December 10, 2024 The Benefits of AI-Driven Simulation in Engineering

The integration of AI with simulation technology is reshaping manufacturing through enhanced predictive insights and process optimizations.

By: Paweł Z. Chądzyński, Senior Director, Strategic Research, Aras

Engineering companies have long operated on the cutting edge, constantly enhancing their capabilities with smarter workflows. Despite their continuous innovation, certain aspects of systems engineering and product development have stayed in traditional siloes. This includes simulations used in R&D and product design.

However, as artificial intelligence (AI) and machine learning (ML) technologies mature and become more accessible, a growing segment of engineering companies are integrating AI with simulations. The combination of AI and simulation technologies is already driving rapid innovation, enabling enhanced product quality, optimized designs, and — most importantly to customers — a reduced time to market.

AI’s Evolving Role in Simulations and Systems Engineering

While traditional simulations are effective, they come with notable limitations. They require extensive expertise, meticulous setup, ongoing management; they’re challenging to adapt to changing requirements; and lack the speed and scalability of AI-driven solutions, as most engineers can only run one complex simulation at a time. They are also difficult to democratize since they often require hands-on assistance from simulation experts.

When AI is strategically incorporated into the simulation processes, it unlocks new advantages in adaptability, scalability, reusability, and affordability. Many of these advantages boil down to the speed with which neural networks and AI algorithms solve problems.

Integrating AI into modern simulation flows allows engineers to update and scale their simulations quickly. They can even run multiple simulations simultaneously to optimize product design and discover new customization opportunities. Engineers may also run several different simulations to explore new possibilities they would have overlooked without predictive and prescriptive AI input. AI-driven simulation enables a scalable, multipronged approach to problem-solving that democratizes simulation process and aids in multidisciplinary product development.

Industry-Specific Applications of AI-Driven Simulation

Certain industries have been slower to adopt AI, but many product-driven industries are already improving their simulations with AI technology:

• Automotive: AI-driven simulation allows automotive engineers to complete virtual tests of vehicle performance, safety, and durability, often simulating real-world conditions. With this approach, automotive companies can identify and mitigate design flaws earlier and faster to avoid developing unnecessary physical prototypes and improve time to market.
• Aerospace: AI-driven aerospace simulations optimize aircraft design, upgrade aerodynamic performance, enhance safety features, and improve customer-facing features on commercial aircraft.
• Medical Devices: AI-driven simulations are used in several unique ways in medicine, including for medical device design optimization, surgical procedure tests, and HIPAA-compliant synthetic data creation.
• Consulting and Professional Services: In professional services companies that develop consumer products, AI-driven simulation creates user-friendly visualizations that customers can review and comment on. This also helps to iteratively update the design and minimize delays in the project timeline.

Developing an Effective Data Management Strategy for AI-Driven Simulation

AI models used in simulation (as opposed to simulation models as such) are only useful if trained with accurate, well-organized, relevant, and abundant data. Organizations must prioritize data management strategies that ensure data accuracy, consistency, and accessibility across all data sources for the best possible results.

While data management tasks like data collection, cleansing, and quality testing are important, many businesses overlook the most important component of data management: Identifying a system for organizing data, adding context, and making data understandable for AI models and users. Particularly in the product design phase of a manufactured product, it’s important to organize data, data relationships, design models, and other simulation assets through a digital thread infrastructure. This strategy will improve traceability for future simulations while providing the information necessary to make smart simulation decisions now.

Preparing Workflows and Workforces for AI-Driven Simulation

AI is quickly reshaping the role of engineers and the workflows in which they operate. While this shift leads to uncertainty, there are several steps business leaders can take to empower their employees and step into AI adoption smoothly:

• Invest in data management, digital thread, and cloud technology: Several emerging technologies help engineering teams manage their datasets, AI models, and other project resources effectively. Consider investing in digital thread, product lifecycle management, data integration, data governance, and other scalable platforms to keep your team organized and operational.
• Tailor workforce training: Training for AI-driven simulation should be customized to each relevant role within the company. Training should not only focus on how to use AI in product development and simulation but also on why and how AI will make workers’ daily routines more efficient.
• Apply “zero trust” practices to AI usage: While AI technology introduces many new efficiencies, it can make mistakes and incorrect assumptions. It’s important to train all users to look for errors in both training data and AI-driven simulation outputs so performance issues can be corrected.

For the long-term success of AI-driven simulation and adoption, it’s essential to empower your workforce to continuously develop adaptive thinking and a strong understanding of AI tools. Over time, this foundation will equip them with the skills to harness AI as a core driver of future design innovations.

About the Author:
Paweł Z. Chądzyński serves as Senior Director of Strategic Research at Aras Corporation, where he leads the strategy and direction for advanced solutions on the Aras PLM Platform. His focus areas include Requirements Management, Model-Based Systems Engineering (MBSE), Simulation, Application Lifecycle Management (ALM), and interdisciplinary design methodologies.

Prior to joining Aras, Paweł held management and technology leadership roles at Cadence, PTC, and OHIO-DA, a successful high-tech software startup. He earned a master’s degree in technology management and a bachelor’s degree in electrical engineering and computer science from NYU Polytechnic School of Engineering.

Subscribe to Industry Today