Describe › Standalone plots

Control chart

A Shewhart control chart plots measurements against run order with a centre line at the mean and warning (±2σ) and action (±3σ) limits.

What is Control chart?

Control charts are the foundation of statistical process control. The centre line is the in-control mean; the ±2σ warning limits flag points that warrant attention (about 5% expected by chance); the ±3σ action limits flag points where the process is almost certainly out of control (~0.3% expected by chance).

Western Electric rules extend the simple ±3σ test with patterns (8 points on the same side, 2 of 3 beyond ±2σ, etc.) — not implemented here, but the chart shows enough that you can spot them visually.

When should I use Control chart?

  • Method validation, calibration drift detection, repeated QC sample monitoring.
  • Process-stability assessment over time.

What data does it need?

One numeric measurement + optional run-order column.

What does it report?

Line chart with centre + warning + action limits + count of out-of-control points.

How do I interpret the result?

A single point outside ±3σ is usually enough to investigate. A run of 7+ points on one side of the centre line or a monotone trend of 7+ points are early-warning patterns that suggest a slow drift even when no single point exceeds the action limits.

See also

References

  • Shewhart (1931). Economic Control of Quality of Manufactured Product.