Advanced intelligent database frameworks for yield enhancement in semiconductor fabrication
i-manager's Journal on Electrical Engineering · 20.03.2026 · 19 (3) · 1
DOI: 10.26634/jee.19.3.2651 ↗Kurz gefasst
Fachartikel in der Zeitschrift i-manager's Journal on Electrical Engineering, online erschienen am 15.03.2026. Verfasst von Padmaja Pulivarthy und 1 weiteren Autor:innen. Die Arbeit nennt 14 Literaturangaben.
Abstract (Originalsprache)
Semiconductor fabrication process is a complicated activity and entails intricate operations with noisy sensor information of high-dimensionality, and fault detection and yield enhancement is complicated. The conventional rule-based monitoring systems and statistical process control (SPC) methods are not effective to address these complexities and frequently results in the unnoticed process variations and massive loss of yield. This study introduces a machine learning-based system to predict faults and improve yield, and these issues are handled in this study through a smart, data-driven framework. The system architecture will comprise of a data ingestion module, preprocessing and feature handling module, machine learning engine and prediction and decision-support module. Data used: Semiconductor manufacturing sensor data of a benchmark data set with multivariate process data, high-dimensionality, missing data, binary quality data (pass/fail), and so on. The preprocessing module deals with the missing values, normalization, scaling and feature selection. The machine learning engine uses the classifiers of Random Forest and XGBoost, model training, hyperparameter optimization, and techniques of dealing with class imbalance. Fault prediction method is step-by-step method which entails reception of data, preprocessing, model training, fault prediction, and yieldrelated insights generation. The yield enhancement decision module takes the prediction as its input and takes high-risk cases to give feedback on fine-tuning processes and preventive maintenance, and finally gives the recommendations to enhance the yield. It has been demonstrated that the suggested framework is superior in comparison with the traditional techniques and helps improve the accuracy of fault detection and promote the yield rate.
Zitieren
APA
Pulivarthy, P., & Erugula, A. R. (2026). Advanced intelligent database frameworks for yield enhancement in semiconductor fabrication. i-manager's Journal on Electrical Engineering, 19(3), 1. https://doi.org/10.26634/jee.19.3.2651
Vancouver
Pulivarthy P, Erugula AR. Advanced intelligent database frameworks for yield enhancement in semiconductor fabrication. i-manager's Journal on Electrical Engineering. 2026;19(3):1. doi:10.26634/jee.19.3.2651
BibTeX
@article{Pulivarthy2026,
author = {Pulivarthy, Padmaja and Erugula, Anji Reddy},
title = {Advanced intelligent database frameworks for yield enhancement in semiconductor fabrication},
journal = {i-manager's Journal on Electrical Engineering},
year = {2026},
volume = {19},
number = {3},
pages = {1},
doi = {10.26634/jee.19.3.2651}
}
EUVIMED