|Now, imagine for a moment if that duplicate stranger were actually you—the same behaviors, manner of speech, memories, foibles. Everything. Pretty strange, wouldn’t it be?
Well, in the world of the digital manufacturing enterprise, such a notion is not strange at all. It is quite real.
Enter the digital twin
Consider this. Suppose you were to start with an object that some individual or company uses every day. Now, let’s say you were able to configure this object with many sensors that capture massive, cumulative amounts of real-world data about the state of that object in near real-time—from its beginning to the present moment. Then, you would have the basis of a digital twin. And, of course, the same idea could apply not only to an object in use, but also to the process that manufactured the object. But more about that later.
Indeed, a digital twin can be defined, fundamentally, as an evolving digital profile of the historical and current behavior of a physical object or process that helps optimize business performance. That profile is, in practical terms, a digital “image” of the object or process on a visual display in near real-time that reflects its measured history.
Digital twins and deployed objects
Still, the question remains—what practical purpose does a digital twin serve, and why should anyone care? At its most basic, the digital twin of a single deployed object could provide its user with critical information about the condition of that object in the field. Such information might prompt some kind of intervention by the owner.
But let’s extrapolate that out to a broader scale. What if the manufacturer of that object maintained a digital twin of each of many of the same type of deployed object? Such cumulative real-world data across many digital twins of the same type of object—when analyzed properly—could yield compelling insights into actual performance that may warrant change in product design or manufacturing process, or both.
There are many practical examples of operational digital twins across a variety of circumstances. Digital twins of complex deployed assets like mining trucks and jet engines, for instance, can yield important insight into wear and tear and a full range of stresses as the asset is used in the field. A digital twin of a wind farm, on the other hand, can provide awareness of operational inefficiencies. Other examples of operational digital twins abound, cutting across industrial and consumer applications.
Digital twins in the manufacturing process
So far, we have focused on the digital twin of the deployed object. But the digital twin of the manufacturing process also appears to offer an especially powerful and compelling application. In this context, thousands of sensors distributed throughout the factory floor can collectively capture data across a wide array of dimensions—from behavioral characteristics of the productive machinery and works in progress to environmental conditions within the factory—forming the basis of the digital twin. But in this example, the digital twin serves as a virtual replica of what is actually happening on the factory floor in near real-time, as opposed to a single deployed object (or many deployed objects) as discussed earlier.
The twin payoff
The payoff occurs when, over a period of time, the analyses that the digital twin performs uncover trends in the actual performance of the manufacturing process that vary from the ideal range identified for tolerable performance. Such comparative insight could provide previously unavailable insights that could trigger the organization to change their manufacturing processes—leading to improved efficiencies, safer conditions, or averted malfunctions, among other outcomes.
Whether based on deployed objects or an entire factory floor, the digital twin may allow companies to realize significant value in wide-ranging areas, from reduced defects to improved operations—and beyond. Digital twins can afford manufacturers a digital “footprint” that lets them understand not only how the product performs in the field, but also to possibly understand performance for the system that made the product.
All of this, of course, is just the tip of the iceberg. There can be many applications of the digital twin across the product life cycle. In a recently published paper, Industry 4.0 and the digital twin: Manufacturing meets its match, Deloitte practitioners provide a much more in-depth look at digital twins including the way they can be created, how they could drive value, their typical applications in the real world, and how a company can prepare for the digital twin planning process.
To be sure, a digital twin is not a panacea by any means. Although computing storage and process costs have declined appreciably in recent years, a digital twin initiative could still require substantial investment. Companies may be wise to start small and monitor progress and return. Still, digital twins have often proven their worth in real world applications in providing insights into product design, manufacture, and deployment in ways previously unknown.
At the end of the day, quite unlike your doppelganger, the digital twin is not a stranger and is much more than skin-deep. It seems to be driving value on the journey toward better products and more efficient processes.