Guides
The questions people type into a search box at four in the afternoon with a number in front of them they do not trust. Each one answers the question first, says what the answer cannot tell you, and ends at a free calculator that runs it on your readings.
What is Cpk?
Cpk is the distance from your process mean to the nearer specification limit, measured in units of three sigma. Here is what it means, what a good value is, and the three ways it lies to you.
Read → CapabilityCpk vs Ppk — what the gap between them is telling you
Cpk uses within-subgroup sigma, Ppk uses the standard deviation of every reading. When they disagree, the disagreement is the finding — not a rounding artefact to be argued away.
Read → VariationCommon cause and special cause variation
Every process varies. The question is whether the variation is the process being itself or something happening to it — and answering it wrong is how well-meant intervention makes a process worse.
Read → Control chartsWhat is a Shewhart chart?
Shewhart chart is the proper name for what most people call a control chart. Here is where it came from, why the limits are at three sigma, and which of the family you need.
Read → Quality toolsHow to make a Pareto chart in Excel
Excel has had a built-in Pareto chart since 2016, and it is fine until you need it to behave. Here is the built-in route, the manual build, and the two mistakes that send an improvement team at the wrong defect.
Read → Comparing processesBox and whisker plot in Excel
Excel draws box plots natively, and the box it draws may not match the one your customer drew from the same numbers. The reason is the quartile definition, and there are three of them in common use.
Read → DistributionsNormal distribution curve in Excel
NORM.DIST draws the curve, and the scaling step is the one everybody misses. Then the harder question — a curve over a histogram is a picture, not a test of normality, and normality matters far more for Cpk than it does for a control chart.
Read →Prefer to skip the reading? The free tools compute all of it on your own data, with no signup and nothing stored.