What does it mean for service when long trains get even longer?

When railroads first figured out that operating longer trains was an easy way to make more money, firemen were still tossing wood into the bellies of their 4-4-0 steam locomotives. Since then the development of larger, more powerful locomotives was done with one goal in mind: Pulling more tonnage with a single crew. So today’s Class I railroad trend toward ever longer trains is nothing new.

What is new is the zeal with which the Class I railroads are embracing longer, heavier trains that allow them to move tonnage with far fewer people and locomotives.

To see why the 14,000-footer is all the rage, just glance at the cost savings from running longer trains. Union Pacific saved $268 million in operating expenses last year by boosting train length. Tack on the savings from increased locomotive productivity, which is a byproduct of bigger trains, and the combined tally comes to $487 million. That’s not chump change.

In all, average train length on UP has grown by 30%, to a shade over 9,000 feet, since it adopted Precision Scheduled Railroading in October 2018. The other U.S. Class I systems running the PSR playbook – CSX Transportation, Norfolk Southern, and Kansas City Southern – have increased train length, too, following in the footsteps of Canadian National and Canadian Pacific. And even BNSF Railway, the only big road that has not jumped on the PSR bandwagon, is not immune to the trend. 

The longer train movement shows no signs of abating. During their earnings calls last month each of the publicly traded Class I systems said they plan on running even longer trains this year to further boost productivity.

What you don’t hear is how moving tonnage on fewer but longer trains affects service. Railroading is, after all, a service business. So it’s fair to ask: Are we headed back to the drag freight era and the hold for tonnage approach to operations, with the resulting toll on service reliability? 

The disadvantages of operating 14,000-foot trains are well understood. They take longer to build and to put in a yard at destination. They take longer to get up to track speed, if they make it at all, and take longer to stop. They might fit in only some passing sidings, which can complicate operations. If there’s a problem en route, like getting flagged by a wayside defect detector, the conductor may have to walk a mile or two to find the problem, then hoof it back to the head end. This ties up a main line, with delays rippling outward from the troubled train. And it means more grade crossings may be blocked, which is of increasing concern to communities, politicians, and regulators.

Yet the long-train trend – when combined with other significant operational changes – does appear to offer some service and operational benefits. 

First, if you’re moving your traffic in fewer trains, that reduces the number of meets in single-track territory and helps ease congestion on busy double-track routes. Some railroads have seen average transit times drop by as much as 25% between crew change points simply by having fewer meets.

Second, blending unit train traffic into the merchandise network helps ensure there’s enough tonnage to support a daily merchandise train. Before the shift to long trains, Norfolk Southern on some corridors might have scheduled a daily merchandise train, an auto train that ran five days per week, and steel and grain shuttles that each ran three days per week. But with volume as variable as it is, the merchandise train might get annulled 10% of the time. By folding the automotive, steel, and grain traffic into the merchandise network there’s enough volume to support a 12,000-foot train that runs every day. This also can reduce dwell time at the originating terminal.

Third, railroads are doing more pre-blocking of traffic at origin, which avoids a stop at a local yard that can consume 24 hours. They’re also relying more on block-swapping traffic en route, which can save 10 hours or more compared to switching or shoving a car over a hump. A long train might carry five or six blocks and pick up or set out blocks three times en route to its final destination. This creates an incentive to run trains every day because if a railroad annuls a scheduled train, it has consequences down the line: Those mainline work events don’t occur, and blocks up to 3,000 feet long get stranded.

So in theory the long-train phenomenon is not a hold for tonnage approach, under which the railroad won’t run a train until enough cars accumulate to hit whatever the set tonnage threshold may be.

The numbers bear this out. At UP, for example, cars travel 15% further per day, terminal dwell is down 24%, and average train speed is up 9% since it adopted PSR. And an analysis by Rick Paterson, a railroader-turned-analyst at Loop Capital, shows that compared to long-term averages dating to 2015, traffic over the past three months through January is moving faster on CP, CSX, and UP and is moving at around the long-term average at BNSF, KCS, and NS.

But in practice, of course, your long-train mileage may vary. Some chief operating officers have cracked down on local operating people who originate trains under 5,000 feet. After getting dressed down by headquarters, you can be sure the folks in the field are holding tonnage in the yard until they can send out a train long enough to keep the boss happy.

You can reach Bill Stephens at bybillstephens@gmail.com and follow him on twitter @bybillstephens