Are digital twins overhyped? Explore alternate strategies for manufacturers may pay off better.

Let’s first agree on terminology. A digital twin is a virtual representation of a physical asset, bringing together the cyber and physical worlds. It allows you to digitally see a robot or machine as it works, whether it’s through data output or through a three-dimensional model. By comparison, a digital thread is the genealogy of a product and everything that went into its manufacture—from the design stage to manufacturing to the customer environment to end of life—allowing you to track all the subcomponents of your equipment or process.

Fundamentally, the difference between a digital twin and digital thread is time. A digital twin represents a moment in time, showing you what’s happening now, so you can model various situations to make better decisions. Whereas a thread shows you the digital DNA of a product or process, providing insight over time to digitally track subcomponents over their lifecycles.

In my view, it’s important to realize that your digital twin and thread may be a byproduct of your digital strategy, not the end goal. As you connect, digitalize and optimize your operations, you can create your digital twin and thread.

In my experience, manufacturers looking to get intelligence out of their systems need tools to clear the clutter. However, there are new and better ways to get data and enable more predictive capabilities. Digitalization provides a competitive edge—a new ability to win. Regardless of the tools you use, it’s important to focus on where you’ll create value, how you’ll do that and what it’ll look like.

Three steps to maximize the impact of digital transformation (no, a digital twin isn’t one of them)

Your digital strategy is about improving manufacturing competitiveness, not achieving the latest industry trend. The first step is to connect and acquire data, bringing it forth in a meaningful way. When it comes to optimization, here’s how that plays out:

• Step 1: Visualize your data. Once you have data, you need a way to see what it’s telling you. You can only consume and react to a relatively limited amount of data. What do you need to see and how often do you need to see it? The specifics of what you need will invariably be unique to your application. You’ll need to determine whether you need incremental data points over milliseconds or hours or just when you reach a certain threshold. When you’re driving, for example, you likely want to see your fuel usage over hours or weeks, whereas you’ll need to know how fast you’re going at any given time.
• Step 2: Apply predictive elements through analytics Not just visualizing data (or watching things happen) but enhancing that data with simple analytics. The natural progression is to move from visualizing data to see what the data can tell you about what may happen next. You are not going to be always looking at data about your processes or equipment, so you need to apply analytics. This will help you get a sense of what will likely happen next.
• Step 3: Utilize a feedback loop and machine learning to improve performance even further. How can you put control back to machines so that they optimize themselves to avert overheating, downtime or other problems? You need to create a feedback loop so that machines are communicating, applying analytics and dynamically make adjustments without people having to intervene. If your analytics are telling you that your equipment may overheat in an hour, there can be a feedback loop in place to slow the motor down, back off a process, run at a lower speed or apply another solution to allow your machines to self-correct and avert downtime. This feedback loop needs to support dynamic scenarios, finding ways to react and course correct in real time.

Real time data is foundational and transformative for manufacturing optimization.

What will you do with it? Typically, you start with a human understanding of what that data is showing you. Then, move on to computational analytics and then to machine learning, which provides a feedback loop.

The point is to keep your equipment running, minimize downtime and enable automation. This doesn’t work with just of a single machine. It takes an interconnected system, moving materials in and out while ensuring 100% quality is met. This creates an ecosystem of automation.

Starting your digitalizaton efforts with implementing a digital twin or thread may not play a significant role in creating this ecosystem of automation and may be difficult to achieve. Most manufacturers are dealing with aging infrastructure, differing levels of connectivity and intelligence from vastly different components used in manufacturing lines. So the cost and time it takes to create a digital twin or thread may not ever payoff.

Improving manufacturing competitiveness

Digital transformation is surely about updating critical legacy equipment. That’s a key area where we’re seeing digitalization applied--for our customers and in our own operations. If vintage equipment is part of the critical path to your manufacturing, then you need to ensure high throughput and reliability, no matter how old that equipment happens to be.

Overlaying digital ensures operational competitive advantage comes through. As I previously mentioned, creating a digital thread/twin is not the end game. Monitoring and data analytics are where it’s at.

That’s exactly what the recent research shows; updating or replacing legacy equipment to be more digitally enabled and monitoring plant equipment were leading use cases for digital transformation across industrial operations. Whether or not you need a digital twin or thread depends on the manufacturing advantage that you’ll get from it.

Digitalization is about manufacturing competitiveness and new abilities to win

Unsurprisingly, the Manufacturing 2023: Digital Transformation and Energy Transition report shows that digital transformation is relatively mature among manufacturers. Nearly all manufacturers have a digital transformation strategy. About half of manufacturers are actively executing on it, with 48% among continuous process industries and 55% among discrete manufacturers.

Overall, industrials are grappling with common issues. The top three concerns across the sector were the same sticky issues: lack of skilled workers, supply chain issues and (old) age of plant machinery. And digital transformation can help. At Eaton, we’re applying automation to work that is dirty, dangerous and/or dull. Keeping people safe. And addressing the shortage of workers, to keep people employed in work they want to do. Digitalization and automation can help enhance safety, validate quality and have a direct impact on workforce efficiency and productivity, all while improving people’s experience on the job.

And while digital twins and threads may be on the path in your digital journey, they’ll likely be a byproduct of your efforts that make a much larger impact.

No matter what challenges you’re solving, now is the time to put your digital strategy into action. It’ll be foundational to your ongoing competitiveness.