The State of the Rails – CN Back from the Brink; PSR Recoverability
The Canadian National Mini-crash
Canadian National gave us a scare last week, with a mini-crash that pushed terminal dwell to a three-year high and velocity to a low we’ve never seen. The data reflected the week ending February 21, and the sequential deterioration from the prior week was also stunning: velocity slowed 15% and dwell increased by 23%. We now have data for the subsequent week, ending February 28, and it thankfully recovered what was lost the prior week, resulting in the V-shaped spikes in the charts below. This is good news, but what caused the mini-crash?
According to CN IR, it was entirely a cumulative and compounding winter weather effect, with persistent February cold and snow in the Western and Eastern regions, as well as the US Midwest. There was also some flooding in the South. As a result, CN had to activate is winter operating plan/tier restrictions on most February days, which means shorter trains, insufficient capacity, and growing backlogs (which are now being worked down). The Western region continues to advance its recovery; the Southern region (US) is in pretty good shape now; and the Eastern Region, which was hit with snow and cold temps last weekend, has been a bit slower to recover.
For perspective, CN’s winter operating plan has three tiers. Tier 1 becomes effective when temperatures drop below -13 deg.F and CN limits intermodal and bulk train length to 10,000 feet. Tier 2 kicks in when temperatures drop below -24 deg.F and further restricts train length to 8,500 feet; and Tier 3 starts at -33 deg.F and caps train length at 6,700 feet.
In the winter of 2023/2024, through the end of February, CN had a total of 21 days under Tier restrictions (10 Tier 1, 11 Tier 2) and that has more tripled to 63 days in the current winter, including a much higher proportion of Tier 2 days (26 Tier 1, 37 Tier 2). The current winter also began tormenting CN’s network much earlier, with Tier restrictions in both November and December 2024, whereas there were none in these months in the prior winter. As measured in Tier days, this winter has been CN’s most challenging since 2020/21, which included the last big polar vortex.
The bottom line here is that after four months of Tier restrictions the compounding effects on train length and capacity seems to have built up and finally detonated the mini-crash in the third week of February. Let’s at least be thankful CN has sufficient power and crews, because deficits on either might have turned this one-week blip into a months-long problem.
Why PSR Networks Have Slow Recoverability
We recently met with a retired Class I operations executive, who had a very interesting take
on the recoverability of modern PSR networks. Here it is in his own words:
Following a service interruption, such as a major derailment or weather outage, why is it more
difficult and longer to restore service in an operation that tries to minimize operating ratio?
Because there are fewer human resources.
To restart a train that has been stopped for several hours or days for an outage with its crew
removed, a fresh 2-person crew must be driven to the train. The engineer stays with the
locomotive while the conductor walks the train to visually inspect whether brakes will set and
release. When the train was “tied-down”, hand brakes had to be set on a sufficient number
of cars to keep the train from rolling away. The one conductor must climb up each car, one
at a time, and release the hand brakes. This may involve 10 or 20 or more cars, depending
on the size of the train and grade. If they exist, supplemental mechanical and/or supervisory
resources can assist in the visual brake inspection and releasing hand brakes. But remember
that on most Class I’s, “PSR” reduced mechanical and engineering headcount as much as train
& engine crews. The resources are no longer there, so all this takes longer now.
Another reason it’s slower is that trains are longer.
It just takes more time to walk or even drive a longer train that has been tied-down for a major
outage. It takes more hand brakes manually set and then released to hold a bigger train. And
the likelihood is greater that the train will need to be “cut” (separated) for road crossings.
Recoupling a cut adds another level of work and time to activate a tied-down train. Picture
one person on the ground at the separation talking by radio with the engineer to back up
the front portion of the train to couple-up again. Then you pull forward to test the coupling.
Then you hook up the air-brake hoses and pump the entire train full of air again to operate
the brakes. A longer train takes longer to pump-up. And THEN you do the walking or driving
brake inspection described in the prior paragraph.
Finally, consider how this would (not) work in a fully autonomous operation.
Addressing Counter-arguments
Push-back from diehard PSR advocates usually includes the following:
- Push-back: These issues are offset or partially offset by fewer trains (less traffic congestion) in some/many corridors. My Response: Are we trying to shrink or grow? Citing lower volumes as a virtue seems self-defeating. That should NOT be one of the goals of PSR.
- Push-back: Autonomous train operations would be supported with much greater on-line resource (human) support who could go to the problem and deal with it. My Response: a) You’re giving back some portion of savings, maybe most of it; and b) Relying entirely on roving resource support assumes people can drive to a stopped autonomous train that has a problem. Significant portions of the US rail network are not equipped with such access. Accessing an autonomous train in such a situation would require a helicopter. Just add that to the recovery time.
- Push-back: Done right, PSR does not need to degrade service. My Response: a) You are describing the sunny-day operation where nothing goes wrong, and b) What do the customers say? We’re pretty deep into PSR at this point. Are rail customers happy with it and rewarding the railroads with lots more traffic? Where’s the proof statement that this strategy can take us forward?
Conclusion
These methods of rail operation do not support making service more reliable and faster. In
order to pivot to growth, the industry must come to grips with this reality.
Our thanks to our contributor for an insightful analysis.
