Apparel Quality Management-II

Title: Applying Lean Six Sigma to Improve Garment

Production Processes
A Case Study on Waste Reduction and Efficiency
Mid-Term Assignment

Submitted to: Dr. Jaikishan Sambharia

Submitted by: Shambhavi P Kendole
 Introduction Results and
Discussion
Content:
Table of

Problem statement & Conclusion

Objectives

Methods Used in References

Study

RESEARCH
METHODS
Introduction
The fashion industry is highly competitive and quality-driven.
Customers expect products with zero defects and high consistency.
Garment industries often face issues like:
[Link] stitching
[Link] finishing
[Link] issues
[Link] Garment operates on a made-to-order system, but still faces quality problems.
Problem Statement
The garment production process at Jhono Garment is experiencing a significant
number of product defects, leading to increased costs, time losses, and reduced
product quality, indicating the need for effective quality improvement methods

Objectives
To determine defect levels and sigma value of garment products
To identify major waste factors in production
To analyze causes of product defects
To provide improvement recommendations
To enhance overall product quality
Methods Used in Study
The study uses:
Lean Six Sigma →Reduce waste + defects
FMEA (Failure Mode Effect Analysis)→ Identify risks
Fuzzy Logic → Improve decision accuracy
These methods are combined to systematically identify problems and prioritize improvements.
Research Methods(DMAIC Approach)
1. Define
Identify the problem areas in production
Select Critical to Quality (CTQ) factors (Visual, Size, Stitching, Neatness)
Map the production process (OPC)
Focus: What needs improvement?

2. Measure
Calculate defect levels
Use control charts to check process stability
Identify seven types of waste
Compute sigma level (process performance)
Focus: How big is the problem?

3. Analyse
Find root causes of defects using:
Fishbone Diagram
FMEA (risk analysis)
Research Methods (DMAIC Approach)
Apply Fuzzy logic to prioritize problems more accurately
Focus: Why is the problem happening?

4. Improve
Suggest solutions using 5W1H (What, Why, Where, When, Who, How)
Example: Improve folding process to reduce defects
Focus: How to fix the problem?

5. Control
Monitor improvements
Ensure process stays stable and consistent
Maintain quality over time
Focus: How to sustain improvement?
Results and Discussion

1. Define Phase
Operational Process Chart (OPC)
Jhono Garment produces T-shirts, polo shirts, and
sleepwear
The production process for all three is almost
similar
Only difference:
Polo shirts include an additional button attachment
step
OPC is used to map the complete production flow

OPC T-Shirt
Critical to Quality (CTQ) – Brief
Identified through owner interview and
customer requirements
4 key quality factors:
Visual (appearance)
Size (accuracy)
Stitching (quality of seams)
Neatness (finishing & packaging)

CTQ Tree
2. Measure Phase
1. Control Chart
This study uses Minitab 18 software to construct control charts, specifically P-charts, as
illustrated in Figures 3–5

[Link] Chart T-shirt [Link] Chart Polo

 3. Control Chart Night dress

As shown in Figure 1, only T-shirts are out of control, so the analysis is focused on T-shirt production.
2. Identify Seven Waste
Based on the OPC for T-shirt products, the production process is described in Table
 The results of value stream mapping show that the Value Added
Ratio (VAR) is 73%. Waste identification was carried out through
observation and interviews with the owner and production
workers, and the seven types of waste are presented in Table

Identify seven waste

Process Chart T-Shirts
1. Overproduction 2. Inventory
Producing more than required Excess raw material or finished goods
Result: Not found Result: Not found
Because company follows made-to-order Materials are used as per order, no
system storage issue

3. Defects 4. Transportation
Faulty or damaged products Unnecessary movement of materials
Result: Found (major waste) Result: Found
Causes: Issue:
Rework cost (₹5,000–₹10,000 per defect) Warehouse is 5 meters away,
Time loss (10–30 minutes) increasing time

5. Motion 6. Waiting
Unnecessary movement of workers Idle time between processes
Result: Not found Result: Not found
Workers’ movements are efficient Parallel production system avoids
delays
 7. Overprocessing
Extra or unnecessary processing
Result: Found (minor)
Some processes are slightly inefficient

Among the 7 wastes, defects were the most critical, followed by transportation and overprocessing,
while others were minimal due to the made-to-order system

3. Sigma Value Calculation

Sigma value was calculated using DPU and DPMO to measure the defect level and overall process
efficiency
Sigma value is used to measure process quality and performance
It shows how many defects occur in production
Steps Involved:
DPU (Defects per Unit)Measures defects per product
DPMO (Defects per Million Opportunities)Standardizes defects per million units
Sigma ValueIndicates overall process capability
Results from Study:
DPU = 0.00610
DPMO = 6,097
Sigma Value = 4.006

Around 6,097 defects per 1,000,000 products

Process is at moderate quality level (4 Sigma)
3. Analyze Phase
Purpose: To find the root cause of defects, not just the symptoms

1. Defects Identified
From data analysis, the main defects in T-shirts were:
Messy stitching
Torn fabric
Dirty garments
Messy packaging
These defects directly affect quality and customer satisfaction

2. Root Cause Analysis

The study used a Fishbone (Cause & Effect) Diagram to identify causes.
Causes were grouped into 5 categories:
[Link]
[Link]
Lack of experience
No proper folding method
Carelessness
No standard procedure
Fatigue

[Link] [Link]
Old machines Poor lighting
Poor maintenance Narrow workspace
Dust
[Link]
Poor quality raw material (sometimes from
customers)
3. FMEA (Failure Mode Effect Analysis)
Each defect was evaluated based on:
→
Severity (S) How serious the defect is
Occurrence (O)→ How often it happens
→
Detection (D) How easily it can be detected
Formula:
RPN = severity × occurrence × detection.
Helps to:
Rank defects based on risk
[Link]
The Fuzzy FMEA (Failure Mode and Effect Analysis) is a critical part of the Analyze phase in
Lean Six Sigma, specifically used to prioritize risks by removing the ambiguity often found
in traditional RPN (Risk Priority Number) scores. While traditional FMEA relies on a simple
multiplication of Severity, Occurrence, and Detection, Fuzzy FMEA uses mathematical
logic to better reflect expert judgment.

1. The Core Components: SOD Inputs

The process begins by scoring each failure mode on a scale of 1–10 across three variables:
Severity (S): The impact of the defect on the customer.
Occurrence (O): The frequency with which the failure happens.
Detection (D): The likelihood of catching the defect before it reaches the customer
2. Fuzzy Membership Functions
Instead of treating a score (like "7") as a single fixed point, Fuzzy FMEA treats it as a range
defined by Membership Functions. This study uses two types of curves to map these values:
Triangular Curve: Used for Low (L) and High (H) categories.
Trapezoidal Curve: Used for Very Low (VL), Moderate (M), and Very High (VH) categories.

Example: Mapping a Score of 7 (High)

For an occurrence score of 7 (Category H), the fuzzy formula used is:
This resulting value (0.75) represents the degree of membership in that category.

3. Calculating the Fuzzy Risk Priority Number (FRPN)

The FRPN is determined through a three-step logical process:
Fuzzification: Converting the crisp SOD scores into fuzzy sets using the curves above.
Fuzzy Inference (Rules): Experts define "If-Then" rules (e.g., "If Severity is VH AND
Occurrence is VH AND Detection is M, then FRPN is VH").
Defuzzification: Converting the fuzzy result back into a single priority number (FRPN)
between 0 and 1000.
4. Comparison: RPN vs. FRPN
The most detailed part of the Analyze phase is seeing how Fuzzy Logic changes the priority
of defects.
Detailed Priority Table (T-Shirt Case Study)
The comparison of RPN and FRPN shows
some differences in ranking, but both
methods identified D3 (untidy packaging)
as the top priority for improvement
4. Improve Phase
The primary goal of this stage is to address the root causes of defects identified in the
earlier FMEA analysis (Table 10) by proposing targeted solutions.

1. 5W1H Analysis Framework

To determine the specific factors causing defects in T-shirt production, the 5W1H tool
(What, Who, Where, When, Why, and How) was applied.

2. Target Defect: Messy Packaging

The analysis focused on the Packaging Workstation during the folding process. Four key
root causes and their corresponding solutions were identified:
Root Cause (Why) Proposed Solution (How)

Rush: Workers rushing through Supervision: Increase oversight of the

folding. folding process.

Human Error: Mixing clothing sizes Standardization: Separate plastic

with wrong plastic bags. sizes and apply clear labels.

Inconsistency: Workers rely on Tooling: Use folding molds for

"instinct" to fold. uniform and neat results.

Resource Allocation: Increase the

High Workload: Negligence due to
number of workers in the folding
overcapacity.
section.
5. Control Phase
The Control Phase ensures that the improvements made (the folding molds) are effective and
sustainable over time.
1. Implementation of Folding Molds
Based on the Fuzzy RPN analysis, the highest priority for repair was identified as untidy packing
(D3).
Action: Researchers introduced folding molding tools to replace manual "instinct-based" folding.

2. Results: Before vs. After

Messy Packaging
Period Condition Production Amount
Defects

May 29 – June 2 Manual Folding 360 units 8 Defects

June 5 – June 9 With Folding Mold 360 units 0 Defects

Before the use of folding molds, 8 packaging defects were recorded during the production
period of May 29 to June 2. After implementing folding molds from June 5 to June 9, no defects
were observed. This demonstrates that folding molds are highly effective in reducing untidy
packaging defects at Jhono Garment.
 Conclusion

The sigma value for Jhono Garment products was determined using control charts, where T-shirts were
identified as exceeding the control limits. The analysis shows that the defect rate reaches 6,097 defects
per 1,000,000 units, corresponding to a sigma level of 4.006, indicating moderate process performance.
Among the seven wastes, defects were identified as the most dominant, occurring most frequently in the
production process. Each defective product requires a repair cost ranging from IDR 5,000 to IDR 10,000,
with a repair time of approximately 10 to 30 minutes, depending on the defect type.
To address this issue, an improvement solution was proposed by introducing folding molds in the product
folding process. The implementation of this solution resulted in a 100% reduction in untidy packaging
defects during a one-period production trial, proving its effectiveness in improving product quality at Jhono
Garment.
However, the study has limitations, as it does not include a comprehensive waste analysis. Therefore,
future research is recommended to use the Waste Assessment Model (WMA) for better identification of
waste, supported by tools such as the Waste Relationship Matrix (WRM) and Waste Assessment
Questionnaire (WAQ).
Additionally, the Fuzzy FMEA calculations in this study were performed manually. For more accurate and
efficient analysis in future studies, the use of software such as MATLAB is recommended.
References
[Link]
Thank
You!