Thought Leadership

AI-Driven Product Design

There can be no doubt that products are getting more complicated. For example, compare smartphones today with the cellphones of the early 90s and it’s easy to see how far product design has come at every level. While smartphones represent one of the most prominent cases of increasing design complexity you would be hard pressed to find a product that hasn’t become more complex over the last 3 decades. Cars, appliances, clothes, the list is endless. This is driven by several factors including: consumer demand for the best, the greenest, and the most innovative products, the truly global mass market allows for goods of greater complexity to be manufactured cheaper, and improvements in IC design and miniaturization allow any device to be made “smart” with an inexpensive chip and a few sensors. Add all these things together and it’s easy to see how anything and everything we use in our daily lives is becoming a complex piece of advanced technology.

We all appreciate the benefits these improvements bring to everyday life, either directly or through enhancements in manufacturing driving down costs and improving quality. But the demand this places on designers and engineers is a heavy one. While the product design space has limitless potential for innovation in functionality, design, and materials the capacity for designers to manually balance all these competing conditions is anything but limitless. As systems get increasingly interconnected and complex, even small changes can have a large ripple effect. Now, the difference between the best product design and “good enough” could be a few very small decisions that would statistically only come to light after hundreds or thousands of design iterations, something that is not feasible for human designers to do.

Likewise, in the emergent field of additive manufacturing, parts can be created in arbitrary shapes with extreme precision that would have been impractical or even impossible using traditional subtractive methods.  Again, the challenge is finding the best design, in a domain with infinite possibilities, how is a designer supposed to pick out the best one? Well, it turns out AI might be the solution.

Even as products have become more complex, the tools used to create them have as well. And thanks to the rapid advances in AI over the last decade, design tools are starting to get smarter too. Through the power of AI, a designer can simply input their requirements and allow the AI model to automatically generate and check hundreds or even thousands of designs before presenting the best ones to the designer for validation. All of a sudden, what was previously impossible becomes very possible, AI handles the long and mundane task of creating numerous different designs based on the input parameters freeing designers to focus on tuning the few resulting best designs. This has the added benefit of uncovering and offering up “unconventional” designs, which is to say designs that wouldn’t have been modeled and tested by hand since they fall far outside standard practices but still fulfill the criteria the AI uses to design and test models. While this is more of an edge case, it is one I find fascinating since it holds the possibility of creating truly avant-garde designs by merging the strengths of designer and machine.

Of course, as exciting as the prospect of having an AI design something completely from a list of parameters is, the more practical benefit currently is its ability work with incredibly complex systems that would bog down a conventional workflow. This is exactly what Renault did to help overcome the challenges of designing automatic manual transmissions (AMTs) for its cars. AMTs are difficult to design since they require a balance between 3 different systems: the electro-mechanical gear shift, electronic sensors, and embedded software to control the engine. It can take up to a year of extensive trial and error to define the systems requirements, design the mechanics, develop the software, and validate the system. In an effort to streamline this process, Renault turned to Siemens Amesim software which relies on AI to provide an intuitive drag-and-drop system to graphically create a model of the AMT which the engineers then use to predict the way various different configurations will behave and perform, allowing necessary refinements to be made early in the design stage, instead of later on when they would be more costly in both time and money. Using this technology, Renault has been able to cut AMT development time almost in half.

Renault is one of many companies starting to leverage AI to optimize traditional product design work, but as additive manufacturing continues to grow in prominence it won’t be long before we start to see AI take center stage in design. As I mentioned earlier, AM allows for parts to be created in any number of novel shapes which can be difficult for human designers to even conceive of but with AI, parts can be optimized by use case, expected forces, weight, size, and any number of other criteria, truly creating the “best solution for the job” for any scenario.

The push for better, smarter products drives product designers to create increasingly complex and innovative products in a process that might even be called evolution. But evolution doesn’t happen in a vacuum, it requires both external pressure and the capacity to adapt to it. To keep pace with these demands for product evolution, designers and engineers will need to rely on AI to keep up with the increasingly complex and interconnected systems that make up even the simplest products in modern society. So, does this mean an AI will be fully responsible for designing your next smartphone? Probably not. But it does mean the team of engineers who do design it will be relying on AI to help deliver the fastest and most innovative device to date.

To read more about how AI is transforming how we design and manufacture products check out this whitepaper: Link


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.

Spencer Acain

Leave a Reply

This article first appeared on the Siemens Digital Industries Software blog at https://blogs.stage.sw.siemens.com/thought-leadership/2022/02/22/ai-driven-product-design/