Cookie settings

We use several types of cookies on this website to provide you with an optimal online experience, to increase the user-friendliness of our portal and to constantly improve our communication with you. You can decide which categories you want to allow and which you do not want to allow (see "Custom settings" for more information).
Name Usage Duration
privacylayerStatus Agreement Cookie hint1 year
Name Usage Duration
_gaGoogle Analytics2 years
_gidGoogle Analytics1 day
_gatGoogle Analytics1 minute
_galiGoogle Analytics30 seconds



What are the meaning and purpose of capability indices?

08 April 2013: Stephan Conrad

Principle of maximum information

While developing the ISO 22514 series of standards, it is about time to think about the meaning and purpose of a capability index once again.

You often face capability indices for the first time when evaluating the number of defective parts. You can find countless tables in technical literature specifying fractions nonconforming expected for each specific capability index. This mainly superseded calculation method based on fractions nonconforming is still contained in ISO/TR 22514-4; however, only with reasonable restrictions. Almost 15 years ago, this method was even a standard in some parts of the automotive industry.

But why using quite an abstract capability index instead of “ppm”, many people asked themselves soon though. In case of capability indices exceeding 1,67 considerably, estimating fractions nonconforming is rather comparable to gazing into a crystal ball. The benefit was highly doubtful indeed. This capability index only provides information about the process output but does not indicate anything about the properties of the process or the process behavior. Moreover, completely different process qualities and process structures might lead to absolutely identical results (see figure 1). Consequently, even this capability index did not help to discover the factors to be optimized in order to reach full potential.

Other calculation methods calculate the capability index from distribution parameters and provide considerably more information...