"Good/good judgment on the measured value."

The concept is simple and straightforward.
By sampling multiple normal values, a "normal range" is statistically determined. If it is within that range, it is normal; otherwise, it is abnormal.
If they are within that range, they are normal; if not, they are abnormal.
In FlexInspector, this is done for 2-dimensional arrays, but it is a concept that can be applied to "good/good judgment of measurement values" in general. I have not heard much about it...
I have not heard much about it, but I am sure there are many places where it could be used.