For Digital Twin Technology, Configuration Can’t Be an Afterthought

For Digital Twin Technology, Configuration Can’t Be an Afterthought
For Digital Twin Technology, Configuration Can’t Be an Afterthought

Digital twins have been hyped as one of the next big things–especially for the industrial space. In fact, research by McKinsey found that 70% of C-suite level tech executives at large companies are exploring and investing in digital twins.
While adoption is increasing, there’s still a lot of confusion in the marketplace about what this technology is and how to use it effectively. Digital twins can be an important component of overall digital transformation efforts, but it can’t just be thrown in the mix without strategy to back it up.
For digital twins to be effective, it’s important to understand the two sides of the coin–configuration and performance–and the differences between them.

Understanding what digital twin really is

Put simply, a digital twin is a virtual representation of a physical product or asset that reflects the real-time configuration and that can trace maintenance activities and operational performance that have occurred to the product over time. This includes decisions made during the engineering and manufacturing processes.
It’s different from a digital model, which reflects a view of a product to support design and engineering scenarios but doesn’t reflect the final product’s reality or how it evolves once it is in service. A digital twin configuration of a manufactured product can be used as the interface to explore the product data and the decisions made on the product. It helps teams explore how their customers use a manufactured product, how that product operates, and how and when it is maintained. It also can be used to understand what components have failed.
Gaining this insight creates opportunities for future product design enhancements. It can yield suggestions on software updates to improve operational efficiency to support unique customer use cases or develop unique product configurations to support new markets or use cases emerging from customer use.

Two important digital twin elements

There are two segments to digital twin technology: configuration and performance. For example, the performance side includes time series data streaming back from sensors where predictive analytics capabilities are used. The current as-running asset configuration—called the digital twin configuration—is equally important and provides the context for interpreting, analyzing and simulating time series data.
Digital twin configuration is the start. It underpins everything else, and it’s how you understand what has happened to the physical product over time. And the performance side is what helps you understand the future of the product. The bottom line is that you can’t truly have a digital twin until you have a manufactured product in the first place.

Both elements are necessary

For many vendors, there’s a mismatch when they try to compare what’s happening to a physical product out in the world to its digital representation. They’re probably looking at a model that doesn’t have all the information; it might not have the product’s history. If it’s a model, it’s not an exact representation, which could produce poor results. And that turns into bad operational decisions if they lack the exact context. That’s why there’s a need for the digital twin configuration.
It’s a big mistake to focus too much on the performance side, but this is where a lot of the focus and investment have been—through pilots and deployments for a combination of capabilities, like cloud storage, data lakes, gateways, AI and machine learning, analytics, and dashboards. But often, these pilots don’t scale, or the value degrades over time when deployed because the configuration of the individual asset or system of assets is not current, leading to poor operational insights. 
Without this configuration, you lose the individual story each product has to tell. The configuration of every product in operation can differ, from vehicle to vehicle, machine to machine and ship to ship. The longer something has been out in the field, the more it changes. To effectively analyze the sensor data, you need accurate context. Digital Twin Performance cannot happen in isolation; instead, it requires the individual Digital Twin Configuration of the managed asset.
Building a Digital Twin configuration creates an opportunity to bring modeling and simulation techniques to the operations phase of an asset. For instance, if your organization is focused on aircraft design, these capabilities would enable you to improve the accuracy of your operations and maintenance decisions to better optimize your fleet and support future improvements in next-generation aircraft.

Moving forward for success

Digital twin technology holds a great deal of promise. But it has to be done with a full understanding of the importance of both performance and configuration–without over-rotating on performance alone. While performance-driven investments such as cloud storage, data analytics and AI have received significant attention, neglecting the digital twin configuration can lead to poor operational insights and failed deployments.
To harness the true potential of digital twins, it is crucial to acknowledge that configuration provides the context necessary for accurate interpretation and analysis of real-time data. By striking the right balance between performance and configuration, businesses can unlock the full value of digital twin technology and drive successful digital transformation efforts.

About The Author

Jason Kasper joined Aras Corporation in April 2017 and is the director, Product Marketing, with his primary focus on supporting Digital Transformation, Digital Thread and Digital Twin go to market programs to convey their importance in supporting innovative strategies within Product Lifecycle. Jason has more than 20 years of experience in working with customers to develop enterprise software solutions for industries such as Discrete Manufacturing, Oil & Gas, Mining & Metals, Utilities and Transportation. 
Before joining Aras Corporation, Jason was an Analyst at LNS Research, focused on Asset Performance Management, Internet of Things and the Digital Twin. Prior to this, Jason held senior level positions at Meridium (GE Digital), IBM, ABB and Aspen Technology dedicated to the development of industry solutions for asset intensive organizations with a primary focus on Enterprise Asset Management and Asset Performance Management markets. Jason holds a BS in Management from California State University, Chico and an MBA from the D’Amore-McKim School of Business at Northeastern University.

Xem Thêm: Hệ thống MES

Trí Cường

Translate »