We, the railroads, have established certain principles of safe train operations that we wish were absolute, positive, inviolable. We apply one such principle in the enforcement of train separation: the principle that no two train movement authorities can be allowed to overlap. We guard, or to use the railroad term, protect,that principle with timetable schedules, operating rules, manual and automatic block signal systems and cab signals. We enforce that protection with automatic speed control and what we call Positive Train Control. We want this principle to be positively enforced, except when we—not don’t—but can’t.
There are times, every day on every railroad, where we have to require trains to follow one another within the same section of track. We substitute the skill, intelligence, alertness and most of all, sight distanceof the train crews for the positive principle.
That gets us to another principle that we bend, flex, or mitigate—yeah that’s the ticket “mitigate”—the principle that we don’t rely on sight distance to maintain safe train operations, since a) trains operate at speeds that exceed the sight distance available when a stop is required, and b) sight distance is variable, subject to disruption, never standard.
We are compelled by the very existence of the railroad to accept a certain level of risk, the risk intrinsic to putting objects of great mass in motion at varying speeds over a fixed guideway. Only when the railroad is not functioning is it without risk. “All trains are stopped. All signals are red, the railway is perfectly safe,” my British colleagues like to say, with more than a pint of British sarcasm.
The task therefore is not to proclaim a “zero tolerance” for risk. The task is to manage that risk, which means first to accurately measure risks in particular and as an aggregation of vulnerabilities, and devise methods for reducing the individual and combined threats.
Which gets us to this, NTSB’s report on the fatality of an Amtrak Northeast Corridor roadway worker on April 24, 2018, near Bowie, Md. (downloadable below). The NTSB report targets the obvious vulnerabilities in the method of on-track protection known as Train Approach Warning (TAW), a method that relies on “lookouts”—also known as gang watchmen and flaggers—to provide advance warning to the workers occupying a track of a train’s approach. The lookouts have to position themselves so that the line of sight provides sufficient warning so that all the workers can reach a designated place of safety 15 seconds before the train reaches the work zone. That’s not “clear the track in 15 seconds,” but be clear of the track 15 seconds before the train reaches the location.
This means a line-of-sight contact distance depends on the maximum authorized train speeds, size of the work gang, tools being used, location of the place of safety, and the geography—curves, hills, buildings, bridges. It’s supposed to be science, but it’s a quantum science, with a number of possible realizable states, and we’re using real people—not Schrödinger’s theoretical cat.
The obvious complication: What happens when in order to maintain the proper line of sight, or in a loss of “situational awareness,” the watchman places himself or herself on an in-service, active track? Short version: Who watches out for the watchman?
Now, the “more usual use” of TAW is for the protection of workers performing work on the railroad where the work on that track is unprotectedby the train dispatcher, who could grant the workers exclusive track occupancy. This means no trains or equipment are authorized to enter the track, now designated a work zone, without permission of the roadway worker in charge of the gang, or a designee. When using TAW in these circumstances, the track remains in service, no information is provided to trains or the train dispatcher of the location of the work, and no speed restrictions are issued for trains on adjacent tracks passing the work zone. Does that make you feel safe?
Can it be done safely? Yes, but that’s not the question. We established the real questions already. What is the particular risk of this method and how is that risk amplified in the aggregate of these methods throughout the system? I think the risk in particular and in combination is unacceptable.
The incident on Amtrak’s NEC near Bowie, Md., did not employ TAW in this “usual” application. The site of the actual work was Track 2, and that work zone was protected by exclusive track occupancy, which requires no lookouts or watchmen. The watchmen were assigned to provide warning for trains operating on the tracks adjacent to the work zone, Tracks 1 and 3. Since the work zone itself was protected, the “15 seconds before” requirement is, more or less, moot. The gang watchmen were likely there simply because the MAS on the adjacent tracks is 125 mph. A train passing at 125 mph creates quite a slipstream.
It is possible that the watchmen might have been responsible for requesting foul time when the work itself might have fouled an adjacent track, but that responsibility should fall to—and only to—the Roadway Worker-in-Charge (RWIC).
Railroads in the U.S. have adopted measures to protect an in-service track that might be fouled by work on an adjacent track. GCOR’s (General Code of Operating Rules) participating railroads (led by BNSF and Union Pacific) utilize Form B protection, which requires the communication through bulletin order, or track warrant, of the placement of fixed flags on specific tracks at specified distances from the track area requiring protection. The “approach” flags (yellow over red) have to be placed 2 miles to the rear (prior to reaching) of the work zone, which itself is identified by a fixed red flag. Trains passing the yellow-red flags must approach the red flag prepared to stop and obtain permission from the RWIC to transit through the area. The RWIC will or will not authorize the movement, specifying any necessary speed restriction.
Other railroads establish working limits on adjacent trackss through bulletin order or track warrants identifying the location of “approach working-limits stop signs” and “working limits stop signs.” Again, the signs have to be placed the appropriate distance to allow deceleration to zero velocity, and again the trains have to obtain permission to transit the work zone from the RWIC.
Historically, Amtrak has not employed a similar method for train movements on adjacent NEC tracks, and that is due to the higher speeds and greater density of train operations. The higher speeds translate into a long required distance for deceleration, which turns into a significantly greater time operating at the reduced speed, and a significant obstacle to meeting schedule, particularly when there might be five or six such work zones between Washington D.C. and New York Penn Station.
At 125 mph, the required distance for a passenger train to reach zero velocity, when making a service brake application, is approximately 9,500 feet (1.8 miles). More important to train performance is the acceleration rate, to recover to 125 mph after stopping. Amtrak’s ACS-64 locomotives are advertised, when hauling 18 coaches, to achieve 125 mph in eight minutes, so you can see the impact on schedule performance that each authorized but unscheduled stop can have.
Now, I oppose the use of TAW as a form of roadway worker protection pretty much across the board. In the past, I thought I could find exceptions allowing for TAW in single-track areas where the MAS does not exceed 40 mph. Such railroad territory is pretty much “dark” territory—unsignaled, where train movements are authorized by timetable schedule, or a written movement authority, or a combination of both.
But things are different today, and the thing that makes the difference today is called PTC. The authorized PTC systems, including Alstom’s ACSESII on the NEC, allow the train dispatcher to identify, communicate and enforce locations for temporary speed reductions as and where necessary or desired. Similar to the Wabtec ETMS-based systems, ACSESII employs digital data radios to communicate the temporary speed restriction to the ACSES subsystem computers aboard each locomotive. No speed signs are required; no working limits on adjacent tracks need be established.
After establishing positive protection on track removed from service, the train dispatcher can set temporary speed restrictions for trains operating on tracks adjacent to the work areas. The train dispatcher can select the locations, the times and the allowable speed when passing the work zone in consultation with the RWIC. Even with such consultation, I don’t think a passing speed greater than 40 mph should be allowed.
Is this perfect? No, it’s not even positive, meaning, regardless of the speed reduction, a fatal accident is possible as we have not remedied the root cause—the fragile, and elastic, nature of the human attention span. But until we can do that, we have to manage the risk we’re forced to accept. We have to “compromise.” The main way we compromise on a railroad is through controlling the maximum authorized speed.
DOWNLOAD THE NTSB REPORT:
David Schanoes is Principal of Ten90 Solutions LLC, a consulting firm he established upon retiring from MTA Metro-North Railroad in 2008. David began his railroad career in 1972 with the Chicago & North Western, as a brakeman in Chicago. He came to New York in 1977, working for Conrail’s New Jersey Division. David joined Metro-North in 1985. He has spent his entire career in operations, working his way up from brakeman to conductor, block operator, dispatcher, supervisor of train operations, trainmaster, superintendent, and deputy chief of field operations. “Better railroading is 10% planning plus 90% execution,” he says. “It’s simple math. Yet, we also know, or should know, that technology is no substitute for supervision, and supervision that doesn’t utilize technology isn’t going to do the job. That’s not so simple.”