AI4I 2020: Industrial IoT Sensor Data for Predictive Maintenance

This dataset is a gold standard for Predictive Maintenance research in Industrial IoT (IIoT). It consists of 10,000 data points stored as rows with 14 features in columns, generated to simulate a real-world milling machine operation.

Dataset Features:

• UID & Product ID: Unique identifiers ranging from L (Low quality) to H (High quality) variants.
• Air Temperature [K]: Generated using a random walk process tailored to standard deviation metrics.
• Process Temperature [K]: Modeled closely with air temperature plus standard operational heat generation.
• Rotational Speed [rpm]: Calculated from a power of 2860 W, overlaid with normally distributed noise.
• Torque [Nm]: Normally distributed around 40 Nm with standard deviation constraints.
• Tool Wear [min]: Tracks the gradual degradation of the milling tool over time.

Target Variables (Failure Modes):

The dataset allows for binary classification (fail/no-fail) or multi-class classification based on specific failure types:

• TWF: Tool Wear Failure.
• HDF: Heat Dissipation Failure.
• PWF: Power Failure.
• OSF: Overstrain Failure.
• RNF: Random Failures.

Use Cases:

Ideal for training Machine Learning models on Binary Classification (Predicting machine failure) and Anomaly Detection in smart factories.

Source: UCI Machine Learning Repository - Licensed under CC BY 4.0.