How AI will enable DevOps for hardware
The ability to rapidly change and adapt is one of the hallmarks of the software industry, allowing it, through strong DevOps practices, to achieve an incredible pace of innovation without sacrificing on quality or robustness. By contrast, the process of designing and building physical products has traditionally been much slower, due unavoidable constraints such as building prototypes and running computationally expensive simulations. Now, however, artificial intelligence could offer a path to enabling the same level of rapid innovation for hardware designers that software engineers have long enjoyed.
In a recent podcast, guests Boris Scharinger, Senior Innovation Manager and Technology Strategist for Siemens Digital Industries, and Dr. Justin Hodges, Senior AI/ML Technical Specialist in Product Management for Siemens Digital Industries Software continued a conversation on how AI is redefining the product design process. AI offers new approaches to validating product designs, encoding information and supporting cross-pollination between teams which, in the future, will support a more rapid pace of innovation and broader scope of design space exploration.
Check out the full episode here or keep reading for some of the highlights from that conversation.
The key to faster simulation
The industrial metaverse is shaping up to be an aggregation platform where thousands of digital twins come together to form a real, true-to-life simulation in a way that simply isn’t possible today, Boris explains. Currently though, simulations, and by extension digital twins, focus on individual components or elements of a system rather than the entire system itself, in no small part due to the cost of running high-fidelity simulations.
Simulation being a targeted tool applied in specific scenarios and with a limited scope makes sense given the cost in time, money and computational resources they require. However this also limits the ability for designers to explore a broad design space or optimize a design all the way to the end of the manufacturing process. But by intelligently leveraging AI it’s possible to accelerate these processes, leveraging ROMs (reduced order models) and AI-based models to work alongside existing simulation methods to reach proven results in a fraction the time.
AI models can offer nearly instant inferences on any and every part of the design process, allowing ideas to validated quickly and teams to check results and transfer information across disciplines in a way that simply isn’t possible using traditional methods, says Justin. Leveraging AI technology to foster a stronger connection between different teams and accelerate the flow of data will only be the first step in AI’s journey to reshape design space exploration though.
AI reimagines design space exploration
Returning to the idea of rapid innovation, by combining the safe to test in and highly accurate digital twin with quick to query AI-enhanced simulations, hardware design can benefit from the same fast iteration process that software has long enjoyed. By combining a no-risk virtual testing environment and highly accurate, low-cost simulations available within the industrial metaverse, Boris says companies will be able to achieve something like DevOps for hardware, potentially allowing for hundreds or thousands of iterations on a design to be tested in the same time that a traditional design process would have been able to test a dozen or for continuous optimization to take place, even while a product is being manufactured.
DevOps for hardware has many exciting implications for improving both the efficiency and sustainability of the product design process. Designers would be able to get feedback on the manufacturing feasibility of a design change immediately, before a single prototype is built. Based on a digital design alone, the entire production process could be simulated allowing for virtual commissioning of a production line while the product itself was still being designed. Optimizations for different machines, or part and material suppliers could be quickly developed and validated to reduce waste and improve sustainability in a way that wouldn’t be possible using traditional methods.
All these advancements are possible because AI can greatly reduce the cost of traditionally slow and expensive steps in the design space exploration process, allowing for rapid and continues adaptation and innovation of a product design. Compared to the design space exploration of today, a more AI-driven approach helps support he ideal of continuous optimization, allowing for greater flexibility in both the design and manufacturing processes.
Smart applications of AI will be vital in reshaping the way companies approach design space exploration as well as making sure next generation products are designed and manufactured to meet the needs of tomorrow. Products must be designed faster and smarter while also supporting more sustainable production methods. While traditional methods can achieve excellent results, to truly meet these goals, companies must be ready to embrace cutting edge technologies, like AI, to meet and exceed the demands of next generation design.
To learn more, check out the full podcast here.
Siemens Digital Industries Software helps organizations of all sizes digitally transform using software, hardware and services from the Siemens Xcelerator business platform. Siemens’ software and the comprehensive digital twin enable companies to optimize their design, engineering and manufacturing processes to turn today’s ideas into the sustainable products of the future. From chips to entire systems, from product to process, across all industries. Siemens Digital Industries Software – Accelerating transformation.
