NORMA eResearch @NCI Library

Automated Detection of Semi-Conductor Wafer Map Defects Using Machine Learning Techniques

Ward, Michael (2022) Automated Detection of Semi-Conductor Wafer Map Defects Using Machine Learning Techniques. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
PDF (Master of Science)
Download (1MB) | Preview
[thumbnail of Configuration manual]
PDF (Configuration manual)
Download (837kB) | Preview


The process of manufacturing and creating silicon wafers for the development of chip sets is a particularly meticulous task. Full automation is used in many semiconductor manufacturing facilities. This results in minimal human intervention as batches of wafers move from processing tool to processing tool using robotic systems such as overhead vehicles (OHVs). This ensures the wafers have a low contamination possibility and ensures a high level of efficiency with tool-to-tool time (T2T). As is the case with many processes that are fully automated, issues can occur where certain parts of the robotic systems can become misaligned. This then has the potential to cause damage to the wafers in various ways. When a wafer undergoes an analytical process where the wafer is scanned for particles, the analytical tool will publish a wafer map that contains the location of the particles. Technicians manually review these wafer maps to check for possible issues but due to the sheer amount that is generated, many issues go unnoticed until it is too late. The aim of this research is to create a system and model using machine learning that can automatically detect and classify issues as soon as the wafer maps are generated. Due to the nature of the system, many machine learning classification models were researched and reviewed for the needed functionality and the speed. Support Vector Machine (SVM) and neural network sequential classification models are used due to “One vs One” approach being a good option and the high accuracy rate of the sequential model. A graphical user interface then alerts any stakeholders of excursions related to the equipment from the data generated on the analysed wafer maps.

Item Type: Thesis (Masters)
Iqbal, Zahid
Uncontrolled Keywords: Defects; Wafer Maps; machine learning; semiconductor industry; classification; SVM; CNN; Sequential Neural Networks
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
T Technology > TK Electrical engineering. Electronics. Nuclear engineering
Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Tamara Malone
Date Deposited: 27 May 2023 11:53
Last Modified: 27 May 2023 11:53

Actions (login required)

View Item View Item