In eXpress, test weighting definitions are comprised of five individual parameters that collectively define the way in which the weighting is to be calculated. By manipulating these parameters (using the Weighting Parameter Dialog), the Analyst can specify that a wide variety of methods for optimizing test selection in the diagnostics. The five parameters are:
Weighting 1: for each test, a Test limit shall be calculated by summing the failure probability for all suspect functions for which a malfunction can be observed using that test. The highest weighting will be assigned to the test with the highest calculated Test limit. Normalized weighting Test limits will be multipled by 100 as they are combined with other test weightings.
Weighting 2: for each test, a Test limit shall be calculated by counting the number of suspect functions for which a malfunction can be observed using that test. The highest weighting will be assigned to the test with the highest calculated Test limit. Normalized weighting Test limits will be multipled by 50 as they are combined with other test weightings.
The net results of these two test weightings is that diagnostics will perform first those tests that are most likely to detect a failure within the set of functions currently under suspicion. As a secondary criteria (that will mostly take effect when two tests are almost equally likely to fail), diagnostics will first perform the test that can detect failures within the largest number of functions currently under suspicion.
