Statistics & SPC — Cheat Sheet

Descriptive stats, normal distribution, Cp/Cpk, control-chart limits and Six-Sigma DPMO.

Descriptive

Meanx̄ = Σx / n
Sample variances² = Σ(x−x̄)² / (n−1)
Std deviations = √s²
RangeR = x_max − x_min
Std errorSE = s / √n

Normal distribution

Z-scorez = (x − µ) / σ
Empirical rule68 / 95 / 99.7% within 1 / 2 / 3σ
Confidence (95%)x̄ ± 1.96·σ/√n

Process capability

Cp(USL − LSL) / 6σ
Cpkmin[(USL−µ), (µ−LSL)] / 3σ
TargetCpk ≥ 1.33 (≈ 4σ); 1.67 for safety-critical
Pp/Ppksame, using long-term σ

Control charts (X̄–R)

X̄ limitsX̄̄ ± A₂·R̄
R chartUCL = D₄·R̄, LCL = D₃·R̄
n=2A₂=1.880, D₃=0, D₄=3.267
n=4A₂=0.729, D₃=0, D₄=2.282
n=5A₂=0.577, D₃=0, D₄=2.114

Six Sigma

DPMOdefects / (units × opps) × 10⁶
Yield(1 − DPMO/10⁶) × 100%
6σ level3.4 DPMO (with 1.5σ shift)

Formula reference — verify against the governing standard. Not a substitute for engineering judgment.