Samsung Electronics DX Division Accelerates Digital Twin Transition
Catalyst for the 2030 'AI Autonomous Factory' Vision

Samsung Electronics is introducing "digital twin" technology—which replicates real-world products in virtual spaces for simulation—into its pre-development processes. By establishing high-performance computing (HPC) infrastructure in the mass production sectors of home appliances and smartphones, previously considered exclusive to high-cost manufacturing industries like automotive and aerospace, the company is seen as establishing a key artery for its declared transition to the "AI Autonomous Factory by 2030."


On June 15, the Device eXperience (DX) division of Samsung Electronics announced that, to reinforce development competitiveness based on digital twin technology, it had rapidly set up 517 HPC servers equipped with high-performance central processing units (CPUs) at its Sangam-dong data center in Seoul. Full-scale service began this month for mechanical and circuit development personnel.

"15-Day TV Test Completed in 2 Days"...Samsung Electronics Deploys 517 HPC Servers in Sangam View original image

Digital twin is an advanced technology that creates a virtual twin in a computer environment and simulates a variety of possible scenarios to predict outcomes. It dramatically reduces the time and costs wasted on repeatedly creating physical prototypes and preemptively eliminates on-site errors. For this reason, it is a core infrastructure in which global manufacturing companies are actively increasing investment.


The newly established in-house HPC infrastructure at Samsung Electronics is approximately 5.8 times faster in processing speed than previous systems and can handle nearly six times more virtual verifications. The performance improvements are clearly demonstrated in the numbers. For example, TV drop tests that previously required 15 days of repeated physical trials can now be completed in just two days, while washing machine drop tests have been reduced from 15 days to five days.


In particular, the scope of verification is greatly expanded to areas that were previously impossible to tackle due to physical time and cost constraints. In the "all-angle drop verification" for smartphones, which measures damage by dropping devices from every possible angle, as many as 700 cases can be simulated in the virtual environment and verified in just one day. Long-term diaphragm (rubber packing) testing for washing machines, which once depended on lengthy physical tests, can also be completed within two days.


This virtual verification system will be rolled out simultaneously across all business units in the DX division. Starting with drop tests in the Mobile eXperience (MX) division for smartphones, the pre-verification processes for all major product lines—including TV drop and heat tests in the Visual Display (VD) division, collision tests for washing machines and robot vacuum cleaners in the Digital Appliances division, and radio unit (RU) heat dissipation tests in the Network division—will be fully virtualized.


Industry insiders are paying close attention to Samsung Electronics' decision to build its own ultra-large infrastructure in-house, rather than relying on outside cloud services. This approach secures the safety of processing core assets such as product design drawings and sensitive quality verification data internally without concerns of external leaks, and also allows for flexible 24-hour response to large-scale analysis demands.


There is also analysis that the operation of this HPC infrastructure is the final piece in completing the strategy for transitioning to the "AI Autonomous Factory by 2030," declared by Samsung Electronics in March. An AI autonomous factory is an intelligent manufacturing infrastructure where AI agents and robots autonomously understand and make decisions throughout the entire process, from materials intake to production and shipment. While the newly operational HPC service is responsible for the digital twin in the "development" phase, the AI autonomous factory covers the digital twin of the "manufacturing" site. As the precisely accumulated simulation data from the development stage connects seamlessly to the manufacturing process, a "full-stack AX (AI transformation)" encompassing the entire product life cycle is completed.



An industry official commented, "HPC services will become the starting point for establishing the digital twin as a routine work method in the development field. As more virtual verification data accumulates, both accuracy and applicability will expand, further accelerating Samsung Electronics' digital twin transition toward the AI Autonomous Factory by 2030."


This content was produced with the assistance of AI translation services.

© The Asia Business Daily. All rights reserved. Unauthorized AI training and use prohibited.

Today’s Briefing