Can digital twins revolutionize electronic design and test?
“Connecting design and test is the nirvana of what many companies are trying to achieve in order to deal with the complexity of the products being designed today.”
In one sentence, Keysight’s Chris Mueth and Duane Lowenstein have outlined the promised land of many engineering departments. It’s a place where design, test, and build processes are linked into a continuously improving cycle that spans organizations and accelerates time to market. And at its center lies digital twins — enabling project teams to get it right the first time by predicting how designs will fare under test, ensuring first-pass success when validating prototypes, and minimizing costly rework and schedule overruns.
But as easy as it is to imagine paradise, getting there is a lot harder.
What is a digital twin?
To quote the article: "By its general definition, a digital twin is a real-time virtual digital representation that mimics a physical object or process."
Traditionally, digital twins have performed predictive analyses on existing products and systems. Think about it this way. Imagine you've built a fighter jet, and you want to know how the rigors of supersonic speeds will affect its performance at various altitudes. Digital twin technology enables you to simulate variables, assess potential impacts, and make critical course corrections — all before takeoff. But that’s not all. Once the prototype rolls off the production line and flies into service, onboard sensors can relay data back to the digital twin to reflect wear and tear over the plane’s lifecycle. A steady stream of data ensures the digital twin remains an identical virtual model of its physical counterpart — creating a continuous feedback loop that enables engineering teams to perform accurate analysis throughout the duration of its mission.
Digital Twins Can Transform Design and Test Strategies
While digital twins have been traditionally employed to analyze existing products and systems, Chris and Duane have a different perspective. In their paper, they asked a simple question:
“What if you built a digital model first and then used digital process threads to link the digital twin with physical system tests?”
Read the full article to hear what the authors have to say on this: