Control Chart
A diagram for ongoing process monitoring based on samples taken over time, the core tool of statistical process control.
The control chart belongs to statistical process control (SPC) and is used to detect shifts or instability in a manufacturing or service process early, before scrap or rework results from it.
Its structure follows a fixed principle: samples are taken at regular intervals, a statistic such as the mean or range is calculated, and plotted as a point on the time axis. The chart includes a center line as well as upper and lower control limits, typically set three standard deviations from the mean, sometimes supplemented by tighter warning limits.
As long as the plotted points scatter randomly within the limits, the process is considered in control, showing only normal, random variation. If a point crosses a control limit or a noticeable pattern appears, such as a sustained trend in one direction, that points to a systematic, avoidable cause that needs to be investigated.
The practical benefit lies in early intervention: deviations become visible before defective parts are produced. This reduces rework and scrap and provides an objective basis for decisions instead of relying on gut feeling.
Practical Example
For a turned shaft with a target diameter of 20 mm, a sample of 5 parts is taken every 30 minutes. After several hours, the mean chart shows a slow trend toward the upper control limit. The team recognizes increasing tool wear in time and replaces the tool before any parts are produced outside tolerance.
How Leanshift Helps
Making problems visible before they become expensive is exactly the mindset behind the control chart, and it matches how Leanshift, as a development workshop, builds improvement on data instead of guesswork.
Frequently Asked Questions
What is the difference between random and systematic variation?
Random variation is normal, unavoidable noise within the control limits. Systematic variation has a specific, usually avoidable cause, such as tool wear or an incorrect machine setting, and requires intervention.
How often are samples taken for a control chart?
It depends on the process. Common intervals range from several samples per hour for critical characteristics to one sample per shift for stable processes.
What happens when a point crosses a control limit?
The process is considered out of control. An immediate root-cause investigation follows before production continues, often supported by tools such as the Ishikawa diagram or 5-Why.