High TechMay 22, 2024

Why virtual twins are the dream for semiconductor development

Developing semiconductors with virtual twins, not digital ones, could be a key differentiator in industry success.
Avatar Shoshana Kranish

The Biden administration recently announced that it will disperse $280 million to companies to develop digital twins of semiconductors. The funding comes after a yearslong shortage of chips, of which semiconductors are a significant component and which are necessary for almost every piece of technology we use on a daily basis. The administration’s push for developing a workaround to a persistent problem is innovative in and of itself, but virtual twins can take things a step past where their digital counterparts fall short. 

Virtual twins replicate physical objects just like digital twins do, but they take it further, simulating the processes and interactions within an entire system or ecosystem. This could be a game-changer for the semiconductor industry and beyond, offering a more holistic approach to understanding and solving complex challenges. By incorporating real-world data in real-time, virtual twins could provide insights into production processes and supply chain dynamics, and even predict future disruptions, offering a more comprehensive solution to the semiconductor shortage and other critical issues facing technology sectors today.

Here’s why virtual twins, not digital ones, would be a true game changer for semiconductors (and really everything else). 

What’s a virtual twin, anyway? 

Digital twins are 3D representations of physical objects. They’re static and relatively limited in the scope of what you’re able to do with them. You can toggle a 3D rendering, move it around, inspect its elements and conduct a simulation. But such a model is siloed. You can’t understand its interactions with other objects or the effects that certain changes might have on it, like temperature or pressure. It exists in a vacuum, so to speak. 

“Digital twins are legacy technology,” explained John Maculley, business strategy consultant for high-tech at Dassault Systèmes. 

That is, they’re useful for some things, but in general, are an outdated choice when it comes to adopting innovative methods to solve persisting problems.

Virtual twins, on the other hand, comprise not just a digital model of an object or process, but the entire environment in which it exists. A more sophisticated option, virtual twins can encompass as much as you want them to. That includes not just the soccer ball and a simulation of someone kicking it, but a detailed map of how that soccer ball gets made, from the supply chain that sources its raw materials to the manufacturing methods that are used to create it. The scope is vast when it comes to virtual twins, and, simply put, there’s so much more you can do with them. 

Virtual twins for semiconductors 

While it’s clear that virtual twins are more advantageous for sophisticated operations than their digital counterparts, they’re also a superior option for semiconductors specifically.  

Semiconductors are essentially a building block of electronic technology. They’re responsible for controlling electrical flows in devices like TVs, smartphones and computers, and are what enables these tools to be fast and powerful and, increasingly, small. With so many use cases, they’re a foundational piece of technology in our lives today. 

Semiconductor manufacturing requires highly specific and intricate processes. The involved equipment must be precise, with no room for deviations in any aspect of the production process. The workers who run the machinery need to be highly skilled and specialized. The labs where the semiconductors are made are extremely expensive to build, operate and maintain. This doesn’t even scratch the surface of what’s involved; in short, however, semiconductors require a high level of detail, skill, investment, sophistication and exactitude. Producing them at scale, which is what the current climate demands, is beyond tricky. 

A manufacturing lab - semiconductors - Dassault Systemes blog
Labs and manufacturing facilities where semiconductors are developed could benefit from virtual twins

Virtual twins as a differentiator 

The Biden administration’s funding announcement indicates a hope that a digital, home-grown solution will present a fix for the current problems that the country is experiencing. It’s part of the CHIPS Act, which was initially published in 2022 in an effort to – among other things – reduce the effects felt from supply chain disruptions. Weather, the COVID-19 pandemic, geopolitical shifts and a variety of other factors caused a slowdown in the production of semiconductors and chips. While the supply of these materials went down, their demand didn’t. This trend impacted industries we touch every day, with automotive, consumer electronics and wireless communications chief among them. 

Virtual twins can prevent similar problems from cropping up again. Regardless of what issues persist, be they meteorological or geopolitical in nature, having the entirety of the semiconductor lifecycle in a virtual setting can support the agility needed to succeed. A virtual twin can house procurement of raw materials specifications, refinement processes, manufacturing set-ups, machinery specifications, shipping and sorting and more. Coupled with AI, virtual twins can sift through data at lightning speed, catching even the most minute changes or errors and addressing them in real-time, a capacity digital twins lack. 

“Companies that transition from legacy digital twins to virtual twin experiences are going to quickly realize they’ve got an entirely new universe of simulation capabilities and a deeper understanding of the underlying physics driving their product decisions,” Maculley said. “Their AI strategies and roadmaps become clearer, and their ability to react to market changes increases exponentially.”

Learning from their own ecosystem and making adjustments as needed, enabling accuracy and efficiency while also expanding the potential for future production are all key differentiators between digital and virtual twins for semiconductors. Implementing cutting-edge technology, rather than relying on legacy systems, can usher in a new era of innovation and advancement that Biden’s funding aims to accomplish. 

The first steps toward creating a virtual future 

Recognizing the need for an innovative solution marks a pivotal step towards progress. However, the true challenge lies in pinpointing the exact solution that will steer the United States onto the correct path. By channeling investments into the development of virtual twins rather than settling for mere digital replicas, we can achieve a comprehensive resolution to the complex problems at hand. 

In February, Dassault Systèmes inked a deal with Purdue University to do exactly that; together, the two sides will offer a hands-on program for students to learn how to leverage virtual twins to explore new solutions for research and development, sustainability and workforce readiness and adaptability in the semiconductor industry. This approach not only enhances our understanding of systems and processes but also significantly improves decision-making and innovation. 

Furthermore, focusing on semiconductors as a starting point for this transformative shift is both strategic and necessary. Given their critical role in virtually all modern technology, semiconductors are and will remain in high demand. By strengthening our capabilities in semiconductor technology, we not only bolster our technological infrastructure but also ensure our competitiveness in a rapidly evolving global landscape. In essence, a commitment to nurturing advancements in virtual twin technology, coupled with a focus on semiconductor innovation, represents a forward-thinking strategy that promises to secure a prosperous and technologically advanced future for the United States.

Stay up to date

Receive monthly updates on content you won’t want to miss


Register here to receive a monthly update on our newest content.