Back to the Mar-Apr 2021 issue

From Science Fiction To Railroad Systems – How Technology Improves Rail Safety

By Lydia Bjorge

Speeding train and caution sign
These trackside detectors can generate 35 million messages per day.

From drones and sophisticated machine vision systems to artificial intelligence, technology is revolutionizing the railroad network. Technology solutions are imperative to railroads’ goal of an accident-free future in communities across Minnesota and the U.S.

Imagine you’re driving and, without warning, a tire blows. Or, it’s dark and you hit a giant car-crunching pothole. Driving hazards like these are not uncommon. But if you could, wouldn’t you want to know when to expect them, so you could avoid them altogether?

Using artificial intelligence

Private freight railroads are already applying predictive technologies using artificial intelligence (AI) to find defective track and freight car components, both of which could potentially contribute to train incidents. Thanks to the speed of today’s computers and “big data” — the massive amounts of information being collected — AI is not only improving safety, but also asset utilization, service, and operational efficiency.

Just like online retailers use AI to suggest your next pair of shoes based on past purchasing habits, railroads are leveraging AI to recognize patterns and better predict issues with equipment and track components, so corrective actions can be taken before the failure.

Imagine knowing days, months, or a year in advance that your tire might blow, or months, a year, or five years in advance that a pothole is likely to form in a specific location. AI helps anticipate and find the problems, and rail employees can work to prevent or fix them.

The advancements in technology not only make railroads safer as they run through your city, but they make railroads better able to deliver Minnesota goods to the worldwide economy.

The wheels on the train

Cities have a variety of vehicles using their roads and bridges — both vehicles owned by the city and those owned by residents and visitors. In the same way, railroads have their own cars, as well as other railroads’ cars and customer-owned rail cars all traveling their network.

Broken railRegardless of rail car ownership, railroads want to ensure, for the safety of the community and continuity of operations, that all cars and their wheels are safe.

Railroads have long used sensors (thermal, acoustic, visual, and force) positioned along the track to detect freight car wheel defects. Thermal/infrared scanners, for example, can indicate the brakes are sticking and the components are overheating, potentially leading to a broken wheel.

Railroad data scientists, in cooperation with rail mechanical experts, use machine learning to evaluate wheel issues and identify conditions that are likely to cause an incident. Technology helps railroads find and repair the wheel conditions, likely reducing incidents and service interruptions.

A newer technology — a machine vision system — is used by railroads to identify cracks and breaks in wheels. Similar to how a child learns to spot a cat or dog by looking at pictures, AI models are learning to detect broken/cracked wheels by analyzing hundreds of thousands of images per day.

Keeping the ‘potholes’ at bay

Imagine if you could inspect every inch of your most traveled roadways at two, three, or even four times the amount of inspections they receive today — with millions upon millions of data points? Railroads are using technologies like ultrasound, radar, and machine vision systems that look deep inside the rail to find tiny flaws imperceptible to the human eye.

One example is the track geometry cars used to identify anomalies in tracks, measuring every foot for wear, track alignment, elevation in curves, gage (distance between rails), and other track geometry measurements. These geometry cars travel across the rail system to register track wear and tear by sending millions of data points back to the railroad.

The amount of data collected allows railroads to use AI to analyze hundreds of millions of bytes of information that help drive track maintenance — finding the “potholes” before they happen. This allows railroads to anticipate potential issues that could cause derailments months or years before they become a safety concern.

Shared goal of safety

The strength and safety of the nation’s infrastructure is important to the health of cities and growth of the national economy. Freight trains are often the forgotten side of freight infrastructure, chugging day in, day out across cities to deliver goods — hauling one-third of U.S. exports or serving nearly every sector of the economy.

Railroads’ $26 billion in annual investments into their own privately owned network generate a ripple effect across thousands of businesses and communities nationwide. Technology, a large part of that investment, has made safer Minnesota rails and communities.

Lydia Bjorge is executive director of public affairs with BNSF Railway (www.bnsf.com). BNSF is a member of the League’s Business Leadership Council (www.lmc.org/sponsors).