Example 13.8 Applying Tests for Special Causes

[See SHWCEX1 in the SAS/QC Sample Library]This example illustrates how you can apply tests for special causes to make charts more sensitive to special causes of variation. Twenty trucks of the same model are inspected, and the number of paint defects per truck is recorded. The following statements create a SAS data set named Trucks3:

data Trucks3;
   input TruckID $ Defects @@;
   label TruckID='Truck Identification Number'
         Defects='Number of Paint Defects';
   datalines;
B1  12    B2  4    B3  4    B4   3
B5   4    D1  2    D2  3    D3   3
D4   2    D9  4    M2  9    M6  13
L3   5    L4  4    L7  6    Z1  15
Z2   8    Z3  9    Z7  6    Z9   8
;
run;

The following statements create a chart and tabulate the information on the chart. The chart and table are shown in Output 13.8.1 and Output 13.8.2.

ods graphics on;
title1 'c Chart for Paint Defects in New Trucks';
title2 'Tests=1 to 4';
proc shewhart data=Trucks3;
   cchart Defects*TruckID / tests      = 1 to 4
                            testlabel1 = 'Test=1 Signaled'
                            testlabel2 = 'Test=2 signaled'
                            odstitle   = title
                            zonelabels
                            tabletests
                            tablelegend;
run;
ods graphics off;

The TESTS= option requests Tests 1, 2, 3, and 4, which are described in Tests for Special Causes Only Tests 1, 2, 3, and 4 are recommended for charts. The TESTLABEL1= and TESTLABEL2= options specify the labels for points where Tests 1 and 2 are positive. The TESTFONT= option specifies the font for the labels indicating points at which the tests are positive.

Output 13.8.1 Tests for Special Causes Displayed on Chart

Output 13.8.2 Tabular Form of Chart

c Chart for Paint Defects in New Trucks

Tests=1 to 4

The SHEWHART Procedure