← back to the reports
DR-007 Passenger express · Amtrak, Northeast Corridor 2015

Amtrak 188 — A Throttle Opened Into the Wrong Curve, Eight Dead

overspeedpassenger2010s
Killed
8
Railway
Amtrak (Northeast Corridor)
Service
Northeast Regional 188
Status
Driver

Summary

On the night of 12 May 2015, Amtrak Northeast Regional train 188 entered the Frankford Junction curve in the Port Richmond section of Philadelphia at 106 mph, more than twice the curve's 50 mph speed restriction, and derailed. Eight passengers were killed and more than 185 people were taken to hospitals, several critically injured, from a train carrying about 250 occupants. It was the deadliest accident on the Northeast Corridor in decades and the event that forced the long-delayed completion of automatic speed enforcement on America's busiest passenger railroad.

The train was a scheduled evening service from Washington, D.C., to New York City: one ACS-64 electric locomotive and seven Amfleet passenger cars, with a single locomotive engineer at the controls. There was no mechanical defect, no track failure, no obstruction. In the seconds before the derailment the engineer advanced the throttle and accelerated hard, as though the train had already cleared the sequence of speed-restricted curves that follow the departure from 30th Street Station. It had not. The locomotive and all seven cars left the rails; three cars overturned, and one was crushed and torn open.

The National Transportation Safety Board investigated and adopted its findings on 17 May 2016 in railroad accident report RAR-16/02. The Board's probable cause was operational and human: the engineer accelerated to 106 mph entering a 50 mph curve because he had lost situational awareness, his attention most likely diverted by radio chatter about a nearby SEPTA commuter train that had made an emergency stop after being struck by a projectile. The Board found that the engineer believed he was at a point on the line where higher speed was authorized. It identified the absence of a positive train control (PTC) system on that stretch of track as a contributing factor — a safeguard that, the Board concluded, would have prevented the accident had it been active.

The legal arc ran for nearly seven years. The engineer, Brandon Bostian, faced charges that included involuntary manslaughter and reckless endangerment; after a series of dismissals, reinstatements, and appeals, a jury acquitted him on all counts on 4 March 2022. Amtrak, which accepted responsibility for the derailment, reached a $265 million settlement of victims' claims in 2016.

Timeline

12 May 2015, evening
Departure
Northeast Regional 188 leaves Washington, D.C., for New York, an ACS-64 locomotive hauling seven Amfleet cars, with about 250 people aboard.
~21:00
Radio traffic builds
As 188 works north out of Philadelphia, the radio carries reports of a SEPTA commuter train that has made an emergency stop after a projectile struck it; a dispatcher and crews discuss the incident.
Final ~1.5 miles
Premature acceleration
Approaching Frankford Junction — the last of the speed-restricted curves leaving 30th Street Station — the engineer opens the throttle and accelerates, consistent with a belief that the curves are already behind the train.
12 May, ~21:23
Into the curve at 106 mph
Train 188 enters the four-degree left-hand curve, restricted to 50 mph, at 106 mph.
Seconds later
Emergency brake, then derailment
The engineer applies the emergency brake; the train derails at about 102 mph. The locomotive and all seven cars leave the rails; three cars overturn.
12–13 May
Casualties confirmed
Eight passengers are killed; more than 185 people are transported to area hospitals, with multiple critical injuries.
13 May 2015
PTC pressure renewed
Within a day, the absence of automatic speed enforcement on the northbound track becomes the central public question; Amtrak moves to complete PTC on the corridor.
17 May 2016
NTSB adopts the cause
The Board adopts RAR-16/02: the engineer's acceleration into the curve from a loss of situational awareness was the probable cause; the lack of PTC was a contributing factor.
October 2016
Civil settlement
Amtrak, having accepted responsibility, settles passenger and family claims for $265 million.
December 2016
PTC operational
Positive train control is reported active across the Amtrak-owned Northeast Corridor.
4 March 2022
Acquittal
A jury acquits engineer Brandon Bostian on all counts, including involuntary manslaughter, ending the criminal case.

The Corridor and the Curve

The Northeast Corridor is the spine of American intercity passenger rail — a high-density, electrified, multi-track main line linking Washington, Philadelphia, New York, and Boston. North of Philadelphia's 30th Street Station, the route threads a series of curves before it can open up to higher track speeds. Frankford Junction is the sharpest of them: a four-degree left-hand curve where the maximum authorized speed is 50 mph, even though trains are permitted far higher speeds on the tangent track immediately beyond it. The geometry demands that a train slow for the curve and then accelerate once through it. The transition is unforgiving, because the difference between the curve's safe speed and the line speed beyond it is large.

In May 2015 that transition was not protected by any system that could physically intervene if a train approached too fast. Positive train control — a technology that uses location data, track sensors, and onboard computers to enforce speed limits and stop a train that an operator has failed to control — was mandated by Congress in 2008 but had not yet been activated on the northbound track at Frankford Junction. The southbound track through the same area had already been equipped with automatic civil-speed enforcement. The northbound side, the side train 188 was on, had not. The protection existed in concept and in adjacent infrastructure, but not where it was needed that night.

Train 188 itself was modern and sound. The ACS-64 locomotive recorded its own speed, throttle, and brake data, and that recorder, recovered intact, established the central facts beyond dispute: the train was doing 106 mph entering a curve posted at 50, the engineer opened the throttle in the approach, and the emergency brake came only in the final seconds, far too late to bleed off the excess speed before the rails curved away beneath the train.

Three Seconds From a Mistaken Position

The Board reconstructed the engineer's actions and the environment he was working in. In the minutes before the derailment, the radio carried an unfolding emergency involving a SEPTA regional rail train that had been struck by a thrown object — a rock or similar projectile that hit the train, prompting an emergency stop. The traffic on the channel concerned that incident: the SEPTA crew, the dispatcher, the response. The NTSB concluded that this chatter diverted the engineer's attention at the critical moment, and that the diversion caused him to lose track of where on the corridor his own train was.

The consequence was a position error. The engineer accelerated as if train 188 had already passed the last of the speed-restricted curves and reached the stretch where higher speed was authorized. The data showed the throttle opening roughly a mile and a half out, accelerating the train through the high-90s and past 100 mph, into the very curve he believed was behind him. The train entered the 50 mph curve at 106 mph. He applied the emergency brake, but only seconds remained; the train was still doing about 102 mph when it derailed. There was no time for the brakes to matter.

The Board was careful about what it did and did not find. There was no evidence of impairment by alcohol or drugs, no evidence of a medical event, no evidence that fatigue was the operative factor, and no evidence the engineer was using a personal electronic device at the time — his cell phone had not been in use during the relevant minutes. The mechanism the Board landed on was attentional: a competent, qualified engineer whose situational awareness collapsed because his attention was pulled to a nearby emergency, leaving him to act on a mistaken mental model of where he was. That is an operational human-factors finding — a driver error in the sequence of train control — not a defect in the locomotive or the track.

The Board's Verdict and the Failsafe That Was Absent

The NTSB adopted RAR-16/02 on 17 May 2016. Its probable cause stated that the derailment resulted from the engineer's acceleration to 106 mph as he entered the 50 mph Frankford Junction curve, due to his loss of situational awareness, likely because his attention was diverted to the emergency involving the nearby SEPTA train. The Board found that the lack of a positive train control system was a contributing factor, and its investigators were direct that PTC, had it been operating on that track, would have automatically enforced the curve's speed limit and prevented the accident.

That second finding reframed the disaster. A loss of situational awareness by a single operator is, in human-factors terms, a foreseeable failure mode — people lose track of position, especially when an emergency competes for their attention. The point of an engineered failsafe is precisely to catch the human error that will eventually occur. The corridor had the technology specified, mandated, and partly installed; it simply had not been switched on at Frankford Junction northbound. The Board's verdict therefore carried two layers: the proximate operational error that caused the train to be where it should not have been at the speed it should not have held, and the systemic gap that left that error uncaught.

Among the Board's recommendations were measures to accelerate PTC implementation and to address the human-factors vulnerabilities that the accident exposed, including the management of distraction and the resilience of operators' situational awareness on high-density passenger lines. The recommendations reinforced what the corridor's owners were already being compelled to do.

The Five Factors

01
Situational awareness as a single point of failure
The entire chain reduced to one operator's mistaken belief about where his train was. When safe operation depends on a person continuously holding an accurate mental model of position and speed, any disruption to that model — fatigue, distraction, a competing emergency — can produce a catastrophic action. Awareness is not a reliable safeguard on its own.
02
Distraction from a legitimate emergency
The chatter that diverted the engineer was not idle; it concerned a real incident on a nearby train. Operationally relevant communication can be as distracting as irrelevant noise. Systems must assume that operators will, at times, have their attention pulled toward genuine emergencies and must not depend on undivided attention to stay safe.
03
Steep speed transitions demand protection
Frankford Junction paired a sharp 50 mph curve with much higher authorized speed just beyond it. Where the gap between a restriction and the surrounding line speed is large, the margin for a position error is small and the penalty severe. Such transitions are exactly where automatic enforcement earns its place.
04
Mandated safeguards delayed are safeguards absent
Positive train control had been required since 2008 and was installed on the adjacent southbound track, yet it was not active where train 188 ran. A safeguard that exists on paper, or one track over, does not protect the train that needs it. The interval between mandate and activation is itself a hazard.
05
Modern recorders make the verdict unarguable
The locomotive's event recorder fixed the speed, throttle, and braking with precision, removing speculation about what the train did. Robust, recoverable data turns an investigation from a contest of theories into a reconstruction of fact — and makes the operational nature of the cause impossible to evade.

Aftermath

The derailment broke the political stalemate over positive train control on the Northeast Corridor. Amtrak completed and activated PTC across the stretch it owns, with the system reported operational by the end of 2016; Frankford Junction, the curve that had no automatic speed enforcement on 12 May 2015, was brought under enforcement. The accident became the standard example cited for why the congressional PTC mandate could not be allowed to slip further, and it sharpened the national conversation about distraction and human-factors resilience in train operation.

The civil resolution came first. Amtrak accepted responsibility for the derailment and, in 2016, settled the claims of the injured and the families of the eight dead for $265 million. The criminal track was far longer and more contested. The engineer, Brandon Bostian, was charged with offenses including eight counts of involuntary manslaughter, numerous counts of reckless endangerment, and causing a catastrophe. The case moved through dismissal, reinstatement, and appellate review before reaching trial, and on 4 March 2022 a jury acquitted him on all counts. The legal outcome turned on whether his conduct rose to criminal recklessness; the NTSB's safety finding — that a loss of situational awareness, uncaught by an absent failsafe, caused the deaths — stood independent of the verdict.

Lessons

  1. Do not let safe operation rest on an operator's unbroken situational awareness; engineer in an automatic enforcement layer that physically prevents overspeed where the geometry is unforgiving.
  2. Treat operationally relevant radio traffic and nearby emergencies as predictable sources of distraction, and design control systems that remain safe when an operator's attention is, inevitably, pulled away.
  3. Activate mandated safeguards on the track that needs them, not just the adjacent one; a partially deployed failsafe leaves the gap that the next accident finds.
  4. Place automatic speed enforcement at the sharp transitions — where a low-speed curve abuts a high-speed line — because that is where a position error becomes a derailment.
  5. Equip vehicles with robust event recorders and recover them; precise data on speed, throttle, and braking is what separates a defensible finding from conjecture.

References