Title: Machine learning approaches towards digital twin development for machining systems

Authors: Krzysztof Jarosz; Tuğrul Özel

Addresses: Department of Industrial and Systems Engineering, Manufacturing and Automation Research Laboratory, Rutgers University, Piscataway, New Jersey 08854, USA ' Department of Industrial and Systems Engineering, Manufacturing and Automation Research Laboratory, Rutgers University, Piscataway, New Jersey 08854, USA

Abstract: Machine learning (ML) and artificial intelligence (AI) have experienced an increased degree of applications associated with Industry 4.0. Their effective utilisation is elevated with readily available computational power and computerisation of production processes toward digital twin development. This paper begins with a review of the use of ML and AI Methods in machining applications, using examples from open literature, discussing the future perspectives for further utilisation of ML and AI techniques within the scope of machining, both in terms of research and industrial applications. Examples of computer-aided production (CAP) systems are presented and compared with a discussion on how ML and AI can be applied to improve applicability and performance of already established software solutions. Additionally, a software solution for numerically controlled (NC) toolpath optimisation is shortly presented. Finally, incorporation of machine learning method in a CAE software solution developed by the authors is discussed along with a case study.

Keywords: virtual machining; production; advanced manufacturing system; CNC; computer numerical control; machine learning; artificial intelligence; digital twin; Industry 4.0.

DOI: 10.1504/IJMMS.2022.124922

International Journal of Mechatronics and Manufacturing Systems, 2022 Vol.15 No.2/3, pp.127 - 148

Received: 31 Aug 2021
Accepted: 21 Mar 2022
Published online: 16 Aug 2022 *