Quality Improvement Methods
Quality technique · See it on the map
Guide only — Choosing and applying an improvement method is a judgment call.
Formal, repeatable programs that drive down defects or variation across many measured cycles over time.
When to use it
A recurring quality problem important enough, and stable enough in its pattern, to be worth measuring, addressing in a structured cycle, and re-measuring — chronic rework, repeat defects across projects, or a process whose output varies more than the client will tolerate.
When to avoid it
A single project's one-off defect — fix it and move on, that's problem solving or root cause analysis, not a program. Also avoid standing up Six Sigma-level rigor (belts, control charts, statistical process control) for a problem a one-week fix would resolve; the program overhead will exceed the problem and it will not get finished.
Steps
What it produces
- A measured before/after comparison against a defined baseline.
- A standardized process change, or a documented and rejected candidate.
- A repeatable cycle set up to run again.
Common pitfalls
- Launching a full DMAIC-weight program for a problem a single fix would have resolved — the program itself becomes the overhead.
- Skipping the baseline measurement, so 'improvement' can't actually be shown.
- Treating one PDCA cycle as finished rather than as one loop in an ongoing process.
Worked example
Larch & Rowe sees drywall finish rework on roughly one in six jobs. Rather than a company-wide quality initiative, the ops lead runs a lightweight PDCA cycle: Plan — require a moisture check before mudding on the next ten jobs; Do — run it; Check — rework drops to one in fifteen; Act — make the moisture check standard practice on every job and retire the old sign-off sheet that didn't include it.
Source
- PMBOK-6 §8.2
Where it comes from: this technique is named by the PMBOK Guide, 6th edition.