Six Sigma Methodologies

Why This Matters

Six Sigma is how modern organizations systematically eliminate defects, reduce costs, and deliver consistent quality. For this course, you need to distinguish between improving existing processes versus designing new ones, understand how data-driven decision-making works, and recognize which tools solve which problems.

Six Sigma operates on multiple levels: strategic frameworks (like DMAIC and DMADV) guide the overall approach, while tactical tools (like control charts and FMEA) execute specific steps within those frameworks. Don't just memorize acronyms. Know when to apply each methodology and how the tools connect. Understanding the why behind each tool will help you tackle any scenario-based question.

---

Strategic Frameworks: DMAIC vs. DMADV

These two methodologies form the foundation of Six Sigma. The critical distinction: DMAIC fixes what exists; DMADV builds what doesn't yet exist.

DMAIC (Define, Measure, Analyze, Improve, Control)

DMAIC is your go-to framework when an existing process is broken or underperforming. Its five sequential phases create a closed-loop system:

1. Define the problem, scope, and project goals. This often includes writing a project charter and identifying stakeholders.
2. Measure the current state of the process using data. You're establishing a baseline so you can quantify how bad things are right now.
3. Analyze the data to identify root causes of defects or poor performance. This is where tools like fishbone diagrams and the 5 Whys come in.
4. Improve the process by developing and implementing solutions that target those root causes.
5. Control the improved process to sustain gains over time, typically using control charts and standardized procedures.

The Control phase is what separates Six Sigma from ad-hoc problem solving. Without it, improvements decay and old habits return.

DMADV (Define, Measure, Analyze, Design, Verify)

DMADV applies when you're designing a new process or product from scratch to meet Six Sigma quality levels from day one.

• Design replaces Improve: instead of fixing problems, you're proactively engineering quality into the process before launch
• Verify replaces Control: the final phase ensures the design meets customer requirements through testing and pilot runs before full implementation

---

Customer-Centric Tools: Capturing What Matters

Six Sigma starts and ends with the customer. These tools translate subjective customer needs into objective, measurable specifications.

Voice of the Customer (VOC)

VOC is the input-gathering mechanism that captures what customers actually want. It uses surveys, interviews, focus groups, and feedback analysis. VOC helps teams distinguish between must-have features and nice-to-have enhancements based on customer value.

The raw data from VOC feeds directly into CTQ analysis.

Critical to Quality (CTQ)

CTQ translates vague VOC data into measurable targets. For example, a customer saying "I want it to be reliable" becomes a specification like mean time between failures $\geq$ 10,000 hours. A customer saying "I want fast delivery" becomes order-to-shipment time $\leq$ 48 hours.

• Identifies key quality attributes that determine customer satisfaction
• Guides improvement priorities so teams focus on what actually matters to customers, not internal assumptions

---

Diagnostic Tools: Finding the Problem

Before you can fix anything, you need to understand what's happening and why. These tools help teams visualize processes and identify root causes rather than just treating symptoms.

Process Mapping

Process mapping creates a visual representation of a workflow, documenting every step, input, output, and decision point. This makes bottlenecks, redundancies, and handoff problems visible to the entire team. It also serves as a communication tool, creating shared understanding among stakeholders who may have very different mental models of how work actually flows.

Root Cause Analysis

Root cause analysis goes beyond symptoms to identify why defects occur, not just what went wrong. Two key techniques:

• 5 Whys: You keep asking "why?" at each answer until you drill down to the fundamental cause. For example: Why did the machine stop? → A fuse blew. Why? → Insufficient lubrication. Why? → The pump wasn't working. Why? → The shaft was worn. Why? → No maintenance schedule. The root cause is the missing maintenance schedule, not the blown fuse.
• Fishbone (Ishikawa) Diagram: Categorizes potential causes into branches (commonly: Man, Machine, Method, Material, Measurement, Environment) so teams can systematically explore all possibilities rather than fixating on the first cause that comes to mind.

Eliminating root causes provides permanent fixes rather than temporary patches.

Value Stream Mapping

Value stream mapping tracks materials and information from supplier to customer across the entire value chain. Its defining feature is that it categorizes every activity as value-adding, necessary non-value-adding, or pure waste. This is a staple of Lean Six Sigma and is essential for identifying improvement opportunities across whole processes, not just individual steps.

---

Statistical Tools: Data-Driven Decision Making

Six Sigma's power comes from replacing gut feelings with data. These tools help teams measure performance, detect problems early, and optimize through experimentation.

Statistical Process Control (SPC)

SPC uses statistical methods to monitor process performance in real time and detect variations before they cause defects. The core concept is distinguishing between two types of variation:

• Common cause variation: Normal, inherent randomness in the process (e.g., slight temperature fluctuations in a factory). You address this by fundamentally redesigning the process.
• Special cause variation: Unusual events that signal something has changed (e.g., a new batch of raw material, a miscalibrated tool). You address this by investigating and removing the specific anomaly.

A process affected only by common cause variation is called stable or "in control." That doesn't mean it's performing well, just that it's behaving predictably.

Control Charts

Control charts plot measurements over time with a centerline (the process mean) and upper and lower control limits, typically set at $\pm 3\sigma$ from the mean. When data points fall outside these limits, or when non-random patterns appear (like seven consecutive points on one side of the centerline), the process is signaling a special cause that requires investigation.

Control charts are the foundation of DMAIC's Control phase, keeping improved processes on track.

Design of Experiments (DOE)

DOE is a systematic testing methodology that plans experiments to efficiently determine relationships between input factors and output responses. Unlike changing one variable at a time, DOE tests multiple variables simultaneously. This is more efficient and reveals interactions between factors that one-at-a-time testing would miss.

For example, if you're optimizing a manufacturing process, DOE might test temperature, pressure, and speed settings together. You might find that high temperature only causes defects when combined with high speed, an interaction you'd never catch by testing each variable in isolation.

---

Risk Prevention Tools: Stopping Problems Before They Start

The best quality strategy prevents defects rather than detecting them. These proactive tools help teams anticipate failures and design them out of processes.

Failure Mode and Effects Analysis (FMEA)

FMEA is a proactive risk assessment that systematically identifies potential failure modes, their causes, and their effects before problems occur. Each potential failure gets scored using the Risk Priority Number (RPN):

$\text{RPN} = \text{Severity} \times \text{Occurrence} \times \text{Detection}$

Each factor is rated on a scale (typically 1–10), so RPN ranges from 1 to 1,000. Higher RPNs get addressed first. Watch out for a common point of confusion: a high Detection score means the failure is hard to detect, which makes it more dangerous. A score of 10 on Detection means you almost certainly won't catch it before it reaches the customer.

Poka-Yoke (Error-Proofing)

Poka-Yoke designs processes or devices that make errors impossible or immediately obvious. The key principle: assume people will make mistakes, and engineer those mistakes out of the system.

Common examples include USB-C ports that fit in any orientation, checklists that force sequence compliance, and sensors that stop machines when parts are misaligned. These mechanisms eliminate reliance on human vigilance, which is always unreliable over long periods.

---

Continuous Improvement Culture: Sustaining Excellence

Six Sigma isn't a one-time project. These methodologies create organizational habits that drive ongoing improvement.

5S Methodology

5S is a workplace organization system built on five principles:

1. Sort: Remove unnecessary items from the workspace
2. Set in order: Organize remaining items so they're easy to find and use
3. Shine: Clean the workspace regularly to spot abnormalities (a leaking machine is obvious on a clean floor)
4. Standardize: Create consistent procedures for the first three steps so every shift does them the same way
5. Sustain: Maintain discipline over time through audits and habits

5S serves as a foundation for other improvements. A disorganized workplace masks problems and creates waste. By making the workspace visual and orderly, abnormalities become immediately visible.

Kaizen

Kaizen is a continuous improvement philosophy emphasizing small, incremental changes made by all employees rather than dramatic overhauls by specialists. Frontline workers often have the best insights into process problems, and Kaizen harnesses that collective intelligence. Individual improvements may be modest, but accumulated gains over time create transformational results.

Lean Six Sigma

Lean Six Sigma is a hybrid methodology combining Lean's focus on speed and waste elimination with Six Sigma's focus on quality and variation reduction. Lean targets the eight wastes (often remembered by the acronym DOWNTIME: Defects, Overproduction, Waiting, Non-utilized talent, Transportation, Inventory, Motion, Extra processing). Six Sigma targets variation and defects through statistical analysis. Most modern implementations blend both philosophies rather than treating them separately.

---

Quick Reference Table

---

Self-Check Questions

1. A manufacturing company discovers their defect rate has increased on an established production line. Which framework should they use, DMAIC or DMADV, and why?

2. Compare and contrast VOC and CTQ: How do they work together, and what happens if a team skips the VOC step and jumps straight to defining CTQs?

3. Which two tools would you use together to first identify potential failures and then prevent them from occurring? Explain how they complement each other.

4. A process is stable (in control) but producing results far below optimal levels. Which tool, SPC or DOE, would be most appropriate for improvement, and what's your reasoning?

5. A company wants to implement Six Sigma but struggles with employee buy-in and sustainability. Which methodologies specifically address cultural and organizational aspects of continuous improvement, and how do they differ in approach?