These terms refer to the theories surrounding the variation exhibited within a process or system.
The terms ‘common’ (Deming) or ‘chance’ (Shewhart) causes represents variation that can be attributed to what are considered normally or naturally occurring, unknown random causes. Variation of this type is considered to be relatively constant, or ‘predictable’ within limits (aka control limits) and thus a given process or system exhibiting only common cause variation actually qualifies itself to be called a ‘controlled’ process or system. Common cause variation is the ‘long term’ rate of variation exhibited by a stable process or system. If the level of common cause variation is considered as ‘unacceptable’, the process or system itself must be changed in order to reduce such levels of variation.
However, not all variation can be attributed to common causes, thus giving rise to the notion of ‘assignable’ (Shewhart) or ‘special’ (Deming) causes of variation. Special cause variation is treated as an isolated incident and is typically investigated and corrected locally (i.e. a malfunctioning machine). Special cause variation is considered separately of and apart from common cause variation. Special cause variation is best thought of as a ‘short term’ spike or aberration (aka a control level threshold is exceeded) which requires immediate analysis and correction in order to return the process or system its’ stable or controlled state.
No process or system is considered perfect. A certain degree of variation (aka natural variation) should be expected in any process or system. Therefore, all processes and systems should include measurements to detect variation, practices to prevent variation and the means to deal with normally or naturally occurring common cause variation.
Although this brief explanation makes variation sound straightforward, determining the exact or root cause of any given variation is rarely a simple and straightforward investigation. Saving Changes...
Common cause variation and special cause variation are different concepts. Their origins and effects are different. Common causes are the usual, historical, quantifiable variation in a system, while "special causes" are unusual, not previously observed, non-quantifiable variation. Special causes are unanticipated, emergent or previously neglected phenomena within the system.