This is a long order, but that’s what Jaff said is that artificial intelligence (AI) technology can help accurately capture data and guide engineers through product design and development.
A November 2020 McKinsey survey found that more than half of companies adopted AI in at least one activity, and 22% of respondents reported that at least 5% of their company’s expense earnings were responsible for AI. And in the manufacturing case, 71% of respondents saw an income of 5% or more with AI adoption.
But that was not always the case. Jaff said that once “rarely used in product development”, AI has experienced evolution. Today, technologists known for innovation in AI, such as Google, IBM, and Amazon, have set new standards for the use of AI in other processes, such as engineering.
“AI is a committed and exploratory field that can significantly improve the user experience for engineers designing, as well as gather relevant information about the development process for specific applications,” said Catherine Wickert, Director of Industrial Solutions at Siemens Industry Software. Says.
The result is a growing appreciation of technology that promises to simplify complex systems, bring products to market faster, and drive product innovation.
Simplification of complex systems
Renault is a perfect example of the power of overhaul AI in product development. In response to growing consumer demand, the French automaker is equipping an increasing number of new car models with an automatic manual transmission (AMT) that behaves like an automatic transmission but allows drivers to shift gears electronically using a push-button command.
AMTs are popular among consumers, but designing them can present the ultimate challenge. This is because the performance of the ATM depends on the operation of three distinct subsystems: an electronic-mechanical actuator that transfers gears, an electronic sensor that monitors the status of the vehicle, and software embedded in the transmission control unit that controls the engine. Due to 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 system as a whole.
In an effort to simplify the ATM development process, Renault returned to Siemenster Amesim software from Siemens Digital Industries Software. Simulation technology relies on artificial neural networks, AI “learning” systems modeling loosely in the brain. Engineers simply drag, drop, and attach icons to create a model graphically. When displayed as a sketch, the model illustrates the relationship between all the components of an AMT system. Instead engineers can make predictions about AMT’s behavior and effectiveness and make the necessary refinements early in the development cycle to avoid late-stage problems and delays. In fact, by using transitions as a virtual engine and stand-in during hardware development, Renault has been able to cut ATM development time by almost half.
Speed without sacrificing quality
So, very effectively rising environmental standards are persuading Renault to rely more on AI. To meet rising carbon dioxide emissions standards, Renault is working on the design and development of hybrid vehicles. However, hybrid engines are more developed than a conventional car with a single power source. This is because hybrid engines require engineers to perform complex fights, such as maintaining the required energy balance from multiple power sources, choosing from multiple architectures, and examining the effects of transmission and cooling systems on a vehicle’s power performance.
“To meet the new environmental standards for a hybrid engine, we must rethink the architecture of petrol engines,” said Vincent Talon, head of simulation at Renault. The problem, he adds, is that carefully examining “hundreds of different actuators that can affect the final outcome of fuel intake and pollutant emissions” is a long and complicated process, complicated by rigorous deadlines.
“Today, we obviously don’t have time to evaluate different hybrid powertrain architectures,” Talon said. “Rather, we need to use an advanced method to handle this new complexity.”
For more information about AI in industry applications, see www.
Download Full report.
This content was produced by Insights as a custom content of MIT Technology Review. This is not written by the editorial staff of MIT Technology Review.