Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/336723 
Year of Publication: 
2025
Citation: 
[Journal:] Journal of Intelligent Manufacturing [ISSN:] 1572-8145 [Volume:] 37 [Issue:] 2 [Publisher:] Springer US [Place:] New York, NY [Year:] 2025 [Pages:] 829-848
Publisher: 
Springer US, New York, NY
Abstract: 
Semiconductor manufacturing systems are highly complex due to intricate processes and material flows. Operating these systems efficiently remains a significant challenge, particularly under the growing demands for operational excellence and cost reduction. Current approaches often rely on extensive manual modeling, which slows down production planning and adaptation. To address these challenges, we propose a data-driven methodology for Automatic Simulation Model Generation (ASMG), enhanced by machine learning techniques. This fully automated pipeline extracts and processes production data (lot tracking information and resource states) to generate simulation models without manual intervention. A machine learning technique called equipment emulation captures complex tool behaviors and mitigates issues with noisy or incomplete data. Validation in two real-world semiconductor production environments, covering over 300 days and showing an accuracy within 5–7% for throughput and uptime, demonstrates the method’s ability to produce precise models. By reducing the time and expertise required for model creation, this ASMG method facilitates agile digital twin implementations and enables faster, more responsive production planning.
Subjects: 
Digital twin
Machine learning
Automated modeling
Semiconductor manufacturing
Data mining
Simulation
Persistent Identifier of the first edition: 
Creative Commons License: 
cc-by Logo
Document Type: 
Article
Document Version: 
Published Version

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.