MACHINE LEARNING · PREDICTIVE MAINTENANCE

Jet Engine Hospital

An end-to-end predictive-maintenance system for NASA C-MAPSS turbofan engines.

Role Machine Learning / Development
Year 2026
Datasets FD001 · FD003 · FD004
Stack Python · Scikit-learn · Streamlit
Jet Engine Hospital dashboard overview

The problem

Predict remaining useful life, identify near-term failure risk, detect anomalous behavior, quantify uncertainty, and turn those signals into an interpretable maintenance action.

Sensor Data Feature Engineering RUL Regression Risk Classification Anomaly Detection Conformal Interval Decision Policy Dashboard

Unified dashboard

Jet Engine Hospital maintenance status Jet Engine Hospital cycle evidence Jet Engine Hospital RUL trajectory Jet Engine Hospital sensor trajectory Jet Engine Hospital thresholds and anomaly analysis Jet Engine Hospital model metadata

Three operating scenarios

FD001 provides the foundation, FD003 introduces multiple fault modes, and FD004 adds multiple operating conditions and multiple faults. The final interface presents a unified workflow across the three datasets.