Control Phase: SPC, Standard Work, and Sustainment
What the exam tests
Control tests whether you can hold the gains: SPC chart selection, control limit calculation, out-of-control rules, standard work, control plans, and response plans.
Core rules
- Match the chart to the data: Xbar-R for subgrouped continuous data with small subgroups (n = 2-9), Xbar-S for larger subgroups, I-MR for individual measurements, p or np for defectives, and c or u for defects.
- Control limits come from process variation, usually plus or minus 3 standard errors. Specification limits come from the customer; never plot specifications as control limits.
- Out-of-control signals include a point beyond a limit, a run of seven on one side, six consecutive trending points, and other Western Electric or Nelson rules.
- A control plan names the metric, method, frequency, owner, and reaction when a signal occurs.
- Standard work documents the current best method with cycle time, work sequence, and required inventory; audits verify adherence, not just document existence.
Worked example
An Xbar-R chart uses subgroups of 5, Xbarbar = 20.0, Rbar = 0.5, and A2 = 0.577. The Xbar limits are 20.0 plus or minus 0.577 x 0.5 = 20.0 plus or minus 0.2885, giving UCL 20.29 and LCL 19.71. A subgroup mean of 20.6 is above the UCL, so the chart signals special cause. The reasoning: investigate that subgroup and the process condition at that time; do not recalculate limits to absorb the point.
Traps
Using specification limits as control limits; recalculating control limits after every signal; using an Xbar-R chart for defect counts; and writing a control plan with no named reaction owner.