Why AI/ML based Alerts needed?
RDPMS AI Model
The RDPMS AI Model is designed to analyze historical data trends across various railway signaling assets like point machine,signal,Track and their associated attributes. Leveraging machine learning and statistical trend analysis, the model provides predictive maintenance alerts that help prevent failures before they occur.
Predictive Alert Insights – Explains alert causes, trends, and possible failure points. Asset Health Checks – Answers natural language queries about current or historical asset performance. Maintenance Guidance – Shares procedures and checklists based on alert or asset type.
Model Properties
Analyzes time-series data patterns to detect anomalies, degradation trends, or early signs of failure—triggering alerts before issues impact operations.
Data is processed individually per asset and attribute, allowing for precise diagnostics (e.g., Point Machine → Motor Current,motor Votage Operating Time, Gap Sensor and Track Feed In current/voltage, relay end current/voltage).
Instead of relying solely on static thresholds, the AI adapts based on the historical behavior of each attribute to reduce false positives and enhance accuracy.