Published:
October 25th 2009
Source: Joan Trossman Bien (miller-mccune.com)
Printer
friendly versionPositive Train Control —
technology that adheres to the simple premise that there should only
be one train to one track — is still years away from full
implementation in the U.S.
When passengers hear the cry of
"All aboard!" they rarely give any thought to whether they will
arrive safely at their destination. There have been many advances in
railway safety, and, when compared with other means of
transportation, the railroad safety record is stellar.
On Sept. 12, 2008, the 222 passengers who were riding Metrolink No.
111 from downtown Los Angeles to the Ventura County Moorpark train
station probably felt that way. After all, the idea that those who
were responsible for running the train could abandon all caution and
fail to follow basic operating procedures, risking the lives of all
on board, was unthinkable.
But engineer Robert Sanchez, who alone controlled the movement of
No. 111, was completely absorbed in texting friends. He ran two red
lights and failed to yield to the oncoming freight train using the
same single track; his train smashed head-on into a Union Pacific
freight train just outside the suburban Chatsworth station. He never
even applied the brakes. The force of the collision shoved the
locomotive straight into the first passenger car, derailing the rest
of the train. Twenty-five people died, including Sanchez, and 135
people were injured in the mangled wreck. None of the crew aboard
the Union Pacific train died.
Human error has been and remains a major, yet entirely avoidable,
cause of some of the worst train wrecks in U.S. history.
Even back in the mid-19th century, two days after the Great Train
Wreck of 1856, The New York Times ran an editorial that blamed the
railroads and said that trains going in opposite directions should
never share a single track.
Now, one year after the Metrolink catastrophe — and 152 years after
the New York tragedy — laws have been passed, money has been
allocated, and a few small changes have been made in operating
procedures. But the one system that would have prevented this
catastrophe, "positive train control," known as PTC, will not be
fully implemented before 2012. That is three years earlier than the
Rail Safety Improvement Act of 2008, which mandates PTC on all
passenger rails in the United States.
Is it really that simple? Policymakers say so.
In testimony at a California State Senate hearing on rail safety
about a month after the crash, U.S. Sen. Dianne Feinstein said, "A
technology called positive train control is in place on other rail
systems to prevent human error from causing fatal disaster, but
trains in California currently don't have it. If positive train
control had been in place on Metrolink on September 12th, I believe
25 people would still be alive today."
A year later, in announcing the appropriation of funds for a new
railroad safety program, her colleague Sen. Barbara Boxer was quoted
in a release saying, "Facts revealed by the National Transportation
Safety Board investigation indicated that PTC could have prevented
the tragic crash of a Metrolink commuter train and a freight train
in Chatsworth, California last year."
In Development For Nearly A Century
The federal government in 1922 originally mandated some form of
automatic train stopping, yet, to this day, many passenger and
commuter lines are without any meaningful and effective form of
protection from simple human error. Trains are legally permitted to
remain without positive train control until Dec. 31, 2015, the
deadline mandated in the Rail Safety Improvement Act of 2008.
According to the Federal Railroad Administration, PTC systems are
"integrated command, control, communications and information systems
for controlling train movements with safety, security, precision,
and efficiency. PTC systems will improve railroad safety by
significantly reducing the probability of collisions between
trains," danger to railway workers and derailments caused by
excessive speed.
By utilizing the technology of global positioning satellites, PTC
would automatically override dangerous train movements, remain
continuously updated on train locations and stop a train if the crew
was incapacitated. For the railroads, the agency writes, the
benefits are "improved running time ... higher asset utilization,
and greater track capacity."
And there are working systems out there. "Pilot versions of PTC were
successfully tested a decade ago, but the systems were never
deployed on a wide scale. Deployment of PTC on railroads is expected
to begin in earnest later this decade."
All
of the train disasters cited in the sidebar, at right, could have
been prevented by some system of automatic train control that would
override human error. Given PTC's benefits and track record, why is
it taking so long for such a system to be installed and operating on
all railroad tracks in the United States?
Wait Until 2015
There have been many different automatic train-stopping systems in
existence going back to the early part of the 20th century. Europe
has had a form of automatic train control operating in some
countries, including Great Britain, Germany and France, since the
1930s.
Under the Transportation Act of 1920, the Interstate Commerce
Commission required 49 railroads to have an operating train-stop or
train-control system on at least a portion of their passenger train
routes. What happened to keep that from coming to fruition?
In the 1980s, the Association of American Railroads developed
detailed plans for an advanced train-control system using digital
radio communication and microprocessor controls. And in 1987, the
Burlington Northern Railroad in tandem with Rockwell International
developed a similar system using GPS called Advanced Railroad
Electronics System.
In 1991, teams from Harvard Business School twice studied the $350
million Burlington Northern system. The outsiders were enthusiastic
— Harvard described their "zealous advocacy of the project" — that
the system would save the freight hauler lots of money. But company
executives at the then-cash-strapped company were leery about the
actual savings and feared their traditional system couldn't adapt to
PTC.
Subsequent studies have back up the business case for PTC. The
consultancy Zeta-Tech Associates studied PTC for the railroad
administration in 2004 and reported its results "suggest that the
railroad industry should carefully consider the opportunity
presented by PTC technology, especially in view of its ongoing
shortage of line capacity and the need to increase the return on
invested capital." Both systems, it said, provide "significant
business benefits to the freight railroads, as well as unquestioned
safety benefits through positive enforcement of movement
authorities."
After the 2008 Chatsworth crash, the magazine Design News published
a series of articles on PTC titled, "Railroad Safety at What Price?"
Then-editor John Dodge recalls being stunned by Burlington's actions
years before.
"There was a proof of concept [for their PTC system] and
[Burlington] tried it out, and it worked. It was a $350 million
rollout. You can talk to all of the people in the railroad industry,
and they will rationalize it, but for more than 20 years there has
been a viable safety system that BNSF has been sitting on."
"Those people who lost their lives in L.A. didn't have to lose their
lives," Dodge said. "That was my view. Railroads are extremely
risk-averse and very regimented. Everything is about moving freight
at a lower cost, and that is how it has been for more than 150
years."
And yet, Metrolink still has no automated system of any sort on its
trains, leaving primary responsibility for the safety of hundreds of
passengers each trip to the engineer. In the Chatsworth case,
engineer Sanchez, was known by supervisors to text or call friends
while running the train alone but no action was taken to stop the
prohibited behavior.
Feinstein told the California Senate Hearing on Rail Safety, "Not a
single mile of California track has modern collision avoidance
Positive Train Control systems — though these systems are in place
on more than 3,100 miles of American track. ... In the past 10
years, the National Transportation Safety Board has investigated 52
rail accidents where the installation of a positive train control
system would likely have prevented the accident."
When Will Passengers Be Safer?
PTC has been on the National Transportation Safety Board's "Most
Wanted List for Transportation Safety Improvements" since 1990. The
NTSB has emphasized that such systems are especially needed where
passenger trains and freight trains share a single track.
And the concept isn't totally foreign to U.S. tracks: Amtrak has a
system fully implemented in the Northeast Corridor between
Washington and Boston, while Alaska and New Jersey are actively
developing their own versions.
In the case of the Metrolink, small safety improvements have been
initiated, according to the president of the Metrolink
Transportation Board, Keith Millhouse. However, most of the new
safety measures have yet to be fully implemented. They include:
placing a "Second Set of Eyes," or a second trained engineer in the
cab, although only a small percentage of the trains actually have
this in action; inward-facing video cameras, which are facing
serious resistance from the railroad workers' unions; ordering new
passenger cars and cabs that would decrease injuries during a crash,
the delivery of which has been delayed; and changing the operating
company from Connex to Amtrak, which will not happen until the end
of the current contract next year.
Today, Metrolink officials were to announce that video cameras have
been installed on all locomotives in the Metrolink system at a cost
of $1 million. That included two inward-facing video cameras, the
presence of which is intended to prevent engineers from using cell
phones, texting or having unauthorized personnel inside the cab
while operating the train.
However, these cameras are not being monitored in real time, but the
video will be downloaded each day for random review. Therefore, the
cameras cannot directly prevent a crash similar to the 2008, train
wreck. Only the possibility of an engineer fearing that he will be
identified after the fact for breaking Metrolink rules will
translate to possible accident prevention.
One more system that is about to be activated is called "automatic
train stop."
"When the train passes a device, a signal is sounded in the cab that
requires the operator to acknowledge the signal that is generated,"
Millhouse explained. "If the operator fails to acknowledge the
signal then the train will be slowed down. We're just at the point
where we're ready to turn on the system."
But automatic train stop is not a new system. "It is somewhat
outdated technology," Millhouse said, and it has limitations.
Between money, technology and geography, Millhouse said Metrolink
can't install and implement PTC by itself.
"We have the largest and most densely congested operating system in
the country. Within the Metrolink operating system we have Metrolink,
Amtrak, Union Pacific, Burlington Northern Santa Fe and the Coaster
system in San Diego, so any equipment that is west of Chicago has
the potential to enter this area. We have been proceeding at
Metrolink to equip all of our locomotives and trains by 2012."
Union Pacific has declared that it will implement PTC by 2012, three
years before the federal mandate.
Additionally, the major railroads, Union Pacific, Suffolk Northern
and Burlington Northern Santa Fe, reported they have reached an
agreement on establishing interoperability standards for PTC.
Despite that agreement, on Aug. 25, Burlington Chief Executive
Officer Matt Rose told Bloomberg News that it would be extremely
difficult for his company to meet the 2015 deadline. Calling the
Railroad Safety Act of 2008 "heavy-handed," he said, "This is just
one of those examples of regulation gone awry where there will be
unintended consequences."
He asked that lawmakers allow the carriers to implement PTC only on
its busiest lines.
"Implementing PTC on the specified routes by 2015 will be a
logistical, technical and financial challenge," a Burlington
spokesperson told Miller-McCune. "The Rail Safety Transportation Act
of 2008 is an unfunded mandate. However, BNSF has always said it
will make all reasonable efforts to comply with the law."
Rose said PTC will cost the railroads $10 billion and that
Burlington would prefer to install the new technology on its own
timetable. In its 2004 report, Zeta-Tech wrote, "PTC is a large
investment by any measure. A cost of $1.3 billion to $4.4 billion
might seem daunting to an industry with gross revenues of only $35
billion. However, the projected annual savings of $2 billion to $3.6
billion provides a rapid payback period."
Epilogue
On Sept. 10, Metrolink ran a test to see how much safety improvement
has been made on its tracks in the year since the Chatsworth
tragedy. Operators intentionally turned what would normally be a
green light during rush hour to a red light. The train blew right
through it and the engineer, who is required to call out the signals
to the conductor, apparently called out the red light as being a
green light.
So whether the railroads will actually move forward in the future
with an effective and technologically advanced PTC system or whether
human error will continue to have the final word on passenger
safety, will be determined only by the legal deadline of 2015.