A Control Chart is a graph used to study how a process changes over time. It is a run chart with the addition of statistically calculated upper and lower control limits. Its primary purpose is to determine if a process is stable and in a state of statistical control, or if there is special cause variation that needs to be investigated.
Key Aspects of a Control Chart
- Statistical Control – It is the primary tool used to determine if a process is stable and predictable.
- Common vs. Special Cause Variation – It helps distinguish between normal, inherent process variation (common cause) and variation from specific, assignable events (special cause).
- Control Limits – These are the key feature, typically set at +/- 3 standard deviations from the mean. They represent the expected range of variation.
- Rule of Seven – A common guideline suggesting that if seven or more consecutive data points fall on one side of the mean, it indicates a non-random pattern that should be investigated.
How to Read a Control Chart
| Component | Description | Purpose |
|---|---|---|
| Center Line (Mean) | The average of the historical process data. | Represents the central tendency or average performance of the process. |
| Upper Control Limit (UCL) | A horizontal line plotted above the mean, typically at +3 standard deviations. | Defines the upper boundary of expected random variation. |
| Lower Control Limit (LCL) | A horizontal line plotted below the mean, typically at -3 standard deviations. | Defines the lower boundary of expected random variation. |
| Data Points | Individual measurements plotted in sequence. | Show the actual performance of the process over time. |
A process is considered “out of control” if a data point falls outside the control limits or if non-random patterns (like the Rule of Seven) are observed.
Example Scenarios
Manufacturing
A factory uses a control chart to monitor the diameter of a manufactured bolt. If a data point falls above the UCL, it might indicate a machine is out of calibration (a special cause), prompting an immediate investigation.
IT Service Desk
An IT department tracks the time it takes to resolve support tickets. A control chart helps them see if their resolution time is stable and predictable. If a new point falls below the LCL (meaning an unusually fast resolution), they might investigate to see if a new, positive technique was used that can be replicated.
Why Control Charts Matter
- Prevents Overreaction – They stop managers from reacting to normal, random process fluctuations (common cause variation) as if they were real problems.
- Provides Process Stability Insights – It is the best tool to know if your process is stable and predictable, which is a prerequisite for any meaningful process improvement.
- Identifies Problems Objectively – It uses statistical data to signal when a problem has occurred, removing guesswork and opinion.
- Drives Data-Driven Decisions – It provides a clear, visual basis for making decisions about when to intervene in a process and when to leave it alone.
See also: Run Chart, Seven Basic Quality Tools, Quality Management, Statistical Process Control (SPC).