3.2 Six Sigma Process Performance Metrics & COPQ

Six Sigma Process Performance Metrics & COPQ

Quick Answer: Six Sigma process performance metrics quantitatively measure defect density, yield efficiency, throughput velocity, and the financial cost of quality deficiencies. Core defect calculations include Defects Per Unit ($DPU = D / U$), Defects Per Opportunity ($DPO = D / [U \times O]$), and Defects Per Million Opportunities ($DPMO = DPO \times 10^6$), where 3.4 DPMO defines Six Sigma capability with a 1.5-sigma shift. Yield analysis contrasts First Pass Yield (FPY) with Rolled Throughput Yield ($RTY = \prod Y_i$), exposing the hidden factory of unrecorded rework. Quality economics are evaluated through the Prevention-Appraisal-Failure (PAF) cost-of-quality model, whose two failure categories make up the Cost of Poor Quality (COPQ), and the 1-10-100 rule of thumb. Independent CSSYB study guide by OpenExamPrep.

The Role of Quantitative Metrics in Six Sigma

A fundamental premise of Six Sigma is that process improvement requires rigorous quantification. Subjective impressions cannot justify capital investments or isolate operational failure modes. Six Sigma provides mathematical formulas to evaluate defect density, calculate yield across multi-step processes, analyze throughput velocity, and translate operational waste into financial terms through the Cost of Poor Quality (COPQ).

Yellow Belts must master these core formulas, perform standard calculations, and interpret their practical operational significance.

---

Defect Metrics: DPU, DPO, and DPMO

To measure defects objectively across processes of varying size and complexity, Six Sigma establishes three interrelated metrics:

1. Key Definitions

• Unit ($U$): The individual item, product, or transaction inspected (e.g., a purchase order, invoice, or machined part).
• Defect ($D$): Any specific flaw, blemish, or failure to meet a customer specification. A single unit may contain multiple defects.
• Defective: An entire unit containing one or more defects, rendering it non-conforming.
• Defect Opportunity ($O$): The number of independent, critical-to-quality opportunities for an error to occur on a single unit.

2. Core Formulas

• Defects Per Unit (DPU): $$DPU = \frac{D}{U}$$
• Defects Per Opportunity (DPO): $$DPO = \frac{D}{U \times O}$$
• Defects Per Million Opportunities (DPMO): $$DPMO = DPO \times 1,000,000$$

3. Worked Numerical Example

Scenario: A commercial purchasing department audits 500 purchase orders ($U = 500$). Each order contains 4 critical fields representing potential defect opportunities: vendor ID, part number, quantity, and unit price ($O = 4$). During the audit, examiners uncover a total of 30 defects ($D = 30$) across the batch.

Step-by-Step Calculation:

1. Total Opportunities: $$\text{Total Opportunities} = U \times O = 500 \times 4 = 2,000$$
2. Calculate DPU: $$DPU = \frac{30}{500} = 0.06 \text{ defects per unit}$$
3. Calculate DPO: $$DPO = \frac{30}{2,000} = 0.015 \text{ defects per opportunity}$$
4. Calculate DPMO: $$DPMO = 0.015 \times 1,000,000 = 15,000 \text{ DPMO}$$

Under standard normal distribution tables incorporating a 1.5-sigma shift, 15,000 DPMO corresponds to approximately a 3.67-sigma performance level.

---

Yield Metrics: First Pass Yield, RTY & The Hidden Factory

While defect metrics measure failure density, yield metrics measure operational efficiency and product throughput.

First Pass Yield (FPY)

Traditional final yield simply divides acceptable units exiting the final workstation by total units started, concealing intermediate rework and repairs.

First Pass Yield (FPY) measures the proportion of units that complete an individual process step defect-free on the first attempt without rework:

$$FPY = \frac{\text{Good Units Completed Without Rework}}{\text{Total Units Entering Step}}$$

Rolled Throughput Yield (RTY)

In multi-step workflows, defects compound across operations. Rolled Throughput Yield (RTY) calculates the probability that a unit passes through an entire sequence of $n$ independent steps defect-free without any rework:

$$RTY = Y_1 \times Y_2 \times Y_3 \times \dots \times Y_n = \prod_{i=1}^n Y_i$$

Worked Example: The Hidden Factory

Scenario: A manufacturing line consists of three sequential process steps:

• Step 1 (Placement): Yield $Y_1 = 0.98$ (98%)
• Step 2 (Soldering): Yield $Y_2 = 0.95$ (95%)
• Step 3 (Testing): Yield $Y_3 = 0.92$ (92%)

Calculation:

$$RTY = 0.98 \times 0.95 \times 0.92 = 0.85652 \approx 85.65\%$$

Operational Significance: Although each individual station reports strong first-pass yield (>90%), the cumulative rolled yield is only 85.65%. The remaining 14.35% represents the "hidden factory"—unplanned labor, floor space, tooling, and overtime spent disassembling, repairing, and re-inspecting non-conforming units. The hidden factory inflates overhead costs and conceals true process vulnerability.

---

Operational Time Metrics: Cycle Time, Lead Time & Takt Time

Process velocity and responsiveness directly impact defect rates and customer satisfaction:

• Cycle Time: The actual elapsed time required to complete one operational cycle of work on a single unit from start to finish.
• Lead Time: The total elapsed time from customer order placement until final delivery, including processing, delays, and transportation.
• Takt Time: The operational pace required to match the rate of customer demand: $$\text{Takt Time} = \frac{\text{Available Net Production Time}}{\text{Customer Demand Rate}}$$ Takt time sets the operational drumbeat; producing faster creates overproduction waste, while producing slower causes delivery delays.

---

Cost of Poor Quality (COPQ) - The PAF Model

Quality costs represent the total financial burden incurred because products or services may not meet customer expectations. Building on Joseph Juran's cost-of-quality work, Armand Feigenbaum grouped these costs into the PAF (Prevention, Appraisal, Failure) model. Prevention plus appraisal is the cost of good quality; internal plus external failure is the cost of poor quality (COPQ); all four together are the total cost of quality.

1. Cost of Good Quality (Conformance Costs)

• Prevention Costs: Upfront investments to design quality into products and processes, preventing non-conformances before they occur. Examples include operator training, design reviews, robust engineering, supplier audits, and Poka-Yoke mistake-proofing.
• Appraisal Costs: Expenses incurred to inspect, test, and audit materials and processes to detect errors before shipment. Examples include receiving inspection, in-process testing, final inspection, gauge calibration, and quality audits.

2. Cost of Poor Quality (Non-conformance Costs)

• Internal Failure Costs: Costs resulting from defects caught inside the organization prior to customer delivery. Examples include scrap, rework labor, re-inspection, downgraded products, and machine downtime caused by defective components.
• External Failure Costs: The most financially damaging quality costs, occurring when defects escape to the customer. Examples include warranty claims, customer complaints, field service calls, product recalls, returns, legal liabilities, and lost goodwill.

The 1-10-100 Rule

The 1-10-100 Rule is a rule of thumb, not a measured constant, that illustrates how the cost of a defect escalates the later it is found:

• $1 invested in Prevention stops the defect from occurring.
• $10 is spent finding and correcting the defect internally (appraisal plus internal failure costs).
• $100 (or more) spent on External Failure resolves warranty claims, customer churn, and recalls after shipment.

Six Sigma initiatives shift organizational focus from reactive inspection and failure firefighting to proactive prevention, which lowers total COPQ over time.