Transition from atoms to bits in manufacturing reduces waste in time, energy, and materials By Dr. Michael Grieves

As a 21st-century concept, the Digital Twin captures the idea that we are moving work from the physical world into the virtual one. It moves toward using bits, which are getting cheaper all the time, in place of atoms, which are getting more expensive. We are replacing expensive physical time, energy, and material with much cheaper digital information.

The Digital Twin is the concept driver behind Manufacturing 4.0. It moves us from a functional-centric approach to a product-centric one. It integrates manufacturing as a critical aspect of product creation, so that how to manufacture products is not an after-thought once a product is designed and engineered.

With the Digital Twin approach, our ideal is to design the product virtually, test the product virtually, manufacture the product virtually, and support the product virtually. We want to do this in an integrative, holistic fashion across functional disciplines. Only when we get the product to its desired functionality do we physically manufacture it.

This approach will increase both efficiency, which is reducing cost and resources, and effectiveness, which is producing quality products that will perform the functions their users require.

Components of the Digital Twin 

The conceptual model of the Digital Twin is shown in Figure 1. The Digital Twin model consists of three components: a) physical space and products that we have historically worked with; b) the new virtual, digital space and products; and c) the connection between the physical and virtual. This connection is commonly referred to as the “Digital Thread.”

Because the Digital Twin concept spans the entire lifecycle of products consisting of create, build, operate/sustain, and dispose, there are different Digital Twin types. The three types of Digital Twins are the Digital Twin Prototype (DTP), the Digital Twin Instance (DTI), and the Digital Twin Aggregate (DTA) (Figure 2).

The Digital Twin Prototype is all the information we need to build physical products. The DTP originates first. In its ideal form, it is the information needed for all the products that we can build.

Second, the Digital Twin Instance is the representation of an actual physical product instance that we have built. The DTI remains connected to its physical form and behavior throughout the entire physical product instance’s life. Products that are of high value and operational complexity required DTIs. We need Digital Twins for airplanes, oil rigs, and pacemakers. We don’t need DTIs for paper clips.

Finally, the DTA is a representation of all the products that were already built. By collecting this information and aggregating it, we can do such things as failure prognostication and machine learning.

The manufacturing process of building products requires the DTP, the information necessary to produce a specific product, and produces the DTI, the information of what and how a product was actually manufactured. The manufacturing process itself consumes the DTP in order to improve product quality and supplies the DTI that can then be used for the life of that particular product.

When it comes to managing factories and their production processes, manufacturers use the DTIs of their machines to gain visibility into factory operations at every tick of the clock. It uses the DTA of those machines to not only reduce downtime, but to predict machine failures so that potential problems can be remedied before failure ever occurs. The Digital Twin is a concept that enables Manufacturing 4.0.

The Digital Twin supports three of the most powerful tools in the human knowledge tool kit: conceptualization, comparison, and collaboration.

Information as Replacement for Physical Resources 

The value of the Digital Twin relies on a premise that is implicitly well known but often not articulated well: The premise is that information is a replacement for wasted physical resources, i.e., time, energy and material. As the left side of Figure 3 shows, we can take any task and divide it into two categories. The lower category in green represents the minimal use of resources in performing this task. If we were omniscient and omnipotent, this is the amount of resources that we would expend in completing our task. The upper part in red represents resources over and above the minimum amount of resources that we should expend.

Since we are neither omniscient nor omnipotent, we generally spend a substantial amount of resources more than necessary. In today’s vernacular, the lower green area would be referred to as “lean manufacturing.” The intent of all lean manufacturing methodologies is to eliminate waste and produce products with the minimal amount of resources. Since we live in a capitalist society, we translate the physical resources of time, energy, and material into monetary cost. We can calculate the costs to do the specific task, especially manufacturing tasks.

The right side shows the impacts of utilizing information in the performance of our task. Information is a substitute or replacement for the physical resources of time, energy, and material. However, it is only a replacement for those resources that are wasted. As the figure shows, the amount of physical resources that we need to expend in performing the task in the most efficient and effective manner stays the same. What changes is that an increasing amount of wasted physical resources are replaced by information.

This is the ideal. In reality there will always be some wasted physical resources. As on the left side, we can calculate the cost of performing the task by replacing wasted physical resources with information. Unfortunately, it is a little more difficult since by its very nature information does not have a unit of measurement. There is no ability to say we have used X units of information in performing the task. However, we can get at a proxy of the cost of information by analyzing the costs involved in producing, storing, and retrieving information. The proxy costs for these are such things as hardware, software, IT resources, and similar costs. If we are buying software as a service, the cost of information is more readily available.

So in order for us to use information as a replacement for wasted resources, we need to keep this in mind: The cost of information needs to be less than the cost of wasted resources over all the times that we perform the task. This means for simple one-time tasks, it makes no sense to create an information system to replace using trial and error and, therefore, likely wasting resources. However, when there are repeatable tasks on a continuous basis, using information as a replacement for wasting resources makes great sense. This is the essence behind smart manufacturing, or Manufacturing 4.0: using Digital Twins as a replacement for wasted resources.

Conceptualization, Comparison, and Collaboration 

The Digital Twin capability supports three of the most powerful tools in the human knowledge tool kit: conceptualization, comparison, and collaboration. These capabilities are a major enabler of sustainability. This is especially true in manufacturing where consistency and repeatability are critically important.

Conceptualization: Computers process differently than humans. Computers perform step-by-step

processing that builds information structures from the bottom up. People try to build conceptual, visual models that they change and modify as they collect data.

Prior to the advancement of Digital Twins, people have struggled with building their conceptual model from fragmented data and reports. Especially in the manufacturing environment, people have built conceptual models of what is happening from report data. Even more problematic, because the data is so fragmented and siloed, different people can and do easily build different conceptual models that conflict with each other. This leads to not building consensus on problems and/or working at cross purposes.

The Digital Twin is a conceptual model, so that its visualization can now be a shared visualization. Instead of trying to build up individual conceptual models, the Digital Twin allows a shared understanding and for drilling down into the data. Shared conceptualizations lead to increased efficiencies on the factory floor (see figure 4).

The Digital Twin, with its inherent emphasis on conceptual visualization, allows all manufacturing participants to see the same conceptualization. All participants can see the factory operations and anomalies. System alerts can draw participants to the issues that need attention.

Comparison: Comparisons are one of the most fundamental decision tools humans possess. We spend almost all waking moments assessing our situation, comparing it to what we want our situation to be, and then taking action to close the gap between where we are and where we want to be. For manufacturing, it’s critical to perform this process continuously in order to keep that gap as small as possible.

With the Digital Twin of our factory updated in real time, we have all the information we need so that, instantaneously and simultaneously, factory personnel can have a holistic view of their operations.

While much attention in Manufacturing 4.0 is given to reducing the time from an adverse event to remediation, correlating past failures with leading sensor indicators using DTAs should allow us to predict equipment failures and remediate them before they occur.

Collaboration: One of the key aspects of Manufacturing 4.0 is collaboration. In order to produce products more effectively and efficiently, functional information silos with duplicative and inconsistent information need to be eliminated. This needs to occur not only within the organization but across the organization’s supply network.

The Digital Twin enables this by providing a product-centric view that everyone in the organization populates and/or consumes information from. Because information is highly granular, meaning that we do not need to share all or nothing like we do with physical things, we can share specific pieces of information with suppliers so that they can do their tasks more efficiently and effectively.

As importantly, suppliers can transmit back the information from their own Digital Twins, allowing manufacturing customers to see those products as they are taking form. This can be done well in advance of the actual physical product arriving at the loading dock. This would allow manufacturers to react early to supplier problems well before defective products arrive and potentially lead to a crisis.

Within the organization, the Digital Twin enables functional areas to collaborate to produce better products at reduced costs. With exponential increases in computing technology, collaborating in virtual space with a Digital Twin will allow for the different functional areas to have input into product creation. Fast cycle times of design, engineer, build, and test of Digital Twins will provide visibility and continuity of design intent.

The Value of the Digital Twin 

This leads back to the underlying premise of the Digital Twin: information as a replacement for wasted physical resources. Moving work from the physical world to the digital one can improve sustainability by reducing waste and identifying potential efficiency improvements. We can be more effective in developing products that perform to user expectations.

The capabilities of conceptualization, comparison, and collaboration enabled by the Digital Twin make an integrated, holistic approach to product development a reality. Manufacturing 4.0 moves within reach. The 21st century promises to be very different for the manufacturing industry compared to the 20th century. The Digital Twin will drive that difference. M