As with severity and occurrence ratings, you can override the detection rating mapping mechanism using a checklist attribute. A detection map attribute must be set up as follows:
|
It must be named “DetectionMap”
· It must be of type Checklist
· It must be associated with the Design
· It must be “Mutually Exclusive”
· It must have values assigned
· It must have one entry for each desired Detection rating (on a 10-point scale) to which detection percentages will be mapped
· For each entry, assign the minimum detection percentage (from 0 to 100) that is to be mapped to that entry
|
This detection map will result in the detection rankings 1, 5, 8 & 10, with the numbers intentionally skewed to emphasize non-detectable failures (75% detectability is only mapped to a 5, for instance). Using this detection map, the detection ratings in the calculated FMECA would appear as follows:
Notice that the partially-detected failure is now mapped to a detection rating of 8 (rather than 6).
The second way of calculating a RPN detection rating in eXpress is the “rating by attribute” approach. Use this approach when the detection rating should be based (at least in part) on the category of test (e.g., BIT class) that detects the failure.
|
This approach requires two checklist attributes—one per test to specify the category, and at the design level to specify how the different test attributes (categories) map to detection ratings.
In this example, three BIT types have been defined for tests—IBIT, PBIT & SBIT. The assigned values are not necessary; they are used to control the order in which the values are listed when tests are included in multiple categories (the highest value is listed first).
|
Create an attribute named “DetectionRating” and select Integer as the attribute type. Set up the attribute so that it is only associated with Failure Effects.
Use this attribute to separately define the value to be used for the RPN detection rating for each failure in the FMECA.
Attribute values must be assigned to object failure effects (not design effects). Detection ratings will thus be associated with the failure for each line in the FMECA, rather than the end item effects.
Although this approach allows you to assign any detection rating to any failure, it does not validate the entries. It is up to the analyst to ensure that the assigned detection ratings are valid and within range.
When included in the FMECA, the detection rating attribute will be used (instead of the built-in detection rating calculation) when calculating the Risk Priority Number.
In the example FMECA setup dialog picture at right, the “DetectionRating” attribute is being used to supply the detection rating.
Notice that a double backslash has been inserted between the words “Occurrence” and “Rating” in the name field at the bottom of the dialog. This will cause the two words to appear on separate lines in the FMECA header. Alternatively, you can insert a space to see the two words on a single line in the header.