Digital twin technology evolved initially for production and engineering with Industry 4.0. But now it has entered into various fields like healthcare, aviation, etc. The digital twin models the state of a physical entity or process. The DT continuously predicts the status of defects or failures by closed loop optimization. Due to big data storage capabilities, data fusion algorithms, and artificial intelligence algorithms, DTs are capable of assessing any hidden patterns or unknown correlations, and with self-healing mechanisms, predictive maintenance approach is used to overcome them. In the healthcare applications, DTs of patients and medical devices virtually replicate the patient's physical characteristics or changes in body for correct diagnosis and treatment.
TopIntroduction
With the advent of technologies like cloud computing, big data analytics, Artificial intelligence, efficient machine and deep learning algorithms, and their requirement in the field of manufacturing, robotics, system engineering, healthcare, etc., the digital twin (Elayan, 2021) has gained attention. This chapter details the definitions of Digital Twin, its characteristics, design implications, application areas, open issues, and challenges along with future scope in conclusion.
Digital twin definitions
DTs can be defined as machines or computer-based models that simulate, emulate, mirror, or make a twin out of physical entities, processes, or features of human beings. A DT is a living, evolving and intelligent model that is the virtual entity of a physical entity or process. The Digital Twin models the state of a physical system by the data collected from sensors and represents this digitally. DTs bridge the gap between the physical and digital world where the predictions for the future are made by enabling the past and present processes.
Digital twin characteristics
The DT characteristics include
- i.
High-dimensional data handling i.e., coding, analysis, and fusion for integrating the data from various sources to provide consistent and accurate information.
- ii.
Feature extraction and selection of continuously acquired big data which is performed by using closed-loop optimization i.e., uninterrupted exchange of data between virtual and physical entities.
- iii.
The capabilities like self-adaptation and self-parameterization resembling the physical twin by modular approach.
- iv.
Predictive analytics for future statuses prediction for any changes like failures in the product life cycle.
- v.
Proper ontologies and data coding techniques utilized to provide feedback for the exchange of data.
- vi.
Interaction interfaces access information such as the status of physical twins under various scenarios from modeling and simulation applications.
Design Implications
The DT primarily demonstrates the need for a socio-technical collaborative approach to the design process. It requires all end-users or domain experts to collaborate and apply the knowledge to solve to satisfy requirements or specifications. The communication gaps between various organizations need to be addressed by framing Human Work Interaction Design (HWID) for Cognitive Work Analysis. The end-users or domain experts are needed to modify and extend DT system features, End-User Development (EUD) techniques, and tools for implementation. Where EUD allows the end-users to utilize Information and Communication Technology (ICT) - related domains by utilizing socio-technical environments, methods or techniques, tools, etc to create, modify, extend and test digital artifacts.
Figure 1. The lifecycle of DT (B. R. Barricelli, 2019)
Figure 1 shows the lifecycle of the DT Process where four phases exist i.e., Design, Development, Operational, and Dismissal phases. During the Design Phase, the DT Object prototype is designed and adapted to suit technical requirements (Laaki, 2019). These prototypes aid in obtaining the operating object along with its product in the development phase. The DT is paired with the object for continuous interaction during the operational phase. The final phase follows it where DT object dismissal takes place (A. M. Madni, 2019).