Destructive tests are tests that damage or destroy the part that is tested. Because destructive tests change the part, or even destroy it, replication is not possible. For example, automotive crash tests are destructive tests because the vehicle is destroyed beyond testing again.
Conversely, non-destructive tests do not damage the part. Thus, with non-destructive tests, operators can make repeated measurements on the same part.
If you use destructive testing during a gage R&R study, you must be able to assume that all parts within a single batch are identical enough to claim that they are the same part. If you are unable to make that assumption then part-to-part variation within a batch will mask the measurement system variation.
If you can make that assumption, then your choice of using a crossed or a nested study for destructive testing depends on your measurement process. If all operators measure parts from each batch, use a crossed or expanded study. If each batch is measured by only one operator, use a nested or expanded study. Whenever operators measure unique parts, you have a nested design.
If you need to include more factors than Operator and Part, have fixed factors, a mixture of crossed and nested factors, or an unbalanced design, you must use an expanded study.
For more information on crossed, nested and expanded studies, go to What is a gage R&R study?