P‐44: Control Chart Pattern Recognition Using Preprocessing Based on DTW and 1D‐CNN for Anomaly Equipment Detection

This study proposes a Control Chart Pattern Recognition (CCPR) model that integrates a Dynamic Time Warping (DTW)‐based preprocessing method and a 1D‐Convolutional Neural Network (1D‐CNN). The model enhances anomaly detection by identifying unique patterns that deviate from similar equipment's behavior. Experimental results demonstrate its effectiveness in handling insufficient time‐series data, achieving high classification accuracy, and reducing unnecessary alerts. This SPC‐AI approach improves operational efficiency and reliability in complex production processes, such as display manufacturing.