AI-Driven IoT Sensor Platform for Predictive Maintenance
AI-powered IoT sensors predict escalator and elevator failures in advance, improving passenger comfort and availability. The system reduced maintenance costs by up to 30% and increased availability by 2%, contributing to Metro Istanbul’s digital transformation.
About this Project
In the rail systems sector, maintenance operations are typically performed after failures or at fixed intervals, often leading to service interruptions and high costs.
Through this project, an AI-driven IoT sensor system was implemented to continuously monitor escalators and elevators. The equipment’s health status is analyzed in real time, enabling planned maintenance before a failure occurs.
By analyzing historical data, the system predicts potential breakdowns, reducing maintenance costs and minimizing downtime and service disruption.
Key Highlights
The project transforms traditional reactive and periodic maintenance into a data-driven predictive model.
IoT sensors continuously collect vibration, temperature, and current data, which are analyzed by AI algorithms to assess equipment health in real time. The system automatically detects anomalies and calculates failure probabilities, providing early alerts to maintenance teams.
This prevents unplanned downtime, optimizes resource use, and enables data-based maintenance planning. The approach establishes a scalable digital maintenance culture within Metro Istanbul’s operations.
Facts & Figures
- With the AI-based IoT sensor system, maintenance costs decreased by 30%, availability improved by 2%, and failure durations were reduced by 40%.
- Station accessibility and passenger comfort improved noticeably.
- The solution is scalable across lines.
- Hardware and deployment costs are recovered in about 3 years, driven by maintenance savings and reduced service interruptions.
