The goal of this thesis is to design, implement, and evaluate a sensor-based maintenance bot that uses Large Language Models (LLMs) to support predictive maintenance and decision-making. The bot should be capable of retrieving, analyzing, and reasoning over time series sensor data as well as unstructured maintenance-related documentation (e.g., technical manuals, incident reports). The result is a unified system that assists technicians and engineers in diagnosing issues, suggesting preventive actions, and retrieving relevant information in natural language. This work is conducted in collaboration with the Fraunhofer Institute for Production Technology (IPT) as part of the research initiative “Generative AI for Production and Business Operations“, aiming to explore practical applications of generative models in manufacturing and industrial operations.
Thesis Type |
|
Status |
Running |
Presentation room |
Seminar room I5 6202 |
Supervisor(s) |
Stefan Decker |
Advisor(s) |
Michal Slupczynski |
Contact |
slupczynski@dbis.rwth-aachen.de |
Predictive maintenance has become an essential element in Industry 4.0, aiming to reduce downtime and improve operational efficiency by using sensor data to predict potential failures. While time series forecasting techniques have been widely explored, these approaches are limited to numerical pattern analysis and lack integration with the vast amount of unstructured domain knowledge stored in textual form.
Recent advancements in Large Language Models (LLMs) have shown strong performance in natural language understanding, information retrieval, and even basic reasoning. When combined with retrieval-augmented generation (RAG) techniques and domain-specific embeddings, LLMs offer a promising path to support hybrid analysis of structured (e.g., sensor data) and unstructured data (e.g., manuals, logs, documentation).
-
Basic knowledge of machine learning, particularly time series forecasting and natural language processing.
-
Familiarity with LLMs (e.g., GPT, LLaMA) and vector-based retrieval techniques.
-
Experience with Python-based ML frameworks (e.g., PyTorch, HuggingFace, scikit-learn).
-
Interest in applications of AI in industrial or manufacturing domains.