Workshop Materials — IFAC 2026 LLMs and Agents | UBC DAIS Lab

Welcome

Introduction

Welcome to the workshop on LLMs and Agents for Process Industries. This page is your single entry point to everything you'll need for the hands-on session and beyond — starter notebooks, datasets, slide decks, and curated reading.

Our aim for the day is modest but concrete. You'll build a couple of small LLM-agent prototypes on realistic process-engineering tasks, deliberately break them to see where they fail, and leave with enough intuition to evaluate claims about this technology rather than take them on trust.

You don't need deep Python experience or prior exposure to LLMs. The morning sessions cover the necessary background, and the afternoon exercises are designed to be completed collaboratively. Observers are welcome — code if you like, watch if you prefer.

The code and datasets will also live on a dedicated GitHub repository once it's published (tentatively github.com/daisubc/ifac2026-workshop) — the link here will update automatically when the repo goes live.

Before you arrive

Checklist

Bring

A laptop (required — you will work directly with LLM APIs on Python notebooks during the session).
Ability to connect to the workshop Wi-Fi.

Provided on the day

Jupyter notebooks and starter code.
Access to LLMs for the duration of the session.
Sample datasets: simplified P&IDs, alarm lists, maintenance logs.
Offline backup — pre-recorded demos and local notebooks using open-source models — in case of connectivity issues.

Resources

Notebooks, slides, and data

Placeholder entries below — they'll turn into live download / Colab links when the repository is published. Bookmark this page and re-visit on the morning of the workshop.

Notebook

Exercise 01 — Prompt engineering on a simplified P&ID

Prompt a model to interpret a simplified P&ID and critique its control structure suggestions.

Coming soon

Notebook

Exercise 02 — Build a minimal fault-diagnosis agent

Combine an LLM with tool calls to query sensor data and search maintenance logs.

Coming soon

Notebook

Exercise 03 — Probe where agents fail

Numerical reasoning errors, hallucinated equipment names, overconfident recommendations.

Coming soon

Slides

Morning deck — Foundations + Applications I & II

Covers LLM and agent fundamentals, Application I (Process Design), Application II (Simulation).

Coming soon

Slides

Afternoon deck — Application III + Wrap-up

Autonomous Control & Fault Diagnosis, research-priorities synthesis.

Coming soon

Dataset

Sample data — simplified P&IDs, alarm lists, maintenance logs

Small curated datasets used across the three hands-on exercises.

Coming soon

GitHub repository

Full source for notebooks, data, and environment setup will live here. The repository is being prepared; the link will go live ahead of the workshop.

github.com/daisubc/ifac2026-workshop

Reading list

Key papers

A short, focused list if you'd like to prime yourself before the workshop. All are linked directly to arXiv or publisher copies.

Balhorn, L. S., Caballero, M., Schweidtmann, A. M. (2024). Toward autocorrection of chemical process flowsheets using large language models. Computer Aided Chemical Engineering, Vol. 53, 3109–3114. Elsevier. Preprint: arXiv:2312.02873
Du, W., Yang, S. (2025). The potential and challenges of large language model agent systems in chemical process simulation. Frontiers of Chemical Science and Engineering, 19(10), 99.
Gill, M. S., Vyas, J., Markaj, A., Gehlhoff, F., Mercangöz, M. (2025). Leveraging LLM agents and digital twins for fault handling in process plants. arXiv:2505.02076
Hirtreiter, E., Schulze Balhorn, L., Schweidtmann, A. M. (2024). Toward automatic generation of control structures for process flow diagrams with large language models. AIChE Journal, 70(1), e18259.
Qaid, H. A. et al. (2024). FD-LLM: Large language model for fault diagnosis of machines. arXiv:2412.01218
Schulze Balhorn, L. et al. (2024). Graph-to-SFILES: Control structure prediction from process topologies using generative artificial intelligence. arXiv:2412.00508
Venkatasubramanian, V. (2019). The promise of artificial intelligence in chemical engineering: Is it here, finally? AIChE Journal, 65(2), 466–478.
Vyas, J., Mercangöz, M. (2024). Autonomous industrial control using an agentic framework with large language models. arXiv:2411.05904
Woo, T., Kim, S., Tariq, S. et al. (2025). Leveraging generative AI and large language models for process systems engineering: a state-of-the-art review. Korean Journal of Chemical Engineering, 42, 2787–2808.
Xia, Y., Dittler, D., Jazdi, N., Chen, H., Weyrich, M. (2024). LLM experiments with simulation: Large language model multi-agent system for process simulation parametrization in digital twins. IEEE ETFA (Best Paper Award).
Xia, Y., Jazdi, N., Zhang, J., Shah, C., Weyrich, M. (2024). Control industrial automation system with large language model agents. arXiv:2409.18009

Questions before the workshop? Email bhushan.gopaluni@ubc.ca.

Hands-on notebooks, slides, and reading list

Enter the workshop passcode

Introduction

Checklist

Bring

Provided on the day

Notebooks, slides, and data

Exercise 01 — Prompt engineering on a simplified P&ID

Exercise 02 — Build a minimal fault-diagnosis agent

Exercise 03 — Probe where agents fail

Morning deck — Foundations + Applications I & II

Afternoon deck — Application III + Wrap-up

Sample data — simplified P&IDs, alarm lists, maintenance logs

GitHub repository

Key papers