Manufacturing MCP

Proyecto de agente de IA basado en MCP (Model Context Protocol) para explorer los datos de equipos de fabración y analizar el riesgo de fallas.

Conjanto de datos

Este proyecto utiliz el AI4I 2020 Predictive Maintenance Dataset. Es un conjunto de datos sintético creando para simular los datos de mantimiento preventivo de entornos instriales realies. Contiene 10,000 observaciones de funcionamiento de equipos y 14 columnas. No hay valores perdidos.

Cada observación incluye el grao del producto, las temperaturas del aire y del proceso, la velocidad de rotación, el torque, el tiempo de use de la herramenta y, junto a elles, si se ha been lost. and el tipo de falla. Se puede sintetir a él scenarios of predictive maintenance like `Therefore it can be used in the following predictive maintenance scenros:

• Supervisión of los estado of los equipos (Monitoring)

• Clasificación del riesgo de fallo (Classification)

• Detección de condiciones anómalas (Anomaly Detection)

Columnaos principales

El ssignificar and local of each child columnas ** can at datos columna document.

Muestra de datos

At below three original observas showing first 3 rows of the original CSV.

Related MCP server: Sparkplug MCP Server

Configurationación del entorno de development

Este proyecto utiliza Python 3.12 o higher. In the root of the project under WSL Ubuntu 24.04, active the virtual environment and install the development dependencies.

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

Cree a local .env file in reference to .env.example. The actual .env that contains the real API Key is not included in Git.

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

After making code changes, run tests, linting and format check.

pytest
ruff check .
ruff format --check .

Currently the Python package is insrc/manufacturing_mcp, the tests are in tests. Application settings are loaded from environment variables and from .env file in src/manufacturing_mcp/config.py.

PostgresSQL execution

Run PostgresQL container with Docker Compose in the background.

docker compose up -d postgres
docker compose ps

If the postgres service state is healthy, everything is normal. To connect directly to PostgreSQL, use the following command.

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

To end the connection, enter \q. To stop the container, use the following command.

docker compose down

PostgreSQL data is preserved in the Docker volume postgres_data, so it remains intact even when the container is stopped or recreated.

Sources and copyright

• Project code: Copyright © 2026 Johyeongseob. Distributed license 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, but when using it you should text in an appropriate manner.