Statistical process control (SPC) is a tool based on mathematical and statistical principles. It keeps an already optimised process in its optimised state through continuous monitoring and targeted corrections. Various types of control chart are its main tools.
It is important to understand its role: SPC cannot directly improve a process through fundamental changes. The method is used in ongoing production, or series production, after the process settings have been established. Major changes to the process are no longer envisaged at this stage. Instead, the aim is to correct minor deviations and identify opportunities for lasting improvements.
Variation as the statistical basis
Statistical process control is based on the observation that differences in the quality characteristic under consideration occur both during the manufacture of components and during their subsequent measurement. The actual measured value therefore does not necessarily correspond to the intended target value, such as a dimension specified on a drawing. This deviation of a quality characteristic from its target value is known as variation.
This understanding is particularly important in metallography and materials testing: measured values are never absolutely constant but are subject to systematic and random influences. Statistical evaluation of this variation enables a reliable assessment of process stability.
Distinguishing random and systematic influences
Two types of influence can cause variation: random and systematic. Random influences consist of many small individual influences that are constantly present. They are stable over time and their combined effect is predictable, but they cannot be selectively controlled. This is known as natural variation.
Systematic influences, by contrast, can usually be traced to one or a few dominant causes. They occur irregularly and make the process unstable and unpredictable. Unlike natural variation, the causes of systematic influences can be identified and eliminated.
Statistical process control helps with this. Its main task is to monitor process variation continuously and distinguish clearly between random deviations and those caused by systematic influences.
Control charts as the central tool
To carry out statistical process control, samples are taken from the ongoing process, measured precisely and the results recorded on a control chart. These charts also include marked warning and action limits.
Important: these limits are not tolerances. They represent the statistical limits of natural process variation. A sample result outside the action limits is a clear indication of a systematic influence. Immediate intervention is then required to restore process stability.
Conclusion: from process monitoring to quality-led production
The aim of statistical process control is to maintain a statistically controlled process that is capable of delivering quality. Continuous monitoring using control charts, evaluation of measurement results and appropriate corrective measures form a closed control loop.
The manufacturing process is therefore not merely monitored, but actively controlled. The result is quality-focused assurance integrated into production, a key factor in reproducible results and sustained competitiveness in demanding technical environments.
