It’s A Train Wreck?

The cable news networks are off on their usual “Who’s To Blame?” penchant in reporting tragic news, and in the midst of the palaver over the Pennsylvania Amtrack wreck there’s the usual lack of context.

For example, has the media told us there were 220 derailments in 2012, 191 in 2013, and 228 in 2014? [FRA]  Or, when looking at accident or incident causation we have the following information at hand? [FRA]

Cause	2012	2013	2014	2015
Track	112	91	87	101
Human	98	96	132	112
Equipment	39	33	47	44
Signal	12	9	8	6

What is interesting about this abbreviated table is the the track and human failure seems to predominate, while equipment and signal failures account for substantially less than the  first two factors.  (Miscellaneous factors are not included in this summation.)

There’s another way to observe train accidents, by state, and we find the following among the states with the highest number of accidents:

State	2012	2013	2014	2015
California	17	20	13	18
Illinois	24	33	50	38
Minnesota	10	9	4	3
North Dakota	5	4	6	4
Nebraska	13	7	8	11
Ohio	14	10	15	14
Texas	35	22	31	37

*This summary doesn’t include crossing accidents.

The Federal Railroad Administration also keeps records of fatalities and injuries, and this is what garners most attention from the news media. Again, we can look at the record over the last four years, noting that those numbers for 2015 are preliminary.

Fatalities	2012	2013	2014	2015
California	19	12	18	25
Florida	4	8	7	4
Georgia	6	4	5	3
Louisiana	3	2	1	6
New York	4	7	1	12
N. Carolina	4	8	5	5
Ohio	7	3	5	4
Pennsylvania	3	4	3	3
Massachusetts	3	1	0	0
Texas	11	7	4	18

*PA will add 8, those fatalities in the Amtrak accident near Philadelphia, May, 2015.

What we might expect to find are a higher number of fatalities in the Northeast Corridor, where ridership includes commuting,  but the reports indicate higher numbers generally in California and Texas.  If the accident numbers (in terms of passenger travel) appear to be increasing, so is the ridership, as shown from BTS (interactive) figures from 2000 to 2015.  The light blue line doesn’t include seasonal adjustment.

The general trend shown by the unadjusted and adjusted numbers indicates more people using rail transportation since 2000. In calendar year 2006 the FRA reported 602,280,892 passengers transported; the number of passengers transported as of the end of calendar year 2014 was 694,507,965.

 And, where is the ridership?  Where we would expect it.  Region 1 reported 369,467,363 transported, Region 2 reported another  83,444,579.  Region 7 reported 99,512,195; Region 6 reported 56,260,844.  Region 4 reported 41,748,653. Region 5 reported 32,943,327.  Region 3 reported 5,938, 814; and, Region 8 reported the lowest passenger traffic at 5,149,686.

What Use Can We Make Of The Numbers?

First, if we take a look at the ridership numbers (which don’t include local transit services) and the reported fatalities,  it’s reasonably clear that passenger travel is remarkably safe. Certainly safer than travel by private automobile.

Secondly, we can question the popular opinion offered to explain the political inaction on public transportation funding – like for Amtrack – that low funding is because most of the country isn’t using passenger rail service.  Granted that Regions One and Two (Northeast Corridor) are reporting the highest usage, but quite obviously Region Seven (California) and Region Four (the upper Mid West) are contributing significant ridership numbers.  The assertion that low ridership may equate to low support could only true for the Southeast and the Northwest.

Third, if we look at accident causation, it’s worth repeating that track issues and human error are the most common.  Therefore, while it’s useful to speak of new and better gadgets for passenger train safety – and we should be applying the best technology we can devise for passenger safety – it’s also true that suggestions like putting a second person in the cab, or promoting better track maintenance should be part of the larger conversation.

Fourth, if we focus down on the human error factor, we should note the 2006 FRA study of the fatigue factor:

“As part of the study, researchers analyzed the 30-day work-schedule histories of locomotive crews preceding approximately 1,400 train accidents and found a strong statistical correlation between the crew’s estimated level of alertness and the likelihood that they would be involved in an accident caused by human factors, FRA said.”

We can delve into the details to substantiate this conclusion:

The risk of human factors accidents was elevated at any effectiveness score below 90 and increased progressively with reduced effectiveness. At an effectiveness score ≤ 50, human factors accidents were 65 percent more likely than chance. Human factors accident risk increases reliably when effectiveness goes below 70, a value that is the rough equivalent of a 0.08 blood alcohol level or being awake for 21 hour following an 8-hour sleep period the previous night. Below an effectiveness score of 70, accident cause codes indicated the kinds of operator errors consistent with fatigue, confirming that the relationship between accident risk and effectiveness was meaningful. [FRA]

If preventing the next accident is our major concern then addressing the fatigue factor is crucial, and yet we have a situation in which railroad employees and management are at odds over the safety rules. For example, SEPTA (the Southeastern Pennsylvania Transportation Authority) wanted to relax fatigue abatement rules in 2014:

“Regional Rail engineers have asked federal regulators to require SEPTA to follow a safety rule designed to limit fatigue. SEPTA wants the Federal Railroad Administration to renew a waiver that the transit agency has had from the work rule for two years. The Brotherhood of Locomotive Engineers and Trainmen asked the federal agency to deny SEPTA’s request and hold a public hearing on the issue, citing accidents at other railroads caused by fatigued engineers.” [Phl.com]

And the reason for the waiver request?  That can be safely predicted:

“SEPTA said the waiver was “in the best interests of the riding public from both a service (more employees available for duty to address service demands) and economic standpoint (reduced labor costs by eliminating a potential need to hire additional employees).”

“Maintaining tight controls on labor expenses and operating expenses is one way SEPTA manages to fulfill that obligation [to operate efficiently],” SEPTA said in its request for the waiver extension.

“SEPTA estimates one additional crew costs approximately $150,000 annually, so even one new employee could cost SEPTA hundreds of thousands of dollars in labor expenses in a relatively short period of time.”[Phl.com]

In short, to address demands for service it was deemed better (from an economic standpoint) to have fewer workers working longer hours, in spite of the FRA report eight years before demonstrating the decline in the probability of passenger safety.  Is the “mission” of SEPTA to function “efficiently” or “safely?”

Fifth, returning to the gadget fixation there’s an abiding American proclivity to believe that we can apply tools to fix human problems.  It’s one of our basic strengths – we have a problem, we invent a tool to mitigate or solve it.  The news media has been abuzz about the Positive Control Train System, but while we have this tool in the box it’s not been applied universally.  The recent wreck provides a case to the point:

“In 2008, Congress ordered the installation of what are known as positive train control systems, which can detect an out-of-control, speeding train and automatically slow it down. But because lawmakers failed to provide the railroads access to the wireless frequencies required to make the system work, Amtrak was forced to negotiate for airwaves owned by private companies that are often used in mobile broadband.

Officials said Amtrak had made installation of the congressionally mandated safety system a priority and was ahead of most other railroads around the country. But the railroad struggled for four years to buy the rights to airwaves in the Northeast Corridor that would have allowed them to turn the system on.” [NYT]

This junction of private vs. public concerns was literally a wreck waiting to happen? Could Congress have made the wireless frequencies immediately available to the railroads? Probably yes. Would they buck the powerful lobbying interests of the mobile broadband providers? It doesn’t look as if they did.

Sixth, in the interests of not perpetually “fighting the last war,” or focusing too narrowly on factors associated with one instance of a rail system problem, we need to be cognizant of the other common factor in derailments and related accidents – the track inspections.  Perhaps it’s time for a report along the lines of the Deep Dive study conducted on the Metro North system, which called for Metro North to “create a plan for the use of advanced (track) inspection technology, ensure track is maintained to Metro North Track Standards, collaborate with labor unions to increase the availability of off hours maintenance time, improve training programs for track inspection and maintenance, and to analyze train schedules to determine whether there is sufficient time for inspection and maintenance of track.”

It’s not reassuring to find out that Minnesota has only one track inspector for its 4,500 mile of track. [MST]  Pennsylvania has 5,600 miles of track and “There are currently six PUC railroad inspectors who each focus on a specific discipline (track, operating practices, hazardous materials, grade crossing and motive power and equipment). PUC inspectors work in close coordination with FRA inspectors to ensure safe train movements throughout the entire state.” [PPUC]  Please tell me there are more than 6 people working on inspections!  The recent controversy over the Oil Trains moved New York to conduct an examination:

“Governor Andrew M. Cuomo today announced completion of the latest round of targeted crude oil tank car and rail inspections, which uncovered 100 defects, including eight critical safety defects that require immediate action. The inspections were the latest part of the Governor’s push to protect New Yorkers from the potential dangers associated with the transport of crude oil by freight rail companies. State and federal teams examined 704 crude oil tank cars and approximately 95 miles of track in these inspections.”  [Gov.NY]

Perhaps the Amtrack accident might move the states to engage in comprehensive reviews of track safety?

Finally, we need to address some philosophical and political problems.  What is the “mission” of our transportation system, especially with regard to our railroads?  As usual, there are more questions than answers.

Is the mission, to replicate the focus as illustrated by SEPTA  on “management efficiency,” or is the mission to provide safe, reliable, and modernized transportation for passengers and freight?  Have we down-sized employment levels of engineers and track inspectors to a point where we are being penny wise and pound foolish?  Are we requiring adherence to the best safety standards or simply accepting what is “economical” at the moment?

Have we placed impediments to modernization such as the implementation of Positive Control systems? Have we also jeopardized other safety considerations and systems by emphasizing privatization at the expense of public safety and the economic benefits of infrastructure improvement?

Are we perceiving our transportation system and infrastructure as essential to the economic well being and growth, or are we limiting our vision to the quotidian arguments between labor and management?

The national conversation about railway infrastructure and its management needs to be far broader than the current fixation on how to prevent wrecks.  That train left the station long ago.