SCHAUMBURG, Ill. — The power of data, or more accurately, of well-used data, may not have explicitly been the topic of BNSF Railway CEO Katie Farmer’s presentation at the Midwest Association of Rail Shippers’ Winter Meeting. But it was a recurring theme with several subjects she touched on as part of her morning keynote address today (Wednesday, Jan. 15) on the conference’s first day.
Data played a part in Farmer’s look ahead to BNSF’s Barstow International Gateway intermodal and logistics project; in discussing the railroad’s program of automated track inspection; and in efforts to improve information on arrival times for shippers.
“I’m really excited about what we’re doing in Barstow,” Farmer said. “And not just because of the capacity … What I’m really excited about is the conversations we’re having with our customers, ocean carriers, drayage companies, the largest importers in the country … What we’re talking about is not only capacity, but how we’re going to be interchanging data with each other.
“To give you an example, if I’m a large importer and I have a box on a vessel, wouldn’t it be cool if while we have a box on the vessel, we have all the information coming to us and we can make decisions, the customer can make the decision while that box is on the vessel about, do I want to go IPI [inland point intermodal]? Do I want to stay local? Do I want to transload it? We have that information, we [can] plan to work in our intermodal hub around that.
“That’s the kind of conversations that we’re having, and that’s how we’re going to continue to adapt and evolve the logistics park concept,” she says.
Regarding track inspection, BNSF has been working to extend a Federal Railroad Adminstration waiver for its automated inspection program — an effort slowed by the FRA’s desire to make those waivers more difficult to obtain [see “FRA aims to tighten rules …,” Trains News Wire, Oct. 28, 2024]. But at the same time, Farmer said, it has been working to develop an inspection system mounting cameras and lasers on locomotives in revenue service.
“It allows us to look, real-time, at foot by foot by foot of track for any structural integrity issues … It allows us to take that information, send it to Fort Worth, and quickly use machine vision and artificial intelligence to identify, more quickly than ever, if there’s an issue with that piece of track. That allows us to take our track inspectors and move them from finders to fixers, and to be able to remediate that track issue and not take large windows of time that interrupt that service,” Farmer says.
On another front, Farmer talked about the railroad’s work to refine ETAs for shippers, noting it has been “a pain point” for customers.
“And so a couple of years ago we hired 20 data scientists and operations research folks — really bright folks — and they went to work on solving several problems,” she says. “One of those is what we can do to improve the ETA’s on our railroad.”
The aspiration, she explained, is to achieve the sort of continuously evolving arrival information available from parcel shippers. As an example, she cited a recent online purchase.
“The minute I placed the order, UPS sent me a tracking number and said you can expect to see this on Jan. 14,” Farmer says. “Over the next week that ETA kept getting refined, and the day before it came, I got a message that said you can expect your shipment between 3:15 and 5:15 on your doorstep.
“That’s the concept we’re trying to use here. Traditionally, the rails will give you a static ETA. What we’re saying is is take all the historical information, use the predictive tools that we have — our base transportation system; Movement Planner, which is what our dispatchers use; hub planner; throughput planner, which is what our folks use to plan work internals — take all of those predictive analytics, along with the historical information, and continue to give a predictive ETA.
“Because what we believe is that it’s not so much where your shipment is, but the confidence level you have of when your shipment is going to get you. Because then you can plan your crews to unload; you can plan your inventory better.”
A pilot program with a few carload and grain shippers, she said, had shown the railroad could, with 85% to 90% accuracy, provide an ETA within a window of plus or minus 4 hours.
“I’m incredibly encouraged with the progress we’re making on this,” she says. “We’re not there yet, but we’re going to broaden the pilot out.”
