Still more questions than answers in the wreck of Amtrak 188

The NTSB released its preliminary findings into the Amtrak 188 derailment on February 1st, and to be perfectly blunt, we still know about as much as we did the day after the crash.  We know the “how” of the derailment — that the train went into the curve at Frankford Junction at 106 mph instead of the 50 mph speed limit — and we knew that on May 13th, but we are still no closer to knowing the “why”.  The NTSB report rules out virtually every form of mechanical failure, so focus has descended on Amtrak 188’s engineer, Brandon Bostian.

However, the NTSB’s initial safety recommendation, asking for inward-facing cameras in locomotive cabs, essentially admits that there is no way of ever knowing exactly what happened in the cab of Amtrak locomotive #601.  This is distressing, since there are two major theories for what happened, with diametrically opposing conclusions as to fault.

The first theory is that Bostian zoned out or lost situational awareness, and deliberately accelerated, thinking that he was somewhere else.  In this event, engineer error is the major proximate cause of the crash, and Bostian is primarily responsible.

The second theory is that, possibly in response to his locomotive getting hit with a rock (as frequently happens in North Philadelphia, and had just happened to a 30th Street-bound SEPTA train nearby), Bostian flinched in such a way as to render himself dazed or unconscious.  (There is no way to medically determine if the concussion Bostian sustained happened either just before, or during, the crash.)  In that case, Bostian is innocent, and the primary responsibility rests on Amtrak for the failure of its safety precautions (the combination of the lack of either automatic train control (ATC) or positive train control (PTC) on the northbound tracks of the NEC at Frankford Junction, and the failure of the locomotive’s deadman/alerter system to recognize Bostian’s incapacity in time to prevent disaster).

However, with Bostian’s memory, rendered unreliable by that concussion, as the only record of what happened, we have no way of knowing which of these two cases actually applies.  That is no fault of Mr. Bostian’s; similar uncertainty has reigned in other cases, where the engineer did not survive to be interviewed by the NTSB.  With personal-injury lawyers circling, the choice before the civil courts is a guessing game, where if they get it right, it will be a tragedy, and if they get it wrong, it will be a dreadful miscarriage of justice.  As it stands, because this question will never be resolved to anyone’s satisfaction, it is almost certain that Bostian will never operate a train again.

Perhaps a more interesting question, and one that is conspicuously absent from the NTSB report, is whether the FRA regulations concerning the survivability of passenger cars played any role in saving lives.  American railroad equipment is infamously overweight, loaded up with steel armor to meet a buff strength requirement, so that it does not deform in a low-speed crash.  Despite this, business class car 81528 on Amtrak 188 was smashed to flinders, to the point where first responders didn’t even realize it had once been an 85 foot long steel tube with people inside.  Moreover, passengers still died in the coach class cars behind the business class car, despite those cars’ much less damaged condition.  The rest of the world uses a crash-energy management (CEM) standard for passenger car safety (which encourages safety features similar to the crumple zones in modern automobiles), which makes cars signginicantly lighter weight, which in turn improves both everyday performance and braking in emergencies.  It isn’t clear that CEM would have made for a less deadly crash at 106 mph, but the unwillingness of the NTSB’s Survival Factors Group to even entertain the question is disturbing, especially when the FRA’s process to adopt CEM as a standard in the US is over a year late and possibly in limbo.  Even a finding that the two standards would have had equivalent results would be a very big deal, since the buff strength standard adds enormous capital and operating costs to American railroads.

One change that looks like it’s for the better, but is not, is the installation of an ATC restriction to 45 mph on the northbound NEC at Frankford Junction.  As a short-term stopgap, this is fine, but should be refined or removed once PTC is fully deployed, since the ATC system has no way of displaying or enforcing the true 50 mph speed limit on those tracks.  The redundancy does not actually help, and the delay from the speed restriction is already significant.

Lastly, also left unaddressed is what exactly the NEC is still doing with a 50 mph curve at Frankford Junction.  Straightening the curve is bound to be expensive, but addressing this anachronism would be far more effective, and for less money, than many of the proposals on the table to improve high-speed rail on the NEC.