Enhancing Fabric Defect Detection for Everest Textile Co., Ltd.
About the Customer
Everest Textile Co., Ltd. is a prominent textile manufacturer based in Taiwan, renowned for its commitment to high-quality products and sustainable practices. With a reputation for excellence, the company sought to overcome challenges in fabric defect detection to enhance product quality and operational efficiency.
- Company Name: Everest Textile Co., Ltd.
- Industry: Textile Manufacturing
- Headquarters: Taiwan
- Established: 1995
- Core Specializations: Fabric production for apparel, home textiles, and industrial applications
- Mission: To deliver exceptional textile products through innovative technology and sustainable practices.
- Vision: To be a global leader in the textile industry known for quality, innovation, and sustainability.
Challenges
Everest Textile Co., Ltd. faced several significant challenges in their fabric defect detection process:
- High Defect Rates:
- Challenge: The company experienced elevated fabric defect rates, including discoloration, irregular texture, and structural flaws.
- Impact: High defect rates resulted in increased rework, waste, and customer dissatisfaction, affecting overall product quality and operational costs.
- Manual Inspection Process:
- Challenge: Fabric inspection was performed manually by quality control inspectors who visually checked for defects.
- Impact: The manual process was time-consuming, prone to human error, and inconsistent, leading to missed defects and quality issues.
- Inefficiency in Defect Identification:
- Challenge: The existing process lacked the capability to swiftly and accurately identify and classify defects.
- Impact: Inefficient defect identification led to delays, increased production costs, and reduced customer satisfaction.
- Limited Automation:
- Challenge: There was minimal automation in the defect detection process, limiting scalability and consistent quality maintenance.
- Impact: The lack of automation contributed to slower production cycles and higher operational costs.
Project Objectives
To address these challenges, the following objectives were defined for the project:
- Develop an Advanced Fabric Defect Detection System:
- Objective: Implement a sophisticated solution using machine learning and computer vision technologies to automate and enhance defect detection.
- Implement Real-Time Defect Detection:
- Objective: Enable real-time analysis of fabric rolls to promptly identify and classify defects.
- Improve Quality Control:
- Objective: Enhance quality control processes through automation and data-driven insights.
- Streamline Operations:
- Objective: Automate workflows and provide actionable insights to improve overall operational efficiency.
Approach
Flivo AI employed the following approach to develop and implement the fabric defect detection system:
- Data Consolidation:
- Action: Collected and integrated fabric images and defect data into a centralized repository.
- Goal: Provide comprehensive visibility into fabric quality and defect patterns.
- Machine Learning Model Development:
- Action: Developed a custom machine learning model to identify and classify various fabric defects.
- Goal: Enhance accuracy in defect detection and reduce manual inspection errors.
- Computer Vision Integration:
- Action: Installed high-resolution cameras and integrated computer vision technology to capture and analyze fabric surfaces in real-time.
- Goal: Automate defect detection and classification during production.
- Automated Defect Detection System:
- Action: Implemented an automated system for real-time defect analysis, including alerts and severity classification.
- Goal: Improve defect identification speed and accuracy.
- Quality Control Dashboard:
- Action: Developed a user-friendly dashboard to monitor defect detection results and generate reports.
- Goal: Provide real-time insights and facilitate data-driven decision-making.
- Training and Support:
- Action: Provided training to the quality control team and ongoing technical support.
- Goal: Ensure effective use of the new system and address any issues promptly.
Project Highlights
- Timeline: 12 weeks (approximately 3 months) from project kickoff to application production.
- Team: A small, agile team of app developers and data engineers.
- Scope: Initial implementation at 7 global sites.
- Data Volume: Monitored and analyzed data from 80,000 suppliers and 3 million item details.
- Integration: Data from 10 different systems, including common ERPs.
Results
15-20%
reduction opportunity across sourcing costs.
300+
global facilities implemented the system
100%
Significant improvements in defect detection
Benefits
- Improved Product Quality:
- Benefit: Enhanced defect detection accuracy, leading to higher product quality and reduced defect rates.
- Increased Operational Efficiency:
- Benefit: Streamlined defect detection process through automation, reducing inspection time and costs.
- Enhanced Quality Control:
- Benefit: Better oversight and control over fabric quality through real-time monitoring and actionable insights.
- Competitive Advantage:
- Benefit: Strengthened market position by adopting advanced technology and improving product quality.
Proven results in weeks, not years
Exec. Briefing
Get insights into Flivo AI’s capabilities, enterprise AI best practices, and highest-value use cases.
Exec. Briefing
2 Hours
Technology
Assessment
Get insights into Flivo AI’s capabilities, enterprise AI best practices, and highest-value use cases.
Technology
Assessment
2-3 Days
Production
Trial
Get insights into Flivo AI’s capabilities, enterprise AI best practices, and highest-value use cases.
Production
Trial
8-12 Weeks
AI Application
Deployment in Production
Get insights into Flivo AI’s capabilities, enterprise AI best practices, and highest-value use cases.
AI Application
Deployment in Production
3-6 Months
