Case Study
KMRL — SIH Solution
End-to-end system for metro data analysis, issue detection, and predictive maintenance.
Manual reporting time -60%
Anomaly detection accuracy 85%
Dashboard load under 2 seconds
Better metro authority decisions
Overview
A complete end-to-end solution for Kochi Metro Rail Limited. The system analyzes 1,000+ daily metro operation events, automates issue reporting, and predicts maintenance needs using ML models.
Key Features
1,000+ real metro data points per day
Automated issue detection 85%
ML-based maintenance prediction
Centralized station dashboard
Incident categorization under 300ms
Real-time REST API integration
Challenges
Handling incomplete or inconsistent real-world metro data required cleaning, normalization, and predictive modeling while ensuring system reliability.
Results & Impact
01Manual reporting time -60%
02Anomaly detection accuracy 85%
03Dashboard load under 2 seconds
04Better metro authority decisions
Tech Stack
PythonFlaskMachine LearningPostgreSQLStreamlitREST APIs
Links
By the Numbers
Manual reporting time -60%
Anomaly detection accuracy 85%
Dashboard load under 2 seconds