Use Nonnormal Capability Analysis for Multiple Variables to compare the capability of multiple process variables, or multiple groups within one process variable, based on a nonnormal distribution. Using this analysis, you can do the following:
To perform the analysis, you must specify a lower or upper specification limit (or both) to define your process requirements. You must also select a nonnormal distribution to fit your data. The analysis evaluates the spread of the process data in relation to the specification limits. When a process is capable, the process spread is small relative to the specification spread. This analysis can also indicate whether your process is on target. In addition, it estimates the proportion of product that does not meet specifications. However, it does not provide estimates of within-subgroup variation or potential capability.
For example, a manufacturer uses two machines to package frozen food. The specifications for the weight of the frozen food packages are 31 ± 4 oz. The manager compares the process capability of each machine by fitting the largest extreme value distribution to the nonnormal data.
To perform nonnormal capability analysis for multiple variables, choose .
If you do not know which nonnormal distribution best fits your data, use Individual Distribution Identification before you perform this analysis.
If you do not know whether your process data are in control or whether they can be evaluated using a specific nonnormal distribution, use Nonnormal Capability Sixpack to evaluate these requirements before you perform this analysis.
If want to obtain estimates of within-subgroup variation and potential capability, in addition to overall capability, use Normal Capability Analysis for Multiple Variables to transform your nonnormal data and analyze process capability based on a normal distribution.