SC QUALITY GUILDAttribute capability
SCQUALITYGUILD.COM/TOOLS/ATTRIBUTE-CAPABILITY.HTML
Enter units inspected, defective units and defects by subgroup to get binomial and Poisson capability, DPMO, Z.bench with and without the 1.5 shift, and p and u charts to check stability.
ON THE EXAM · CSSGB II.E.1, III.F.3, III.F.4, VI.A.3 · FREE · NOTHING YOU TYPE LEAVES YOUR BROWSER
Q-BotClick for an explanation of this tool.For educational use. This page is a study aid. The method behind it has far more depth and detail than one page and one tool can cover, and what you see here is not everything the method offers or everything this site offers. For full understanding and application, use the governing standards and texts, and seek further education, training or qualified advice. Results depend on the data you enter, so check them before relying on them.
Capability when the data are counts
Cp and Cpk need a measured characteristic and specification limits. When the output is pass or fail, or a count of nonconformities, capability is stated as a rate instead. Two models apply:
- Binomial (defectives): each unit is good or bad. Capability is the average proportion defective p̄, often quoted as PPM defective or as yield, 1 − p̄.
- Poisson (defects): a unit can carry several defects. Capability is defects per unit, DPU. Dividing by opportunities per unit gives DPO, and DPO × 1,000,000 is DPMO. Under the Poisson model the share of units with no defects is e−DPU.
Z.bench and the 1.5 shift
To put an attribute rate on the same scale as variables capability, convert it to a standard normal value: Z.bench = Φ⁻¹(1 − p), the Z that leaves a tail of area p. Because the data span the whole study, this is a long-term figure. The Six Sigma convention adds 1.5 to report a short-term sigma level, which is how 3.4 DPMO becomes six sigma. The tool labels both. The shift is a convention, not a measured property of your process, so always say which figure you quote. The sigma level and DPMO calculator shows the conversion table in both directions.
Stability comes first
A capability figure predicts future output only if the process is stable. The tool draws a p chart for defectives and a u chart for defects, with limits that widen for smaller subgroups. Points outside the limits are special causes to remove before the rate means anything. Use 20 or more subgroups where you can.
Common traps
- Inflating opportunities. Counting every conceivable way a unit could fail lowers DPMO and raises the sigma level without any change the customer would notice.
- Mixing defects and defectives. A unit with three defects is one defective.
- Quoting a rate from very few failures. The confidence interval shows how little a handful of defectives pins down.
On the CSSGB exam
Expect to compute DPU, DPO, DPMO and yield from a short scenario, convert DPMO to a sigma level with a table, and know that the table includes the 1.5 shift. Questions also test the difference between a defect and a defective.