Ten90 Solutions LLC

So...here we are again.  Yesterday, January 26, 2016, The New York Times pre-published on its website an article on the derailment of Amtrak 188 last May at Frankford Junction.

It's a big article, scheduled for the magazine section of the upcoming Sunday edition, 6500 words, plus pictures and charts.   The article contains any number of errors and mistakes, but of course, when dealing with the media, errors and mistakes aren't important, because it's all about entertainment, not information. 

How could I say such a thing, about the great gray eminence that is The New York Times?  How could I be so churlish when the article presents so clearly, fairly, and balanced, the great suffering that this accident has inflicted on so many, including the locomotive engineer? 

Actually, it's not that hard.  I'm in the business. The difference between information and entertainment is that the latter, entertainment, requires a suspension of disbelief on the part of the reader or the viewer-- known in the entertainment biz as "the audience." 

The former, information, is produced by the application of disbelief; by the use of critical inquiry to the stories, narratives, interpretations that surround every incident.   The article on The New York Times website trades, in fact, on the readers' ignorance of the evidence established, the information distilled by the inquiry to date.  

The article wants to present a story, a narrative, of a tragedy, an overdetermined tragedy, in which multiple circumstances combine and overtake the best intentions, expertise, and skill of a locomotive engineer.  Some of those circumstances are the stressful nature of the job, the reduced rest between train runs, the unavoidable frailty of the human operator, budget constraints, the risks of operating trains through inner city neighborhoods, etc. It's clear where the author of this piece is going-- to a rendezvous with destiny, that classical condition, where the controlled is sabotaged by the uncontrollable and the cosmos once again disabuses the human animal of its pretenses at mastery. 

Makes for good drama. 

The problem with drama, with entertainment, is that the narrative rarely fits the mundane nature of the facts, the pedestrian qualities of the circumstances, the quotidian characteristics of the quotidian life.  

The problem with drama is that it obscures the real information that holds the key to preventing repeats of the incident itself. 

Accuracy is essential requirement for accident investigation.  For drama, it's not necessary. 

So dramatically, the article claims "Engineers must be on the lookout not just for pedestrian tresspassers on the rails, and errant cars and school buses and trucks at busy road crossings, but also for orange jacketed work crews." 

Undramatically, these are not the primary obligations of locomotive engineers.  By the time a locomotive engineer operating a train at authorized speed on the NEC spots a trespasser, it's already too late to do anything other than sound the horn, if there's even time to do that.

Bluntly, undramatically, striking a trespasser, which no one wants to occur, does not threaten the 200-1000 passengers on board the train.  The safety of those passengers is the primary obligation. All other concerns are secondary. All locomotive engineers are taught that. 

Undramatically, on the section of the NEC between DC and New York Penn Station, there are no grade crossings where trucks and cars and buses cross.

Undramatically, when work crews are given exclusive use of a track, and may obstruct the movement of trains on an adjacent track, the exact locations where that work is being performed is transmitted to all train and engine crews, so that "surprises" do not occur.  

Where work crews are on track, under their own protection, meaning the work crew accepts the responsibility for clearing the track in advance of an approaching train without requiring additional warning or protection from the train dispatcher, the crews are working in exactly that manner-- under their own protection, and the locomotive engineer of a train authorized to use that track has very limited responsibility, and few actions that he or she can take if and when the work crew fails to protect itself.  

Dramatically, the article says:  "Elsewhere in the country, long-haul engineers might be able to bring a train up to speed, level the master controller and skate for a hundred miles without making an adjustment.  On the Northeast Corridor, even a moment's inattention can be catastrophic." 

Less dramatically, a moment's inattention can be catastrophic to the  operation of any train on any section of track on any railroad.  That there are passengers on passenger trains is what makes the risk of inattention seem more catastrophic than that risk when only freight trains are involved.  It is precisely not the route complexity of the NEC that heightens the risk of accident, as that route complexity is mitigated by cab signal/speed control, the reduced length and weight of the train, and the scheduled nature of the service.

Dramatically, the article states: "In much of Asia and Europe, engineers are protected by a technology known in the United States as positive train control, or PTC.  Connected by digital radio waves or GPS signals, PTC transponders in the track maintain constant contact with computers in the cabs of oncoming trains...Amtrak has been working on its own in-house version of PTC, called Advanced Civil Speed Enforcement System, or ACSES, for almost a decade.  But owing to insufficient funding and a row with the FCC, which Amtrak said had been slow to approve the use of requisite radio bandwidth, its actual implementation has been piecemeal.  At the time of the accident, large portions of the Northeast Corridor, including Frankford Junction were not online.  Practically speaking, that meant engineers were working with no safety net." 

Less dramatically, none of the passenger service in Asia or Europe utilizes what has been designated by the US Congress in the Rail Safety Improvement Act of 2008 as PTC.  Various systems of train control, and advanced automatic train control exist on railways throughout the world.  But except for 40 miles of railway in Panama, some tracks in Brazil,  and some rail lines in Iraq (courtesy of the US DOD), PTC is not used outside the US.

The European Union has named its train management system the European Rail Traffic Management System (ERTMS) and the enforcement and train control module of that system is called the European Train Control System (ETCS).  There are various "levels" of enforcement anc control, levels 1, 2, and 3, with level 3 corresponding to the CBTC "moving block" concept.  Last time I checked Level 3 is not in operation anywhere in Europe. 

Level 2, which corresponds closely to US cab signal/speed control systems enhanced to enforce  positive stops, and civil speed restrictions, is operable on certain sections of EU countries' railways, and that system requires transponders, called balises, to communicate from "track to train." 

The PTC system adopted by most railroads outside the Northeast Corridor, and all the Class 1s and their tenants, does not require in-track transponders, but utilizes wireless digital radio communication to authorize and restrict movement. 

Amtrak is indeed using ACSES to satisfy the functional requirements for PTC as mandated by the RSIA, because Amtrak had installed an initial version of ACSES on portions of its road well before 2008. 

Yes, radio bandwidth is an issue now, but it was not an issue for the original version of ACSES.  That original version required in track transponders to enforce speeds around curves and over bridges and did not require additional radio bandwidth outside the band already assigned to rail operations. 

The enhanced version is backwards compatible with ACSES 1, meaning that the speed limit on the curve at Frankford Jct. could have been enforced through use of transponders communicating with the ACSES equipped locomotive on #188 prior to the installation of the PTC compliant version of ACSES.  

Even absent the application of ACSES 1 to sections of track, Amtrak engineers were never "working with no safety net."  They were working with cab signal/speed control systems designed to maintain safe train separation.  They were working with timetable special instructions that identified areas where train speed was restricted.  They were trained to know these areas without referring to the timetables.  They were tested on their knowledge, and certified.  And re-tested and re-certified.

Dramatically, the article quotes the head of FRA: "If a train is traveling in an area where PTC isn't in place and working as a backstop, you've got a situation where an engineer has to execute everything perfectly every hour every day every week.  All the time.  Because the slightest, smallest lapse can mean disaster." 

Less dramatically, that's what the job of locomotive engineer is--  no matter what happens, always to control the speed of the train. That's what the job always has been. 

That's what the job will always be, because PTC, like all train control systems can fail, and then the human operator must deploy his/her skill, training, knowledge, vigilance to accomplish the single task she or he was hired to accomplish-- to move the train safely, at the proper speed, as determined by the operating rules and procedures of the railroad.

Dramatically, the article proclaims:  "The spring of 2015 was a fractious time for Northeast Corridor engineers.  In March, over the objections of the unions, Amtrak had cut the time engineers were allowed between daily runs, from an average of roughly two and a half hours to, in many case, 90 minutes or less.  Amtrak employees foresaw disaster: 'Forcing shorter breaks, day after day, between runs increases fatigue related risk and the potential for loss of focus," Edler a member of the Brotherhood of Locomotive Engineers and Trainmen, wrote in an open letter to Amtrak brass. 

According to other engineers, Bostian's "couplet" on May 12...was one of the routes affected...The new route had his Acela 2121 arriving at Union Station at 4:53 pm and, after a change of trains, pushing off at 7:10 PM onboard 188's Cities Sprinter.  If all went well, that would leave him roughly two hours in Washington. "

Less dramatically, the devil's in the details.  Was indeed the 2121/188 schedule changed?  Was the time between trains reduced?  If so, by how much?  Look, the author claims prior to Amtrak's unilateral change to the crew assignments, the average time between runs was "two and a half hours or so."  What do we have as the scheduled time in May for 2121/188 turnaround?  Two hours and 17 minutes.   Is anyone seriously claiming that a scheduled reduction of 15 minutes or so, if there was even that,  increases the risk of a fatique-based accident?

Dramatically, the article reports:  "But just minutes after the Acela 2121 left New York, Bostian encountered a problem with his 'cab signals'...Bostian was forced to reduce the top speed on the 2121, his eyes fixed on every wayside signal for indications of trouble ahead.  It was an onerous task.  He arrived at Union Station half an hour behind schedule." 

Less dramatically-- See, I told you train control systems fail.  And this is precisely what the locomotive engineer is hired, trained, certified, checked, to do.  This is the task. 

This is precisely what Bostian found so fascinating about working as a locomotive engineer.  That precise condition, where and when the locomotive engineer must employ his/her skills is what it's all about.  If the trains could be run by drones, believe me, I'd be the first one to advocate installing joysticks in Operations Control Centers and then installing robots to operate the joysticks. 

Less dramatically, we have a situation where all the noise about reduced turnaround time proves to be just that--noise, irrelevant to the circumstances of the incident as we do not, because we cannot, write crew assignments based on the anticipation of equipment failure.  

Dramatically, the reporter notes that two trains had been struck by objects (presumably thrown) in the vicinity of Frankford Junction, and notes: "According to an assistant conductor onboard Amtrak 188, Bostian heard the hot track warning and replied that he had been hit by something as well." 

Less dramatically, this tale told by the assistant conductor, of Bostian reporting being struck by an object, was shown to be without supporting evidence, as recordings of radio transmissions showed no such communication by 188's engineer.  It's been 8 months since that tale was refuted, why bring it up again?

We need to keep this in mind a bit later when the reporter quotes an accident investigator who is convinced he knows why the accident occurred.  

Drama upon drama, the article continues:  "In car 3, the derailment was registered in a variety of ways.  Josh Gotbaum...heard a screeching sound as the traction motors lost their grip...." 

Less dramatically, I don't think that's what he heard, since the cars themselves don't have traction motors, the locomotive provides the tractive effort, and car 3 was at least 160 feet to the rear of the nearest set of traction motors.  

Now, dramatically, back to the stoning:  "There are two main schools of thought on what may have caused Bostian to lose his bearings.  The first takes into account the rockings in the vicinity and Bostian's own reported account of this train being struck by a large object, his forehead wounds and a small pocked dent on the left sde of the windshielf-- a dent of the kind typically produced by a rock.  'To me, it's pretty clear what happened,' Richard Beall, the longtime accident investigator, told me. 'Bostian's got the throttle open to get the train up to speed. A projectile hits the windshield.  Now the windshields on these locomotives are thick, but that impact is going to be out of nowhere and scary.  As a human, you've got a tendency to duck. But he ducks into the dashboard and smacks his head, knocks himself out.  And by the time he's back up, and he's reoriented himself, it's 'Oh, crap.' "

Less dramatically, if as the reporter reported, Bostian radioed that he too had been struck by an object, then that transmission has to be after the object hit the windshield, and consequently, Bostian did not "duck into the dashboard" and knock himself unconscious. He was awake,  cognizant of his surroundings and his responsibilities to report the stoning as required by the rules. (Note--We don't call them them "dashboards." We call them control stands).

Secondly take a look at the control stand on an ACS-64.  To hit your head on any part of that control panel from the engineer's seat, you would have to bend from the waist, forward and down.  I mean seriously down, which is consistent with a rapid deceleration.  But Bostian was accelerating, and continuing to accelerate.  I don't know how flexible Bostian is, but unless he's an Olympic level gymnast,  circus performer,  advanced yoga instructor, or Plastic Man, he's not going to be able to reach the flat surface of the control panel with his forehead. 

Event recorder data from the locomotive will tell us how long this period of "unintended acceleration," in reality, overspeeding, lasted.   We will also be able to determined if the locomotive engineer took any action from the time of initial overspeeding to the time of emergency braking to satisfy the on-board alertness warning system and prevent that system from automatically applying the brakes.

Finally, we get the coda of this tragedy, the moral of this fable, and a fable has truly been spun.  Quoting FRA Administrator Sarah Feinberg, the reporter writes:  'My hope,' Feinberg told me recently, "is that the derailment was our long overdue wake-up call that we need PTC-- that we owe it to passengers and rail staff to have it online."

PTC was mandated for installation on all passenger lines, and Class 1 railroad line segments exceeding a specific annual gross-ton mileage and handling hazardous commodities 7 years ago by an act of Congress after an accident that took more lives than this. 

If Congress doesn't want to enforce the law, it's Congress' privilege to withdraw the law, amend the law, change the requirements of the law, but in all cases, wake-up calls have nothing to do with it.

David Schanoes

January 27, 2016

"Shakes? Me too. It comes with the business."

"I''m not in the business, Mr. Deckard.  I am the business."

-- Blade Runner