SYE6047 Chap.3 Six Sigma Variation and Defect Metrics
Six Sigma Variation and Defect Metrics
Variation is the operating problem
Six Sigma links customer requirements to the distribution of process output. The mean describes location and the standard deviation describes spread, but neither alone establishes performance. A process can be centred with excessive variation or tightly grouped around the wrong target.
Improvement therefore requires a defined characteristic, a trustworthy measurement method and an explicit relationship between the observed distribution and specification limits.
Units, defects and opportunities
A unit is the item or transaction being studied. A defect is one failure to meet a requirement, and a unit may contain several defects.
An opportunity is a defined chance for a defect of the relevant type to occur. Defects per unit divides total defects by units. Defects per million opportunities divides defects by units times opportunities per unit, then scales by one million.
The denominator must come from the operational definition, not from a desire to obtain an attractive rate.
Spread and specification width
The potential capability ratio compares the total specification width with six estimated standard deviations. It describes the room available if the process is stable and reasonably represented by the estimate.
It does not measure centring because moving the mean while holding the spread fixed leaves the ratio unchanged. Nor does it replace a control chart: an unstable process does not have one dependable distribution to compare with requirements.
Interpretation after arithmetic
Every metric needs a sentence that preserves its denominator and assumptions. DPU is defects per observed unit, not defective units.
DPMO supports comparisons only when opportunities are defined consistently. A capability ratio compares spread with tolerance under stated conditions; it does not prove customer satisfaction, control or future performance. The correct sequence is define, measure, calculate and qualify.
What this chapter covers
- 01
Location and spread
- 02
Common and special variation
- 03
Units, defects and defectives
- 04
Opportunities per unit
- 05
DPU calculation
- 06
DPMO calculation
- 07
Specification width and six sigma spread
- 08
Limits of potential capability
Worked example · free
Keep the DPMO denominator honest
- 3Calculate defects per unit.
- 4Calculate defects per million opportunities.
- 3Interpret the rates and state a denominator risk.
Key terms
- Defect
- One observed failure to meet a defined requirement.
- Defective Unit
- A unit containing at least one defect, regardless of the number of defects.
- Opportunity
- A defined chance for the defect type under study to occur on one unit.
- DPU
- The observed number of defects divided by the number of units.
- Potential Capability
- A comparison of specification width with six estimated standard deviations for a stable process.
Six Sigma Variation and Defect Metrics FAQ
Why are defects and defective units different?
One invoice can contain several errors but is only one defective invoice. DPU counts all defects, whereas a fraction defective counts units with at least one defect. Mixing them changes both numerator meaning and statistical model. Write the unit and counting rule before calculating so repeated defects on one item cannot silently become several defective items. Preserve both raw counts when a problem reports them.
Can two processes with equal DPMO be treated as equally good?
Only if opportunities, defect definitions, sampling and customer consequences are comparable. The scaled rate removes volume but cannot repair inconsistent definitions or reflect severity. Always inspect how the opportunity count was constructed.
Why check stability before capability?
Capability presumes a distribution that meaningfully represents future process behaviour. Special-cause signals indicate that location or spread changes over time, so one pooled ratio may hide distinct operating conditions.
Exam move
Create a calculation card with four labelled lines: defects, units, opportunities per unit and total opportunities. On every practice problem, circle whether the question concerns defects or defectives. For capability questions, write stable? and centred? beside the formula before interpreting the result.
Rework the same data after changing the unit definition and observe which rates move; this makes denominator sensitivity concrete. Invent one invalid opportunity count and explain exactly how it makes DPMO look better without changing output. Keep raw counts beside every scaled rate so another analyst can reconstruct the calculation.
For spread questions, sketch two distributions with the same mean but different sigma, then two with the same sigma but different means. Use those sketches to explain why potential capability addresses width and not centring. When comparing periods, check whether inspection intensity, opportunity definitions, product mix and sigma estimation stayed fixed.
Your final sentence should contain the metric's noun and denominator, such as defects per defined opportunity, followed by one assumption. This prevents a correct number from becoming a false statement about defective units, customer satisfaction or statistical control.
