Control Chart Maker

Paste measurements or defect counts to draw X-bar & R, X-bar & S, I-MR, p, np, c and u control charts with the centre line, UCL, LCL and out-of-control points. Use it in SPC to check whether a process is stable.

How to use

1. Choose the "Chart type": "Measured values" for measurements, "Counts" for defectives or defects.
2. Paste values from a spreadsheet into "Data". For X-bar charts each row is one subgroup; for p and u charts, give the count and the sample size in two columns.
3. If needed, turn on rules 2 to 8 under "Out-of-control rules", or use "Fix the limits from the first subgroups" or "Show specification limits".
4. Check "Centre line and control limits" and "Out-of-control points", and export with "Save PNG", "Save SVG" or "Copy limits".

How it works

• Limits are the centre line ± 3 standard deviations of the plotted statistic, from the standard constants (A2, D3, D4, A3, B3, B4 and others). A lower limit below zero is shown as 0.
• With unequal sample sizes, p and u charts get stepped limits.
• Rule 1 (a point beyond 3σ) is always on; rule numbers follow Nelson. Rules 5 to 8 apply only to X-bar and individuals charts.

Questions and answers

Which control chart should I use?

For measured values, use X-bar & R (subgroups of 2 to 10) or X-bar & S, and I-MR for single values. Use p or np for defective items, c or u for defects.

How are UCL and LCL calculated?

On an X-bar & R chart, the X-bar limits are the grand mean ± A2 × R-bar, and the R limits are D4 × R-bar and D3 × R-bar. For subgroups of 5, A2 is 0.577.

Does an out-of-control point mean the process has a problem?

It means there is a reason to look for a cause at that time. A signal alone does not prove that the process has changed.

What is the difference between control limits and specification limits?

Control limits show how the process behaves; specification limits are what the product must meet. An X-bar point is an average, so it can be inside the specification while single items are not.