Sigma Flashcards

During a DMAIC project reviewing officer dispatch response times, the team calculates a Cp of 1.45 but a Cpk of 0.78. Which conclusion is MOST accurate?

Answer: The process has sufficient spread potential but is significantly off-centre from the target

Cp measures the potential capability (spread relative to spec width) while Cpk accounts for process centering. A Cp of 1.45 indicates the process spread could fit well within specification limits, but a Cpk of 0.78 (below 1.0) reveals the process mean is significantly shifted toward or beyond a specification limit. This gap between Cp and Cpk is a classic indicator of a centering problem — the variation is manageable but the process is aimed incorrectly. The fix is centering, not redesign.

A Six Sigma Black Belt facilitating an OACP crime-reduction project is conducting a Gauge R&R study on a newly introduced evidence-classification rubric. The study returns a %Gauge R&R of 22%. How should the team proceed?

Answer: Consider the measurement system marginally acceptable, document the risk, and investigate sources of variation before proceeding cautiously

Gauge R&R thresholds are: under 10% = acceptable, 10–30% = marginal (may be acceptable depending on application importance and cost of repair), over 30% = unacceptable. At 22%, the measurement system falls in the marginal zone. For a high-stakes public-safety application like evidence classification, the team must investigate reproducibility vs. repeatability contributors, document the risk, and determine whether the marginal system is adequate for the decision being made — not simply accept or reject it outright.

A police service's Six Sigma team plots an X-bar and R chart for daily 911 call-handling time. Eight consecutive points fall on the same side of the centreline but all remain within the upper and lower control limits. What does this pattern indicate?

Answer: A special-cause (assignable-cause) signal is present, indicating a non-random shift in the process mean

Western Electric (Nelson) run rules identify non-random patterns even when all points remain within control limits. Eight or more consecutive points on the same side of the centreline is a classic special-cause signal (Rule 2 / Nelson Rule 2), indicating a sustained shift in the process mean. Staying inside the control limits does not guarantee common-cause variation — the probability of eight consecutive points on one side by chance alone is less than 0.4%, making this a statistically significant, actionable signal.

An OACP-affiliated service wants to build a new digital records-management system and is deciding between DMAIC and DMADV. The current paper-based process has a defect rate of 34% and analysts confirm it is fundamentally incapable of reaching the 3.4 DPMO target even if optimised. Which framework applies, and why?

Answer: DMADV, because the existing process cannot be brought to the desired sigma level through improvement alone and a new design is required

DMAIC (Define–Measure–Analyse–Improve–Control) is used to improve an existing process. DMADV (Define–Measure–Analyse–Design–Verify), also called Design for Six Sigma (DFSS), is used when a process does not exist or when the existing process is so fundamentally incapable that incremental improvement cannot achieve the required performance level. A 34% defect rate with confirmed structural incapability is precisely the trigger for DMADV — attempting DMAIC in this scenario wastes resources on a process destined for replacement.

During the Analyse phase of a project targeting wrongful-charge error rates, a team constructs a fishbone (Ishikawa) diagram and then runs a multi-vari study. The multi-vari chart reveals that the largest source of variation occurs BETWEEN shifts rather than within shifts or between individual officers. What type of variation does this represent, and what is the primary investigative implication?

Answer: Temporal (time-to-time) variation — the team should investigate what systematically differs between shifts, such as supervision, training cohort, or handover protocols

Multi-vari studies partition variation into three families: positional (within-unit), cyclical (unit-to-unit within a time period), and temporal (across time periods or groupings such as shifts). Dominant between-shift variation is temporal variation, pointing investigators toward factors that differ systematically by shift — such as supervision quality, shift-specific briefing content, differing training cohorts, or handover protocol gaps. Chasing individual officer differences (unit-to-unit) would misdirect resources given where the variation actually lives.

A Green Belt presents a completed Six Sigma project claiming a 1.8-sigma improvement in a use-of-force reporting cycle. A reviewing Black Belt notes the project relied entirely on a before/after comparison with no control group and that the improvement period coincided with a province-wide policy change. Which validity threat MOST undermines the claimed sigma improvement?

Answer: History effect — an external event concurrent with the intervention confounds attribution of the improvement to the Six Sigma project

The history effect is a classic internal validity threat in quasi-experimental designs: when a significant external event (here, a province-wide policy change) occurs during the same period as the intervention, it is impossible to determine whether the observed improvement resulted from the Six Sigma project or from the external event. Without a concurrent control group unaffected by both the intervention and the external change, the claimed 1.8-sigma gain cannot be attributed to the project. This is why rigorous Six Sigma projects include control conditions or at minimum document and statistically isolate confounding events.