🎓 CSSBB Quest › Learn

Process capability in DMAIC: baseline vs entitlement

In the Measure phase, you calculate the baseline capability — the current performance of the process as it actually runs. This becomes the "before" number you will compare against your Improve-phase results. See the CQE pages on Cp vs Cpk vs Cpm → and Sigma level & DPMO → for the underlying math. This page focuses on the DMAIC framing.

Baseline vs entitlement

Two capability benchmarks matter in a CSSBB project:

Baseline capability: Current long-term performance, measured over enough time to capture all sources of variation (shift changes, raw material lots, equipment warm-up). Use Ppk (long-term index), not Cpk, for the baseline.
Entitlement: The best short-term performance the process has already demonstrated — what the process is capable of when everything goes right. Measured with Cpk on a short, stable run under ideal conditions. Entitlement is a practical ceiling for what the current process can achieve without fundamental redesign or capital investment.

Baseline (Ppk) = min[ (USL - X̄) / (3σ_overall), (X̄ - LSL) / (3σ_overall) ] Entitlement (Cpk) = min[ (USL - X̄) / (3σ_within), (X̄ - LSL) / (3σ_within) ] σ_overall = standard deviation of all data points σ_within = short-term within-subgroup σ (from R-bar / d₂ or S-bar / c₄)

Baseline Sigma level

The DMAIC project goal is often stated as a Sigma level increase, not just a capability-index target. The long-term Z (distance from the mean to the nearest spec limit in long-term σ units) is Z_LT = 3 × Ppk. By the Six Sigma convention, the reported (short-term) Sigma level adds the 1.5σ shift: Sigma level = Z_LT + 1.5. So "Six Sigma" means Z_LT = 4.5, which gives 3.4 DPMO (one-sided normal tail).

Sigma level (reported, with 1.5σ shift)	Z_LT	Ppk ≈ Z_LT/3	DPMO
3σ	1.5	0.50	66,807
4σ	2.5	0.83	6,210
5σ	3.5	1.17	233
6σ	4.5	1.50	3.4

A short-term Cpk of 2.00 (Z_ST = 6) corresponds to a 6σ process; after the assumed 1.5σ long-term shift it performs like Ppk = 1.50. The 1.5σ shift is a convention, not a law — state it when you report a Sigma level.

What if the process is not normally distributed?

Use transformation (Box-Cox, Johnson) or non-normal capability indices (Cnpk based on percentiles). See Capability: non-normal →

Measure tollgate: capability checklist

MSA acceptable per AIAG MSA (4th ed.) guidelines: %GRR under 10% acceptable, 10–30% may be acceptable depending on the application, over 30% unacceptable; ndc ≥ 5
Data collected over a representative time period (all shifts, multiple lots)
Normality tested (histogram, Anderson-Darling or Shapiro-Wilk)
Baseline Ppk (or attribute defect rate) documented
Baseline Sigma level calculated
Entitlement documented (best observed short-term performance)

Test yourself in 2 minutes

Try 5 free CSSBB practice questions on process capability in DMAIC.

Take the free CSSBB diagnostic

More: DMAIC overview · Project charter · VOC & CTQ · SIPOC · MSA in DMAIC · Process capability · Data collection planning · Graphical analysis · Root cause analysis · Multi-vari analysis · Hypothesis testing · Regression · DOE · FMEA · Define tollgate · Change management · Financial benefits · Team dynamics · Improve tollgate · Piloting & scaling · Poka-yoke (BB) · Solution selection · Control plan · SPC in DMAIC · Sustaining gains

CSSBB Quest by Bizarre Act. Original study material; not affiliated with ASQ. CSSBB is a certification of ASQ. Terms · Privacy