AI-ENABLED FAULT IDENTIFICATION IN ELECTRONIC CIRCUITS FOR SEMICONDUCTORS INDUSTRY

Abstract

In semiconductor manufacturing, ensuring high yield rates is critical for optimizing production efficiency 
and minimizing costs. However, the vast number of signals collected from sensors and process 
measurement points often contain a mix of relevant information, noise, and irrelevant data, making it 
challenging for engineers to identify the key factors affecting yield. Feature selection techniques are 
instrumental in addressing this challenge, as they help identify the most relevant signals that significantly 
impact yield.  In conventional semiconductor manufacturing, engineers are inundated with an extensive 
array of signals, making it cumbersome and time-consuming to pinpoint the critical factors influencing 
yield excursions. This data overload often results in suboptimal process efficiency and increased 
production costs. Traditional approaches were not effectively distinguished between useful information 
and noise, leading to inefficient troubleshooting and reduced yield rates. Additionally, manual feature 
selection processes are labour-intensive and was not uncover complex causal relationships between 
variables, limiting their effectiveness in enhancing semiconductor manufacturing operations. To 
overcome the limitations of the conventional approach, this study proposes the use of artificial 
intelligence-based feature selection techniques. By leveraging sophisticated algorithms, the proposed 
system will rank features according to their impact on semiconductor manufacturing yield. These 
techniques will not only streamline the identification of crucial variables but also unveil causal 
relationships within the data, providing a deeper understanding of the production process. The application 
of cross-validation ensures robustness and reliable evaluation of feature relevance for predictability using 
error rates. The goal is to empower engineers with a more efficient and data-driven approach to 
semiconductor manufacturing, resulting in increased yield rates, reduced production costs, and shorter 
learning cycles. The preliminary results presented here demonstrate the potential of this approach, 
highlighting the promise of artificial intelligence in revolutionizing semiconductor manufacturing 
optimization.