The challenge of driverless trucks

Some of you don't see my columns in Trains Magazine. But even if you do, it’s not a format that can accommodate give-and-take between readers or between you and me. So I am reproducing my February 2017 essay here and inviting your comments.

Fred

What should the railroad industry do when technology threatens to upend it? This is the question of the hour, and it concerns not some new way to run trains, but a new way to run trucks, without drivers. Not all of you share my concern, but I put you on notice: This is the future, and it’s not a pretty place. Today these vehicles are making test runs, meaning it’s only a matter of time until the future becomes now. Matt Rose, executive chairman of BNSF Railway, believes the first wave of autonomous trucking will arrive by 2020-2022, in the form of truck platooning. The question you should be asking: If railroads want to remain meaningful, what should they be doing today?

It’s a hard question to answer, because autonomous trucks are only one threat hanging over this centuries-old industry. Coal had been the financial backbone of railroading, and now that backbone is broken. Intermodal traffic is held out as the growth engine that could replace the revenues lost from coal. But it’s not happening now, and for it to ever happen you must provide shippers with service that is relatively fast, utterly dependable, and truck-competitive in price.

So back to the question of what to do. It would obviously help the economics of railroading to have one-person crews. In 2014, BNSF and SMART, the former United Transportation Union, negotiated a deal one step away from that objective. It provided for engineers only on the locomotive, and for a conductor to assist all trains in a designated part of the crew district by truck. SMART members voted it down handily.

Just to make sure nobody tried that stunt again, former UTU official Joe Szabo made it his business in 2014, while he still headed the Federal Railroad Administration, to initiate a rulemaking that would require two people aboard trains. That rule got lost in the White House because FRA could cite no evidence that two people on the locomotive make a train any safer than one. But it resurfaced last year and remains in the regulatory pipeline. It boggles the mind that the same Department of Transportation that is investing billions to enable driverless trucks wants to tie the hands of railroads in that respect.

And one more variable: According to Rod Case, who heads the rail team for consulting giant Oliver Wyman, one in four U.S. freight trains encounters an unplanned delay, primarily recrews, unscheduled work, and crossing protection failure, in that order. That compares to, say, one in 3,000 freights on the Russian railways. In short, a lot stands in the way of the kind of reliability that leads to new business.

These trends — autonomous trucks versus regulated two-person rail crews and demands for reliability versus frequent service failures — appear to have our highly profitable and widely acclaimed freight railroad industry ready to slide into oblivion.

But I’m not ready to concede, nor should the railroads. Consultant Case presents these cost comparisons: Today, if rail costs to move a container 2,000 miles are indexed at 100, highway costs are 131. (A 31 percent cost advantage hides a lot of late deliveries.) A three-unit platoon with drivers in all cabs would bring the truck index down to 121, a three-unit platoon with a single driver would drop the index to 96, and a driverless truck would be indexed at 79.--Fred W. Frailey

It’s clear that reducing the crew size to one person won’t by itself make rail costs competitive with a driverless truck. On the other hand, if you stand pat, the game is going to end badly. BNSF’s Rose says he has stayed in touch with unions representing his on-board employees. “Three years ago there were a lot of deniers,” he says. “A change in attitude is occurring.” As for the FRA’s pending crew-size rule, Rose says this: “The drafts we’ve seen don’t meet the giggle test of cost-benefit analysis. We have the right to sue, and we will win.”

Fuel accounts for 20 percent of all rail costs, and a switch from diesel to liquefied natural gas could take a chunk out of that cost. Maybe railroads should contract with companies like Hulcher Services to assist engineer-only trains that get need assistance en route (bye-bye conductors). But chipping away at this cost bucket or that while largely running a railroad the same old way won’t get us there. Let me pose four what if’s.

What if crew costs no longer determined train lengths or schedules? What if there were no crew costs? The Australians are closing in on driverless train technology. Instead of one scheduled  train per day between city pairs, there might be two or three or four.

What if railroads made it their goal to be highway-competitive over shorter and shorter distances and then did whatever it took, including technological breakthroughs, to make that possible? This is where the money lies; why leave it on the table?  Rail market share today over distances of less than 500 miles is 7 percent, according to 2013 estimates.

What if railroads doubled or quadrupled asset utilization? Instead of one turn a month from boxcars, they got two or four? The Canadian roads are good at getting more from less, but U.S. carriers not so good.

What if railroads thought of themselves as part of their customers’ supply chains, instead of just companies that run trains, and wormed their way inside the decision-making process of how to route freight? This is what Canadian National has started to do, and what intermodal trucker J.B. Hunt Transport already does, to the benefit of BNSF in particular.

These aren’t my ideas. They’re out there, begging for answers.