P‐40: Virtual Yield Platform for Yield Simulation: AI‐EDA Co‐Design Practice to Boost Display Panel Yield
SID Symposium Digest of Technical Papers · 2026-06 · 57 (1) · 1826-1829
DOI: 10.1002/sdtp.70001_496 ↗Kurz gefasst
Fachartikel in der Zeitschrift SID Symposium Digest of Technical Papers, online erschienen am 11.09.2026. Verfasst von Jing Ba und 5 weiteren Autor:innen. Die Arbeit nennt 6 Literaturangaben.
Abstract (Originalsprache)
With the intensification of competition in the panel industry, display panels need to pursue “higher yield.” This thesis will delve into the essence of yield and explore the core strategies for improving yield. Based on yield loss caused by random defects, a yield prediction model is developed, a co‐optimization simulation method for design and manufacturing is established, and AI is integrated to achieve output of design hotspots or design specifications. Drawing on the IC industry, a random yield prediction simulation method for the panel industry based on “random defect density D 0 + critical area Aca” is proposed: using big data from the production line to fit the particle size‐density distribution; completing the simulation and integration of intra‐layer open/short and inter‐layer short critical area curves to obtain the sensitive area Aca of the design to particles; then using the Poisson distribution yield model to predict the Array segment electrical test yield, and after calibrating D0 with the actual product measurement results, blind tests were conducted on 65/75‐inch products, with intra‐layer and inter‐layer yield prediction deviations about 2%. By establishing an Aca‐Yield loss matrix, high‐risk layout hotspots can be quickly located, design weaknesses can be identified, and design revisions can be guided to improve product yield. This yield simulation solution can quantify random defect losses in advance and provide a practical AI + EDA design simulation solution for yield improvement in the display industry.
Zitieren
APA
Ba, J., Tan, Z., Zhao, B., Guo, R., Pan, W., & Mao, J. (2026). P‐40: Virtual Yield Platform for Yield Simulation: AI‐EDA Co‐Design Practice to Boost Display Panel Yield. SID Symposium Digest of Technical Papers, 57(1), 1826-1829. https://doi.org/10.1002/sdtp.70001_496
Vancouver
Ba J, Tan Z, Zhao B, Guo R, Pan W, Mao J. P‐40: Virtual Yield Platform for Yield Simulation: AI‐EDA Co‐Design Practice to Boost Display Panel Yield. SID Symposium Digest of Technical Papers. 2026;57(1):1826-1829. doi:10.1002/sdtp.70001_496
BibTeX
@article{Ba2026,
author = {Ba, Jing and Tan, Zhiwei and Zhao, Bin and Guo, Ruiqi and Pan, Weiwei and Mao, Jun},
title = {P‐40: Virtual Yield Platform for Yield Simulation: AI‐EDA Co‐Design Practice to Boost Display Panel Yield},
journal = {SID Symposium Digest of Technical Papers},
year = {2026},
volume = {57},
number = {1},
pages = {1826--1829},
doi = {10.1002/sdtp.70001_496}
}
EUVIMED