Jongseok completed Predictive Maintenance: Industrial ML for Fault Detection

Built a compact rolling-bearing fault detector end to end: engineered time-domain features, used envelope-spectrum energy around supplied defect frequencies, trained a shallow decision tree on training bearings only, and evaluated it on a held-out set with catch-rate and false-alarm rate. The final memo defended a 0.60 alarm threshold with explicit cost reasoning and correctly localized the fault to the outer race from the dominant BPFO signature.

• Leakage-Free Feature Engineering
• Honest Held-Out Evaluation
• Cost-Based Operational Decision-Making
• Interpreting Defect Frequencies