In a recent collaborative webinar hosted by SimScale, CDFAM, and NAVASTO, experts David Heiny, Duann Scott, and Matthias Bauer delved into the transformative role of AI in computational design. The discussions were rich with insights on how AI is reshaping the engineering landscape, especially in simulation-driven design. Here are the top five highlights from the session on AI-powered computation design (link to webinar below).
1. The Data Dilemma: Quantity vs. Quality
One of the most frequently asked questions in the realm of AI is, “How much data do you need?” Matthias Bauer, CTO of NAVASTO, provided a nuanced answer: “It’s not just about the quantity but the quality and structure of the data. For complex automotive use cases, starting with as few as 30 to 50 samples can be sufficient to train a useful model. However, the focus should be on continuously improving these models with new data, ensuring that the data is well-organized and that engineers are aware of its importance.” This approach transforms AI from a black-box tool into a valuable resource that grows in accuracy and utility over time.
2. Machine Learning’s Strength: Interpolation Over Extrapolation
A key technical insight shared by Bauer was the distinction between interpolation and extrapolation in machine learning. “AI excels at interpolation—predicting within the range of the data it has been trained on. However, when it comes to extrapolation—predicting outside that range—the results can be unpredictable.” This highlights the importance of providing users with tools to assess whether they are operating within a reliable prediction zone. Engineers need to understand these limitations to make informed decisions about when to trust AI predictions and when to revert to traditional simulations.
3. The Power of Fast Feedback Loops
Duann Scott, founder of CDFAM, drew an interesting analogy between the evolution of cameras and the impact of AI on design processes. “Just as digital cameras shortened the feedback loop for learning photography, AI-powered tools are drastically reducing the time it takes to iterate and improve designs.” With immediate insights from AI predictions, designers can quickly test, modify, and optimize their designs, leading to faster innovation and higher-quality products. This rapid iteration is a game-changer, enabling engineers to experiment more and push the boundaries of what’s possible.
4. Democratizing AI for Engineers
One of the most significant takeaways from the webinar was the notion that AI doesn’t need to be the exclusive domain of data scientists. As David Heiny, CEO of SimScale, pointed out, “Engineers can effectively implement AI-based solutions without requiring deep expertise in data science. With the right tools, such as SimScale’s cloud-native simulation platform, engineers can harness AI’s power to enhance their design processes, making AI a practical tool rather than a theoretical concept.”
5. AI as the Future of Integrated Design Systems
Looking ahead, the panelists discussed how AI will evolve from being a tool for solving specific problems to becoming an integral part of the entire design ecosystem. The vision for the next five to ten years is a fully connected engineering process where AI leverages vast amounts of previously untapped data, enabling engineers to access insights from every corner of their organization. This shift will empower teams to make informed decisions earlier in the design process, leading to more innovative and optimized products.
The Future is Now
The consensus from the webinar is clear: AI is not just a future possibility—it’s a reality as we speak. With tools like SimScale and NAVASTO, engineers can start leveraging AI today to accelerate their design processes, make more informed decisions, and ultimately create better products. The time to embrace AI in computational design is now, and those who do will be well-positioned to lead in the next wave of engineering innovation.
On-Demand Webinar
If the above highlights caught your interest, there are many more to see. Watch the on-demand Engineering Leaders Series webinar from SimScale on Preparing Your Teams for AI-Powered Computational Design by clicking the link below.
