When the practitioner samples, the design must reflect what the test is for and how the compliance activity behaves, the sample must be large enough to bring sampling risk down to a low level and chosen so that each unit could be picked and the result represents the population, and any selected item on which the procedures cannot be applied is treated as a deviation.
The graph holds this control, the 0 it maps to, and the evidence behind each claim, over MCP and REST.