Memory Usage Reduced to One-Tenth, Accuracy Improved
Priority Application to In-House Manufacturing Process Innovation
"Development Efficiency Expected to Improve Significantly"
LG Electronics has developed a technology that reduces the quality prediction time of finished products by up to 99% compared to the existing method by utilizing artificial intelligence (AI) technology.
On the 27th, LG Electronics announced that it has developed a technology that shortens the development period and improves production efficiency by reducing repetitive verification time during the product development process. Previously, quality prediction before prototype production took 3 to 8 hours each time, but with the introduction of this AI technology, quality can be predicted within 3 minutes without separate simulation.
In particular, LG Electronics explained that compared to existing deep learning-based AI technology, the learning time was reduced by more than 95%, and memory usage was reduced to about one-tenth. At the same time, the accuracy of quality prediction improved by more than 15%. Another distinctive feature is that the prediction results are visualized in 3D form, allowing developers to intuitively review them. LG Electronics also plans to launch a web-based service platform called 'Eng.AI' so that developers can utilize this technology in their work.
LG Electronics plans to first apply this AI technology to its own manufacturing process innovation. It will be sequentially introduced in the design of home appliances such as refrigerators and washing machines, as well as TVs and automotive electronic components. It also plans to apply it to the smart factory solution business to strengthen manufacturing consulting and external order competitiveness.
Using the web-based service platform 'Eng.AI' to predict the quality of refrigerator parts, provided by LG Electronics
This year, the external order scale of LG Electronics' Smart Factory Solution at the Production Technology Institute is expected to reach around 400 billion KRW, an increase of more than 30% compared to the previous year. LG Electronics plans to expand its customer base not only in the home appliance industry but also in various industries such as semiconductors, pharmaceuticals, and chemicals, aiming to grow this business into a trillion-won scale by 2030.
Currently, the smart factory solution is focused on ▲ production system design, monitoring, and operation ▲ equipment and process management based on big data and generative AI ▲ industrial robots. The newly added AI-based quality prediction technology will further strengthen product development and production element competitiveness.
Since last year, LG Electronics has been conducting joint research with Professor Park Chan-young’s team at the Korea Advanced Institute of Science and Technology (KAIST) to study verification technology for streamlining the product design process. Related papers will be submitted to the International Conference on Machine Learning (ICML), a prestigious academic conference in the AI field, to receive academic verification of the technology.
Jung Dae-hwa, President of LG Electronics' Production Technology Institute, said, "We provide optimal solutions to customers at every stage of the smart factory," adding, "Through AI-based verification technology, the product development cycle will be shortened, and development efficiency will also be greatly improved."
© The Asia Business Daily(www.asiae.co.kr). All rights reserved.
