The product design receives an AI revision


It’s a big job, but according to Zapf, artificial intelligence (AI) technology can help capture the right data and guide engineers through product design and development.

No wonder a McKinsey survey from November 2020 found that more than half of companies have adopted AI in at least one role, and 22% of respondents say that AI accounts for at least 5% of their company-wide revenues. In manufacturing, 71% of respondents saw sales growth of 5% or more with the introduction of AI.

But that has not always been the case. In the past “rarely used in product development”, AI has seen a development in recent years, says Zapf. Today, technology giants like Google, IBM and Amazon, known for their innovations in AI, have “set new standards for the use of AI in other processes,” for example in engineering.

“AI is a promising and exploratory area that can significantly improve the user experience for designers and gather relevant data in the development process for specific applications,” says Katrien Wyckaert, Director of Industry Solutions at Siemens Industry Software.

The result is a growing appreciation for a technology that promises to simplify complex systems, bring products to market faster, and drive product innovation.

Simplify complex systems

A perfect example of AI’s ability to revamp product development is Renault. In response to increasing consumer demand, the French automaker is equipping a growing number of new vehicle models with an automated manual transmission (AMT) – a system that behaves like an automatic transmission but allows the driver to shift gears electronically with the push of a button.

AMTs are popular with consumers, but developing them can be daunting. This is because the performance of an AMT depends on the operation of three different subsystems: an electromechanical actuator that shifts gears, electronic sensors that monitor vehicle status, and software that is embedded in the transmission control module that controls the engine. Because of this complexity, it can take up to a year to define the functional requirements of the system, design the actuator mechanics, develop the necessary software and validate the overall system.

To optimize the AMT development process, Renault turned to the Simcenter Amesim software from Siemens Digital Industries Software. The simulation technology is based on artificial neural networks, AI learning systems that are loosely modeled on the human brain. Engineers simply drag and drop symbols and connect them to graphically create a model. When the model is displayed as a sketch on a screen, it shows the relationship between all the different elements of an AMT system. In return, engineers can predict the behavior and performance of the AMT and make any necessary improvements early in the development cycle to avoid problems and delays later. By using a virtual engine and gearbox to replace hardware development, Renault has succeeded in almost halving the AMT development time.

Speed ​​without sacrificing quality

The emerging environmental standards are also causing Renault to rely more heavily on AI. To meet emerging carbon dioxide emissions standards, Renault has worked on the design and development of hybrid vehicles. However, the development of hybrid engines is far more complex than for vehicles with a single energy source, such as e.g. B. a conventional car. This is because hybrid engines require engineers to perform complex tasks, e.g. B. balancing the power requirements of multiple energy sources, choosing from a variety of architectures, and examining the effects of transmissions and cooling systems on the energy efficiency of a vehicle.

“In order to meet new environmental standards for a hybrid engine, we have to completely rethink the architecture of gasoline engines,” says Vincent Talon, Head of Simulation at Renault. The problem, he adds, is that carefully examining “the dozen different actuators that can affect the bottom line of fuel economy and pollutant emissions” is a lengthy and complex process made difficult by rigid schedules.

“Today we clearly don’t have the time to carefully evaluate different hybrid powertrain architectures,” says Talon. “Rather, we had to use an advanced methodology to deal with this new complexity.”

You can find more information on AI in industrial applications at

Download the full report.

This content was created by Insights, MIT Technology Review’s custom content area. It was not authored by the editorial staff of MIT Technology Review.


Steven Gregory