Blog

Insights on statistical process control, quality engineering, and data-driven manufacturing

All alarm-fatigue anderson-darling-test apqp assumption-free-statistics cluster-detection confidence-intervals control-charts cpk data-driven-manufacturing distribution-analysis distribution-fitting dpmo egdf eldf entropy-methods entropy-statistics first-pass-yield homogeneity-testing iatf-16949 lognormal-distribution measurement-system-analysis non-normal-data normal-distribution ppap process-capability process-drift process-monitoring process-stability quality-4-0 quality-control quality-standards sample-size-determination sigma-level six-sigma small-sample-statistics statistical-concepts statistical-process-control subgroup-analysis weibull-distribution yield-analysis

The Distribution Fitting Trap: Weibull, Lognormal, or None of the Above?

Distribution fitting replaces the normality assumption with a different guess. With typical sample sizes, Weibull, lognormal, and gamma all pass goodness-of-fit tests — giving different Cpk values. The distribution fitting step that should fix your analysis becomes its own error source.

weibull-distribution lognormal-distribution distribution-fitting anderson-darling-test distribution-analysis

Mar 13, 2026·

Alex Dyba