articleSix Sigma

Why Sigma?

Avatar photo By John Knotts Updated on August 19th, 2024

I was asked a question that I had never been asked before: “Why Sigma: Why is only Sigma in Lean Six Sigma Yellow and Green & Black belt? Why not use Alpha, etc…?”

Honestly, no one had ever asked me that question before. So, I did some research and thought I would share it on a Gemba Academy blog.

Why Sigma?

The choice of “Sigma” in the Six Sigma program is rooted in its statistical significance rather than its position in the Greek alphabet. Sigma (σ) is a statistical term that represents the standard deviation of a process or set of data. Standard deviation measures the amount of variation or dispersion from the mean (average). In the context of Six Sigma, a high sigma value means a large variation, while a low sigma value indicates that data points are closely clustered around the mean.

The Six Sigma methodology aims to improve processes by reducing variation and defects. By focusing on Sigma, the program emphasizes the importance of achieving consistent, high-quality outputs. A process that operates at “Six Sigma” quality produces only 3.4 defects per million opportunities, indicating an extremely high level of performance. Using Sigma provides a precise, quantitative measure of process capability and quality. This precision allows businesses to set clear, measurable goals for improvement and to track progress objectively.

Sigma is a universally recognized symbol in statistics and quality control. Its use in Six Sigma leverages this broad recognition and the established body of knowledge around statistical process control. Motorola popularized the term “Six Sigma” in the 1980s as part of its quality improvement initiatives. The choice of Sigma aligns with its focus on statistical analysis and data-driven decision-making.

Why Not Alpha, etc…?

Choosing “Alpha” or another Greek letter would not have conveyed the same statistical meaning and precision as Sigma, making it less suitable for a program centered on reducing variation and improving quality through statistical methods. If the Six Sigma program had been named using “Alpha” instead of “Sigma,” the conceptual framework and focus of the methodology would likely be quite different. Alpha often denotes the level of significance in hypothesis testing, representing the probability of rejecting a true null hypothesis (Type I error). A methodology named “Six Alpha” might emphasize minimizing Type I errors in quality control and decision-making processes.  The methodology might concentrate on improving decision accuracy and reducing false positives in process assessments. The goal could be to achieve a high level of confidence in quality control decisions.

Instead of focusing on standard deviation and process variation, the metrics might involve confidence intervals, p-values, and other statistical measures associated with hypothesis testing and decision accuracy. The goals might include reducing the frequency of incorrect decisions about process quality or improvements and ensuring that processes are not incorrectly deemed to meet quality standards when they do not. The approach might involve more rigorous testing and validation of processes to ensure that any observed improvements are statistically significant and not due to random variation.

Alpha-related Changes:
  • Training programs might focus on statistical hypothesis testing, confidence level assessments, and methods to reduce Type I and Type II errors (the latter representing the probability of not rejecting a false null hypothesis).
  • The DMAIC (Define, Measure, Analyze, Improve, Control) framework might be adjusted to incorporate steps specifically aimed at hypothesis testing and validation, such as formulating and testing hypotheses about process improvements.
  • The terminology would likely shift from terms like “defects” and “variation” to terms like “error rates” and “confidence levels.”

In summary, a quality improvement methodology based on “Alpha” would likely focus on the accuracy and reliability of decision-making within processes, rather than directly on reducing process variation and defects as in Six Sigma.

I hope this helps you in conveying this program statistically to people.

Aside from what we have across Gemba Academy, here are some helpful references that provide more detailed explanations and context for the points discussed above:

  • “Introduction to Statistical Quality Control” by Douglas C. Montgomery. This textbook provides a comprehensive overview of statistical methods used in quality control, including Six Sigma.
  • The American Society for Quality (ASQ) website offers resources and articles explaining Six Sigma principles: ASQ Six Sigma.
  • “The Six Sigma Handbook” by Thomas Pyzdek and Paul Keller. This book is a key resource for understanding the Six Sigma methodology, metrics, and tools.
  • Motorola Solutions’ historical context and application of Six Sigma: Motorola Six Sigma.
  • “Statistics for Business and Economics” by Paul Newbold, William L. Carlson, and Betty Thorne. This textbook covers hypothesis testing, significance levels, and confidence intervals.
  • Khan Academy’s resources on hypothesis testing: Khan Academy Hypothesis Testing.
  • “Quality Control and Improvement” by Amitava Mitra. This book provides insights into quality control techniques, including statistical process control and hypothesis testing.
  • NIST/SEMATECH e-Handbook of Statistical Methods, which covers various statistical methods and their applications in quality control: NIST/SEMATECH Handbook.
  • “Lean Six Sigma: Combining Six Sigma Quality with Lean Speed” by Michael L. George. This book discusses Six Sigma training and certification.
  • The International Association for Six Sigma Certification (IASSC) provides information on Six Sigma certification levels and training: IASSC.

Have something to say?

Leave your comment and let's talk!

Start your improvement training today.