Cpk = 2.00: Six Sigma Capability

What kind of process actually achieves Cpk 2.00, and when does that target belong on a print?

• Typical PPM: ≈ 0.00099 PPM at the nearest limit; ≈ 0.00198 PPM only for a perfectly centered, two-sided process
• Sigma equivalent: 6σ

Common automotive practice uses 1.33 for established production and 1.67 for some initial or launch studies, but the controlling value is always the applicable customer-specific requirement, drawing, Control Plan, or internal standard. Do not treat these values as universal pass/fail rules.

1. Verify the 2.00 requirement in the CSR or design record; do not assume.
2. Confirm Cp ceiling before setting a Cpk target — Cpk cannot exceed Cp.
3. Run a long-window Ppk study; short-term Cpk 2.00 is easier than sustained Ppk 2.00.
4. If achievable only in the short term, plan the SPC reaction to protect the customer.

• Requiring Cpk 2.00 on a characteristic with a physically bounded tolerance the process cannot deliver.
• Reporting Cpk 2.00 on tiny samples where the confidence interval extends below 1.5.

• Cpk 2.00 is rarely a tuning outcome; it is usually a design decision (looser tolerance or fundamentally different process).
• Motorola's original 6σ definition assumed a 1.5σ shift, which corresponds to Cpk ≈ 1.5 long-term, not 2.00.
• Sustaining Cpk 2.00 typically requires closed-loop control or in-process feedback, not open-loop SPC.

• Do not require 2.00 as an internal default; specify it only when risk justifies it.

• Cpk 2.00 ≈ 6σ ≈ 0.001 PPM at the nearest tail (no shift convention).
• With a 1.5σ shift convention, Cpk 2.00 corresponds to ≈ 3.4 DPMO — but the shift is a modeling choice, not a fact.

Use the free Process Capability Calculator to compute Cp, Cpk, Pp, Ppk, sigma level, and predicted PPM from your measurement data — no sign-up required.

• State the shift convention (0σ vs 1.5σ) with every Cpk 2.00 result.

• ✓2.00 target justified by CSR, safety, or regulatory classification
• ✓Sigma-shift convention (0σ or 1.5σ) declared
• ✓Long-window Ppk confirms the short-term result

• Process condition: statistical stability is required.
• Distribution assumption: approximately normal data.
• PPM convention: nearest specification-limit tail.
• Sigma-shift convention: no sigma shift applied.
• Cpk alone identifies distance to the nearest specification limit; it does not determine total two-sided defects unless centering or both tail distances are known.
• Predicted PPM is a model estimate, not a replacement for observed defect data.