2 months ago
Published [hour]:[minute] [AMPM] [timezone], [monthFull] [day], [year]
- Two associated world papers connected AI-based metrology exertion presented
- Gauss Labs’s innovative exertion improves process power and instrumentality productivity successful semiconductor manufacturing
- Gauss Labs is astatine nan forefront, starring business AI efforts to toggle shape nan manufacturing industry
SEOUL, South Korea, Feb. 29, 2024 /PRNewswire/ -- SK hynix Inc. (or “the company”, www.skhynix.com ) and Gauss Labs announced coming that they participated successful nan SPIE AL* 2024, an world convention held successful San Jose, California, and presented 2 papers based connected nan latest exertion for AI-based metrology.
* SPIE Advanced Lithography + Patterning (SPIE AL): a convention hosted by nan Society of Photo-Optical Instrumentation Engineers (SPIE), nan astir prestigious world nine successful optics and photonics pursuing instauration successful 1955, wherever wide lithography exertion for drafting semiconductor circuits is discussed |
*Metrology: measuring nan beingness and electrical characteristics to guarantee that requirements are met successful nan manufacturing process |
SK hynix has been collaborating intimately pinch Gauss Labs successful various areas to summation semiconductor output and productivity, and nan results of this collaboration are published successful these 2 papers.
In nan insubstantial “Model Aggregation for Virtual Metrology for High-Volume Manufacturing,” Gauss Labs introduces “aggregated AOM*", an algorithm that increases nan prediction accuracy of its AI-based virtual metrology solution, Panoptes VM (Virtual Metrology).
*Adaptive Online Model (AOM): modeling algorithms that reside information changes owed to process authorities changes and support precocious predictability
Since its take successful December 2022, SK hynix utilized Panoptes VM to behaviour virtual measurements connected much than 50 cardinal wafers truthful far, which translates to much than 1 wafer per second. The institution was capable to amended process variability* by 29% acknowledgment to this technology.
*Variability: nan standard of nan value variety of nan products from a definite process. It is important to support a debased process variability successful bid to support nan merchandise value consistent.
The latest algorithm by Gauss Labs, nan aggregated AOM, pools processing machines, chambers, etc. that stock nan aforesaid shape and models them together. According to Gauss Labs, this solves nan situation of information shortage while further expanding nan prediction accuracy.
In its 2nd paper, “Universal Denoiser to Improve Metrology Throughput,” Gauss Labs introduces a “universal denoiser”, which removes random variations (noise) from CD-SEM* images. Measurements from CD-SEM images are taken astatine nanometer scale, truthful it is highly important to region sound successful bid to return measurements accurately.
*Critical Dimension Scanning Electron Microscope (CD-SEM): instrumentality that combines physics and infrared optical microscopes to measurement nan size, shape, and position of semiconductor patterns
Gauss Labs’s cosmopolitan denoiser uses AI to region sound from various types of images astatine once. Through a bid of extended tests pinch SK hynix, Gauss Labs observed image acquisition clip reduced to arsenic overmuch arsenic ¼, compared to accepted technology. Gauss Labs states that this exertion is expected to amended nan productivity of metrology instrumentality by 42%.
Mike Kim, CEO of Gauss Labs, said that his institution is moving connected investigation and improvement for applications of business AI successful real-world semiconductor manufacturing fabs. “We will proceed to motorboat innovative AI-based solutions to revolutionize manufacturing.”
Authors
- Minsuk Shina, Minju Junga, Simon Zabrockia, Doh-Hyung Rob, Hyeon-Kyeong Jeongb, Dongkyun Yim a, “Model aggregation for virtual metrology successful high-volume manufacturing,” SPIE Advanced Lithography + Patterning (2024)
- Yonghyun Kima, Seyun Kima, Hoon Byuna, Sang-Gil Parka, Tae Jong Leeb, Seong Il Leeb, Min Woo Kangb, Il Koo Kima, “Universal denoiser to amended metrology throughput,” SPIE Advanced Lithography + Patterning (2024)
aGauss Labs bSK hynix
“These materials are not an connection for waste of nan securities of SK hynix Inc. successful nan United States. The securities whitethorn not beryllium offered aliases sold successful nan United States absent registration pinch nan U.S. Securities and Exchange Commission aliases an exemption from registration nether nan U.S. Securities Act of 1933, arsenic amended. SK hynix Inc. does not intend to registry immoderate offering successful nan United States aliases to behaviour a nationalist offering of securities successful nan United States.”
About SK hynix Inc.
SK hynix Inc., headquartered successful Korea, is nan world’s apical tier semiconductor supplier offering Dynamic Random Access Memory chips (“DRAM”), flash representation chips (“NAND flash”) and CMOS Image Sensors (“CIS”) for a wide scope of distinguished customers globally. The Company’s shares are traded connected nan Korea Exchange, and nan Global Depository shares are listed connected nan Luxemburg Stock Exchange. Further accusation astir SK hynix is disposable astatine www.skhynix.com, news.skhynix.com.
About Gauss Labs Inc.
Gauss Labs develops innovative business AI solutions and package products to make factories much intelligent by leveraging machine-generated data. It was founded successful August 2020 and now has 2 world locations successful Silicon Valley, CA, and Seoul, Korea. Further accusation astir Gauss Labs is disposable astatine www.gausslabs.ai.
View original contented to download multimedia: https://www.prnewswire.com/news-releases/gauss-labs-and-sk-hynix-publish-the-latest-results-on-ai-based-semiconductor-metrology-technology-at-spie-al-2024-302075185.html
SOURCE SK hynix Inc.
