Digital Twins in Industrial Product Realization - A Literature Study

Abstract: Increased focus on sustainability in conjunction with larger demand for individualized and customized products are presenting the manufacturing industry with new challenges in production cycle management and sustainable processes. Digital Twin Technology is an emerging technology and could be an answer to these challenges. With wide-spread recognition in industry and academia, Digital Twin Technology can play a key role in enabling manufacturers to adapt to changes. M. Grieves first introduced the term “Digital Twin” during a course held at University of Michigan 2003, and described the Digital Twin as a high-fidelity real-time virtual representation of a product or a process (Grieves, 2014). Grieves stated that a Digital Twin consists of three main parts: 1) physical products in real space, 2) virtual products in virtual space, and 3) the connections between the physical and virtual. NASA later defined the Digital Twin more precisely as a “Multiphysics, multiscale, probabilistic, ultra-fidelity simulation that reflects, in a timely manner, the state of a corresponding twin based on the historical data, real-time sensor data, and physical model” (Glaessgen & Stargel, 2012). Research on Digital Twin Technology, and applications of Digital Twins in industry have demonstrated the potential of Digital Twins in industrial production, from design all through manufacturing and shipping. General Electric got promising results from applying Digital Twins to wind farms, while several independent research groups at various universities developed Digital Twins of different products and processes (GE Renewable Energy, 2016) (Vachálek, et al., 2017) (Tao & Zhang, 2017) (Biesinger, et al., 2018) (Moreno, et al., 2017). Nevertheless, there are still many challenges to overcome, one of the challenges being how to properly process the vast amount of data collected from sensors and actuators in industry, and another challenge being how to develop Digital Twins in a generic way (Tao, et al., 2019). Many articles have been written about how Digital Twins can answer these challenges in various ways, but I have found no article which brings the application of Digital Twins throughout the entire manufacturing process to light. This paper aims to do that, and presents the Digital Twin, in the context of the Design, Production Planning and Manufacturing phases of production. The latest findings in academia and industry function as a base for this literature study, and the focus is on applying Digital Twins throughout the manufacturing process to improve current methods, thus achieving more sustainable manufacturing. Several independent research groups found that application of Digital Twins in the industrial production process leads to better use of operational data and more informed, rational decisions (Biesinger, et al., 2018) (Hochhalter, et al., 2014). Vachálek et al. found that applying Digital Twins to a production line shortens and streamlines production cycles, and reduces the time to introduce new products and to detect inefficient settings of processes (Vachálek, et al., 2017). Tao et al. found that one fourth of the most relevant journal and conference articles about Digital Twin Technology published from 2003 to 2018 reported the application of Digital Twins in the context of Product Health Management, which is considerably more than any other area (Tao, et al., 2019). However, in this literature study I have found that the application of Digital Twins in early design phases is promising. The Digital Twin can make data interaction more intuitive for the designer and ensure that customer needs are taken into consideration when developing the next generation of products (Tao, et al., 2019). Moreover, compliance verification can be performed on the Digital Twin instead of on the physical counterpart, resulting in less tedious testing and shorter time-to-market. This enables companies that deploy Digital Twins to be more flexible and receptive to changing circumstances. General Electric is one of the companies who have deployed Digital Twins, and consequently reported a 16% increase in annual energy production at one of their customer’s wind farm through the use of Digital Twins in their Digital Wind Farm initiative (GE Renewable Energy, 2016). Lastly, in this paper I present the current challenges that Digital Twins as a research field and technology is facing, and what data related issues are currently limiting the advancement of Digital Twins.