5.3 Customer Requirements: CTX, CTQ Trees, and Operational Definitions

Broad customer statements such as "I want reliable products" or "Deliver orders quickly" cannot be directly measured or controlled by engineering and operations teams. In the Define phase, Six Sigma practitioners utilize structured deployment tools to translate qualitative Voice of the Customer (VOC) feedback into precise Critical-to-Quality (CTQ) metrics with quantitative targets and specification limits.

---

The CTQ Tree Deconstruction Methodology

A CTQ Tree is a hierarchical diagram that systematically breaks down qualitative VOC feedback into actionable operational metrics. A standard CTQ Tree consists of three progressive levels:

1. Customer Need: The overarching qualitative requirement stated directly in the customer's voice.
2. Quality Driver: The key operational attribute or process performance area that fulfills the customer need.
3. CTQ Metric: A quantifiable measurement specification with explicit targets, Lower Specification Limits (LSL), and Upper Specification Limits (USL).

Sample CTQ Deconstruction Matrix

---

Formulating Rigorous Operational Definitions

An Operational Definition is an explicit, unambiguous set of instructions detailing how a specific variable or metric is to be measured and evaluated. Without operational definitions, different operators interpret metrics inconsistently, introducing measurement system error.

Core Requirements of an Operational Definition

• Metric Identifier: Precise name and unit of measure (e.g., Order Turnaround Time in decimal hours).
• Measurement Boundaries: Exact starting event trigger (e.g., time-stamp when order enters ERP system) and stopping event trigger (e.g., time-stamp when shipping label is scanned).
• Measurement Instrument & Calibration: Standardized device, software log, or gauge utilized (e.g., ERP System Log v2.1, timestamped to nearest second).
• Sampling Procedure: Sample size, frequency, and sampling plan details.
• Decision Criteria: Clear mathematical formula and Pass/Fail threshold.

Operational Definition Template Example

---

Defect Taxonomy & Core Six Sigma Quality Calculations

Six Sigma enforces a precise mathematical taxonomy for defects and yield:

Defect vs. Defective

• Defect ($D$): Any single non-conformance of a unit with a specific requirement or specification limit.
• Defective ($U_d$): An entire unit or service transaction that contains one or more defects. (A single defective unit can contain multiple distinct defects).
• Opportunity ($O$): The total count of potential failure points on a single unit where a defect could logically occur.

Fundamental Six Sigma Equations & Worked Numerical Example

1. Defects Per Unit ($DPU$):

$DPU = \frac{D}{U}$

1. Defects Per Opportunity ($DPO$):

$DPO = \frac{D}{U \times O}$

1. Defects Per Million Opportunities ($DPMO$):

$DPMO = DPO \times 1,000,000 = \left( \frac{D}{U \times O} \right) \times 10^6$

1. First-Pass Yield ($FPY$):

$FPY = \frac{U - U_d}{U}$

1. Rolled First-Pass Yield ($RFPY$):

The probability of a unit passing through a $k$-step sequential process without incurring any rework or defects:

$RFPY = \prod_{i=1}^{k} FPY_i = FPY_1 \times FPY_2 \times \dots \times FPY_k$

Worked Example Calculation

An electronics plant inspects $U = 200$ completed circuit assemblies. Each assembly has $O = 10$ defect opportunities. Inspection reveals $D = 10$ total defects across $U_d = 8$ defective boards.

• $DPU = \frac{10}{200} = 0.05$ defects per unit.
• $DPO = \frac{10}{200 \times 10} = \frac{10}{2000} = 0.005$.
• $DPMO = 0.005 \times 1,000,000 = 5,000$ DPMO.
• $FPY = \frac{200 - 8}{200} = \frac{192}{200} = 96.0%$.

If this process consists of 4 identical sequential steps each operating at $FPY_i = 96.0%$, the Rolled First-Pass Yield is $RFPY = (0.960)^4 = 0.8493 = 84.93%$.

---

Worked Quality Metrics Calculation Example

An assembly line inspects $U = 500$ completed sub-assemblies. Each sub-assembly has $O = 8$ potential defect opportunities. Inspection logs show $D = 20$ total defects found across $U_d = 15$ defective units.

1. Defects Per Unit ($DPU$): $DPU = \frac{D}{U} = \frac{20}{500} = 0.040\text{ defects/unit}$
2. Defects Per Opportunity ($DPO$): $DPO = \frac{D}{U \times O} = \frac{20}{500 \times 8} = \frac{20}{4000} = 0.0050$
3. Defects Per Million Opportunities ($DPMO$): $DPMO = DPO \times 1,000,000 = 0.0050 \times 10^6 = 5,000\text{ DPMO}$
4. First-Pass Yield ($FPY$): $FPY = \frac{U - U_d}{U} = \frac{500 - 15}{500} = \frac{485}{500} = 0.9700 \quad (97.00\%)$