Register now After registration you will be able to apply for this opportunity online.
Large Language Models for Manufacturing
In the manufacturing industry, determining the optimal process parameters is crucial for enhancing efficiency and quality. A vast amount of unstructured data exists in books, academic and industrial publications, and various measurement datasets. Artificial intelligence systems, particularly those us-ing large language models (LLMs), offer promising solutions to integrate and utilize these diverse data sources effectively without human input.
Keywords: Large Language Models (LLMs), Manufacturing Process Optimization, AI in Manufacturing, Process Parameter Recommendations, Machine Learning in Manufacturing, Data-Driven Manufacturing, Artificial Intelligence, Industrial AI Applications
This thesis aims to develop a system based on LLMs, such as ChatGPT or other suitable AI models, to determine optimal parameters for various turning operations. The project will focus on identifying the best combination of system architecture, base model, and data integration methods to achieve high performance. Additionally, the system will incorporate real-time measurements from tool wear and surface roughness as feedback to continually refine and improve parameter recommendations.
This thesis aims to develop a system based on LLMs, such as ChatGPT or other suitable AI models, to determine optimal parameters for various turning operations. The project will focus on identifying the best combination of system architecture, base model, and data integration methods to achieve high performance. Additionally, the system will incorporate real-time measurements from tool wear and surface roughness as feedback to continually refine and improve parameter recommendations.
• Literature Review: Conduct a comprehensive review of the current state-of-the-art LLM applications in manufacturing, with a specific focus on turning operations.
• Data Integration Methodology: Develop a methodology for integrating unstructured data sources, such as books, academic publications, and datasets, into the LLM system's knowledge base.
• Real-Time Data Incorporation: Incorporate real-time machine measurements, such as tool wear and surface roughness, to provide feedback and dynamically tune the LLM-based parameter recommendations.
• System Design and Implementation: Design and implement a system capable of determining optimal parameters for a range of turning operations, incorporating both pre-existing and real-time data.
• Validation: Validate the effectiveness of the developed system by comparing its performance against traditional methods and benchmarks.
• Literature Review: Conduct a comprehensive review of the current state-of-the-art LLM applications in manufacturing, with a specific focus on turning operations.
• Data Integration Methodology: Develop a methodology for integrating unstructured data sources, such as books, academic publications, and datasets, into the LLM system's knowledge base.
• Real-Time Data Incorporation: Incorporate real-time machine measurements, such as tool wear and surface roughness, to provide feedback and dynamically tune the LLM-based parameter recommendations.
• System Design and Implementation: Design and implement a system capable of determining optimal parameters for a range of turning operations, incorporating both pre-existing and real-time data.
• Validation: Validate the effectiveness of the developed system by comparing its performance against traditional methods and benchmarks.
Dr. Markus Maier, +41 44 556 58 38, markus.maier@inspire.ch
Ruben Zwicker, +41 44 556 58 84, ruben.zwicker@inspire.ch
Dr. Markus Maier, +41 44 556 58 38, markus.maier@inspire.ch