Exploring AI and Advanced Simulation in Digital Engineering: Insights from Industry Leaders

The recent “Pulling the Digital Thread” panel, Harnessing the Power of Simulation and AI for the Future of Engineering, brought together four influential figures to discuss the intersection of artificial intelligence (AI) and advanced simulation in digital engineering. I moderated the discussion in which we delved into how AI and simulation reshape engineering processes and accelerate product development across industries.

Thank you to our panelists for bringing their unique expertise to the conversation: David Long from INCOSE, Tom Martinek from Ansys, Jan Paul (JP) Stein from McKinsey, and Matteo Nicholich from Aras.

Benefits of AI in product development

Tom Martinek, a veteran in product development at Ansys, began by sharing his perspective on how AI is increasingly integrated with traditional simulation tools. “AI and machine learning are enabling us to achieve the same results we get from physics-based simulations, but in a much faster and more efficient way,” Tom explained. He highlighted that AI has started to predict product behaviors without relying solely on solving complex physical equations, thus speeding up the design process. While emphasizing that he’s not an AI expert, Tom pointed out how AI augments the portfolio of simulation technologies to make product development more efficient.

David Long, representing the systems engineering viewpoint, explained how AI adds another layer of intelligence to complex engineering problems. “Systems engineering is based on the idea that the collective intelligence of many is greater than the singular intelligence of one,” David stated. He highlighted that AI could assist multidisciplinary teams by identifying interactions and early solution concepts that may not have been apparent otherwise. The goal, according to David, is to democratize access to systems engineering, making it more approachable for all engineers through the support of AI.

Deep learning surrogates and faster simulations

Jan Paul Stein from McKinsey provided insights into how AI-based technologies, particularly deep learning surrogates, can partially replace classical simulation methods. “Deep learning surrogates are a fascinating technology that allows partial replacement of classical simulations by training AI models based on full data from simulations,” Jan Paul noted. He elaborated on how these models enable faster inference and allow for more efficient exploration of optimal design solutions. This approach offers companies the chance to cycle through multiple designs quickly and achieve more refined outcomes. According to Jan Paul, deep learning surrogates represent a fascinating new technology allowing for partial replacement of classic simulations based on Large Language Models (LLMs) for training AI.

JP also highlighted the role of AI in enhancing time to market for various industries. “In the past, improving product performance was the main driver. But now, the focus has shifted to reducing time to market,” he said. He explained how industries, particularly automotive, increasingly use AI to streamline the engineering process and meet competitive market pressures.

Bridging traditional optimization and AI

Matteo Nicolich shared his insights on the evolution of optimization methods in product development. He explained the balance between traditional optimization techniques and modern AI-driven approaches. “Conventional methods focus more on local optimization, while AI-based models can explore larger and more complex spaces,” Matteo pointed out. He mentioned that AI’s adaptability and exploration capabilities make it suitable for addressing nonlinear, multi-objective problems in product development.

However, Matteo also emphasized the importance of accuracy, stating that while AI models can handle larger design spaces, they may sacrifice precision. “Combining AI with conventional models helps achieve the right balance between exploration and accuracy,” he explained. Matteo sees AI as a valuable tool that can facilitate more informed decision-making during the design phase by reducing repetitive tasks and allowing engineers to focus on innovation.

Challenges of adoption and the human factor

The panel also touched upon the challenges of adopting AI and simulation technologies at scale. One of the key takeaways was the need to engage top management, middle management, and engineering teams to ensure successful adoption. “Middle management is where change often goes to die,” David Long noted, emphasizing the importance of aligning incentives and ensuring that middle managers understand how new technologies will benefit them and their teams.

Jan Paul highlighted that successful adoption requires creating a competitive spirit within teams and showcasing success stories to inspire others. He mentioned that many companies are still in the early stages of using AI for digital engineering, with most applications being proof of concepts rather than fully scaled initiatives. “We need to create forums where teams can share best practices and learn from each other,” Jan Paul said, adding that fostering a collaborative culture is key to driving widespread adoption.

Matteo also stressed the importance of starting small and focusing on practical use cases to enhance engineers’ day-to-day work. “Think small and start with use cases that truly help the end user in their daily tasks,” Matteo advised. He believes starting with manageable goals and demonstrating tangible benefits can help build momentum for larger initiatives.

The future of engineering: Continuous innovation

The panelists expressed excitement about the future of AI and advanced simulation in engineering. Tom Martinek mentioned that he is eager to see what new innovations will emerge as AI evolves. “AI is unlocking a whole new realm of possibilities for innovation. I can’t wait to see what will come of it,” he said.

David Long shared a vision of engineering transformation, where AI and simulation allow engineers to focus more on creativity while handling routine tasks more efficiently. He believes that we are witnessing a transformation in engineering processes, moving from siloed approaches to continuous value delivery.

Jan Paul offered an inspiring thought about using AI for the greater good. “Using AI and simulation for sustainability, like designing electric vehicles that consume less energy or washing machines that use less water, is something I’m particularly excited about,” he said. He also touched on the potential for fully connected toolchains that would allow engineers to design optimized products more seamlessly in the future.

Closing thoughts

The “Pulling the Digital Thread” panel provided a comprehensive overview of how AI and advanced simulation are reshaping digital engineering, and explored trending topics including manufacturing, and product lifecycle management from the perspective of a connected digital enterprise.

From accelerating time to market to democratizing systems engineering and enhancing product optimization, AI is poised to revolutionize how we design and build products. However, the journey requires careful adoption, a focus on data management, and an emphasis on human-centered approaches to technology integration.

Let’s keep this conversation going! The panelists encouraged everyone to continue the dialogue on social media, sharing insights and best practices to collectively advance the field of digital engineering.

Interested in hearing more? Watch the on-demand version here.