ACIS 2024 Proceedings
Abstract
The semiconductor manufacturing industry is undergoing a data-driven revolution, driven by advancements in electronic devices and smart technologies. This shift has significantly increased the volume, velocity, and variety of data, enabling enhanced knowledge extraction and process optimization. However, traditional solutions, such as, the Cross-Industry Standard Process for Data Mining, Knowledge Discovery in Databases, and the Team Data Science Process are insufficient for addressing real-time analytics, high-dimensional data, and domain-specific challenges. To bridge these gaps, we introduce a novel framework that combines Explainable Artificial Intelligence with the Design Science Research methodology. Key contributions of this framework include real-time processing capabilities, integration of domain knowledge, and enhanced transparency of Artificial Intelligence (AI) models, ensuring accurate and interpretable decision-making. Demonstrated through wafer map clustering, this framework provides a comprehensive, industry-specific systematic guidance for implementing data mining and AI projects, providing efficient, and easy-to-understand solutions that can improve semiconductor manufacturing.
Recommended Citation
Podder, Itilekha and Bub, Udo, "An Explainable Artificial Intelligence Framework for Improving Semiconductor Manufacturing: A Design Science Research Approach" (2024). ACIS 2024 Proceedings. 88.
https://aisel.aisnet.org/acis2024/88