Enhanced Anomaly Detection with Optimized Deep Learning in Manufacturing System
Paper Title: Enhanced Anomaly Detection with Optimized Deep Learning in Manufacturing System
Journal Name: ICT Express
Abstract: In Manufacturing Execution Systems (MES), identifying irregularities, defects, or unforeseen variations in production metrics, sensor readings, and equipment performance holds critical significance. Various solutions have emerged to address this challenge, leveraging diverse models, including machine learning, deep learning, and neural network models. This research is centered on assessing the efficacy of these models in anomaly detection, with a particular focus on the Optimized-Visual-Geometry-Group named (OVGG16) model, celebrated for its remarkable success rate of 99%. Beyond its outstanding accuracy, the OVGG16 model distinguishes itself with unparalleled anomaly detection capabilities, making substantial strides in industrial quality control. This investigation also explores inventive approaches to anomaly detection in MES, highlighting the pivotal role of anomaly detection rates in evaluating the performance of these models.