Definition
A control chart is the basic tool of statistical process control (SPC): it plots measurements over time with a centre line and ±3σ control limits derived from the natural variability of the process. Points beyond the limits or non-random patterns (trends, runs) signal a special cause that must be found and removed — before defects are produced.
Quality control & SPC
For continuous measurements the X-bar and R (or S) pair is used for subgroups and I-MR for individual measurements; for attributes the p/np chart (proportion nonconforming) and c/u chart (number of defects). Control limits are not tolerances — they describe what the process can do, not what the customer wants from it.
Interpretation follows rules (Western Electric, Nelson): one point beyond a limit, 8 points on one side, 6 points monotonically. For slow mean shifts CUSUM and EWMA charts are more sensitive; for several correlated characteristics multivariate charts (Hotelling T²).
In Statistica
Statistics → Industrial Statistics & Six Sigma → Quality Control Charts offers all Shewhart charts for variables and attributes, CUSUM, EWMA, MA and multivariate charts, with run tests, process phases, alarms and connection to live data. Charts can update automatically in the Statistica Enterprise environment.
Related terms
- Process capability (Cp, Cpk)Capability indices compare process variability with the customer's tolerance limits.
- Standard deviation and varianceVariance is the average squared deviation of values from the mean; the standard deviation (SD) is its square root and has the same units as the data.
- Normal distributionThe normal (Gaussian) distribution is a symmetric bell-shaped distribution described by its mean and standard deviation.
- Six SigmaSix Sigma is a data- and statistics-driven methodology for process improvement.
- Pareto chartA Pareto chart is a bar chart of causes or defect categories sorted from most to least frequent, with a cumulative-share curve.
Knowledgebase guides
FAQ
- What is the difference between control limits and tolerances?
- Control limits come from process variability (voice of the process), tolerances from customer requirements (voice of the customer). Capability indices compare the two.
- How many subgroups do I need to set up a chart?
- Usually 20–25 subgroups in a stable phase (Phase I) from which the limits are computed; only then is the chart used for monitoring (Phase II).
Try it on your own data
Statistica free for 30 days
Full version, no credit card. Or get a pricing estimate in a minute.
Quality control & SPC
Updated: September 2026.