Project case study

Predictive Vehicle Maintenance System

Applied machine learning workflow for anticipating maintenance needs with experiment tracking and a user-friendly web UI.

Key outcomes

  • Trained and evaluated predictive models for maintenance forecasting.
  • Used MLflow for experiment tracking and model versioning.
  • Built a web UI so predictions were accessible outside notebook workflows.

This project helped me understand how to implement the following features:

Multi-Model Classification: LightGBM, XGBoost, and Random Forest models for failure type prediction

Hybrid Time Series Forecasting: LSTM-based engine temperature prediction with intelligent trend analysis for extreme temperatures Multiple Issue Detection: Simultaneous detection of engine, brake, and tire issues

Intelligent Anomaly Handling: Smart anomaly indication that only triggers maintenance when other values are normal

Safety-Critical Temperature Analysis: Emergency detection for dangerous engine temperatures above 120°C