PECB Six Sigma Yellow Belt cheat sheet

Sigma level expresses process capability as defects per million opportunities (DPMO): 1σ ≈ 691,462 DPMO, 2σ ≈ 308,537, 3σ ≈ 66,807, 4σ ≈ 6,210, 5σ ≈ 233, 6σ ≈ 3.4 DPMO (with the conventional 1.5σ shift).

DPMO = (defects / (units × opportunities per unit)) × 1,000,000; it normalizes defect counts so processes of different complexity can be compared.

Rolled Throughput Yield (RTY) = product of the first-pass yields of all process steps; a five-step process each at 90% yield gives RTY = 0.9^5 ≈ 59%, not 90%.

Lean targets the eight wastes (TIMWOODS): Transport, Inventory, Motion, Waiting, Overproduction, Overprocessing, Defects, Skills underutilized; Six Sigma targets variation and defects.

The five Lean principles are: specify value from the customer's view, map the value stream, create flow, establish pull, and pursue perfection.

Cost of poor quality (COPQ) covers internal failure, external failure, appraisal, and prevention costs; the first two are failure costs and are the primary target of improvement.

DMAIC applies to improving an existing process; DMADV/DFSS applies to designing a new process or product that does not yet exist.

Y = f(x): the output Y (CTQ characteristic) is a function of process inputs x; improvement work identifies and controls the critical few x's.

A CTQ (critical to quality) translates a vague customer need into a measurable, verifiable requirement with a defined specification limit.

DMAIC phases in order: Define, Measure, Analyze, Improve, Control; each phase ends with a tollgate review before the next begins.

Define phase deliverables: project charter (problem statement, goal, scope, business case, team, milestones), SIPOC diagram, and voice-of-customer translated into CTQs.

A good problem statement states what, where, when, and how much, with no cause and no solution embedded in it; the goal statement is SMART and measurable.

Measure phase: operational definitions, measurement system analysis (Gage R&R), data collection plan, baseline capability, and process maps.

Gage R&R rule of thumb: under 10% of tolerance study variation is acceptable, 10–30% may be acceptable depending on application, over 30% is not acceptable.

Analyze phase tools: Pareto chart, cause-and-effect (fishbone) diagram, 5 Whys, hypothesis tests, scatter plots, and FMEA for prioritizing failure modes by severity × occurrence × detection.

Improve phase: generate and select solutions, pilot, run DOE or kaizen events, and verify the improvement with the same measurement system used for the baseline.

Control phase: control plan, control charts, standard work, mistake-proofing (poka-yoke), response plan, and handover of the process to the process owner.

Control charts: X-bar/R charts for variable data in subgroups; p, np, c, and u charts for attribute data (defective vs defect, constant vs variable sample size).