PDF Solutions has developed a machine learning model that uses wafer sort parametric measurements to improve yield results by analyzing outlier spatial patterns. The patented method involves obtaining testing data from multiple test sites on a wafer and using neural networks to predict process parameters and yield for each die. GlobalData’s report on PDF Solutions gives a 360-degree view of the company including its patenting strategy. Buy the report here.
Smarter leaders trust GlobalData
Access deeper industry intelligence
Experience unmatched clarity with a single platform that combines unique data, AI, and human expertise.
According to GlobalData’s company profile on PDF Solutions, CMOS gate array designing was a key innovation area identified from patents. PDF Solutions's grant share as of May 2024 was 18%. Grant share is based on the ratio of number of grants to total number of patents.
Machine learning model for imputing process control parameters in dies
A recently granted patent (Publication Number: US11972987B2) outlines a method for optimizing semiconductor manufacturing processes. The method involves obtaining testing data from various test sites on a semiconductor wafer before slicing it into individual dies. A die level map is created, detailing semiconductor features and their locations on the wafer. Each die's testing data and map are input into neural networks, which determine non-linear relationships, impute process parameters, and predict yield for each die. The neural networks are trained on specific data sets to perform these functions efficiently.
Furthermore, the patent includes the identification of the importance of process parameters for yield prediction, continuous learning and updating of neural networks, evaluation of spatial patterns on die level maps, and modeling outlier patterns based on imputed parameters. The method also involves forming test sites on scribe lines between dies and near wafer edges to guide slicing. Additionally, a computer-readable medium with instructions for implementing the method and the use of machine learning tree models for the same purpose are detailed in the patent claims. Overall, the patent aims to enhance semiconductor manufacturing by utilizing neural networks and machine learning models to optimize process control parameters and improve yield performance based on testing data and die level maps.
To know more about GlobalData’s detailed insights on PDF Solutions, buy the report here.
Data Insights
From
The gold standard of business intelligence.
Blending expert knowledge with cutting-edge technology, GlobalData’s unrivalled proprietary data will enable you to decode what’s happening in your market. You can make better informed decisions and gain a future-proof advantage over your competitors.
GlobalData, the leading provider of industry intelligence, provided the underlying data, research, and analysis used to produce this article.
GlobalData Patent Analytics tracks bibliographic data, legal events data, point in time patent ownerships, and backward and forward citations from global patenting offices. Textual analysis and official patent classifications are used to group patents into key thematic areas and link them to specific companies across the world’s largest industries.
