Lean Six Sigma Control Phase
The fifth phase of DMAIC: making the improvement hold after the project team has gone
Control answers the question every sponsor eventually asks: will this still be true in a year? It hands the improved process back to the people who run it, with the means to keep it there — visual and physical controls from Lean, statistical process control to detect drift early, and a control plan that says who does what when the process moves.
It is the phase most often shortened under time pressure, and the reason so many documented improvements quietly disappear. Below is what Control covers in the Lean Six Sigma syllabus, section by section.
📌 Quick navigation
What you actually do in Control
| Step | Output |
|---|---|
| Make the right way the easy way | 5S controls, Kanban signals and mistake-proofing built into the process |
| Decide what to monitor | The characteristic to chart, the sampling frequency and the subgroup |
| Choose the right chart | A control chart matched to the data type and to what you need to detect |
| Set the limits | Centre line and control limits calculated from the process, not from the specification |
| Say what happens when it moves | A control plan and a response plan with named owners |
| Confirm the benefit | A cost benefit analysis against the baseline established in Measure |
The phase ends when the process owner — not the project team — is running the controls and knows what to do when a chart signals.
5.1 Lean Controls
- 5.1.1 Control Methods for 5S
- 5.1.2 Kanban
- 5.1.3 Poka-Yoke (Mistake Proofing)
5.2 Statistical Process Control (SPC)
- 5.2.1 Data Collection for SPC
- 5.2.2 I-MR Chart
- 5.2.3 Xbar-R Chart
- 5.2.4 U Chart
- 5.2.5 P Chart
- 5.2.6 NP Chart
- 5.2.7 Xbar-S Chart
- 5.2.8 CuSum Chart
- 5.2.9 EWMA Chart
- 5.2.10 Control Methods
- 5.2.11 Control Chart Anatomy
- 5.2.12 Subgroups, Impact of Variation, Frequency of Sampling
- 5.2.13 Center Line & Control Limit Calculations
5.3 Six Sigma Control Plans
- 5.3.1 Cost Benefit Analysis
- 5.3.2 Elements of the Control Plan
- 5.3.3 Elements of the Response Plan
Which belt level covers this
Control is examined at every level, but the chart selection in section 5.2 is where the depth increases: Yellow Belt is expected to read a control chart, Green Belt to choose and construct the common ones, Black Belt to handle the full range including CuSum and EWMA and to judge subgrouping and sampling frequency.
| Level | Course | Mock exam |
|---|---|---|
| Yellow Belt | Course | Practise |
| Green Belt | Course | Practise |
| Black Belt | Course | Practise |
Frequently asked questions
Are control limits the same as specification limits?
No, and confusing the two is the classic error. Specification limits come from the customer or the design; control limits are calculated from how the process actually behaves. A process can be in control and still produce out-of-specification output — which is a capability problem, not a control problem.
Which control chart should I use?
It depends on the data type and the subgroup: individual measurements, averages and ranges, proportions and counts all have their own chart. Section 5.2 exists because selecting the wrong chart produces limits that signal either constantly or never.
Why is Poka-Yoke in the Control phase?
Because the most reliable control is one that makes the error impossible rather than detectable. Where mistake-proofing can be designed in, it outlasts any amount of monitoring and training.
What belongs in a response plan?
What to do when the chart signals: who is told, what is checked first, what authority they have to stop or adjust. A control plan without a response plan tells people a problem exists but not who owns it.
Control is short in the syllabus and long in consequences — it is what decides whether the benefit survives. Try the mock exam for your belt level to see how you score on chart selection. Contact us for advice on which level to sit.
The other DMAIC phases
- Define — deciding which problem is worth solving.
- Measure — establishing the baseline and trusting the data.
- Analyze — finding and proving the causes.
- Improve — designing and testing the change.
- Lean Six Sigma — the section overview.
