Quality Control and Six Sigma: Methods for Process Excellence
Quality is not free. It is the most profitable investment a company can make. This insight, articulated by quality pioneer Philip Crosby, captures the fundamental truth that poor quality costs far more than good quality. Defective products require rework, create scrap, generate customer complaints, and damage brand reputation. Quality control and Six Sigma provide the systematic methods for reducing defects and achieving process excellence.
The concept of Six Sigma originated at Motorola in 1986. Engineer Bill Smith defined Six Sigma as a statistical measure of process capability — 3.4 defects per million opportunities. Jack Welch made Six Sigma famous at General Electric, attributing billions of dollars in savings to the methodology. Today, Six Sigma is embedded in organizations worldwide, from manufacturing plants to hospitals to financial services.
The DMAIC Methodology
DMAIC — Define, Measure, Analyze, Improve, Control — is the core problem-solving framework of Six Sigma. It provides a structured approach to process improvement.
Define Phase
The Define phase establishes the project scope, goals, and team. A project charter documents the business case, problem statement, goals, scope, timeline, and team members. The voice of the customer is captured through surveys, interviews, and complaints. Critical-to-quality characteristics — the features that matter most to customers — are identified and quantified.
Measure Phase
The Measure phase collects data on the current process performance. Process mapping creates a visual representation of the process flow. Data collection plans specify what to measure, how to measure it, and how much data to collect. Measurement system analysis evaluates whether the measurement tools are accurate and precise enough for the task.
Baseline capability is established using process capability indices. Cp compares the specification width to the process variation. Cpk accounts for both variation and centering. A Cpk of 1.0 corresponds to three sigma capability — 2,700 defects per million. A Cpk of 2.0 corresponds to Six Sigma capability.
Analyze Phase
The Analyze phase identifies the root causes of defects. Statistical tools reveal relationships between process inputs and outputs. Hypothesis testing determines whether observed differences are statistically significant. Regression analysis quantifies the effect of input variables on output quality.
Root cause analysis tools include fishbone diagrams, 5 Whys, and failure mode and effects analysis. Each defect is traced backward through the causal chain until the fundamental root cause is identified. The statistical process control article covers the analytical tools used in this phase.
Improve Phase
The Improve phase develops and implements solutions to address root causes. Design of experiments systematically varies input factors to determine their effects on output quality. Full factorial experiments test all combinations of factors. Fractional factorial experiments test a subset, reducing cost while still identifying main effects.
Solution selection considers effectiveness, cost, implementation difficulty, and risk. Pilot implementations test solutions on a small scale before full deployment. The lean manufacturing article discusses complementary improvement methods.
Control Phase
The Control phase ensures that improvements are sustained over time. Control plans document the new process, monitoring procedures, and response plans. Statistical process control charts monitor key quality characteristics and signal when the process goes out of control.
Standardized work instructions ensure that operators follow the improved process consistently. Training programs transfer knowledge to the workforce. Periodic audits verify that control plans are being followed and that improvements are holding.
Quality Tools and Techniques
Six Sigma practitioners use a toolkit of analytical and problem-solving tools.
Fishbone Diagram
Also called Ishikawa diagrams, fishbone diagrams organize potential causes of a problem into categories. The typical categories are the 6 Ms — Man, Machine, Material, Method, Measurement, and Mother Nature. Teams brainstorm causes within each category, then prioritize the most likely root causes for investigation.
Failure Mode and Effects Analysis
FMEA identifies potential failure modes in a product or process, their effects, and their causes. Each failure mode is rated for severity, occurrence probability, and detection difficulty. The risk priority number is the product of these three ratings. High RPN items receive priority for preventive action.
Design FMEA addresses product design failures. Process FMEA addresses manufacturing process failures. FMEA is a living document — it is updated as new failure modes are discovered or process changes are made.
5S Workplace Organization
5S — Sort, Set in Order, Shine, Standardize, Sustain — creates an organized, clean, and efficient workplace. Sort removes unnecessary items from the work area. Set in Order arranges tools and materials for efficient access. Shine cleans the workplace and equipment. Standardize creates consistent procedures for the first three S’s. Sustain maintains the discipline through audits and accountability.
Six Sigma Roles and Certification
Six Sigma projects are led by trained and certified practitioners at different levels.
Yellow Belt
Yellow Belts have basic awareness of Six Sigma concepts. They participate on project teams and contribute to data collection and implementation. Yellow Belt training typically requires one to two days of instruction.
Green Belt
Green Belts lead smaller projects or support Black Belts on larger projects. They receive four to five days of training in DMAIC methodology and basic statistical tools. Green Belts spend 10 to 25 percent of their time on Six Sigma projects while maintaining their regular roles.
Black Belt
Black Belts are full-time Six Sigma project leaders. They receive two to four weeks of training covering advanced statistical methods, project management, and change management. Black Belts lead complex projects with high financial impact and mentor Green Belts.
Master Black Belt
Master Black Belts are the highest level of Six Sigma expertise. They train and certify Black Belts, develop improvement strategies, and lead organizational transformation. Master Black Belts typically have five or more years of Six Sigma experience and deep statistical knowledge.
Design for Six Sigma
DFSS applies Six Sigma principles to product and process design, preventing defects before they occur rather than fixing them after.
DMADV Methodology
DMADV — Define, Measure, Analyze, Design, Verify — is the DFSS counterpart of DMAIC. The Define phase establishes design goals consistent with customer requirements. The Measure phase identifies critical-to-quality characteristics and measures current capability. The Analyze phase develops design alternatives and evaluates them against requirements.
The Design phase creates detailed design specifications. Design of experiments and computer simulation optimize the design parameters. Finite element analysis, computational fluid dynamics, and other engineering simulation tools predict product performance under operating conditions.
The Verify phase tests the design through prototypes and pilot production. Verification confirms that the design meets specifications under real-world conditions. Design flaws found during verification are corrected before full-scale production begins, avoiding the high cost of late-stage design changes.
Design scorecards track DFSS progress by measuring design maturity against criteria at each phase. The scorecard identifies gaps that must be addressed before the design proceeds to the next phase. Gates between phases prevent premature progression.
Quality Function Deployment
QFD translates customer requirements into engineering characteristics. The House of Quality matrix maps customer needs — what customers want — against design parameters — how the product will deliver those wants. Relationship symbols indicate strong, moderate, or weak relationships between customer needs and design parameters.
QFD ensures that engineering effort focuses on characteristics that matter to customers. It also identifies conflicts between design parameters — improving one characteristic may degrade another. These tradeoffs are identified and addressed during design rather than discovered during production.
Failure Mode Avoidance
DFSS emphasizes proactive failure prevention rather than reactive failure detection. Design guidelines, lessons learned databases, and standard design practices capture knowledge from past projects. Design reviews at defined milestones verify that failure prevention practices have been applied.
Frequently Asked Questions
What is the difference between Six Sigma and Lean? Six Sigma focuses on reducing variation and defects through statistical methods. Lean focuses on eliminating waste and improving flow. The two methodologies are complementary and are often combined as Lean Six Sigma. Lean addresses process speed while Six Sigma addresses process quality.
How much does a Six Sigma project typically save? The average Six Sigma project saves between 50,000 and 250,000 dollars annually. Well-established programs at large companies report cumulative savings in the billions. The key is selecting projects that align with strategic priorities and have measurable financial impact.
Is Six Sigma only for manufacturing? No. Six Sigma has been successfully applied in healthcare, financial services, software development, logistics, and government. The DMAIC methodology is universal — it works for any process that produces outputs. The tools are adapted to each domain, but the problem-solving approach remains the same.
How long does it take to complete a Six Sigma project? A typical Six Sigma project takes three to six months from project charter to final control plan. Simpler projects may complete in two months. Complex projects involving significant process redesign or technology implementation may take nine to twelve months.
Statistical Process Control — Lean Manufacturing — Production Systems Design