Train safety technology hasn’t quite reached the station

Fatalities add up as cost and complexity delay full implementation of 'positive train control' system

Rescue crews and investigators inspect the site of a May 2015 Amtrak train derailment in Philadelphia . (Photo by Samuel Corum/Anadolu Agency/Getty Images)
Rescue crews and investigators inspect the site of a May 2015 Amtrak train derailment in Philadelphia . (Photo by Samuel Corum/Anadolu Agency/Getty Images)
Posted December 5, 2019 at 5:52am

After years of delays, a railroad safety system that federal regulators say could have prevented some 300 deaths since 1969 is finally close to full implementation — but large gaps remain, with commuter railroads using the system on fewer than half of the tracks required by December 2020.

Overall, the news for supporters of the so-called positive train control system is promising — 92 percent of the 58,000 track miles required to implement the safety system have it installed, according the Federal Railroad Administration, which is overseeing compliance with the law. 

But while Amtrak is using positive train control on 99.8 percent of its track miles and Class 1 freight railroads — which account for 53,718 of the track miles covered under the statute — are operating the system on 95.4 percent of their  tracks, commuter railroads and miscellaneous railroads have a way to go.

As of September, just 41.9 percent of the 3,129 track miles used by commuter railroads and 22.8 percent of the 108 track miles used by other host railroads under the statute had adopted the technology.

Decades before automated vehicles and drones captured the imagination of the traveling public and transportation companies, positive train control — a system that uses GPS, cellular technology, and radio waves to create what is essentially a backstop for train engineers — aimed to eliminate human error from the railway system.

[Lawmakers aim to thwart Amtrak forced-arbitration policy]

The system can, for example, slow a train exceeding the speed limit, prevent two trains from colliding or keep trains from going down the wrong track if a switch is in the wrong position.

While it does not eliminate all railway accidents the positive train control system “certainly eliminates much of the human error-type issues that have been a problem in the past,” said David J. Carol, chief operating officer for the American Public Transportation Association.

Since 2016 the Federal Railroad Administration’s Office of Safety Analysis reported 2,610 railroad accidents in which human factors were a cause. In all, there were 6,849 accidents on railroads during that period.

Transit challenges

Carol, whose organization represents many of the commuter railroads still implementing the systems, said that those lines have often had to wait to implement positive train control until after the freight and short-line railroads that they travel along have implemented the technology.

Others see another headwind.

Allan Zarembski, director of the University of Delaware’s Railroad Engineering and Safety Program, said the slower implementation has more to do with cost: While freight railroads are generally lucrative businesses, commuter railroads “are the ones that are broke. They have no money” to implement the systems.

The federal government in recent years has tried to help, administering more than $2.5 billion to rail operators — not just commuter lines — for positive train control through grants and loans.

Still, Carol said he’s confident the commuter railroads will meet the deadline.

“I’m pleased with their progress,” he said, saying the vast majority of commuter railroads were testing the technology. “They’re basically finding the funds to do this.”

The technology, he said, “is not replacing the locomotive engineer, but it’s making sure the train itself can address safety issues if and when it needs to address them.”

The National Transportation Safety Board began calling for the system as a top priority in 1990. But Congress didn’t mandate its implementation until it passed the 2008 Rail Safety Improvement Act (PL 110-432) in the aftermath of a deadly collision between a commuter and freight train in Los Angeles that killed 25 people.

Deadlines delayed

Congress originally set a Dec. 31, 2015 deadline to implement the technology. But an initially resistant railroad industry, which was largely forced to pay for the costly system, and the technical complexities of installing an interoperable and highly sophisticated new technology, slowed the implementation.

The deadline was first pushed to 2018, then to 2020.

Meanwhile, the fatalities continued: Three people died in a December 2017 crash south of Tacoma, Wash.; a May 2015 crash in Philadelphia killed eight; a February 2018 crash in South Carolina killed two.

In all three cases, the NTSB said positive train control could have helped prevent the crashes.

Even after years of delay, Russell Quimby, a former NTSB investigator, said he doubts all 58,000 miles will be compliant by the end of next year. The problem, he said, isn’t just installing the technology. It’s getting people trained as well making the technology interoperable throughout a spider web of multiple railroads and companies.

Some railroads owned by one company often play host to multiple railway companies. Amtrak, for example, plays host to commuter railways stretching from Washington, D.C. up the northeast corridor.

And Zarembski of the University of Delaware said the technology, though helpful, won’t prevent all accidents: Railways will still have to deal with broken wheels or damaged tracks, for example.

“Positive train control will have a real impact on safety, but it’s not a panacea,” he said.

 Still, he added, “it will prevent a lot of these very high-profile collisions or over-speed accidents.”

While the railroad industry initially resisted the idea, Quimby said it has come around as other modes of transportation — including trucking, a chief competitor — embrace automation. And while a change isn’t imminent, he said he can envision a world with driverless freight rail in the long term.

The industry has fought with organized labor for years over whether to have one or two-person crews in trains. But with the trucking industry exploring driverless technology, he said, it’s possible that a driverless freight train will be the next frontier for the industry — and the new technology could ease the technological aspects of that transition.

“Ironically, this system could allow railroads to remain extremely competitive,” he said.

““

Loading the player...

””