L2M4PC: Large Language Models for Production Control

• Applicant: Dr.-Ing. Marvin Carl May
• Subject Area: Production Systems, Operations Management, Quality Management and Factory Planning
• Term: since 2025
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 558595415

A central challenge of modern production, which is characterized by a large number of variants, the increasing integration of circular strategies and individualization, is the efficient control of the production system in batch size one. At the same time, the volume of data in production and product development is increasing many times over, requiring production planners and controllers to make complex, data-based decisions in less time, whereas traditional, algorithmic production control approaches have difficulties with the effective processing of non-numerical features. This can be automated either by transforming the data into numerical data or by using other algorithms for data processing. For both, large language models have proven to be promising as state-of-the-art methods in the field of artificial intelligence. Due to the great heterogeneity of data in production and the difficulty of modeling production-related data and correlations purely numerically, there is great potential here. However, research into the application of large language models in production planning and control has yet to be carried out. For this reason, a method is being developed in this application to pre-process non-numerical data for intelligent production control using large language models and to make decisions or intelligently derive decisions directly using the large language models. The central research challenge is thus the development and validation of a data-based, large language model (LLM) integrated production control system that makes intelligent, agent-based production control decisions to better achieve logistical, economic and ecological target functions and is validated as a software artifact in example use cases and simulations.

• DFG Programme: WBP Fellowship
• International Connection: USA
• Host: Professor Sang-Gook Kim