Predictive Surface Defect Detection in Particleboard Manufacturing using Defect Tracking Matrix–Principal Component Analysis Framework toward Zero Defect Manufacturing
1Department of Industrial and Systems Engineering, Institut Teknologi Sepuluh Nopember, Indonesia
2Departement of Industrial Engineering, Universitas Panca Marga, Indonesia
*Author to whom correspondence should be addressed:
E-mail: yustina.suhandini@upm.ac.id (YST)
E-mail: yustina.suhandini@upm.ac.id (YST)
Received: June 05, 2025 | Revised: August 14, 2025 | Accepted: December 17, 2025 | Published: December 2025
Abstract
Zero Defects Manufacturing (ZDM) is a proactive quality strategy aimed at preventing defects during production. This study proposes a novel integrated method using the Defect Tracking Matrix (DTM) and Principal Component Analysis (PCA) to predict the sources of surface defects in particleboard manufacturing. The authors evaluated twenty technical attributes and sixteen quality defects. Results showed that duct cleaning, setting blower, screen cleaning, press calibration, and blade sharpening were key contributors to detect patterns. The DTM-PCA framework improves traceability and helps implement ZDM through structured, data-driven analysis in a previously unexplored context.
Keywords
defect tracking matrix; particleboard industry; prediction; principal component analysis; quality control; zero defects manufacturing
Available Repositories
Share Article
Article Metrics
--
Views
--
Downloads
--
Citations
Export Citation
Full Text
