Leveraging LLM Agents and Digital Twins for Fault Handling in Process Plants
Milapji Singh Gill 1 , Javal Vyas 2 , Artan Markaj 1 , Felix Gehlhoff 1 , Mehmet Mercangöz 2 · 2025
Read Leveraging LLM Agents and Digital Twins for Fault Handling in Process Plants by Milapji Singh Gill 1 , Javal Vyas 2 , Artan Markaj 1 , Felix Gehlhoff 1 , Mehmet Mercangöz 2 (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2505-02076.txt.
Description
Industrial process plants still lean on human experts when faults strike, since fixing them needs plant-specific knowledge. Gill and co-authors propose pairing language model agents with a digital twin of the plant. The agents read the plant state and issue control actions, while the twin holds the engineering knowledge they draw on and serves as a test bed for checking their fixes before use.
On a mixing module, the setup ran the unit on its own and cleared a pipe clog after only a few reprompts. The authors present the frame as a path toward plants that handle faults with less human help.
Catalog record
| CREATOR | Milapji Singh Gill 1 , Javal Vyas 2 , Artan Markaj 1 , Felix Gehlhoff 1 , Mehmet Mercangöz 2 |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 38 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2505-02076