Six Sigma: A Comprehensive Guide to Process Excellence

Sourav Ghosh
Operation Analyst
Classic Fashion Apparel Ltd, Jordan
Email: souravghosh090@gmail.com

What is Six Sigma?

Six Sigma is a structured methodology that equips organizations with the tools needed to enhance the efficiency of business processes. By reducing process variability and improving performance, Six Sigma fosters defect reduction boosts profitability, elevates employee morale, and enhances the quality of products and services. The term “Six Sigma quality” refers to well-controlled processes operating within ±3 standard deviations (±3σ) from the centerline on a control chart and within ±6 standard deviations (±6σ) from tolerance limits.

Key Benefits of Six Sigma

1. Improved Customer Satisfaction: Enhanced product and service quality leads to higher customer satisfaction and loyalty.
2. Cost Reduction: Identifying and eliminating inefficiencies reduces waste and operational costs.
3. Employee Engagement: Structured problem-solving methodologies empower employees and improve morale.
4. Data-Driven Decisions: Emphasizes using statistical data for objective and effective decision-making.
5. Increased Revenue: Streamlined processes and better quality products result in higher revenue and market share.

Different types of Six Sigma belt & their responsibilities

Six Sigma employs a hierarchical system of belts, each with specific roles:

• Master Black Belt: Trains and mentors Black Belts and Green Belts, sets strategic directions, develops metrics, and acts as the organization’s Six Sigma consultant.
• Black Belt: Leads problem-solving initiatives and coaches project teams.
• Green Belt: Supports data collection and analysis for Black Belt projects, and leads their own projects or teams.
• Yellow Belt: Participates in project teams and reviews process improvements.
• White Belt: Understands basic Six Sigma principles and may contribute to localized problem-solving efforts.

What are the differences between Lean & Six Sigma?

DMAIC Phases of Lean Six Sigma

1. Define Phase

A. Problem Statement: Clearly articulate the problem using metrics, facts, minimal jargon, and without suggesting solutions. Example:

• Poor: “We have too much inventory and need to implement JIT.”
• Better: “The inventory of unsold cars at ABC Motor Store in Amsterdam exceeds the target of 500 units, averaging 5000 units since 2015, causing space and cash flow issues. Reducing this to the target level is expected to save INR 360 Cr annually.”

B. Project Goal: A clear project goal is required to make a project successful. Otherwise, it will cause problems in different stages of the project.

• Why project goal
  • -Inform others
  • -Obtain approval
  • -Measure success
• The project goal should concentrate on a few measures, which are
  • -What?
  • -How much
  • -By when
• Example:
  • Poor: “Improve inventory problem at ABC Electronics.”
  • Better: “Reduce mobile phone inventory at ABC Electronics by 90% before January 10, 2025.”

C. Project Charter: Outline the project name, business case, goal, team members, scope, schedule, and sponsor details.

D. Critical Path Analysis: Identify tasks essential to project completion and their dependencies to streamline planning and execution.

2. Measure Phase

In a project, there will be many variables that can be measured on various scales.

• Use measurement scales:
  • A. Nominal: Categories without order (e.g., nationality).
  • B. Ordinal: Ordered categories (e.g., education level).
  • C. Parametric: Numeric data (e.g., income).

A. Nominal Measurement Scales

• Values are words instead of numbers.

• Values can’t be put in a low-high order.

• Binary variables are always measured on a nominal scale

Example –

• What is your Nationality?
• What is the Location of the factory?

B. Ordinal

• Values are words or number ranges.

• Values can be put in a low-high order.

Example –

• What is your satisfaction level with the hotel? Rate 1 to 10, where 1 is low & 10 is high.

• What is your highest education degree?

• Options are — No college degree / Bachelor degree / Master degree / PhD degree

C. Parametric

• Values are numbers.

• Values can be put in low-high order.

Example –

• What is your age? …

• What is your monthly net $ income?

Special note;

• Ensure data aligns with the normal distribution and meets 95% or 99% confidence levels.
• Data Collection Plan: Define what data to collect, how to collect it, and ensure it is reliable and valid.
• Process Mapping: Visualize the workflow to identify inefficiencies and opportunities for measurement.

D. Normal distribution & Confidence level

When we collect the data & try to extract something meaningful, we should put the data in normal distribution & that data should show a 95% or 99% confidence level So that it will help us to make some decisions based on the data.

The collected data need to be placed in an X-Y plot & get a bell-shaped curve. This bell shape curve is known as a normal distribution curve. The normal distribution curve data is then needed to justify based on the required sigma level.

Below, we show the normal distribution curve & how to make the required sigma level from it.

There are different sigma level

• Sigma level 2 (68% confidence)

• Sigma level 4 (95% confidence)

• Sigma level 6 (99% confidence)

Example

95% Confidence Interval

Sample 100 soldiers

Hours at the gym per week: 100

99% Confidence Interval

Sample 100 soldiers

Hours at the gym per week: 100

3. Analyze Phase

Aim: Find out the root case(s)

• A. Qualitative Tools:
  • 5 Why Analysis: A method to identify root causes by repeatedly asking “Why?”
• B. Quantitative Tools:
  • Regression Analysis: Examine relationships between variables.
  • Fishbone Diagram (Ishikawa): Categorize potential causes of a problem to pinpoint root issues.
  • Pareto Analysis: Focus on the most significant factors contributing to a problem, following the 80/20 rule.

A. Qualitative Tools

5 Why Analysis

“5 Why Analysis”, also known as “Why Analysis”

Method for Finding Root Cause(S) In the Analysis Phase of DMAIC

B. Quantitative Tools

Single OLS (Ordinary Least Squares regression) regression

• Quantify the effect of 1 independent variable (X) on 1 dependent variable (Y). Y = a + b*X

Multiple OLS regression

• Quantify the effect of several independent variables (Xs) on 1 dependent variable Y. Y = a + b1*X¹ + b2*X² …+ bn*Xⁿ

Single OLS –Mistakes = 11.6  – 0.05 * hours training

Single OLS –Mistakes = 7.84  – 0.63 * years experience

Multiple OLS –Mistakes = 10.72 – 0.28 * years experience – 0.04 * hours training

4. Improve Phase

5. Control Phase

Conclusion

Six Sigma is a powerful methodology that drives operational excellence by minimizing defects and process variability. By adopting the DMAIC framework and leveraging Lean principles, organizations can achieve faster, cost-effective, and high-quality outcomes. The systematic belt system ensures structured learning and expertise, enabling teams to tackle challenges with precision and strategic focus. As industries evolve, Six Sigma remains a cornerstone for continuous improvement and business success.

References

[1]https://asq.org/quality-resources/six-sigma?srsltid=AfmBOoq1yoVPv9Roq lSLptthAUyqNr6bYhKOh46mAfsG8_hLlqZYMEl

[2] https://asq.org/quality-resources/six-sigma/belts-executives-champions?srsltid=AfmBOopPnvs3ohMCCCd5Tn6P52HbXH7Mut-KRC0OudHge-HswgEzWJEo

[3] https://www.udemy.com/course/six-sigma-black-belt-including-ms-excel-application/learn/lecture/12826257#overview

About the Author: Sourav Ghosh is a graduate of Govt. College of Engg. & Textile Technology, Serampore, West Bengal. Where he pursued his Bachelor's degree in textile Engineering. He worked with prestigious companies like Arvind Mills ( Ahmedabad) & Coats India ( Bangalore) in fabric development. He completed his Masters degree from the National Institute of Fashion Technology (New Delhi). Currently, he works as an Operations Analyst at Classic Fashion Apparel Ltd (Jordan).