Control Chart

Control charts are the foundational tool of statistical process control. They display individual measurements or subgroup statistics (averages, ranges, standard deviations) plotted sequentially over time. The center line represents the process mean, and the upper and lower control limits (UCL and LCL) are calculated from the data, typically at ±3 standard deviations from the mean. This corresponds to 99.73% of normal variation, so points outside these limits are very likely caused by something other than random variation.

Control charts detect special-cause variation through several rules: points beyond the control limits, runs of consecutive points on one side of the center line, trends (consistently increasing or decreasing points), and other non-random patterns. When a special cause is detected, the process should be investigated and the root cause identified and corrected. Common chart types include X-bar and R (subgroup averages and ranges), X-bar and S (subgroup averages and standard deviations), Individuals and Moving Range (I-MR), and attribute charts (p, np, c, u).

For calibration management, control charts are valuable for monitoring intermediate check data on reference standards, tracking as-found calibration results over time, and monitoring environmental conditions in calibration laboratories. A control chart of as-found bias at each calibration event can reveal drift trends and predict when an instrument will approach its tolerance limits, supporting proactive calibration interval adjustments. Control charts of laboratory temperature and humidity help ensure environmental requirements are consistently met.