Statistical Quality Control Calculations for Textile Manufacturing

Statistical Quality Control (SQC) is vital for maintaining high standards in textile manufacturing. This article presents key SQC calculations, including defect rate, coefficient of variation, standard deviation, process capability (Cp and Cpk), and control chart limits, tailored for textile processes like spinning, weaving, and dyeing. Each calculation is accompanied by formulas, practical examples, and references to standards such as ASTM and ISO. These tools enable manufacturers to monitor process variability, reduce defects, and ensure product reliability, enhancing overall production efficiency.

1. Introduction to Statistical Quality Control in Textiles

Statistical Quality Control (SQC) employs statistical methods to monitor and control quality in textile manufacturing. By analyzing data from processes such as yarn spinning, fabric production, and finishing, SQC ensures consistency, minimizes defects, and optimizes production. This article details essential SQC calculations, providing textile professionals with tools to assess and improve quality.

2. Key Statistical Quality Control Calculations

2.1 Defect Rate

Purpose: Measures the frequency of defects in textile products (e.g., yarn imperfections, fabric flaws) to assess quality.

$$\text{Defect Rate (\%)} = \frac{\text{Number of Defects}}{\text{Total Inspected Units}} \times 100$$

Example: For 25 defects in 1000 meters of fabric: Defect Rate = (25 / 1000) × 100 = 2.5%

Reference: ASTM D123-17, Terminology Relating to Textiles

2.2 Coefficient of Variation (CV%)

Purpose: Quantifies variability in textile parameters (e.g., yarn weight, fabric thickness) relative to the mean.

$$\text{CV\%} = \frac{\sigma_w}{\text{Mean}} \times 100$$

Where:

• σ_w = Standard deviation of the parameter
• Mean = Average value of the parameter

Example: For yarn weights [200, 202, 198, 201, 199] g, Mean = 200 g, σ_w = √((0² + 2² + (-2)² + 1² + (-1)²) / 5) ≈ 1.414 g: CV% = (1.414 / 200) × 100 ≈ 0.707%

Reference: Uster Statistics

2.3 Standard Deviation of Fabric Weight

Purpose: Measures the variability in fabric weight to ensure uniformity.

$$\sigma_w = \sqrt{\frac{\Sigma (W_i – \text{Mean})^2}{N}}$$

Where:

• W_i = Individual weight measurement
• N = Number of samples

Example: For fabric weights [150, 152, 148, 151, 149] g/m², Mean = 150 g/m²: σ_w = √((0² + 2² + (-2)² + 1² + (-1)²) / 5) ≈ 1.414 g/m²

Reference: ASTM D3776-20

2.4 Process Capability Index (Cp)

Purpose: Assesses whether a textile process (e.g., yarn spinning) meets specification limits.

$$C_p = \frac{\text{USL} – \text{LSL}}{6 \times \sigma_w}$$

Where:

• USL = Upper Specification Limit
• LSL = Lower Specification Limit
• σ_w = Process standard deviation

Example: For yarn count with USL = 22 Ne, LSL = 18 Ne, σ_w = 0.5 Ne: C_p = (22 – 18) / (6 × 0.5) = 4 / 3 ≈ 1.333

Reference: ISO 9001:2015

2.5 Process Capability Index (Cpk)

Purpose: Evaluates process capability while considering process centering.

$$C_{pk} = \text{min}\left(\frac{\text{USL} – \text{Mean}}{3 \times \sigma_w}, \frac{\text{Mean} – \text{LSL}}{3 \times \sigma_w}\right)$$

Example: For Mean = 20 Ne, USL = 22 Ne, LSL = 18 Ne, σ_w = 0.5 Ne: C_pk = min((22 – 20) / (3 × 0.5), (20 – 18) / (3 × 0.5)) = min(2 / 1.5, 2 / 1.5) ≈ 1.333

2.6 Control Chart Limits for X-Bar Chart

Purpose: Establishes control limits to monitor process stability (e.g., fabric weight consistency).

Upper Control Limit (UCL):

$$\text{UCL} = \text{Mean} + 3 \times \frac{\sigma_w}{\sqrt{n}}$$

Lower Control Limit (LCL):

$$\text{LCL} = \text{Mean} – 3 \times \frac{\sigma_w}{\sqrt{n}}$$

Where:

• n = Sample size

Example: For fabric weight Mean = 150 g/m², σ_w = 1.414 g/m², n = 5: UCL = 150 + 3 × (1.414 / √5) ≈ 151.897 g/m²; LCL = 150 – 3 × (1.414 / √5) ≈ 148.103 g/m²

Reference: ASTM E2587-16

2.7 Range Chart Limits

Purpose: Monitors variability within samples using range (R) values.

Upper Control Limit (UCL_R):

$$\text{UCL}_R = \bar{R} \times D_4$$

Lower Control Limit (LCL_R):

$$\text{LCL}_R = \bar{R} \times D_3$$

Where:

• R̄ = Average range of samples
• D_3, D_4 = Control chart constants (e.g., for n = 5, D_3 = 0, D_4 = 2.114)

Example: For ranges [4, 3, 5, 2, 4], R̄ = (4 + 3 + 5 + 2 + 4) / 5 = 3.6 g/m²: UCL_R = 3.6 × 2.114 ≈ 7.610 g/m²; LCL_R = 3.6 × 0 = 0 g/m²

2.8 Yarn Imperfection Index

Purpose: Quantifies yarn defects (thick places, thin places, neps) per kilometer for quality assessment.

$$\text{II} = N_{\text{thick}} + N_{\text{thin}} + N_{\text{neps}}$$

Example: For 50 thick places, 30 thin places, 20 neps per km: II = 50 + 30 + 20 = 100 imperfections/km

Reference: Uster Statistics

3. Practical Applications and Examples

3.1 Yarn Count Control

For a cotton yarn production with:

• Target Ne = 20
• Sample measurements: [19.8, 20.2, 19.9, 20.1, 20.0] Ne
• USL = 21 Ne, LSL = 19 Ne

Mean and Standard Deviation:

Mean = (19.8 + 20.2 + 19.9 + 20.1 + 20.0) / 5 = 20 Ne

σ_w = √(((19.8 – 20)² + (20.2 – 20)² + (19.9 – 20)² + (20.1 – 20)² + (20.0 – 20)²) / 5) ≈ 0.1414 Ne

CV%:

$$\text{CV\%} = \frac{0.1414}{20} \times 100$$

CV% ≈ 0.707%

Process Capability:

$$C_p = \frac{21 – 19}{6 \times 0.1414}$$

C_p = 2 / 0.8484 ≈ 2.357

$$C_{pk} = \text{min}\left(\frac{21 – 20}{3 \times 0.1414}, \frac{20 – 19}{3 \times 0.1414}\right)$$

C_pk = min(1 / 0.4242, 1 / 0.4242) ≈ 2.357

3.2 Fabric Weight Control Chart

For fabric weight samples (5 samples, 4 measurements each):

• Sample means: [150, 151, 149, 150.5, 149.5] g/m²
• Sample ranges: [4, 3, 5, 2, 4] g/m²

X-Bar Chart:

Grand Mean = (150 + 151 + 149 + 150.5 + 149.5) / 5 = 150 g/m²

σ_w = R̄ / d_2 (d_2 = 2.059 for n = 4): R̄ = (4 + 3 + 5 + 2 + 4) / 5 = 3.6 g/m²; σ_w = 3.6 / 2.059 ≈ 1.748 g/m²

$$\text{UCL} = 150 + 3 \times \frac{1.748}{\sqrt{4}}$$

UCL = 150 + 3 × (1.748 / 2) ≈ 152.622 g/m²

$$\text{LCL} = 150 – 3 \times \frac{1.748}{\sqrt{4}}$$

LCL = 150 – 3 × (1.748 / 2) ≈ 147.378 g/m²

Range Chart:

$$\text{UCL}_R = 3.6 \times 2.282$$

UCL_R = 3.6 × 2.282 ≈ 8.215 g/m²

$$\text{LCL}_R = 3.6 \times 0$$

LCL_R = 0 g/m² (for n = 4, D_3 = 0)

4. Summary Table of Key SQC Calculations