Intelligent defect management system for porcelain industry through cyber-physical systems

This paper presents an intelligent defect management system based on cyber-physical systems that orchestrate computational resources with physical systems in the specific environment of porcelain manufacturing. Sensors and computer vision techniques detect defects and faults, an intelligent information processing system based on ontologies and deep learning methods quantify different types of defects in porcelain products and provide information to the decision making system and actuators in the workflow. The system is developed within a project in collaboration with one of the biggest porcelain ware producer in Europe. It focuses on defect causation analysis and provide an advanced management system aiming to facilitate defect measures and rectifications by: (1) collecting and classifying defect data; (2) identifying causations of defect and analyzing its impact; (3) searching and managing defect information by means of knowledge management (KM) techniques, namely ontologies; and (4) developing agile defect control of porcelain ware. The final goal and result of the project is to achieve an intelligent and agile manufacturing system.