I found the silver bullet at WRI

We’re on the verge of something big.

Until just this morning, when I paused to reflect on the 23rd Wheel Rail Interaction conference from Montreal in early June, I thought there was no silver bullet out there for railroading to cure all real and perceived problems; no steam-to-diesel-like transition waiting in the wings.

I was wrong.

Data from more machines and sensing devices will drive railroading changes fast enough to make heads spin — just as in nearly every other industry.

My evidence: Results are just coming in from a long-term test on the New York City Transit Authority’s Flushing Line that show: 1.) the 27.5-mile, 46-million gross ton a year-subway line can be monitored continuously and 2.) there’s enough data to begin tweaking the system this year and measuring the results. NYCTA officials are partnering with investigators from the National Research Council of Canada and a laundry-list of suppliers in the tests that are being paid for by Federal Transit Administration (watch for a story on this in the October issue).

Oh, and Canadian National’s Vice President of Engineering David Ferryman says that the railroad is readying a sensor-laden boxcar to enter revenue service and report back train handling and track conditions continuously, automatically, and indefinitely. Sources attending WRI say one U.S.-based supplier is outfitting three boxcars, one for CN and two other railroads. If these boxcars work flawlessly, as expected, there’s little reason to say that sensors can’t continue to improve, shrink in size, AND lead the way to foot-by-foot continuous freight railroad monitoring.

But wait, there’s more.

The suppliers attending the conference this year are showing more tools for railroaders to use to be more precise and efficient: Albany, N.Y.-based automated wheel-shop maker Simmons Machine Tool Corp. presented their latest as did Spain’s Danobat. The Spanish company proposes an automated freight car repair shop where wheels, axles, trucks, and wheel bearings (among other things) can be replaced by robots assembly-line style in as little as 28 minutes and returned to service. Another supplier was showing off wayside equipment that will detect truck hunting problems on freight cars in motion and flag problem wheels in real time.

WRI organizers wrapped this topic with a bow by a presentation on how railroads will be able to extract information from all of this information. (How? Carefully, and with documentation.)

And then? And then railroads will be able to find patterns in the daily performance of every locomotive, freight or passenger car, wheel or wayside detector. Managers will be able to spot problems more quickly and demand more from suppliers in terms of durability and quality; they'll make quick changes to operations and practices the same hour when weather interferes and only when necessary.

If I had to make an analogy of how far along I think the railroad industry is along the path of using big data, I would say that North American railroads have built their first engine and are checking the valves, compression, and timing and are waiting for the right platform to run the Big Data engine on. No, it won’t be perfect or wonderfully efficient to start. But when railroads get it right, look out.