Past, present, and future research of digital twin for smart manufacturing

Abstract

In the era of the Fourth Industrial Revolution, there is a growing focus on digital twin (DT) in order to advance toward smart manufacturing. Thus, researchers have conducted numerous studies on DT and extensively developed related technologies. There are many studies that apply and analyse DT to actual manufacturing sites for the realization of a smart factory, but it is necessary to clearly consider which part of DT is applied and what function it performs in manufacturing. As such, this study analysed and classified prior literature based on various phases of product lifecycle management, an application field of DT in manufacturing, and the hierarchy level axis of Reference Architecture Model Industry 4.0, the target scope of DT. Accordingly, this study identified research trends in the past and present as well as analysed and identified the major functions of DT (prototyping, pilot testing, monitoring, improvement, and control). Through a gab study on the inadequate aspects of past and present researches, this study proposes directions for future studies on DT and a system architecture that can perform all the functions of DT.

Region‐based fully convolutional networks with deformable convolution and attention fusion for steel surface defect detection in industrial Internet of Things

“…In sixth‐generation (6G) networks‐enabled of industrial Internet of Things (IIoT) [1, 2], many emerging techniques, such as artificial intelligence (AI) [3], deep learning [4], are applied in the industrial applications to improve smart manufacturing [5, 6]. Steel surface defect detection has recently received increasing attention from the military industry, aviation industry and other fields, which is closely related to the quality of industrial production products [7].…”

Digital Twins for the Designs of Systems: a Perspective

“…A specific concept that has received increased attention from the scientific community is the "digital twin". The digital twin concept includes the perspective that a set of physics-based simulation models can be used to model the state of a system, and be combined with dynamically collected data to provide operation support [2,3]. Examples of where the digital twin concept has been applied can be found in aerospace, manufacturing, and automotive engineering [4,5].…”