ABB IRB 1100 Robot: Synthetic Predictive Maintenance Data with 2,000 Instances

2,000 synthetic instances with 215 explanatory characteristics simulate sensor data for an ABB IRB 1100 industrial robot. The dataset was created by Mohamed Lamine Soudani via a sophisticated simulation model and published on figshare in May 2026. It is designed for detecting and classifying failures, including overheating, misalignment, and gear wear across three severity levels.

Use Cases

• Training fault classification models based on simulated electrical, mechanical, thermal, and vibration sensor parameters.
• Benchmarking feature selection methods using the described 215 explanatory characteristics and a reduced subset of 100 features.
• Developing severity-level prediction systems for maintenance prioritization based on the three severity levels per defect type.
• Comparing ensemble tree methods (like LightGBM) against SVM-based approaches for industrial robot failure prediction.

Strengths

• Dataset comprises 2,000 simulated instances, providing a structured basis for model development.
• Includes 215 explanatory features derived from a multisensory simulation model covering six robot axes.
• Results show strong discrimination for specific defects (F1 ≈ 0.97–1.00 for overheating and misalignment).
• Dimensionality analysis confirms the feature set can be reduced by more than 50% without performance degradation.

Limitations

• Data is entirely synthetic, generated via simulation, and requires validation on real-world industrial datasets.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is known, but the specific distribution of defect classes and severity levels is not detailed.

Provenance

Source

Mohamed Lamine Soudani via figshare.

Collection Method

Generated via a sophisticated simulation model of the ABB IRB 1100 robot's six axes and sensors.

Freshness

Last updated 2026-05-22 12:48:59; freshness should be verified.