Manufacturing MCP

This is an MCP(Model Context Protocol) based AI Agent project that explores manufacturing equipment data and analyzes equipment failure risk.

Datasets

This project uses the AI4I 2020 Predictive Maintenance Dataset. The dataset is a synthetic dataset created to simulate predictive maintenance data from actual industrial sites, with a total of 10,000 equipment operating observations and 14 columns. There are no missing values.

Each observation includes the product grade, air and process temperature, rotational speed, torque, tool wear time, and equipment failure status and failure type. Therefore, it can be used in the following predictive maintenance scenarios.

• Equipment status monitoring (Monitoring)

• Failure risk classification (Classification)

• Anomaly condition detection (Anomaly Detection)

Key Columns

The meaning and unit of each column can be found in the dataset column documentation.

Sample Data

Below are the first 3 observations of the original CSV.

Related MCP server: Sparkplug MCP Server

Development Environment

This project uses Python 3.12 or later. Activate the virtual environment at the project root on WSL Ubuntu 24.04 and install development dependencies.

source mcp/bin/activate
python -m pip install --upgrade pip
python -m pip install -e ".[dev]"

Create a local .env file by referring to .env.example. The .env file containing the actual API Key is not committed to Git.

cp -n .env.example .env
chmod 600 .env

Run tests, linting, and formatting checks after making changes.

pytest
ruff check .
ruff format --check .

The current Python package is stored in src/manufacturing_mcp, and the tests are in tests. The application configuration is loaded from src/manufacturing_mcp/config.py using environment variables and the .env file.

Running PostgreSQL

Run the PostgreSQL container in the background using Docker Compose.

docker compose up -d postgres
docker compose ps

The postgres service is healthy if its status is healthy. To connect directly to PostgreSQL, use the following commands.

docker compose exec postgres sh -c 'psql -U "$POSTGRES_USER" -d "$POSTGRES_DB"'

Type \q to quit the session. Use the following command to stop the container.

docker compose down

PostgreSQL data is stored in the postgres_data Docker volume, so it persists when the container is stopped or recreated.

Source and Copyright

• Project code: Copyright © 2026 Hyeongseob. Distributed under the MIT License.

• Dataset: AI4I 2020 Predictive Maintenance Dataset (2020), UCI Machine Learning Repository, https://doi.org/10.24432/C5HS5C

• Related paper: Stephan Matzka, "Explainable Artificial Intelligence for Predictive Maintenance Applications," 2020 Third International Conference on Artificial Intelligence for Industries (AI4I), pp. 69–74, https://doi.org/10.1109/AI4I49448.2020.00023

• Dataset license: Creative Commons Attribution 4.0 International (CC BY 4.0). Sharing and modification of the dataset is permitted, provided that you give appropriate credit to the original author and source.