Enter your data for Nonnormal Capability Analysis for Multiple Variables

Stat > Quality Tools > Capability Analysis > Multiple Variables (Nonnormal)

In This Topic

Enter your data
Fit distribution
Specification limits

Enter your data

Use one of the following procedures to enter data for the analysis, depending on whether you have a grouping variable.

Process data without a grouping (by) variable

If you have multiple columns of process data, without a grouping variable, do the following.

In Variables, enter the columns of numeric data that you want to analyze.

In this worksheet, each column contains the diameter of piston rings for a different shift. To perform a capability analysis for each shift, enter Diameter Shift 1 and Diameter Shift 2 in Variables.

C1	C2
Diameter Shift 1	Diameter Shift 2
74.030	73.992
74.002	74.014
74.019	73.887

Process data with a grouping (by) variable

If you have only one column of process data, and another column to indicate grouping information, do the following:

In Variables, enter the column of numeric data that you want to analyze.
Select By variables, and then enter the grouping column or columns to perform a separate capability analysis for each unique value in the column or each unique combination of values in the columns.

In this worksheet, Diameter contains the diameter of piston rings for both shifts. Shift indicates the grouping variable for the numeric data in Diameter. To perform capability analysis for each shift, enter Diameter in Variables, and enter Shift in By variables.

C1	C2
Diameter	Shift
74.030	1
74.002	1
74.019	1
73.992	2
74.014	2
73.887	2

Fit distribution

From the drop-down list, select a nonnormal distribution to fit your data. To produce a reliable estimate of process capability, the data must follow the distribution that you select. If you are unsure which distribution best fits your data, use Individual Distribution Identification.

For more information on choosing an appropriate distribution for nonnormal data, go to Capability analyses with nonnormal data.

Specification limits

Important

To perform the analysis, you must enter a lower specification limit, an upper specification limit, or both.

Lower spec

To specify the minimum acceptable value for the product or service, enter one value, multiple values, or a column of values. If you enter one value, it applies to all variables or groups. If you enter multiple values or a column of values, the values correspond sequentially to the columns that you entered in Variables or to the groups in the By variables column. If you do not have a lower specification limit, enter *.

Boundary

Select to define a lower specification limit as a boundary only if it is not possible for measurements to fall below the Lower spec value. If a boundary applies to the lower specification limit for some variables or groups, but not others, enter a value of 0 or 1 for each group or variable in the analysis. 1 indicates that the lower spec is a boundary for that variable or group, and 0 indicates that the lower spec is not a boundary for that variable or group.

You can enter one value, multiple values, or a column of values. If you enter one value, it applies to all variables or groups. If you enter multiple values or a column of values, the values correspond sequentially to the columns that you entered in Variables or to the groups in the By variables column.

For example, suppose you enter C1 C2 in Variables, and 3 5 in Lower spec. If 3 is a boundary for C1, but 5 is not a boundary for C2, select Boundary and enter 1 0. The 1 is for C1, and the 0 is for C2.

Upper spec

To specify the maximum acceptable value for the product or service, enter one value, multiple values, or a column of values. If you enter one value, it applies to all variables or groups. If you enter multiple values or a column of values, the values correspond sequentially to the columns that you entered in Variables or to the groups in the By variables column. If you do not have an upper specification limit, enter *.

Boundary

Select to define an upper specification limit as a boundary only if it is not possible for measurements to fall above the Upper spec value. If a boundary applies to the upper specification limit for some variables or groups, but not others, enter a value of 0 or 1 for each group or variable in the analysis. 1 indicates that the upper spec is a boundary for that variable or group, and 0 indicates that the upper spec is not a boundary for that variable or group.

For example, suppose you enter C1 C2 in Variables, and 10 8 in Upper spec. If 10 is a boundary for C1, but 8 is not a boundary for C2, select Boundary and enter 1 0. The 1 is for C1, and the 0 is for C2.

Note

When you define a specification limit as a boundary, Minitab reports the expected capability indices related to that specification limit as missing values (*). Therefore, define a limit as a boundary only if it is impossible for measurements to fall beyond the limit. For example, an upper specification limit of 100% purity is a boundary because exceeding 100% purity is impossible. A lower specification limit of 0% purity is a boundary because falling below 0% purity is impossible.