In a traditional measurement systems study, the same part is measured by one operator multiple times and then the same part is measured by another operator multiple times. This is the “crossed” study that allows for the two components of measurement systems error to be estimated: repeatability and reproducibility. The repeatability comes from the same part being measured multiple times by an operator. The reproducibility comes from the same part being measured across operators.
But what happens when a part is either destroyed or measurably altered during the measurement process? Then the same part cannot be measured more than once. This presents a repeatability problem and completely rules out an estimate of reproducibility. This becomes a “nested” study and falls under the category of destructive testing.
Some examples of destructive testing situations are:
- Tensile Strength
- Bond Strength
- Weld Strength
- Crash Tests
- Explosive Tests
In this video, Vijay Sabale of Learn & Apply, shows how to conduct a destructive analysis in Minitab.
Vijay gives the following guidelines before giving an example and taking the viewer through the analysis.
- Data should be collected in a random order
- Select parts that represent the actual or expected range of process variation
- Part is nested in operator
- Operator and part must be random
You can watch Vijay’s video here.
To learn more about analysis in Minitab using an earlier version, check out the following link.