This dataset provides time-series sensor readings from multiple industrial machines operating on a production line. Designed to support predictive maintenance and equipment health monitoring use cases, it captures realistic operational behavior with both normal and fault conditions.

Data was simulated to represent real-time collection via an IoT-enabled sensor network. Each entry includes a timestamp, machine identifier, raw sensor measurements, and decomposed signal features to reflect complex patterns often associated with early-stage equipment faults.

It is well-suited for developing machine learning models for fault detection, anomaly detection, and forecasting in industrial maintenance contexts.

⭐ Key Features Timestamp: Precise time of each sensor reading.

Machine ID: Identifies the specific machine or production line unit.

Vibration: Measures mechanical oscillations indicating wear or imbalance.

Acoustic Signal: Captures sound emissions related to machine operation.

Temperature: Records operational heat levels in °C.

Current: Electrical current drawn by the machine in amps.

Intrinsic Mode Functions (IMF_1, IMF_2, IMF_3): Decomposed signal components representing underlying patterns in sensor data.

Label: Indicates machine state (0 = healthy, 1 = faulty).