Saving Control Limits

[See SHWMR1 in the SAS/QC Sample Library]You can save the control limits for median and range charts in a SAS data set; this enables you to apply the control limits to future data (see Reading Preestablished Control Limits) or modify the limits with a DATA step program.

The following statements read measurements from the data set Detergent (see Creating Charts for Medians and Ranges from Raw Data) and save the control limits displayed in Figure 15.45 in a data set named Detlim:

proc shewhart data=Detergent;
   mrchart Weight*Lot / outlimits=Detlim
                        nochart;
run;

The OUTLIMITS= option names the data set containing the control limits, and the NOCHART option suppresses the display of the charts. The data set Detlim is listed in Figure 15.49.

Figure 15.49 The Data Set Detlim Containing Control Limit Information
Control Limits for Detergent Box Weights

_VAR_ _SUBGRP_ _TYPE_ _LIMITN_ _ALPHA_ _SIGMAS_ _LCLM_ _MEAN_ _UCLM_ _LCLR_ _R_ _UCLR_ _STDDEV_
Weight Lot ESTIMATE 5 .002909021 3 20.7554 24.4996 28.2439 0 5.42036 11.4613 2.33041

The data set Detlim contains one observation with the limits for process Weight. The variables _LCLM_ and _UCLM_ contain the control limits for the medians, and the variable _MEAN_ contains the central line. The variables _LCLR_ and _UCLR_ contain the control limits for the ranges, and the variable _R_ contains the central line. The values of _MEAN_ and _STDDEV_ are estimates of the process mean and process standard deviation . The value of _LIMITN_ is the nominal sample size associated with the control limits, and the value of _SIGMAS_ is the multiple of associated with the control limits. The variables _VAR_ and _SUBGRP_ are bookkeeping variables that save the process and subgroup-variable. The variable _TYPE_ is a bookkeeping variable that indicates whether the values of _MEAN_ and _STDDEV_ are estimates or standard values. For more information, see OUTLIMITS= Data Set.

You can create an output data set containing both control limits and summary statistics with the OUTTABLE= option, as illustrated by the following statements:

proc shewhart data=Detergent;
   mrchart Weight*Lot / outtable=Dtable
                        nochart;
run;

This data set contains one observation for each subgroup sample. The variables _SUBMED_, _SUBR_, and _SUBN_ contain the subgroup medians, subgroup ranges, and subgroup sample sizes. The variables _LCLM_ and _UCLM_ contain the control limits for the median chart, and the variables _LCLR_ and _UCLR_ contain the control limits for the range chart. The variable _MEAN_ contains the central line for the median chart, and the variable _R_ contains the central line for the range chart. The variables _VAR_ and Batch contain the process name and values of the subgroup-variable, respectively. For more information, see OUTTABLE= Data Set.

The data set Dtable is listed in Figure 15.50.

Figure 15.50 The Data Set Dtable
Summary Statistics and Control Limit Information

_VAR_ Lot _SIGMAS_ _LIMITN_ _SUBN_ _LCLM_ _SUBMED_ _MEAN_ _UCLM_ _STDDEV_ _EXLIM_ _LCLR_ _SUBR_ _R_ _UCLR_ _EXLIMR_
Weight 1 3 5 5 20.7554 22.56 24.4996 28.2439 2.33041   0 9.54 5.42036 11.4613  
Weight 2 3 5 5 20.7554 23.54 24.4996 28.2439 2.33041   0 3.07 5.42036 11.4613  
Weight 3 3 5 5 20.7554 24.35 24.4996 28.2439 2.33041   0 4.65 5.42036 11.4613  
Weight 4 3 5 5 20.7554 25.52 24.4996 28.2439 2.33041   0 5.05 5.42036 11.4613  
Weight 5 3 5 5 20.7554 23.25 24.4996 28.2439 2.33041   0 8.04 5.42036 11.4613  
Weight 6 3 5 5 20.7554 23.01 24.4996 28.2439 2.33041   0 6.33 5.42036 11.4613  
Weight 7 3 5 5 20.7554 24.19 24.4996 28.2439 2.33041   0 2.19 5.42036 11.4613  
Weight 8 3 5 5 20.7554 26.27 24.4996 28.2439 2.33041   0 2.18 5.42036 11.4613  
Weight 9 3 5 5 20.7554 22.31 24.4996 28.2439 2.33041   0 6.08 5.42036 11.4613  
Weight 10 3 5 5 20.7554 22.66 24.4996 28.2439 2.33041   0 8.21 5.42036 11.4613  
Weight 11 3 5 5 20.7554 26.38 24.4996 28.2439 2.33041   0 4.97 5.42036 11.4613  
Weight 12 3 5 5 20.7554 23.01 24.4996 28.2439 2.33041   0 5.05 5.42036 11.4613  
Weight 13 3 5 5 20.7554 23.15 24.4996 28.2439 2.33041   0 5.37 5.42036 11.4613  
Weight 14 3 5 5 20.7554 24.73 24.4996 28.2439 2.33041   0 4.95 5.42036 11.4613  
Weight 15 3 5 5 20.7554 25.86 24.4996 28.2439 2.33041   0 5.58 5.42036 11.4613  
Weight 16 3 5 5 20.7554 23.99 24.4996 28.2439 2.33041   0 4.35 5.42036 11.4613  
Weight 17 3 5 5 20.7554 24.30 24.4996 28.2439 2.33041   0 5.24 5.42036 11.4613  
Weight 18 3 5 5 20.7554 24.88 24.4996 28.2439 2.33041   0 4.87 5.42036 11.4613  
Weight 19 3 5 5 20.7554 25.92 24.4996 28.2439 2.33041   0 7.22 5.42036 11.4613  
Weight 20 3 5 5 20.7554 25.63 24.4996 28.2439 2.33041   0 6.30 5.42036 11.4613  
Weight 21 3 5 5 20.7554 25.17 24.4996 28.2439 2.33041   0 6.52 5.42036 11.4613  
Weight 22 3 5 5 20.7554 26.72 24.4996 28.2439 2.33041   0 5.18 5.42036 11.4613  
Weight 23 3 5 5 20.7554 23.67 24.4996 28.2439 2.33041   0 9.17 5.42036 11.4613  
Weight 24 3 5 5 20.7554 24.92 24.4996 28.2439 2.33041   0 4.62 5.42036 11.4613  
Weight 25 3 5 5 20.7554 24.51 24.4996 28.2439 2.33041   0 5.40 5.42036 11.4613  
Weight 26 3 5 5 20.7554 24.75 24.4996 28.2439 2.33041   0 3.41 5.42036 11.4613  
Weight 27 3 5 5 20.7554 25.10 24.4996 28.2439 2.33041   0 4.76 5.42036 11.4613  
Weight 28 3 5 5 20.7554 25.64 24.4996 28.2439 2.33041   0 3.47 5.42036 11.4613  

An OUTTABLE= data set can be read later as a TABLE= data set. For example, the following statements read Dtable and display charts (not shown here) identical to those in Figure 15.45:

title 'Median and Range Charts for Detergent Box Weight';
proc shewhart table=Dtable;
   mrchart Weight*Lot;
run;

Because the SHEWHART procedure simply displays the information in a TABLE= data set, you can use TABLE= data sets to create specialized control charts (see Specialized Control Charts: SHEWHART Procedure). For more information, see TABLE= Data Set.