Predictive analytics

TECHNOLOGY-DRIVEN TRAIN INSPECTION

Predictive analytics is a key part of our technology-driven train inspection program.

Zone Configuration ‭[1]‬

Content Editor ‭[1]‬

In our industry, we are at the forefront of predictive analytics – using our patented technology, we accumulate Big Data on locomotives, railcars and track infrastructure. As a result of this data, we are able to anticipate issues and take preventative measures before an incident occurs.

CP's patented algorithm has resulted in a 95% reduction in bearing failures during scheduled service, keeping our trains moving quickly and safely across our network.
Our predictive analytics innovation was recognized with a safety award by the Railway Association of Canada in 2018.

Zone Configuration ‭[2]‬

Content Editor ‭[2]‬

Wheel Impact Load Detectors

Our wheel impact load detectors, located on the rails, more precisely measure and produce alerts for overweight or severely imbalanced railcars. CP can now forecast individual freight car wheel life and be more pro-active to avoid service interruptions.

Acoustic Sensors

By using acoustic detectors, we can better predict when bearings are going to fail. The model is able to predict this three months in advance, allowing us to pull and repair these assets before a failure occurs.

Autonomous Track Geometry Measurement System

This system uses a non-contact, laser-based optical measuring system attached beneath a boxcar for near real-time defect detection and it can operate on any train, at any speed, providing notification of critical defects with a link to Geographic Information Systems (GIS) mapping.

The system’s ability to consolidate defect information allows it to predict track deterioration, thus improving service, reducing derailments and unplanned work outages, and allowing for increased planning around track maintenance.

Track Inspection

We also have technology mounted on locomotives to identify possible track defects by measuring vertical and horizontal acceleration and identifying unusual movements. Vehicle track interaction units now analyze more than 600,000 miles of data annually on our network.

All of our track data from new or existing inspection applications is fed into our Track Asset Management platform to provide a consolidated view of track health and leverage big data.

Train Inspection Portal System

We have a high-speed, camera-based train inspection system capable of producing full body high-resolution images of trains at mainline track speed. The system uses multiple algorithms to assist in identifying defective car components.

Undercarriage imaging is a system that uses cameras to inspect the underside of all passing railcars and locomotives enabling CP to identify missing bolts, bent or broken brake rigging, open bottom gates, broken couple systems and draft arrangements.

Automated Bridge Monitoring

This system monitors the conditions of bridge components in real time and sends an alarm in case of any issues. Highly portable and powerful devices capture small movements, vibrations and temperature fluctuations.

The data from the devices can be reviewed or set up to automatically send an alarm if movement is beyond pre-set thresholds.