A knowledge-based Digital Shadow for machining industry in a Digital Twin perspective

• Published in: Journal of manufacturing systems Vol. 58; pp. 168 - 179
• Main Authors: Ladj, Asma, Wang, Zhiqiang, Meski, Oussama, Belkadi, Farouk, Ritou, Mathieu, Da Cunha, Catherine
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.01.2021; Elsevier
• Subjects: Data and knowledge management; Digital shadow; Digital twin; Engineering Sciences; Machining; Other; Data and knowledge management; Digital twin; Digital shadow; Machining; digital twin; machining; data and knowledge management; digital shadow
• ISSN: 0278-6125; 1878-6642
• DOI: 10.1016/j.jmsy.2020.07.018

•The Digital Shadow, as a core component of Digital Twin when there is complex or big data•Combining data analytics and knowledge management•Illustrated and validated on an industrial use-case from the aeronautic machining industry This paper addresses the problems of data management and analytics for decision-aid by proposing a new vision of Digital Shadow (DS) which would be considered as the core component of a future Digital Twin. Knowledge generated by experts and artificial intelligence, is transformed into formal business rules and integrated into the DS to enable the characterization of the real behavior of the physical system throughout its operation stage. This behavior model is continuously enriched by direct or derived learning, in order to improve the digital twin. The proposed DS relies on data analytics (based on unsupervised learning) and on a knowledge inference engine. It enables the incidents to be detected and it is also able to decipher its operational context. An example of this application in the aeronautic machining industry is provided to stress both the feasibility of the proposition and its potential impact on shop floor performance.