A rational subgroup is collected close together in time, from one tool, chamber, spindle or cavity, so the within-subgroup spread captures only common cause noise and any special cause shows up between subgroups. Get it wrong, for example by pooling wafers from four chambers into one subgroup, and the within-subgroup range swells, the X-bar limits balloon and real shifts disappear inside them. Subgrouping is the biggest single design decision in setting up a variables chart.
Those five sites aren't a rational subgroup, center and edge differ by design. Chart the wafer mean as an individual and track uniformity separately.
Also heard as
- subgroup
- rational subgrouping
Related terms
X-bar and R chart
Paired control charts for subgroup averages and subgroup ranges, the workhorse for measured data sampled in small groups.
Control limits
Boundaries on a control chart computed from the process's own variation, marking how far routine noise normally reaches.
Hugging the centerline
Chart pattern where points cluster unnaturally close to the centerline, usually a sign the limits are wrong.
Common cause
Background variation built into a stable process from many small sources, present all the time.