As industries continue to evolve, Artificial Intelligence (AI) stands out not merely as a tool but as a fundamental partner in engineering innovation, reshaping the landscape of modern product development. In a recent podcast with my colleague Markus Olson, our Business Development Lead – DACH region, we discussed how AI/ML technologies have evolved in the engineering world and at Noesis Solutions.

The integration of Artificial Intelligence (AI) in engineering design processes has marked the beginning of a transformative era in the field of engineering. This evolution marks a very important shift towards leveraging AI in product development, enabling the creation of sophisticated, innovative solutions that automate routine tasks and facilitate complex simulations previously unattainable by human capabilities alone.

The Early Days of AI in Engineering

At first, AI was a tool that engineers had adapted to automate their repetitive tasks like material selection and routine stress analysis. We still relied heavily on traditional methods, and AI was more of an assistant than a partner. As the technology rapidly advanced, AI and Machine Learning (ML) techniques were integrated into our processes to optimize our model training algorithms and make simulations faster.

One particular breakthrough for us, was the introduction of Adaptive DOE (Design of Experiments). Traditional DOE methods were effective, but they were flat—they followed the same linear strategy for every problem, regardless of complexity. Adaptive DOE evolved our approach by dynamically adjusting to the problem at hand, ensuring more efficient experimentation. This saved us time and resources while improving accuracy. It was a breath of fresh air. I can tell you firsthand that this was a significant shift—AI wasn't just helping us; it was becoming a key driver in our innovation.

Enter Deep Neural Networks (DNN)

Before deep learning came along, our models were either fast but not accurate enough, or accurate but too slow to handle the increasingly large datasets that engineering demands. As the latest digital engineering technologies and their computing power advanced enough to support DNNs, it was a game-changer for engineers and at Noesis Solutions, we wanted to be at the frontier of this change.

With DNNs, we were finally able to tackle massive datasets and capture the complex, nonlinear relationships between inputs and system responses. This was huge because, until then, managing large datasets efficiently was a significant bottleneck in our process. Now, we could train models more accurately and faster than ever before.

Smart Models: Balancing efficiency and expertise

Another exciting adaption of AI/ML technologies at Noesis Solutions is the introduction of smart models. Before this innovation, the design decisions were heavily biased and dependent on human expertise. This required a deep understanding of statistics and mathematics. But smart models changed that.

With the adaption of ML strategies, smart models were able to automatically select the best type of model for the data at hand. This meant a shift from expertise-driven human decisions to data-driven decisions - with technology bridging the gap between expertise and efficiency. With Smart Models, a broader range of audiences from the product development team were included in the design process, and were enabled to build reliable models, accelerating decision-making and reducing design time. It was incredibly empowering to see how AI has made these processes more accessible to a broader range of engineers and designers.

The Rise of nvision

Future advancements in AI significantly enhanced simulation and modeling capabilities in engineering designs. These improvements allowed engineers to conduct sophisticated virtual experiments on a scale and with a level of detail previously unimaginable.

nvision, an AI-powered surrogate modelling tool by Noesis Solutions, was launched in January 2024 as an open, solver-agnostic, data-based modeling tool. It enables simulation engineers, designers, and product owners to significantly reduce design time by leveraging AI to perform real-time what-if analysis.

Let me give you an example. Figure 1 is the screenshot of the 3D viewer of nvision of a use case related to cooling a battery pack. In one panel (top-left), I had the temperature output field from a traditional simulation that took several minutes to generate. In another (top-right), I had the same output, created in seconds through a model built in nvision. Sure, the AI-powered model wasn’t 100% accurate, but it is very close (bottom panel), and the insights it provided were invaluable. These tools let us understand and explore design spaces more effectively, giving organizations a competitive edge by speeding up the time to market. This leap in simulation technology will drastically reduce the need for physical prototyping, accelerating the design process and fostering the exploration of innovative design concepts.

The Role of Large Language Models (LLMs)

In recent months, the buzz around large language models (LLMs) like ChatGPT has been everywhere. You might wonder, how does that affect us engineers? Surprisingly, more than you might think. Engineering design is a knowledge-intensive process, and LLMs can help us manage and access the vast amounts of information we work with daily.

With our latest virtual assistant tool, Noesis AI Assistant, users are equipped with instant, accurate information, and support on utilizing Optimus, our automation and optimization tool. This tool has a deep understanding of Optimus, covering all its latest features and functions, as well as best practices and troubleshooting techniques of the tool. It provides its users with 24/7 support for queries regarding Optimus, through a secure, easy-to-use interface. Currently, we are expanding this to accomplish tasks that are related to data analysis (figure 2), where users can ask queries related to the results of their simulation analysis that would require users several steps to filter data and link plots before getting the information of interest.

The Future is AI-Driven

AI is shaping the future of how we design and innovate and at Noesis Solutions, we’re expanding the use of AI horizontally across our products and vertically within each product’s capabilities. With the rapid evolution of AI technologies and our strong user-centric vision, we aim to reduce complexity from today’s product development processes and make technologies more accessible to all our users, and creating user experiences

At Noesis Solutions, we are consistently evolving our product and service portfolio to adapt to the latest market and customer needs, and how our customers can leverage the use of AI/ML to gain competitive advantage. For more information, review the latest announcement, webinars, case studies and stay up to date on the latest products and case studies at Noesis Solutions. Watch a free demonstration of the product here and to discover how nvision can accelerate innovation at your organization, request a free non-obligatory consultation here.