Trends of AI in industry part 2: Speeding up the mundane

Every year AI grows more integrated with a wide range of industries, taking on a greater number of increasingly important tasks. In this series I explore the ways different industries are utilizing AI and how, despite their differing approaches and requirements, they are often working to achieve the same goals through the use of AI. It is these trends that will shape the future of AI research and usage going forward, perhaps showing a glimpse of what living and working in the world of tomorrow will look like.

In the first part of this series, I looked at the way AI is reshaping the user experience, especially with respect to professional software. In this second part, I examine the ways AI is speeding up and automating slow and mundane tasks, driving more efficient workflows and helping engineers and designers get back to doing what they were trained to do.

How AI is speeding up simulation

High-fidelity simulations, while crucial to many design processes, are also notoriously slow. However, by leveraging AI and machine learning (ML) different tools are developing methods to speed up and get the most of every simulation run.

Within the Simcenter suite of tools, part of the Xcelerator portfolio, AI/ML-driven reduced order models (ROMs) and surrogate models are trained using a carefully selected sampling of simulation data after which they can be used to infer accurate simulation results in real time across the entire domain of interest. These techniques can greatly reduce the total number of simulations needed during the design process, reducing the amount of time designers are left waiting for runs to finish while also helping accelerate future design cycles by reusing existing models.

The computer aided engineering space is not the only one to benefit from the power of machine learning. In the complex and fast-paced world of chip design, Solido, a part of Siemens EDA, has been taking advantage of ML for nearly 15 years to reduce the number of simulations required to validate a chip design from billions down to just a few thousand.

Thanks to the highly parallel nature of microprocessors and the vast quantities of highly accurate data available from SPICE simulations, chip design and validation were prime candidates to reap the early benefits of ML. By using a smarter approach to simulation, Solido can test more edge cases while areas where failure are rarer can be validated rather than retested without compromising the results. Not only does this approach provide as much as a 1,000,000x reduction in required simulations, it is also the only practical way to achieve the required level of accuracy with how large and complex modern chips have become.

Automation of mundane tasks

When it comes to complex design workflows, there are often a slew of menial tasks that are vitally important, but are boring, repetitive, and time-consuming for users to complete. One example of this is part classification, such as preparing the model of a car for aerodynamic simulation. Each part, down to the smallest screw or washer, must be manually categorized as a “screw” or “washer” or any number of other components. This is slow and repetitive work requiring human eyes and understanding to accomplish – at least until now. Thanks to recent advances in AI, Simcenter has been able to implement fully automatic part classification capable of fully the hundreds or thousands of parts in something like a car in a matter of hours instead of the days of manual labor it previously took.

As AI models grow more sophisticated, an increasing number of these mundane yet complex tasks will be completed automatically with minimal human oversight required. This not only frees expert users to do the skilled work they trained for but allows tasks to be completed in a fraction of the time with potentially fewer errors.

AI clears the path

Nothing it more disruptive to a workflow than having to stop and wait for a slow simulation to run or go back and check if every part in an assembly is correctly labeled. Especially when creating something new, the spark of creativity can be easily lost when waiting for a simulation that will take more than 24 hours to run. With AI, however, this could be a thing of the past with simulation results inferred instantly and parts classified automatically, clearing the obstructions of slow and menial tasks from the path of creativity and design.


Siemens Digital Industries Software is driving transformation to enable a digital enterprise where engineering, manufacturing and electronics design meet tomorrow. Xcelerator, the comprehensive and integrated portfolio of software and services from Siemens Digital Industries Software, helps companies of all sizes create and leverage a comprehensive digital twin that provides organizations with new insights, opportunities and levels of automation to drive innovation.

For more information on Siemens Digital Industries Software products and services, visit siemens.com/software or follow us on LinkedIn, Twitter, Facebook and Instagram.

Siemens Digital Industries Software – Where today meets tomorrow.

Want to stay up to date on news from Siemens Digital Industries Software? Click here to choose content that's right for you

Leave a Reply

This article first appeared on the Siemens Digital Industries Software blog at https://blogs.sw.siemens.com/thought-leadership/2022/12/27/trends-of-ai-in-industry-part-2-speeding-up-the-mundane/