Austin, Texas – 78741.
Research Article
World Journal of Advanced Research and Reviews, 2022, 14(02), 696–706
Received on 02 April 2022; revised on 16 May 2022; accepted on 19 May 2022
As semiconductor technology advances toward smaller nodes, optimizing physical design has become a critical challenge in achieving high performance, efficiency, and scalability. Traditional design methods often fall short of meeting the demands of advanced node technology due to their limited adaptability and efficiency. This paper explores artificial intelligence (AI) and machine learning (ML) techniques tailored for physical design optimization in advanced node semiconductors. By leveraging AI-driven algorithms, including deep learning, reinforcement learning, and hybrid models, this study aims to streamline critical design processes such as placement, routing, and power optimization. The results demonstrate that AI-driven methods significantly outperform traditional techniques, achieving improvements of 13.5% in area efficiency, with a utilization rate of 89.1%, and a total power reduction of 18.8%. Furthermore, signal integrity, measured by Signal-to-Noise Ratio (SNR), improves by 40.8%, reaching 21.4 dB, while routing congestion is reduced to 7.2%. These findings highlight the transformative potential of AI and ML methodologies in addressing the complexities of advanced node design, offering scalable, efficient, and high-performance solutions for modern semiconductor technologies.
Advanced Node Semiconductors; Physical Design Optimization; Artificial Intelligence; Machine Learning in Semiconductor Design; Placement and Routing Optimization; Power and Area Efficiency
Preview Article PDF
Rashmitha Reddy Vuppunuthula. AI and machine learning-driven optimization for physical design in advanced node semiconductors. World Journal of Advanced Research and Reviews, 2022, 14(2), 696-706. Article DOI: https://doi.org/10.30574/wjarr.2022.14.2.0415