SPC in DMAIC Control: Chart Selection and Response
Control chart selection in the DMAIC Control phase — choosing the right chart for the Y, setting control limits, Western Electric rules, and the reaction plan. It sits in Section VIII (Control) of the 2022 CSSBB Body of Knowledge. For chart mechanics, see the CQE pages on control chart selection → and Western Electric run rules →
Choosing the chart
| Data | Subgroup | Chart |
|---|---|---|
| Variables (continuous) | n = 1 | I-MR (individuals and moving range) |
| Variables | n = 2 to about 9 | X̄-R |
| Variables | n ≥ 10 (or variable n) | X̄-S |
| Attribute: defectives (pass/fail units) | constant n | np chart (or p) |
| Attribute: defectives | variable n | p chart |
| Attribute: defects (counts) | constant inspection area | c chart |
| Attribute: defects | variable area / units | u chart |
| Small sustained shifts matter | any | CUSUM or EWMA |
Chart the key Xs (inputs) as well as the Y where you can: an input chart warns before the output goes bad.
Setting control limits
Control limits come from the process (voice of the process), never from the specification (voice of the customer). In Control, calculate them from data collected after the improvement, once the new process is stable; old baseline limits no longer describe it.
Western Electric rules (zone A = 2–3σ, B = 1–2σ, C = within 1σ of the centerline)
- One point beyond 3σ (outside a control limit).
- Two of three consecutive points beyond 2σ on the same side.
- Four of five consecutive points beyond 1σ on the same side.
- Eight consecutive points on the same side of the centerline.
Nelson's rules extend these (e.g. nine in a row on one side, six in a row steadily increasing or decreasing, fourteen alternating up and down). Each added rule raises the false-alarm rate, so use only the rules your control plan names.
Phase I and Phase II
In Phase I you collect about 20–25 subgroups from the improved process, compute trial limits, investigate and remove assignable causes, and recompute. In Phase II the limits are frozen and new points are plotted against them to detect change. The control plan hands Phase II monitoring to the process owner.
Reaction plan
Every chart in the control plan needs a reaction plan (sometimes called an out-of-control action plan, OCAP): who responds to a signal, what they check first, how product made since the last good point is contained, and when to escalate. A chart without a reaction plan is monitoring, not control.
Common exam traps
- In control ≠ capable. A stable process can still make out-of-spec product; capability is a separate check against specs.
- Don't recalculate limits after every signal. Recalculate only after a deliberate, verified process change.
- Don't adjust for common-cause variation (tampering); it increases variation.
- Rational subgroups: units within a subgroup should be made under the same conditions so that variation between subgroups shows up on the X̄ chart.
Try 5 free CSSBB practice questions on this topic. No account needed.
Take the free CSSBB diagnosticMore: DMAIC overview · Project charter · VOC & CTQ · SIPOC · MSA in DMAIC · Process capability · Data collection planning · Graphical analysis · Root cause analysis · Multi-vari analysis · Hypothesis testing · Regression · DOE · FMEA · Define tollgate · Change management · Financial benefits · Team dynamics · Improve tollgate · Piloting & scaling · Poka-yoke (BB) · Solution selection · Control plan · SPC in DMAIC · Sustaining gains
CSSBB Quest by Bizarre Act. Original study material; not affiliated with ASQ. CSSBB is a certification of ASQ. Terms · Privacy