The product design receives an AI revision

It is a huge job, however in keeping with Zapf, synthetic intelligence (AI) expertise can assist seize the proper information and information engineers via product design and improvement.

No surprise a McKinsey survey from November 2020 discovered that greater than half of firms have adopted AI in not less than one position, and 22% of respondents say that AI accounts for not less than 5% of their company-wide revenues. In manufacturing, 71% of respondents noticed gross sales progress of 5% or extra with the introduction of AI.

However that has not at all times been the case. Previously “hardly ever utilized in product improvement”, AI has seen a improvement in recent times, says Zapf. At this time, expertise giants comparable to Google, IBM and Amazon, identified for his or her improvements in AI, have “set new requirements for the usage of AI in different processes,” for instance in engineering.

“AI is a promising and exploratory space that may considerably enhance the person expertise for designers and collect related information within the improvement course of for particular functions,” says Katrien Wyckaert, Director of Trade Options at Siemens Trade Software program.

The result’s a rising appreciation for a expertise that guarantees to simplify complicated methods, carry merchandise to market quicker, and drive product innovation.

Simplify complicated methods

An ideal instance of AI’s capacity to revamp product improvement is Renault. In response to rising client demand, the French automaker is equipping a rising variety of new automobile fashions with an automatic guide transmission (AMT) – a system that behaves like an computerized transmission however permits the driving force to shift gears electronically with the push of a button.

AMTs are widespread with customers, however creating them might be daunting. It’s because the efficiency of an AMT is determined by the operation of three completely different subsystems: an electromechanical actuator that shifts gears, digital sensors that monitor automobile standing, and software program that’s embedded within the transmission management module that controls the engine. Due to this complexity, it might probably take as much as a 12 months to outline the useful necessities of the system, design the actuator mechanics, develop the mandatory software program and validate the general system.

To optimize the AMT improvement course of, Renault turned to the Simcenter Amesim software program from Siemens Digital Industries Software program. The simulation expertise relies on synthetic neural networks, AI studying methods which might be loosely modeled on the human mind. Engineers merely drag and drop symbols and join them to graphically create a mannequin. When the mannequin is displayed as a sketch on a display, it reveals the connection between all of the completely different parts of an AMT system. In return, engineers can predict the conduct and efficiency of the AMT and make any needed enhancements early within the improvement cycle to keep away from issues and delays later. Through the use of a digital engine and gearbox to exchange {hardware} improvement, Renault has succeeded in nearly halving the AMT improvement time.

Velocity ​​with out sacrificing high quality

The rising environmental requirements are additionally inflicting Renault to rely extra closely on AI. To fulfill rising carbon dioxide emissions requirements, Renault has labored on the design and improvement of hybrid automobiles. Nevertheless, the event of hybrid engines is way extra complicated than that of automobiles with a single vitality supply, e.g. B. a standard automotive. It’s because hybrid engines require engineers to carry out complicated duties, e.g. B. balancing the ability necessities of a number of vitality sources, selecting from a wide range of architectures, and analyzing the consequences of transmissions and cooling methods on the vitality effectivity of a automobile.

“With the intention to meet new environmental requirements for a hybrid engine, we now have to utterly rethink the structure of gasoline engines,” says Vincent Talon, Head of Simulation at Renault. The issue, he provides, is that rigorously analyzing “the dozen completely different actuators that may have an effect on the underside line of gasoline economic system and pollutant emissions” is a prolonged and complicated course of made tough by inflexible schedules.

“At this time we clearly do not have the time to rigorously consider completely different hybrid powertrain architectures,” says Talon. “Relatively, we had to make use of a complicated methodology to cope with this new complexity.”

You could find extra details about AI in industrial functions at www.siemens.com/artificialintelligence.

Obtain the total report.

This content material was created by Insights, the customized content material arm of MIT Expertise Assessment. It was not authored by the editorial employees of MIT Expertise Assessment.