If your model is not linked to a diagnostic study, this dialog can be used to review and manually assign tests to test programs. If you select “All Tests Available for Diagnostics” in the “Fill With” dropbox at the top, the dialog will list all tests that are defined in test sets whose usage setting indicates that they are available for diagnostics. The tests are sorted alphabetically by name in the column on the left. In the column on the right, there are dropboxes allowing you to manually select the test program for each test. Although this is nice, the real power of this dialog is unleashed only after you have calculated diagnostics. When a calculated diagnostic study is loaded into memory for the active model, then the Test Program Optimization dialog looks quite different:
By default, when a design is linked to a diagnostic study, the Test Program Optimization dialog will be filled with all tests that are included in the calculated diagnostics, sorted by how likely each test is to be used to diagnose a failure. Before invoking this dialog, develop a diagnostic sequence that represents your preferred order of testing. Then, on this dialog, the Test limits that appear in the Probability column will indicate the relative likelihood that each test will be used to diagnose a failure. In the example above, the top Test limit is 89.91—this is assigned to the test which will always be performed first (its probability is not 100.0 because 11.09 percent of the expected failures are non-detectable). To assign tests to a particular test program, click on the “Assign” button in the lower left corner of the Test Program Optimization dialog. The following assignment dialog will then appear:
