>"Railroads are under government orders to install positive train control by the end of 2018 after the industry lobbied Congress to extend earlier deadlines, citing complexity and cost"
Amtrak receives well over a billion dollars a year in subsidies from the US Federal government[1], they should not be allowed to use cost as an excuse for not modernizing safety.
Also I think if you sight safety as being too complex then maybe you have no business operating a monopoly passenger rail service. Lawmakers are just as much to blame for this tragedy as the incompetence and disregard at Amtrak.
My understanding is that many modern Amtrak locomotives are equipped with PTC. Either way, the problem was that the track was not yet equipped with the necessary transmitters to communicate with the train (and isn't yet mandated to, since the railroad industry keeps lobbying to move the implementation date forward).
In Amtrak's defense, my understanding is that this track was actually owned by WSDOT / Sound Transit. And the mainline track (privately) owned by Union Pacific / BNSF isn't any better. Amtrak owns very little track of it's own on the west coast.
The failure to deploy PTC is a systemic issue with the entire railroad industry.
In Poland we still have the relatively medieval central radio and lights or semaphore system, tiered with rules to override it via radio plus really dumb automated braking system with train central detector, also enforcing speed limits in more advanced versions.
Plain old local moving block systems are widespread but US does not mandate them.
These are all mandatory (esp. the wired one where light or semaphore control is tied to NTC and/or PTC)
And the system is really dumb and electrical, not electronic. It is considered one of Automatic Train Protection systems. (The Polish short is SHP, translated as Automated Train Braking. The other system is Radio-Stop, a dumb radio alert system of local range triggered mostly manually forcing braking of all cabs in range.)
Similar slightly more modern systems are still used throughout Europe. The fact they are dumb (train is only slightly smarter) makes them cheap to deploy. They are mostly supervised or not automated other than the moving block light signal. (Which is often local and can be overridden at low speed only after radio confirmation.)
SHP is placed before speed limits and stations to alert the conductor. And if you pass the second one while the first is not cleared it means the train is runway and automated breaking is applied. The "logic" is in the train.
Compared to this Amtrak style PTC needs a big central computer system and control lines plus a train with an impressive array of sensors with remote data. Similar to ERTMS class 3.
US used to have legacy inductive systems in place, but for dumb cost saving reasons the companies were allowed to nor install them anymore.
Pretty much true for the entire coast - not to mention outright expensive. I looked at a ticket from Portland to la and it was more than a flight and they’d probably stick me on a bus in Bakersfield.
Then it should be broken up and managed differently. The idea that a company with a sanctioned monopoly who receives over a billion dollars a year in subsidies loses money is absurd.
You are basically demanding that the service should not exist. No one will compete for these routes, very few train systems in the world are profitable in themselves (even JR would lose money if it wasn’t for train station real estate).
No I am not "basically demanding" that the service should not exist. There is no reason why smaller regional companies couldn't exist and be profitable, offer better service and provide a compelling alternative to buses.
What I understand, which has at least been consistent with my experience riding many routes in most regions of the country, is that Amtrak is hindered by having to yield to freight, which takes priority in the US.
It's very rare to take Amtrak and not spend what can often be many consecutive hours waiting at a junction for a freight train to pass. Of course freight and passenger trains travel at very different speeds so the problem has no solution that even billions of dollars can solve, and it's only compounded by the vast distances of single-lane track in the US, which serve the purposes of freight needs just fine.
The difference in operation and overhead requirements for each use cannot be overstated. Besides a few routes in the Northeast and the Socal coastline, traveling by Amtrack is a novelty or a mistake. Competing with airlines on any route is far from realistic. I used to take this train between Portland and Seattle before it was beat by a <$20 twice-a-day bus program with good marketing and free wi-fi, relegating the train to being only a bit more comfortable and about $60 more expensive (and now more scary). I've been on >10 hour Amtrak rides which had no more than 2 people per car! When you run a route once every few days, you can't just shift people around like the airlines do. Without the subsidies, Amtrak simply doesn't exist.
> It's very rare to take Amtrak and not spend what can often be many consecutive hours waiting at a junction for a freight train to pass.
Its true that Amtrak often has to yield to freight on a lot of rail, but saying its “rare” to ride without being stuck for “hours” seems a gross exaggeration, or at least starkly different from my own rides. (Up and down west coast, northeast corridor, a couple times across the middle - highly recommended if you have a few days to spare).
Usually it’s 5 minutes of waiting, and for me I don’t recall ever more than 20.
Riding across the entire country (Oakland -> Philadelphia) I probably accrued 30 minutes of waiting.
I could have pressed on it more, but I think the Southern California coastline and the frequented routes in the Northeast are the only exception here.
It sounds like your trip across country was quite lucky. I've never done trips quite as long as coast-to-coast but I've taken some odd routes. I love the train for writing, so I do it a few times a year.
I think that varies on a rail-by-rail basis. I was chatting on 2 meters the other day with someone who does transport with BNSF on the route this train runs and in that case at least freight yields to passenger trains.
Agreed on all points. In this case is there a (e:) more competent passenger carrier that could step in to replace Amtrak? My understanding of the subsidies is that there would be no passenger service without it. Is it feasible to have competitive rail service in the US, does anyone do it?
Amtrak actually makes a profit [1][2] on both Acela and the slower services in the Northeast Corridor, but not anywhere else.
If we suppose that a privatized passenger carrier would want to make a profit, shedding long-distance trains would be wise, but several Amtrak directors have been against doing so time and time again.
If a private carrier were allowed to take over the northeast regional segment and abandon everything else, it might be feasible. Amtrak runs a profit on this segment, but only because Congress foots the capital expenditures. But maybe a private entity could harness cost efficiencies to make it.
Except the lobbying effort to delay the roll out of PTC was done by the Association of American Railroads[1], of which Amtrak is a full member with much influence.[2]
If Amtrak has in fact implemented PTC on all of the tracks they control, I would blame the freight railroads who are the other 20 full members of the AAR.
If Amtrak is also stalling on implementing PTC on their own tracks, yeah, I'd stop the buck at them.
>"If Amtrak has in fact implemented PTC on all of the tracks they control, I would blame the freight railroads who are the other 20 full members of the AAR."
This is exactly the logic everyone has been using to absolve themselves from any blame. 'Yes it was our train and our customers but it wasn't our track.'
If Amtrak is doesn't want to take any responsibility for peoples lives when they offer service on tracks they don't "control" then they should restrict their routes to tracks they do control. People who purchase fares on Amtrak should have a reasonable expectation that the carrier uses routes that meet modern safety standards.
The deadlines were extended for short term financial considerations. The law makers responsible for it should be named. Unlike the executive branch messing up in Flint and killing people the legislative branch can only be made accountable by the people.
Besides being a tragedy, this is a national embarrassment. It's a fucking train on a track on it's maiden voyage. And they crashed it.
It should not be humanly possible to configure the train to run at 80 mph over a section of track rated for 30 mph. The idea that some people say we need billions of dollars to make this work only compounds the embarrassment.
We want Level 5 self driving cars but we can't control a single variable (velocity) in a 1 dimensional problem space?!
Not even PTC. Conductor awareness system would have caught it. Or automated train protection system like used in Europe. (Not related to PTC such as ERTMS) These are dirt cheap and local to the train so there is no problem with deployment. Some use dirt cheap passive electromagnet beacons but are executed inside the train.
US still has these on two lines. For some unfathomable reason the regulators allowed not installing these.
But I have a feeling that the money sucking industry selling a particular solution called "Positive Train Control" has managed to create some sort of undeployable monstrosity. And what we actually need is something completely different which costs virtually nothing to implement.
Nobody would be going to jail... It's a relatively new technology, and legislation is already in place mandating it be universal by the end of 2018. I find difficulty in faulting either Amtrak or Sound Transit for not yet having 100% coverage -- both are actively rolling it out.
Outside of PTC, multiple safety protocols were violated, with responsibility likely falling to the Engineer and possibly Conductor. In systems not equipped with PTC, the engineer and conductor both are responsible for knowing the allowable speed for a given block of track, with the conductor being responsible for "enforcing" the safe operation of the train -- i.e. yell at, or even remove the engineer from the controls and/or initiate emergency braking.
(This is based on the way CSX operated when I was a freight conductor in '05. Obviously a decade later and a different railroad the responsibility hierarchy probably varies a bit, but I imagine it isn't terribly far off)
I still think it's fair to question why it takes so long to deploy new transportation tech in the US. In many parts of Asia the rollout of something like PTC probably would have taken months rather than 10+ years (the act requiring it passed in 2008).
I don't disagree, but to suggest somebody go to jail over a single incident that could have been prevented by a technology already under deployment is equally ridiculous.
In 2005, PTC was a technology found almost strictly in yards and on local or regional lines where a single railroad operated both the rolling stock and track infrastructure. Seeing that in 2017, 12 years later, Amtrak is at 49% coverage on locomotives, and 67% on track coverage, the situation (from a PTC deployment standpoint) seems hardly criminal.
ATP you mean, not PTC. They are complementary. ATP systems ate designed to be dirt cheap to deploy. Somehow most of US track does not have any such support despite extremely low cost and maintenance of these.
Part of ERTMS level 1. Also a similar system is deployed in India.
Poland has an even simpler one, well used inductive ATS. German legacy system is similar. These are even deployed in Canada so not far. Are as old as 1900s.
US has that old inductive system as well. If used right, it can control speed too.
US regulators foolishly allowed the railways to not install this basic protection. Similarly, timed trip stops are not used.
> with the conductor being responsible for "enforcing" the safe operation of the train -- i.e. yell at, or even remove the engineer from the controls and/or initiate emergency braking.
Interesting that you mention this. There was a similar fatal train crash in 2003 [1] - remarkably similar circumstances to this one - in which the driver had a heart attack, resulting in excessive speed entering a curve, derailment, and collision with a rock wall.
The organizational culture at the time considered the driver to be the one firmly in charge of the train. Also, the guard/conductor was not in the forward cabin with the driver, but at the rear of the train, a historical anachronism.
After this crash, there was an update to training and procedures that emphasised the guard's responsibility in braking the train if the driver neglected to or was unable to.
I'd guess that until PTC becomes standard across the country, new lines may be deployed with out it. The lack of PTC certainly doesn't stop them from running trains on pre-existing lines today.
I've spent almost the last decade involved with designing and implementing the messaging system behind PTC. Every one of these disasters makes me wish it was deployed everywhere. But for better or worse, railroads are incredibly change and risk intolerant. Everything goes through extensive testing before being rolled out into the field, which can add several years to any rollout plan.
That this is a problem is just ridiculous. It's on freaking rails! There is absolutely no excuse for this. GPS, Inertial Nav, simply integrating speed over time, it should be absolutely dead simple to calculate exactly where the train is at all times.
The article mentioned crap like differential GPS. Completely unnecessary. You don't need to be that accurate, I mean good grief, +/- 100 feet would be just fine for dealing with such gross overspeed detection.
I can't believe this is actually a problem. Pure politics/bureaucracy. It's certainly not a technical problem.
You don't even need GPS, or an external signal at all. All you need is an odometer.
1) Engineer selects the train's starting position (the station and departure gate) and selects the route.
2) Train software presses the equivalent of a 'trip reset' button that our cars have had forever. Current Position = 0.0
3) The train and the engineer each program a maximum speed. The train determines the maximum speed by selecting the maxSpeed which has the highest Position less than currentPosition. The engineer determines the maximum speed as she normally would.
4) The train's speed must not exceed the lower of the two configured maximum speeds.
> probably runs into issues with accumulated error
Typically there's no single nor double integration required for a wheeled ground vehicle; distance (revolutions times wheel circumference) is the primary measurement, with speed being trivially computed from that plus an accurate clock.
That assumes there's insignificant slippage between the wheel and ground. A reasonable assumption for a rubber tire on a concrete road, with a 20 sq. in. contact patch. Less reasonable for steel on steel with a 2 sq. in. contact patch.
I certainly agree that there are still errors, wheel slip being a major one. However, that does not cause an accumulating/ever-compounding error in speed nor distance travelled in the way that, say, a one-time error from an IMU's accelerometer does.
The lack-of-integration common in wheeled ground vehicles is highly beneficial to long-term speed & distance accuracy; it's a main reason why odometers were accurate for literally thousands of years prior to anyone knowing how to build an equally-accurate aircraft/spacecraft IMU. (The Romans were able to achieve <0.5% errors over hundreds of miles by the first century AD, something that no aircraft IMU was consistently able to achieve until after WWII).
overspeed protection has been standard on most national railways for decades
is there any reason they invented a new system (expensive and slow to rollout) vs. using an existing tried and tested one?
the main UK system (there are two) is nothing to write home about, but it is gloriously simple and was very fast and cheap to deploy: two metal transmitters live in the track bed coming up to a hazard, spaced such that they are 1 second apart at the maximum speed
if the train detector passes over both in less than 1 second then the brakes are triggered automatically
PTC is a lot more than overspeed control, imagine any use case that would be enabled by electronically controlled trains connected to a highly available network.
TPWS was always a stopgap (rolled out in the 90s while ETCS was being developed); as I said it was cheap and easy to deploy, and has prevented many accidents exactly like this one for 25 years
ETCS does what you are describing, and has been live in the UK and Europe since the early 00s
it works, and is slowly being rolled out everywhere replacing TPWS
The Pennsylvania Railroad had the equivalent of positive train control on all their mainlines by the late 50s. It's sad that we don't have it everywhere almost 60 years later
Is someone offering to pay? The reason we do not have a Pennsy anymore is that they went bankrupt, partly because they had to compete with subsidized highways and airports (there are various other reasons).
The fact is that PTC is an expensive way to increase the safety of one of the safest ways people can travel. Automobiles are far more dangerous. If the billions of dollars being spent on PTC were instead used to expand passenger service (and take cars off the road) we would almost certainly save more lives.
Also keep in mind that PTC solves just one of many ways that train wrecks can occur. It is not hard to find examples of other recent fatal incidents that PTC would have done nothing to prevent:
> Is someone offering to pay? The reason we do not have a Pennsy anymore is that they went bankrupt, partly because they had to compete with subsidized highways and airports (there are various other reasons).
I would say more than partially. Large portions of their revenue suddenly shifted to use those subsidized highways and airports. Sudden shocks are hard, even for a company doing financially well like the Pennsy was until the last few years.
My point is more that this technology has largely existed for the past 60 years. Even implemented slowly over the last 30 years, that time would have left us with a system that had most of the benefits of PTC.
> The fact is that PTC is an expensive way to increase the safety of one of the safest ways people can travel. Automobiles are far more dangerous. If the billions of dollars being spent on PTC were instead used to expand passenger service (and take cars off the road) we would almost certainly save more lives.
No argument. I hate cars. Car culture is toxic and causes so many problems.
> Also keep in mind that PTC solves just one of many ways that train wrecks can occur. It is not hard to find examples of other recent fatal incidents that PTC would have done nothing to prevent:
Just because a safety feature wouldn't prevent every accident doesn't mean it shouldn't be used. The question is how much of a benefit does it bring?
I'm not sure I understand your point. Are you saying we should disregard train safety because there are other more dangerous forms of travel?
Travel by rail may be one of the safest methods of travel (citation needed) but the US has much more dangerous rail infrastructure than the rest of the developed world. There doesn't seem to be a good reason for this.
The fatal train crash in this article was on the first run of a newly expanded line. It injured more people than were on the train because it fell on the freeway.
You may be able to cherry pick train crashes that PTC would not have prevented but my understanding is that several recent (last 10 years) fatal crashes would have been prevented by PTC.
My point is that we can spend billions of dollars making trains safer, or we can spend those billions expanding passenger rail to reduce the number of people who drive, something which would save far more lives.
It is a matter of economics. A dollar spent on PTC is a dollar not spend expanding passenger rail service (and the PTC mandate raises the cost of building new passenger lines), and the question is whether or not it is worth the extra safety. Considering how safe rail transit already is, and how much more unsafe driving is, it seems like our money would be better spent taking cars off the road.
Or to put it another way, I am arguing that focusing only on rail fatalities is too narrow and that we should instead focus on travel fatalities in general. Increased access to passenger trains, more frequent and more reliable service, and faster passenger service are all ways to reduce travel fatalities because all the above reduce the number of people who drive. PTC also reduces travel fatalities, but not by as much. If you need a more extreme example, many lives could be saved by banning automobiles altogether, but obviously the costs would be far too high to justify such an approach.
Except in this case we did spend the money on PTC but then didn't bother to turn it on. This fatal crash was the first run of a newly expanded line. Are all new rail line expansions dangerous? If so will trains still be safer than cars if we expanded them enough to carry similar numbers?
I think I understand your argument but I'm not convinced more routes without PTC is safer without seeing numbers.
That train was only 1/3 full, if we added more trains and put more people on those trains would the deaths really be lower? What if we compare to just the alternative of traffic fatalities on interstates between Seattle and Portland? Also this train fell onto the freeway and injured drivers, is that included in traffic statistics (slightly facetious)?
I'm not convinced that banning automobiles entirely would save lives. There are too many practical concerns there. Do ambulances still exist? Would people just drive anyway, perhaps ignoring all laws or driving unsafely to flee police? It's not a compelling argument.
I take issue the most with the attitude that we shouldn't be critical of the safety of trains because there is another method of travel that is worse because that kind of thinking does nothing to make trains safer.
Keep in mind that Amtrak had tested the route and conducted numerous practice runs prior to this accident. None of those practice rules ended with derailments. It is unfortunate that on the first day that this route was operated with passengers, some kind of mistake was made and the train exceeded the speed limit. Obviously there is not much more that can be said about what happened until the NTSB report is published. New passenger services are introduced without incident on a fairly regular basis.
I have no idea how popular Cascades is between Portland and Seattle, or how many people would choose to take that train rather than drive on I-5.
As for fatality numbers, here is one study that tries to extract an apples-to-apples comparison of different transit modes, from 2010 (before PTC had been installed or activated on many routes):
Trains are safer than cars according to those numbers. There may be some nuances involving how fatalities are being reported; for example, whether or not we should count fatalities involving trespassers being struck by trains. Regardless, it seems pretty clear that yes, if more people took trains instead of driving, fewer people would die. If you know of a better way to make the comparison I would love to see it.
I am not suggesting that we should not be critical of train safety. There is a long history of railroads killing people and there are good reasons for many of the safety regulations applied to trains. The reason passenger rail is so safe is that so many regulations are in place. For the same reason, it is hard to see why spending billions of dollars to install one more safety system makes sense, especially when there is a competing mode of transit that is far less safe.
The test runs did not include PTC though. The entire point of requiring PTC is that railroads will not install it themselves because they don't think they need it.
My concern with saying trains should replace cars because they are safer is that trains don't do the same thing as cars and interstates do. Trains can't take me around town but cars are often used for that which skews the numbers. I would like to see the risk in micromorts [1] for travel from Seattle to Portland by car, plane and rail[2].
If I want to go to Portland from Seattle I have several choices. The top three from most to least likely for me are:
1) Drive my car down I-5.
2) Fly.
3) Take the train.
I suspect most people are in a similar situation and I think we can both agree that more people drive their cars to Portland than take the train in any given period of time.
I do not agree with your assumption that just putting more people on trains will magically save lives. Specifically I do not think that putting the people who drive to Portland on a train will necessarily save lives. This is for several reasons:
1) The existing trains do not carry anywhere near as many people as I-5 does so we must drastically increase the number of trains and possibly rails. This would drastically increase the chances of train on train collisions which are uncommon today (but still happen). There could be other issues here such as the availability of qualified operators.
2) Trains are much slower than driving or flying. The train to Portland takes 10.5 hours, by car it takes 3.5 and by plane is an hour in the air plus a couple hours to get to the airport and get through security so call it the same as driving. For trains to compete with this we would have to significantly increase speed which would increase danger.
I'm with you, I would love for there to be better rail options in the United States. I don't think we get there by just scaling up and ignoring the risks and problems. If rail companies can't safely operate their trains with the small amount of current utilization why would we expect it to get better (or even stay the same) with higher volumes? Changes must be made to a lot of aspects of railroads to increase ridership but safety is absolutely one of them.
e: From the linked study it looks like highway driving is one of the most dangerous forms of driving. My point stands that this is more complex than it looks and rail transit does not scale linearly.
Actually, our freight rail system is the best in the world. The problem is there is no money for passenger server because the US is stuck in a car-centric mode of thinking about transportation.
I've worked on two PTC projects, one successful and one failure, after hundreds of millions dollars. The successful one took about 6 years and is still on a trial basis. The failed one had an aggressive schedule but ran out of money after 3 years going nowhere. This thing is not as easy as it seems.
I'm interested in your experience, what makes it so hard? The PTC roll-out is not new but delays seem common, it's obviously hard but why? Once we figure it out in one place can that experience be applied elsewhere easily?
Given the estimate of how long something takes, and dividing it up amongst the people who can do it, and then say... we're going to deliver it in 75% of the optimal time (based on the size of the project - increasing the speed of getting it done by adding people and having a more aggressive schedule)...
Well, for the data set of 20,000 projects there were 750 that were trying to deliver in less than 75% of the optimal time. Of those 750, none were able to be completed in that time.
Granted, that was for software projects. But I suspect that beurocratic projects have similar problems. Need it done faster? Add more people. Have more people? Need more meetings. Have more tasks going on at once? More likely that someone is going to be waiting on another task getting complete because of a bad (overly optimistic) estimate / deadline. Have more people doing nothing at any given time? Spend money faster... and run out of budget all that much faster.
I'm sure there are other complexities added in when there are pieces of iron rather than silicon involved... and safety regulations start chewing up time and reworks.
But overall, the "if it is possible to do, but someone put too low of a cost in the budget or tried to get it done faster than the engineers said it was doable"... well, isn't that bad project management that promised too much and underdelivered... and then got a canceled project? Is there something else it could be?
Yeah, this is really weird to me. They're talking about installing sensors on particular tracks and the like, after saying that it's GPS-based, but... why? Stick a GPS on the engine, load a database of maximum speeds, get instant safety. What'm I missing?
(Yeah, you have to integrate with the engines themselves, but that seems like not a 6-year project, and especially not once it's been designed and tested. It probably also won't account for "stop, there's something else on the tracks", but just speed-checking seems like it would be a massive reduction of risk with very little complexity...)
Allowed speed is not as simple as just where you are; it also depends on the trackside signals and how the track in front of you is lined and/or occupied. For example, if you are 1 mile from an interlocking you might be able to proceed at track speed because you will be going straight through it (clear signal) or you might need to slow down to, I.e., 20 mph as you are going to go through a switch to another track (approach diverging signal). One of the main complexities of PTC is getting all that signaling data onto the locomotive, combined with the route and rules, so that it can actually make the right decision. (Keep in mind that on the US's relatively slow freight railroads "cab signals"; I.e. seeing signal indications in the cab of the locomotive, is relatively uncommon. Signals are generally trackside only and are also often 50+ years old. They work fine, but they're not exactly networked in any modern sense. Updating all this trackside equipment is one of the primary costs of implementing PTC.)
Even an app on the driver’s phone that alerts when they’re exceeding the speed limit would have prevented many of the recent crashes because the drivers weren’t incapacitated - just distracted or confused. Many recent cars also know when you’re driving over the speed limit and display an alert. Not billions of $
> Engineer in fatal Amtrak derailment might have been distracted by trainee in locomotive
> Investigators are looking into whether the Amtrak engineer whose speeding train plunged off an overpass, killing at least three people, was distracted by the presence of an employee-in-training next to him in the locomotive, a federal official said Tuesday.
> The official, who was not authorized to discuss the matter publicly and spoke on condition of anonymity, said investigators want to know whether the engineer lost “situational awareness” because of the second person in the cab.
Just distracted or confused is a real possibility that is being looked into.
There is a publicly visible website referenced in some of these articles that showed the trains speed moments before the crash. I believe they also have speed limit information. I wonder if it's possible to write a simple service that consumes this data, compares live train speeds to speed limits and sends an email to Amtrak. Could it be done in the time it takes to ride the train from Seattle to Portland? Quite possibly.
There are a number of reasons, typical of large projects. Scope creep, mismatched of experience/expertise/organization-setup, unrealistic expectation, insufficient funding, complex vendor/system integration, and long testing cycle.
Funding for PTC inevitably is politically motivated. There are petty projects or requirements tacked onto the main project by various stakeholders, in regardless of technical feasibility.
Railroads are usually operation centric companies focusing on day to day operation. Their expertise is not at designing, engineering, or building things. Outside vendors are brought in to bid on the PTC project. Due to high risk and high liability, only a few companies have the financial backing to put up the surety bond to bid on the project. Sometimes there's only one bidder. The competent ones know the high cost and long time to build the thing and decide to stay put. The less competent one misjudges the time and cost, provides the lowest bid, and becomes the general contractor.
Nobody has a complete set of technologies to implement the entire PTC project. The general contractor has the financial mean but not the technologies. Numerous subcontractors bid on various parts of the project to join the party. The selection of these subcontractors and their technologies sometime are political. The general contractor might not like a sub but is forced to take on the sub because of pressure. A working combination might not be transferable to different railroads due to different vendor/technologies winning the bids.
Pretty much all the vendor technologies are proprietary, and integration among them is a nightmare. Some of them sound good on paper and on promise but fall short when the rubber meets the road. In the failure case I cited above, one fortune 500 company acting as a sub provides a critical part which simply cannot be integrated with the rest of the system. It was too late and too expensive to pick another vendor and the project went into limbo due to running out of funding.
The testing cycle is very long. You are retrofitting a busy daily running railroad live, with people's lives at stake.
It’s totally believable. China and Europe both implemented some form of positive train control, largely because much of their passenger traffic runs on relatively new track designed for passenger traffic. In the US, we’ve not built significant passenger rail track in a century. The only part of Amtrak’s network owned by Amtrak is the northeast corridor. Key parts of that infrastructure date to the 1800s (e.g. the tunnel under Baltimore). Everywhere else, Amtrak runs in freight track.
I came in here to vent about how unsafe our rail systems are. However, our deaths per passenger-mile are well below the EU average, and we're safer than all except UK, Denmark, and Netherlands: https://www.theguardian.com/news/datablog/2013/jul/25/how-sa...
Still, this wreck is sad, and also embarrassing. Our rail system is just not a meaningful transportation network, and is such a missed opportunity. Even small efforts at improving it seem to end in failure, or even disaster.
Reading transportation fatality statistics requires some nuance.
In the realm of railways, most fatalities tend to be suicides causing no additional deaths, which are often analyzed separately. The EU does this [1], while the Guardian article only raises this distinction halfway through.
Once you exclude suicides, there are several other categories that are useful to distinguish: collisions (e.g. between two trains), derailments, accidents at road-crossings, and 'accidents to persons caused by rolling stock in motion' [2]. The share of accidents by category will tell you a lot about a particular country's issues; a relatively large amount of level crossing accidents may suggest that not enough of the rail network may be grade separated (e.g. Poland, Hungary, Romania), while a high share (and absolute number) of collisions or derailments would be a serious cause for alarm.
Also consider this 2012 post [3] which compares collisions and derailments per usage by country between Europe and the US, originally written as a response to that 2011 article.
This is insane to think this happened in 2017 in USA. No other modern industrialized nation suffers these many embarrassing train derailments and with no blame/shame in the end. "Thoughts and prayers" my ass from WSDOT and Amtrak.
It shouldn't take a billion dollars to figure this out. even the simplest techniques like proper railway training that they do in Japan would go a long way.
The PATH subway system between NYC and NJ has been boasting on little ads inside the cars that they are implementing communications-based train control and positive train control. How would positive train control work underground, if it's GPS-based?
If I accurately recall, PATH is rolling out Siemens Trainguard MT; a Communication-Based Train Control (CBTC) system which uses trackside (between-the-tracks) beacons and in-car equipment.
Radar is used to figure out the vehicle's distance from the beacons. The system communicates using a wireless-and-wired network to a control center and keeps track of mobile blocks. Here's a brochure [1]. I'm not entirely sure, but I think GPS isn't used in Trainguard -- which makes sense; the beacons know where they are in relation to the track geometry, and locations only need to be resolved in track-space.
I'm not convinced this is operator error. This is the first run on this route so it's unlikely the engineer was asleep.
2nd Hand copy paste from another part of the web I visit. I couldn't find the original source:
"If the wheels locked up at 81.1 MPH on the 1.4% down grade there would be no way to slow that lightweight train as in 41 seconds it would be into the 30 MPH curve. The engineer would not have time to release the locked-up brakes, apply sand and make a new braking application. Even with electronic brakes, the system does not work that fast. The NTSB should not take a year, like happened regarding the engineer who failed to stop at Hoboken, had failed to blow his horn at many crossing in his trip and the Conductor failed to check on his bad performance during that inbound train trip. That NJ Transit train had an inward facing camera but its recording was kept secret for over a year.
This was a passenger only track and the rail conditions were not likely felt by the engineer as there were no prior locations where he could get a feel of the slipperiness of the rail head. This was a new line and even thought there were many qualifying test trips, it could be likely this was the first morning with just the right dampness, temperature to make very bad rail conditions. Were many test runs made at this early time of morning?
As far back in the 1960, when single Budd Rail Diesel Cars were use on commuter runs with various station stops in an early Fall morning, some morning the rails were so slimy the RDC’s Rollacon anti-slide system would release the brakes when the wheels locked up, and the RDC would fail to stop at the platform and have to back up. One morning this happened at 3stations so when the engineer approached the terminal interlocking with a possible red signal, there was a very slow approach as going past a red signal could not be corrected by a short back up.
Many East Coast electrified commuter railroads are very familiar with bad rail condition caused by ground up leaves on the rail head which creates carbon black and misty water for a very efficient lubricant that can only be eliminated with sand. But electric suburban cars do not have sand. There is an entire industry used to wash the rail head and apply solutions to improve the traction for these trains
Railfans who own and use their own track motor cars (Speeders) become well qualified of the need to test for the rail conditions and learn the distance it would take to stop at any place on their run. A leaf ground-up track can be so slimy that a light track car can fail to get enough traction to go up a 0.10 % grade without spinning their power wheel. You cannot even gain any speed as the wheels will just spin out. And if you’re coming down a 1 or 2% grade, with is very common on many lines, you can go into a slide and wish you had an anchor to throw out. Being on a frozen rail track, and then have some very misty or rain situation, can be worst that being on a highway that was frozen over and then have been rain on over the ice. Sometimes in those cases a car rubber tire will just coast you over to the curb with just the crown on the street. Most cases a rail head can be worse than any highway surface for having traction.
I believe the train 501 wheels locked up and the train slid into the 30 MPH curve at too high a speed.
PS: Positive Train Control will not know of the possible bad rail condition and this wreck may have happened regardless of PTC in effect or not. That system cannot tell which is the worst case rail condition and is designed for the average expectation of conditions. So much for crewless trains, that will never happen. "
There is nothing about the physical world that guarantees a person pays attention because something is important. A common cause of accidents is when people do not pay attention. Train crashes specifically have happened in recent years due to distracted operators. Just because this run was more important than normal does not somehow mean the operator could not have been distracted.
> He's saying that there's things other than driver error or mechanical failure that could have contributed to the crash
Nothing in the article about a “rush to launch”. I would love all safety tech to rollout immediately but it doesn’t and it’s not reasonable to expect no new lines to launch. Pushing to delay installs for cost reasons though still doesn’t sit well.
What do you mean, "expect no new lines to launch"? The article says that the safety systems in question are being installed and will be ready relatively soon, but they chose to start service without them. Delaying a new service for a little while to wait for such things to be installed seems pretty reasonable!
Define "relatively soon". In particular, how likely do you consider a delay? How likely do you consider a delay after that? You could wind up waiting several years, while your "ready to go" line just sits there, eating up maintenance costs but not being used.
Also, define "ready". Is it ready when it's running, but is only 75% likely to actually stop a train if needed because of various bugs? (Note well: I do not know what PTC's bug track record is. I just know that such systems often have bugs that take a while to work out.)
In the real world, projects that wait for everything to be ready before they ship at all are often very late.
On the other hand, you don't ship something without basic safety features. So it all depends on whether you view PTC as a basic safety feature. (After the crash, it's easy to say "obviously it is", but I'm not sure that it's fair to use hindsight that way. PTC was supposed to be a small increment of safety to an already quite safe mode of transportation.)
How likely are delays in other parts of the project? There are plenty of things that could be delayed and would physically prevent train service from running on the line. Is it really such a massive risk to to say that safety equipment must be installed before operations begin, even if the train could potentially make the run without it?
I don't know about that "quite safe" thing. I looked up the stats in another comment in this tree and they look pretty bad for trains.
It seems less reasonable when you compare the safety of train travel to the safety of automobile travel, and when you consider that every time a new passenger route is opened it takes cars off the road. Yes there are train wrecks, and yes, PTC would have prevented this specific wreck, but trains remain one of the safest modes of transit.
Train fatalities tend to be clumped so the stats tend to vary a lot from one year to the next, but it looks like in recent years the typical rate is something like 5 fatalities per 100 million passenger-miles. The car data only gives fatalities per vehicle-mile for some reason, but that will always be a pessimistic number (at least until autonomous cars happen), and that rate is just a bit over 1 fatality per 100 million vehicle-miles.
It looks like cars are quite a bit safer, unless I've misunderstood something. The absolute number of fatalities is far lower for trains, but the number of passenger miles per year is also far lower.
One issue is that the motor vehicle fatalities table seems to include data from buses (that is how I understand the references), which are safer than cars. I tried looking at the sources for the passenger rail data but a quick glance was not very informative.
Here is another study, from 2010, showing some very different numbers:
PTC wouldn't have just prevented this specific wreck, PTC would have prevented like 9 of the last 10 fatal train wrecks in the USA. Every single fatal wreck that happened since Obama pushed back the mandate would have been avoided by PTC.
Let's assume you are right, and that in the past three years PTC would have prevented the majority of fata train wrecks. How many fatalities are you talking about? How many lives should we expect PTC to save over the next decade?
The more important question is this: how many automobile fatalities could be prevented if we took the billions of dollars being spent on PTC and spent it building a new commuter rail line that took hundreds of cars off the road every day?
Make sure to build it up to at least 1950s standards in safety. Many parts of US rail do not feature even basics. They rely on the driver and signals only...
Well, if they hadn't started service on this line they would have continued service on the old line, which also doesn't have PTC, so it's not exactly a net loss. This new line is a sort of bypass meant to replace an older one.
The fact that they safely operated on the old track for years, then had a horrible crash on their very first run on the new track, makes me think it really did make a big difference. It's possible that this was just a horrible coincidence, but it seems more likely that something about the new line made it riskier. Maybe just the fact that the driver is less familiar with it?
Allowing operations to start on a new line before important safety equipment has been installed is also a good way to make that installation happen very late. If you say "you cannot operate until this is done," it has a way of motivating people to get it done. If you say "you can operate now, just try to get it done" then it may take quite a long time to actually happen.
I'm not sure what point you're making. If a new line is more dangerous then isn't that a pretty good argument for delaying its use until the safety systems are installed?
You said that delaying the start of the new line until PTC was installed seems pretty reasonable.
roywiggins replied that the line being replaced also didn't have PTC, so that switching to the new line meant swapping a line without PTC for another line without PTC, which is a net wash.
You replied "It's possible that this was just a horrible coincidence, but it seems more likely that something about the new line made it riskier. Maybe just the fact that the driver is less familiar with it?" I presume that you're still arguing here for waiting until PTC was installed.
I replied. Since you didn't seem to understand my reply, let me try again.
There are lots of things that could make a new line more dangerous. Some of them could be fixed by PTC. Some of them would not. So unless you know that the cause of the crash is something that PTC would have prevented, then "the new line is more dangerous" does not imply "wait for PTC".
Why is the new line more dangerous? Is it because it has a curve? The other line had far more. Is it because the new line had a curve that was lower speed than the rest of the line? It's possible that the other line had so many curves that there never was a place for fast running, and therefore that there never was a curve that you could be suckered into coming into too fast.
Is the new line more dangerous because people aren't familiar with it? Well, that's what months of training runs are for. (They did do months of training runs, didn't they?)
It's possible that the cause is a combination of all of this: A tight curve in the middle of a fast stretch, and engineers unfamiliar with the route ran into the curve too fast. If that's all true, then yes, PTC probably would have prevented the crash.
Also note that all this is hindsight. When making the decision to open the new line, was it obvious that the new line (without PTC) was more dangerous than the old line (also without PTC)? Probably not. If anything, I suspect that the thinking was that the new line would be less dangerous - it wouldn't be cluttered up with a bunch of freight trains, running on a line without PTC, where a mistake could squish a passenger train. So if the thinking in advance is that the new line is at least as safe, and plausibly safer, even without the safety equipment, why would you wait?
That's pure speculation with a million seemingly unfounded assumptions behind it.
My reasoning is pretty simple: since this crash happened on the very first run, and PTC would have prevented it, it looks probable that the danger of PTC-preventible crashes was substantially higher on the new line. I don't know about the reasons or causes or potential mitigations, I'm just looking at the outcome.
It is possible that it's just a horrible coincidence. But that's not the way to bet.
You're right that it's all hindsight. We can't go back in time and push back the opening of the new line or push forward the installation of PTC. What we can do is convert this hindsight into foresight for the next time.
Objection, your honor. Assumes facts not in evidence.
At the moment, it looks likely that your statement is correct. But I've seen at least one source claiming that the speed data (where they got the 81 MPH from) isn't continuous data, and therefore isn't necessarily reporting the speed from when the train entered the curve. Note well: I do not know the truth of this claim. (I also don't know the truth of the 81 MPH claim.)
We'll know if PTC would have prevented it when the NTSB produces some concrete conclusions. Until then, I think you're taking speculation and regarding it as concrete data.
You could be right, though. You could be right in every detail. But I think you're stating your case as if it's certain, and I don't think certainty is warranted yet.
The article does not explain the use of the word "rush" anywhere in the article. The article infers that because the line opened before the PTC system was ready that it was rushed. The article even inserted a disclaimer:
"The agencies did not immediately respond to questions about why they did so [publicize line opening] while the speed-control technology was still months away."
That is lazy reporting. All documents about construction details and timelines are available from the project website.
Amtrak receives well over a billion dollars a year in subsidies from the US Federal government[1], they should not be allowed to use cost as an excuse for not modernizing safety.
Also I think if you sight safety as being too complex then maybe you have no business operating a monopoly passenger rail service. Lawmakers are just as much to blame for this tragedy as the incompetence and disregard at Amtrak.
[1] https://www.cbo.gov/budget-options/2013/44782