You can save process capability indices in an OUTLIMITS= data set if you provide specification limits with the LSL= and USL= options. This is illustrated by the following statements:
title 'Control Limits and Capability Indices'; proc shewhart data=Partgaps; xchart Partgap*Sample / outlimits = Gaplim2 usl = 270 lsl = 240 nochart; run;
The data set Gaplim2
is listed in Output 18.37.1.
Output 18.37.1: Data Set Gaplim2
Containing Control Limit Information
Control Limits with Capability Indices for Gap Width Measurements |
_VAR_ | _SUBGRP_ | _TYPE_ | _LIMITN_ | _ALPHA_ | _SIGMAS_ | _LCLX_ | _MEAN_ | _UCLX_ | _LCLR_ | _R_ | _UCLR_ | _STDDEV_ | _LSL_ | _USL_ | _CP_ | _CPL_ | _CPU_ | _CPK_ |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Partgap | Sample | ESTIMATE | 5 | .002699796 | 3 | 242.087 | 259.667 | 277.246 | 0 | 30.4762 | 64.4419 | 13.1028 | 240 | 270 | 0.38160 | 0.50032 | 0.26288 | 0.26288 |
The variables _CP_
, _CPL_
, _CPU_
, and _CPK_
contain the process capability indices. It is reasonable to compute capability indices in this case, because FigureĀ 18.97 indicates that the process is in statistical control. For more information, see the section OUTLIMITS= Data Set.