Building Reliable Industrial Agents with MCP:
A Hands-on AssetOpsBench Tutorial for AI-Driven Operations

2026 KDD Hands-on Tutorial

Tuesday, August 11, 2026 1:30 PM – 4:30 PM

Location: Venue, Jeju, Korea

Abstract

First abstract paragraph. Introduce the problem: AI agents and LLMs are increasingly applied to industrial asset operations — predictive maintenance, anomaly detection, fault diagnosis, work-order reasoning — yet evaluating their reliability in these high-stakes settings remains an open challenge.

Second abstract paragraph. Introduce AssetOpsBench: what it measures, the tasks and data it covers, the evaluation methodology, and what attendees will learn. Summarize the takeaways and the hands-on component of the tutorial.

Schedule

Section 1: Introduction and Motivation (10 min)
- AI agents in industrial asset operations
- Why benchmarking and evaluation matter
- Overview of the tutorial
Section 2: The AssetOpsBench Benchmark (30 min)
- Task taxonomy and datasets
- Evaluation metrics and protocol
- Baselines and leaderboard
Section 3: Hands-on Session (60 min)
- Environment setup and notebooks
- Running and evaluating an agent

(Break - 10 min)
Section 4: Results and Analysis (20 min)
- Findings across agents and tasks
- Failure modes and error analysis
Section 5: Open Challenges and Future Directions (20 min)
- Generalization and robustness
- Real-world deployment considerations
Q&A

Building Reliable Industrial Agents with MCP:
A Hands-on AssetOpsBench Tutorial for AI-Driven Operations

Abstract

Schedule

Tutorial Materials

Speakers

Building Reliable Industrial Agents with MCP:A Hands-on AssetOpsBench Tutorial for AI-Driven Operations

Abstract

Schedule

Tutorial Materials

Speakers

Building Reliable Industrial Agents with MCP:
A Hands-on AssetOpsBench Tutorial for AI-Driven Operations