For a long time, the way automotive products were designed and made was largely the same. It was a slow, time-consuming process defined by trial and error and physical prototypes. Big projects required many people with many specialized skill sets. And success was not always guaranteed.
Today, however, the industry operates at an extraordinarily accelerated pace. Transformative new technologies like generative artificial intelligence (GenAI) and large-scale additive manufacturing have opened a world of new possibilities — in electrification, autonomous systems, and more —and have shortened design cycles. Products are more reliable and more sustainable. It is an exciting time, but with compressed design cycles and furious competition, engineers and designers are presented with new challenges: How can we implement new technology and meet market demand without compromising performance and quality?
The answer is AI-powered engineering. AI-powered engineering unleashes the power of AI across the entire product life cycle. By seamlessly embedding AI into design and simulation tools, teams can accelerate exploration and innovation. Integrated user-friendly workflows, coupled with on-demand high-performance computing (HPC) resources, empower teams to augment AI with human expertise. Most powerfully, AI-powered engineering workflows can deliver results up to 1,000x faster than traditional physics-based simulation workflows.
Beyond Just Simulation
Automakers and suppliers know the importance of simulation and computer-aided engineering (CAE). But AI-powered engineering encompasses much more than just simulation. Complete AI-powered engineering workflows combine simulation with organization-wide data analytics and HPC to create new, unique, data streams.
AI-powered engineering also allows designers to consider multiple objectives simultaneously. Teams can then use these workflows to optimize vehicles’ aesthetics, comfort, aerodynamics, noise, vibration, and harshness (NVH), and manufacturability.
For example, automakers can train machine learning algorithms to detect NVH issues earlier in the design life cycle. These algorithms, which can be created using low- and no-code tools, can learn to understand the effects of many design variables on NVH performance and analyze results 100x faster than traditional methods. AI-powered engineering solutions can learn from historical and synthetic data and streamline the design and testing cycle — speeding time to market, increasing efficiency, and lowering costs.
Fast Innovation for a Fast Industry
In addition to its power in design and testing, one of AI-powered engineering’s main benefits is the time it saves. OEMs and suppliers know that delays, downtime, and lengthy iteration impacts profitability, and causes teams to miss market opportunities. Luckily, AI-powered engineering tools take a fraction of the time needed for traditional solver simulation while enabling teams to test and explore more designs for better decision-making and better outcomes.
Moreover, AI-driven models can approximate complex physics with minimal computational cost, allowing for real-time optimization without the need to rerun full-scale simulations. This lets engineers test and validate multiple designs in parallel rather than sequentially. And automakers can reuse AI-trained models across different vehicle platforms, eliminating redundant work and unlocking valuable insights from past projects.
For instance, automakers can use data and AI to conquer aerodynamic challenges that previously took a long time using computational fluid dynamics (CFD) tools. In one real-world use case, a manufacturer trained a machine learning algorithm to analyze aerodynamic performance on a large automotive model comprised of more than 2 million elements. Using traditional CFD tools, simulation runtimes regularly exceeded 12 hours. But the runtimes of AI-powered tools trained on historical data were completed in a mere three minutes — with the same level of accuracy as the CFD results. From days to minutes: the promise of AI-powered engineering.
A New Design Paradigm
The ultimate objective of AI-powered engineering is to transform the ways automakers design, test, and manufacture components, systems, and vehicles — to usher in a new era of automotive innovation. This is enabled by next-generation digital twins, which empower organizations to revolutionize the way they conceptualize their entire design and operations process.
Primarily, by creating a connected, dynamic virtual replica of a real-life asset, digital twins help teams rely less on costly, time-consuming physical prototypes. Digital twins also combine simulation, HPC, AI, and data analytics capabilities so teams can evaluate what-if scenarios, enable predictive maintenance, and extend the remaining useful life (RUL) of their products. Moreover, they enable a comprehensive cross-functional system evaluation that eliminates information silos and communication bottlenecks. With all this, organizations and their products become more efficient from concept design to in-service performance to end of life.
With AI-powered engineering and digital twin capabilities, organizations can drive automotive lightweighting, battery innovation, sustainability, and achieve zero-prototype development. For example, teams can use AI-powered engineering workflows to gain real-time insight into electric vehicles’ state of charge (SoC) with over 98 percent accuracy; achieve continuous real-time monitoring of critical battery KPIs without needing physical sensors; predict the state of battery health and RUL in vehicles in service; reduce the weight of crucial components; improve battery effectiveness and thermal management; and so much more. Overall, the aim is simple: creating vehicles and processes that are lighter, more powerful, more efficient, more sustainable, and more affordable.
Doing More with Less
AI-powered engineering is transformative for many reasons, but it all boils down to doing more with less. These solutions give engineers and designers a head start in design as well as equip them with the resources to test iterations much faster than traditional solver simulation. They give teams the power of all prior institutional knowledge cross functionally.
In addition, AI-based solutions deliver high-fidelity results with less computation, fewer resources, and reduced costs — meaning automakers benefit from faster innovation, faster decision-making, and a leaner, more cost-effective development process.
AI-powered engineering is the ultimate value-add technology. It’s superb on the individual level because it empowers engineers in ways that have not been historically possible. And organizationally, AI-powered engineering will reduce development time and costs while increasing products’ sustainability, safety, and time to market.
This article was written by Royston Jones, CTO of Product Design and SVP of Automotive, Altair (Troy, MI). For more information, here .
