z(1−DPMO/1,000,000)+1.5, which helps communicate capability in familiar Six Sigma terms while staying grounded in measured defect data. For very low defect rates, expand inspection volume to reduce uncertainty before declaring a capability improvement.
Category tracking turns measurement into action. Sort categories by count to build a Pareto view, then use cumulative share to target the vital few. If the top two categories contribute 70% of defects, prioritizing containment, root-cause analysis, and verification on those categories yields faster improvement than broad, unfocused fixes. After corrective actions, rerun the tracker and export CSV or PDF to document the before-and-after evidence clearly.
Use the number of defect chances per unit, such as checkpoints, characteristics, or joints inspected. Keep the same OPU definition for the same product family so DPMO comparisons across shifts, lines, and weeks remain consistent and meaningful.
Category rows are optional. If you enter categories, their counts should usually add up to the total defects field. A mismatch can indicate missing categories, double counting, or a total entered from a different inspection record.
Yes. One unit can contain multiple defects, so total defects may exceed units inspected. If you want a ‘defective units’ view instead, record only one defect per unit or add a separate tracker for defective units versus total defects.
It is an approximation derived from DPMO using the normal distribution, then adjusted with a 1.5 shift. Use it as a communication shorthand, not a substitute for control charts, capability studies, or process-specific reliability models.
The Poisson-based approximation assumes defects occur randomly and independently across units. It converts the average defect intensity (DPU) into an expected ‘defect-free unit’ yield. If defects cluster or are dependent, validate with additional sampling and analysis.
Use CSV for spreadsheets, trend charts, and sharing with analytics tools. Use PDF for audits, shift handoffs, and reporting packs. Export after each submit so the file reflects the current results and any updated category Pareto table.
Important Note: All the Calculators listed in this site are for educational purpose only and we do not guarentee the accuracy of results. Please do consult with other sources as well.