Jet-Engine Multi-Sensor and Multi-Mode Failure Data for Bayesian RUL Prediction

A dataset containing multi-sensor time-series signals and time-to-failure event data for multi-mode failure prediction, focused on jet-engine systems. The data supports a hierarchical Bayesian joint model integrating a Cox proportional hazards model, a Convolved Multi-output Gaussian Process, and multinomial failure mode distributions. It was published by Sina Aghaee Dabaghan Fard on the figshare platform in April 2026.

Use Cases

• Predict remaining useful life (RUL) from multi-sensor time-series signals using a Convolved Multi-output Gaussian Process model.
• Model the relationship between multiple failure modes and sensor data via integrated multinomial failure mode distributions.
• Perform robust uncertainty quantification for failure predictions using the hierarchical Bayesian framework and Monte Carlo sampling.
• Validate joint modeling approaches that leverage both multi-sensor time-series data and failure time event data, as demonstrated with the jet-engine case study.

Strengths

• Data is validated through extensive numerical and case studies, as stated in the source paper.
• The dataset is 200.9 MB, indicating a medium scale suitable for detailed model development.
• Supports a unified modeling approach for multi-sensor and multi-mode failure data, a specific methodological strength.

Limitations

• The exact number of rows, sensors, and failure events is unknown, limiting reproducibility assessments.
• Primary focus is a specific jet-engine case study, which may limit generalizability to other industrial systems without adaptation.

Provenance

Source

figshare, authored by Sina Aghaee Dabaghan Fard.

Collection Method

Data is associated with a research paper proposing a Bayesian joint model; likely generated or curated from jet-engine system monitoring.

Time Range

null

Freshness

Last updated in April 2026, indicating recent availability.

Geography

null