Digital Twin

Digital Twins, is a concept that Michael Grieves presented at the University of Michigan in 2002. The Grieves model - thought to effectively monitor the life of a product - considered a real and virtual system they should be connected during the four typical phases of production: development, manufacturing, operation and disposal, which he called the Information Mirror Model.

Honoring his name, both systems had to share information at all times and in real time, as if one were a mirror of the other. The concept was vastly expanded in a 2011 paper, also written by Grieves, and it was only there that the term Digital Twin was used for the first time, more by how the academic explained the model than by the name with which he had baptized it: as If they were digital twins. The concept was more popular, salable and self-explanatory, so the rest of the industry actors adopted it and started using it from then on.

As Grieves explains in a more recent document, in the creation phase the physical system does not exist yet. The system begins to take shape in a virtual space as a Digital Twin Prototype. That is, the Digital Twin precedes the real device. This phenomenon is not new. For much of human history the virtual space where systems were created existed in people's heads. Only in the last quarter of the 20th century could this space exist within the confines of computational digital space.

?This opened a completely new way of creating systems. Before this technological leap, the system should have been physically implemented, first in drafts and blueprints, then in expensive prototypes, ”explains Grieves. The leap in technological capacity of the last decade has allowed simulating these drafts and blueprints to the point where they do not have a noticeable difference with the product in question, beyond the obvious one: that it does not have a physical support.

A digital twin is a digital replica of a living or non-living physical entity. By bridging the physical and the virtual world, data is transmitted seamlessly allowing the virtual entity to exist simultaneously with the physical entity. Digital twin refers to a digital replica of potential and actual physical assets (physical twin) , processes, people, places, systems and devices that can be used for various purposes.The digital representation provides both the elements and the dynamics of how an IoT device operates and lives throughout its life cycle. Definitions of digital twin technology used in prior research emphasize two important characteristics. Firstly, each definition emphasizes the connection between the physical model and the corresponding virtual model or virtual counterpart. Secondly, this connection is established by generating real time data using sensors. 

The concept of the digital twin can be compared to other concepts such as cross-reality environments or co-spaces and mirror models, which aim to, by and large, synchronise part of the physical world (e.g., an object or place) with its cyber representation (which can be an abstraction of some aspects of the physical world). 

Digital twins integrate IoT , AI, machine learning  and software analytics with spatial network graphs to create living digital simulation models that update and change as their physical counterparts change. A digital twin continuously learns and updates itself from multiple sources to represent its near real-time status, working condition or position. This learning system, learns from itself, using sensor data that conveys various aspects of its operating condition; from human experts, such as engineers with deep and relevant industry domain knowledge; from other similar machines; from other similar fleets of machines; and from the larger systems and environment in which it may be a part of. A digital twin also integrates historical data from past machine usage to factor into its digital model.

In various industrial sectors, twins are being used to optimize the operation and maintenance of physical assets, systems and manufacturing processes. They are a formative technology for the Industrial Internet of things, where physical objects can live and interact with other machines and people virtually. In the context of the Internet of things, they are also referred as "cyberobjects", or "digital avatars". The digital twin is also a component of the Cyber-Physical System concept.

By definition "The Digital Twin is a set of virtual information constructs that fully describes a potential or actual physical manufactured product from the micro atomic level to the macro geometrical level. At its optimum, any information that could be obtained from inspecting a physical manufactured product can be obtained from its Digital Twin."Grieves & Vickers (2016)

In other words, Digital Twins are “live” digital representations of devices and processes that make up a factory, connected to the real system they represent through “Cyber-Physical Systems” (CPS). With the live information of the plant, the history of operations and maintenance, and the application of Machine Learning techniques, it is possible to obtain a high-precision model whose behavior closely resembles that of the real system.

Consequently, we achieve a protected and safe environment for experimentation, being able to detect problems before they occur, plan maintenance tasks avoiding unexpected stops, build new more efficient operating scenarios (OEE), develop new business opportunities and new manufacturing plans , or even make future forecasts.

However, the Digital Twin is still a development concept that also presents various technological barriers for adoption in the industrial fabric, there is a technical difficulty to massively monitor and digitize processes in the industry, with a great variety of equipment, isolated legacy systems, field buses, proprietary protocols, as well as a strict industrial integration and automation architecture.

This is linked to the technological complexity that requires going beyond digital representations and moving towards management scenarios of multiple simultaneous digital copies, with greater capacity to evaluate alternative scenarios.

Origin and Types of Digital Twins

Dr. Michael Grieves, Chief Scientist of Advanced Manufacturing at the Florida Institute of Technology, originated the Digital Twin concept. The concept and model of the Digital Twin was publicly introduced in 2002 by Dr. Michael Grieves, then of the University of Michigan, at a Society of Manufacturing Engineers conference in Troy, Michigan. Dr. Grieves proposed the Digital Twin as the conceptual model underlying Product Lifecycle Management (PLM).

The concept which had a few different names was subsequently called the Digital Twin by John Vickers of NASA in a 2010 Roadmap Report.

The Digital Twin concept consists of three distinct parts: The physical product, the digital/virtual product, and connections between the two products. The connections between the physical product and the digital/virtual product is data that flows from the physical product to the digital/virtual product and information that is available from the digital/virtual product to the physical environment. This connection is often called the Digital Thread.

The concept was divided into types later. The types are the Digital Twin Prototype ("DTP"), the Digital Twin Instance ("DTI"), and the Digital Twin Aggregate ("DTA"). The DTP consists of the designs, analyses, and processes to realize a physical product. The DTP exists before there is a physical product. The DTI is the Digital Twin of each individual instance of the product once it is manufactured. The DTA is the aggregation of DTIs whose data and information can be used for interrogation about the physical product, prognostics, and learning. The specific information contained in the Digital Twins is driven by use cases. The Digital Twin is a logical construct, meaning that the actual data and information may be contained in other applications.

The characteristics of digital twin technology

Digital technologies have certain characteristics that distinguish them from other technologies. These characteristics, in turn, have certain consequences. Digital twins has some of these characteristics.

Connectivity

One of the main characteristics of digital twin technology is its connectivity. The recent development of the Internet of Things (IoT) brings forward numerous new technologies. The development of IoT also brings forward the development of digital twin technology. This technology shows many characteristics that have similarities with the character of the IoT, namely its connective nature. First and foremost, the technology enables connectivity between the physical component and its digital counterpart. The basis of digital twins is based on this connection, without it, digital twin technology would not exist. Like described in the previous section, this connectivity is created by sensors on the physical product which obtain data and integrate and communicate this data through various integration technologies. Digital twin technology enables increased connectivity between organizations, products, and customers. For example, connectivity between partners in a supply chain can be increased by enabling members of this supply chain to check the digital twin of a product or asset. These partners can then check the status of this product by simply checking the digital twin.

Also, connectivity with customers can be increased.

Servitization is the process of organizations that are adding value to their core corporate offerings through services. In the case of the example of engines, the manufacturing of the engine is the core offering of this organization, they then add value by providing a service of checking the engine and offering maintenance.

Homogenization

Digital twins can be further characterized as a digital technology that is both the consequence and an enabler of the homogenization of data. Due to the fact that any type of information or content can now be stored and transmitted in the same digital form, it can be used to create a virtual representation of the product (in the form of a digital twin), thus decoupling the information from its physical form.Therefore, the homogenization of data and the decoupling of the information from its physical artifact, have allowed digital twins to come into existence. However, digital twins also enable increasingly more information on physical products to be stored digitally and become decoupled from the product itself.

As data is increasingly digitized, it can be transmitted, stored and computed in fast and low-cost ways. According to Moore's law, computing power will continue to increase exponentially over the coming years, while the cost of computing decreases significantly. This would, therefore, lead to lower marginal costs of developing digital twins and make it comparatively much cheaper to test, predict, and solve problems on virtual representations rather than testing on physical models and waiting for physical products to break before intervening.

Another consequence of the homogenization and decoupling of information is that the user experience converges. As information from physical objects is digitized, a single artifact can have multiple new affordances.Digital twin technology allows detailed information about a physical object to be shared with a larger number of agents, unconstrained by physical location or time. In his White Paper on digital twin technology in the manufacturing industry, Dr. Michael Grieves  notes the following about the consequences of homogenization enabled by digital twins:

“In the past, factory managers had their office overlooking the factory so that they could get a feel for what was happening on the factory floor. With the digital twin, not only the factory manager, but everyone associated with factory production could have that same virtual window to not only a single factory, but to all the factories across the globe.”

Reprogrammable and Smart

As stated earlier, a digital twin makes it possible to make remote adjustments through the digital component of a twin. It enables a physical product to be reprogrammable in a certain way. Furthermore, the digital twin is also reprogrammable in an automatic manner. Through the sensors on the physical product, artificial intelligence technologies, and predictive analytics.  A consequence of this reprogrammable nature is the emergence of functionalities. If we take the example of an engine again, digital twins can be used to collect data about the performance of the engine and if needed adjust the engine, creating a newer version of the product. Also, servitization can be seen as a consequence of the reprogrammable nature as well. Manufactures can be responsible for observing the digital twin, making adjustments, or reprogramming the digital twin when needed and they can offer this as an extra service.

Digital traces

Another characteristic that can be observed, is the fact that digital twin technologies leave digital traces. These traces can be used by engineers for example, when a machine malfunctions to go back and check the traces of the digital twin, to diagnose where the problem occurred. These diagnoses can in the future also be used by the manufacturer of these machines, to improve their designs so that these same malfunctions will occur less often in the future.

Modularity

In the sense of the manufacturing industry, modularity can be described as the design and customization of products and production modules. By adding modularity to the manufacturing models, manufacturers gain the ability to tweak models and machines. Digital twin technology enables manufacturers to track the machines that are used and notice possible areas of improvement in the machines. When these machines are made modular, by using digital twin technology, manufacturers can see which components make the machine perform poorly and replace these with better fitting components to improve the manufacturing process.

Digitization of “deploy & forget” plant

Through the application of IoT and Cyber-Physical Systems (CPS), researchers are working on a proof of concept of a plug & play system to digitize elements in the plant (old and modern machinery, tools, people, products ...) in an agile and minimal way intrusive This system, dubbed "CPS deploy & forget", is specially designed for quick and easy commissioning, without the need for communications and energy infrastructure, and for transparent integration with plant systems through new architectures and protocols. Industry 4.0. The system uses fog-computing techniques to absorb all the complexity involved in the diagnostic and maintenance tasks of these devices and their communication networks, as well as a simplified system for configuration and commissioning. In this way, deploying this system in different locations to carry out specific audits, or continuous monitoring to feed the Digital Twins ceases to be a problem of time and human resources, and the level of plant digitalization can be significantly increased.

Its power lies in the "digitizers", intelligent and autonomous elements that do not require their own communications infrastructure, since the nodes themselves have the ability to self-organize among themselves and solve many of the maintenance problems that until now had to solve the plant staff. These digitizers have a set of integrated sensors to measure different parameters such as temperature, humidity, vibration or energy consumption, but also allow the connection of additional sensors, actuators and interfaces, which extends their use to virtually any operation necessary in industry.

Big Data Analytics for the construction of Digital Twins

A requirement for the construction of a Digital Twin is to use all available data. However, the large amount of data generated by a continuous plant poses a challenge in terms of both storage and processing. This fact requires the application of Big Data Analytics technologies, to efficiently manage real-time data, combined with historical and context information (logistics, sales, warehouse, environment). On the other hand, for the creation of the predictive model that bases a Digital Twin, it is necessary to apply cutting-edge techniques in the field of Machine Learning. Only by introducing both facets, it is possible to create a Digital Twin that provides precise simulations from both its current operations and past and future states.