EPI: Continuous Improvement Methodologies

Continuous Improvement Methodologies tests whether you can select and apply a disciplined improvement approach rather than recite tools. Questions cover PDCA, DMAIC, Lean, Six Sigma, Kaizen, A3 thinking, root-cause analysis, FMEA, Kanban, and Theory of Constraints.

Core rules and concepts

• Match method to problem. PDCA suits iterative, lower-risk experiments. DMAIC suits complex problems with unknown root causes and a need for statistical validation and control. Kaizen suits short, focused waste reduction with the people doing the work. A3 suits structured problem solving and coaching.
• Define the problem in measurable terms: gap, location, time, magnitude, and impact. Avoid embedding a solution in the problem statement.
• Validate root causes with evidence, not voting. Use 5 Whys or fishbone to generate hypotheses, then confirm with data or a designed test.
• Treat Lean as flow and waste reduction, not headcount reduction. Treat Six Sigma as variation reduction for defect and process capability problems.
• Use FMEA to prioritize failure modes by severity, occurrence, and detection, then act on high-risk modes.
• Close with a control plan: owner, measure, response plan, audit cadence, and standardization. Without control, gains decay.

Worked example

An accounts-payable team has recurring invoice errors: 7% of invoices require rework. The sponsor wants a one-week Lean event. The reasoning is that the defect rate is stable, the root cause is unknown, and several handoffs are involved. A one-week event may generate useful ideas, but DMAIC is the better fit: define the defect and cost, measure error types by source, analyze which supplier, template, or approval step predicts errors, improve the validated causes, and control with a pre-submission check and supplier feedback loop. If analysis shows most errors come from one manual rekeying step, a Kaizen event can then remove that step quickly. The method choice follows the uncertainty, not the sponsor's preferred tool.

Traps that cost marks

• Tool-first thinking: running a fishbone because it is familiar, not because it fits.
• Claiming a root cause after brainstorming without confirmation.
• Confusing Lean with layoffs, which destroys the trust needed for improvement.
• Ending at pilot success with no control plan, standard work, or owner.
• Using DMAIC for a simple, reversible change that PDCA could test in a day.