OT: Amtrak Controls

[Aaron O]

I'm not an engineer, I am a know-nothing that likely just doesn't know what I don't know. This is a bright bunch, and I'm guessing there will be someone who can answer this.

I'm not getting why putting safety features on Amtrak would be complex or expensive. It seems to me that all you have to do is have a gauge that reflects the train's speed (or this can be done via GPS), which they already have. Then all you need is a program of some sort that shows the speed limits along the rail and tracks the train's location (GPS)...and if the speed exceeds the limit by a certain amount, the brake is engaged.

2015 cars all have more complicated features than that. My pohone's GPS can do all of this except automatically engage the brake, but I can't imagine that's especially difficult.

So what am I not getting?

[oldpotatoe]

I'm not an engineer, I am a know-nothing that likely just doesn't know what I don't know. This is a bright bunch, and I'm guessing there will be someone who can answer this.

I'm not getting why putting safety features on Amtrak would be complex or expensive. It seems to me that all you have to do is have a gauge that reflects the train's speed (or this can be done via GPS), which they already have. Then all you need is a program of some sort that shows the speed limits along the rail and tracks the train's location (GPS)...and if the speed exceeds the limit by a certain amount, the brake is engaged.

2015 cars all have more complicated features than that. My pohone's GPS can do all of this except automatically engage the brake, but I can't imagine that's especially difficult.

So what am I not getting?

Money, politics, partisanship, technology, money loser company supported by Govt, politicians trying to make a point..........probably others.

[rwsaunders]

Even golf carts and UHaul trucks have speed governors...why can't a train engine?

[Plum Hill]

Cost and technology.
Spend some time searching Positive Train Control.

Even the Swiss railways have screw-ups.

[rnhood]

Money, politics, partisanship, technology, money loser company supported by Govt, politicians trying to make a point..........probably others.

Its been appropriately described as a money burning machine....gouging taxpayers for near $1.5B per year. It does serve a transportation purpose but, no one has been able to put an effective plan in motion to remedy the situation. I guess Nixon had one but unfortunately he resigned before it was implemented.

[oldpotatoe]

Its been appropriately described as a money burning machine....gouging taxpayers for near $1.5B per year. It does serve a transportation purpose but, no one has been able to put an effective plan in motion to remedy the situation. I guess Nixon had one but unfortunately he resigned before it was implemented.

I guess but without that $1.5 billion($ compard to other boondoggles), it wouldn't exist and those people and that freight wouldn't 'get there'. All other successful rail systems in every other place on earth is govt subsidized.

[fuzzalow]

Its been appropriately described as a money burning machine....gouging taxpayers for near $1.5B per year. It does serve a transportation purpose but, no one has been able to put an effective plan in motion to remedy the situation. I guess Nixon had one but unfortunately he resigned before it was implemented.

The projected size of the 2015 Federal budget is 3.9 Trillion. For me, 1.5B that is part of interior transportation infrastructure is not a great deal of money.

I disagree with the use of the term "gouging" but interprete it to be your political editorial bias and not a reference to actual an fraud or malfeasance.

[BSBD]

Cost and technology.
Spend some time searching Positive Train Control.

Even the Swiss railways have screw-ups.

This.

PTC is a hard project, a very hard project.

All of the Tier 1 railroads are going to blow their mandate date, plain and simple.

[Aaron O]

Can we stick to the question and not make this a debate about the merit of funding Amtrak? We have a rail system...regardless of your opinions about subsidizing it, i think we all agree that if we have one, it should be safe.

So why is it difficult and expensive to implement what i'd think are simple, cheap, readily available safety measures? If the answer is it's hard, explain why it's hard...because I'm not seeing it.

[fuzzalow]

I'll take a swing at it. Not my field, passenger rail, but off the topof my head:

Infrastructure, methods and means of management are all predicated on human controls and failsafes based on those human controls. As time, tradition and custom have run and operated rail systems for centuries.

Trains are not high-tech Google cars waiting to cede controls, as referenced in the above paragraph, to a blend of computer run controls with human override inside the engineers cabin. I do not know how integrated computer control is to the running of a train.

Expense, availability and acceptance of safety systems that can be retro-fitted to the entire fleet of rolling stock.

Anything done to the scale of a national rail system will not be done quickly or cheaply. The procurement and bidding processes are fraught with pork barrel politics and other special circumstances that avalanche into a SNAFU because the payoff is enormous and the contract is virtually everlasting.

[Louis]

Just because no one here knows enough about the details of the issue to explain it, doesn't mean that it isn't a legitimately challenging problem.

I believe I read in the NYT that that line has PTC in one direction, but not in the other. Given budget and time limitations and the lack of a magic wand one might wave over the whole system to make it 99.999999999% safe, I'd say that I'm not surprised that an accident happened, whether they had PTC or not.

Furthermore, I bet that within a relative short period of time they will get PTC in both directions, but accidents will continue. Maybe not that exact type of accident, but given how things work in life, stuff will happen.

[makoti]

From the AAR (Association of American Railroads) site:

"Nevertheless, due to PTC's complexity and the enormity of the implementation task — and the fact that much of the technology PTC requires simply did not exist when the PTC mandate was passed and has had to be developed from scratch — much work remains to be done. Despite railroads' best efforts, various technical and non-technical challenges make full development and deployment of PTC by 2015 impossible, as outlined in a March 2014 industry report on the progress of PTC implementation."

It doesn't say what those challenges are, but there must be something(s).
EDIT:
Interesting article on it:
http://www.latimes.com/opinion/opinion-la/la-ol-philadelphia-amtrak-crash-positive-train-control-20150514-story.html#page=1

[shovelhd]

Don't forget the union factor. Less train control by union employees is bad for union employees.

[Aaron O]

I'm going to check what PTC actually is and does.

Fuzz's answer makes sense, I'm probably not appreciating the complexity of adding a control like that to a train made in the 60s. The procurement process being quasi-government probably can have the effect described.

[93legendti]

"According to Amtrak, PTC was installed in the section of track where the Philly accident occurred," a committee source writes in an email to U.S. News. "There have been delays in 'turning it on' associated with FCC dealings and getting the bandwidth to upgrade the radios from 900 MHz to something higher (for more reliability)."

Amtrak's application for the bandwidth needed to use the positive train control system was approved in "early March," an FCC official says.

“The spectrum Amtrak wanted to use in 2011 was owned by someone else," the official says. "It took them three more years to negotiate with private parties to acquire the needed spectrum for the Washington, D.C.-to-New York corridor. Once Amtrak finalized their application, the commission approved it within two days.”

Amtrak President Joseph Boardman said Thursday the technology was installed where the crash occurred, according to The Associated Press, but it had not been turned on because the system needed to be tested further.

http://www.usnews.com/news/articles/2015/05/14/exclusive-train-control-was-installed-turned-off-at-time-of-amtrak-crash-congressmen-say


In 2008, a month after a commuter train and a freight train collided in Chatsworth, California, killing 25 people, Congress passed a law requiring that positive train control be installed by Dec. 31, 2015. But railroads have long complained that complications will prevent them from meeting that deadline.

http://bigstory.ap.org/urn:publicid:ap.org:4271b2ca952348168556f7d234709e 78

[93legendti]

I guess but without that $1.5 billion($ compard to other boondoggles), it wouldn't exist and those people and that freight wouldn't 'get there'. All other successful rail systems in every other place on earth is govt subsidized.

A third option would be to privatize.

"When is a hamburger not just a hamburger? When it costs Amtrak $16 to make. They sell it for $9.50, and taxpayers cover the difference — every time. Then it becomes a glaring symbol for spiraling costs, crippling deficits, and the inherent inefficiencies of big government.

Thirty years ago, the idea of Amtrak losing money on food sales was as outrageous as it is today. (Hungry customers on a moving train with nowhere else to go. How hard can it be?) In fact, it was so outrageous that Congress passed a law against it. The Amtrak Improvement Act of 1981 prohibits the government-owned company from selling food at a loss. Nice try. Today, Amtrak is selling more and losing more than ever before.

This month the Government Accountability Office reported that losses on food service exceeded $80 million last year and totaled $834 million during the past decade. Auditors blamed the staggering losses — most of which occur on Amtrak’s 15 long-distance routes — on waste, theft, and lack of oversight. That’s only a fraction of the total losses from long distance operations, but it’s real money nonetheless."

http://www.bostonglobe.com/opinion/2012/08/12/amtrak-governmental-inertia-sesame-seed-bun/4bpbUOQSjqqOnPmWgxvl1O/story.html

[93legendti]

"Amtrak said it was increasing the use of credit cards for food sales to cut down on cash thefts by employees, reducing staff, creating a better system to track inventory and to collect revenue. It has also set up a three-person loss-prevention unit...

Mr. Alves (Amtrak Inspector General) who issued a report on the problem last year, estimated that theft by Amtrak food service employees could cost the agency $4 million to $7 million annually. According to charts shown by Republican committee staff members during the hearing, Amtrak charges about $2 for a soft drink, but the cost to taxpayers is about $3.40 when labor is included. A $9.50 hamburger on the train costs taxpayers $16, the charts showed. Labor adds nearly 60 percent to food and beverage costs."


http://www.nytimes.com/2012/08/03/us/politics/amtrak-lost-834-million-on-food-in-last-decade-audit-finds.html?_r=0

[FlashUNC]

Interesting that the problems, regardless of the issue always seem to stem to blaming labor.

Train safety? Labor.

Efficiency of the system? Labor.

Cost to the taxpayers? Labor.

Not, say, freight companies and their stranglehold on rail lines starving any real possibility of decent rail in this country.

Its a complex problem. To reduce it to simply "unions are bad" is ridiculous.

[Louis]

But the Republican committee staff members said it, so it must be true.

Interesting that the problems, regardless of the issue always seem to stem to blaming labor.

Train safety? Labor.

Efficiency of the system? Labor.

Cost to the taxpayers? Labor.

Not, say, freight companies and their stranglehold on rail lines starving any real possibility of decent rail in this country.

Its a complex problem. To reduce it to simply "unions are bad" is ridiculous.

[Aaron O]

I really didn't mean this as criticism of Amtrak, or any sort of commentary on their efficiency, labor, subsidies, etc.

I just don't understand the complexity and expense of what seems, on the surface, like a pretty easy solution.

[FlashUNC]

I really didn't mean this as criticism of Amtrak, or any sort of commentary on their efficiency, labor, subsidies, etc.

I just don't understand the complexity and expense of what seems, on the surface, like a pretty easy solution.

As a country, we've not been particularly good about building infrastructure from scratch since, say, Eisenhower and Kennedy were in office.

I'm totally a layman, but from what I understand this kind of safety system -- while mandated -- is a serious engineering and infrastructure challenge, even assuming we'd get all the parties on board with adequate funding to do the job.

We have none of those things right now, and at least in this instance in Philly, it seems to have cost some people their lives.

Anything beyond that is just folks getting out their particular dead horse to beat.

[Aaron O]

As a country, we've not been particularly good about building infrastructure from scratch since, say, Eisenhower and Kennedy were in office.

I'm totally a layman, but from what I understand this kind of safety system -- while mandated -- is a serious engineering and infrastructure challenge, even assuming we'd get all the parties on board with adequate funding to do the job.

We have none of those things right now, and at least in this instance in Philly, it seems to have cost some people their lives.

Anything beyond that is just folks getting out their particular dead horse to beat.
Forgive me...I'm also a layman. I understand that there was some government mandate to implement some safety features within a fixed period of time. I have no idea what those features are...other than that they're apparently expensive. I don't see the need for something expensive, it seems like basic GPS technology would solve this simply and inexpensively.

[FlashUNC]

Forgive me...I'm also a layman. I understand that there was some government mandate to implement some safety features within a fixed period of time. I have no idea what those features are...other than that they're apparently expensive. I don't see the need for something expensive, it seems like basic GPS technology would solve this simply and inexpensively.

From what I've read, it starts to get complicated when you're talking implementing control in the cars, an over-ride to what the engineer is doing, ensuring compatibility across a variety of locomotives, track systems, etc etc.

The guts, to your point, sound eminently feasible. And expensive in normal people dollars? Yes. But in terms of what the federal government spends in a given year? No, not all that much.

[Peter B]

Here are a few things any prudent railway management structure will be dealing with:

Define the system and its functionality (regions, routes, trains, technology, testing, training, maintenance, etc).
Identify and validate the applicable system technologies and component sourcing.
Identify signal carriers, ensure bandwidth access, transmission integrity and security.
Establish and document system and component performance benchmarks.
Ensure cross-compatibility with all existing infrastructure systems, technology, processes and procedures or modify/amend as required.
Develop and document a quality control program to ensure strict compliance with all regulatory and project requirements to ensure the public safety.
Develop and document comprehensive, detailed specifications for functionality and performance for all aspects and components of system.
Develop and document detailed drawings and diagrams for installation, wiring, piping/plumbing, programming and implementation across all train models and types in fleet (based on regional criteria as applicable).
Develop and document testing programs (pre and post implementation).
Develop and document user-training programs.
Develop and document maintenance procedures.
Develop and document written processes and procedures for system operation.
Determine implementation path; which aspects of implementation shall be contracted via competitive bidding, completed using in-house forces and/or out-sourced labor.
Develop bidding documents and procedures for all regions, routes and trains included in program.
Review all reasonably foreseeable legal aspects of system implementation.
Negotiate with labor representatives where required.
Determine roll-out plan(s)--system-wide, regional, phased, etc.
Develop and document implementation schedules.
Develop, document, implement and manage cost estimating and accounting structures.
Ensure adequate staffing.
Develop funding proposals accordingly.
Secure funding.
Negotiate and award contracts.
Manage and administer contracts.
Manage and implement installation across all regions, railways and trains in fleet.
Manage and implement change requirements.
Manage and implement testing, training and maintenance programs.


Remember this is a key transportation system subject to the Transportation Security arm of the Department of Homeland Security regulation/oversight. In FY2013 Amtrak moved some 31.6 million passengers. Every day, an average of more than 86,000 passengers ride more than 300 Amtrak trains, serving more than 500 destinations in 46 states, the District of Columbia and three Canadian provinces on more than 21,300 miles of routes, with more than 20,000 employees.

This is not a maker-based project you'll pull off with an Arduino and a few sensors, relays and interfaces although that is the essence of what will ultimately run the system.

[fkelly]

Automated controls will be good some day, as they can be tested and implemented. There is a plan in place for that.

That said: why does no one mention 110 mph in a 55 mile zone as a factor. If this is confirmed the engineer should be going to jail for life for murder.

I'm pretty sure there was a fairly recent crash in the Westchester area where it was 80 something mph in a 30 something zone.

It's crazy that this goes on. The trains have black boxes already. Don't they record speed and location? Does any one monitor the data, even retroactively and take action on engineers who are speeding. When I upload ride data to ridewithgps from my $400 Garmin it shows my speed at every point on every ride. Why can't Amtrak do that and have a program that automatically reads the data for all completed routes and flags the speeders for action? This would not cost billions or require wireless spectrum or anything else complicated. If engineers knew they were going to get flagged, fined and even fired every time they exceeded the speed limit I bet that behavior would disappear pretty quickly.

[oldpotatoe]

Can we stick to the question and not make this a debate about the merit of funding Amtrak? We have a rail system...regardless of your opinions about subsidizing it, i think we all agree that if we have one, it should be safe.

So why is it difficult and expensive to implement what i'd think are simple, cheap, readily available safety measures? If the answer is it's hard, explain why it's hard...because I'm not seeing it.

Actually in people moved, way safer than the airlines.. This the first type, derailment, loss of life in something like 25 years, in the US. 1993. And in terms of people and equipment, way bigger than the airlines or truck/highway 'system'.

But I agree, some sort of speed limiting technology is needed.

http://www.infoplease.com/ipa/A0001450.html

[93legendti]

Automated controls will be good some day, as they can be tested and implemented. There is a plan in place for that.

That said: why does no one mention 110 mph in a 55 mile zone as a factor. If this is confirmed the engineer should be going to jail for life for murder.

I'm pretty sure there was a fairly recent crash in the Westchester area where it was 80 something mph in a 30 something zone.

It's crazy that this goes on. The trains have black boxes already. Don't they record speed and location? Does any one monitor the data, even retroactively and take action on engineers who are speeding. When I upload ride data to ridewithgps from my $400 Garmin it shows my speed at every point on every ride. Why can't Amtrak do that and have a program that automatically reads the data for all completed routes and flags the speeders for action? This would not cost billions or require wireless spectrum or anything else complicated. If engineers knew they were going to get flagged, fined and even fired every time they exceeded the speed limit I bet that behavior would disappear pretty quickly.
A forumite above scolded us for daring to mention labor. Apparently that's off limits.



"Amtrak’s largest expense is labor, salary, and benefits, which cost over $2 billion in 2014. Maintaining fully-staffed trains on infrequently-traveled routes has contributed to high labor costs, but the pay rate of Amtrak’s employees raise its costs substantially. The average onboard employee made $41.19 an hour on Amtrak in 2012, while railroads that contracted out services to private companies paid their employees $7.75 to $13.00 an hour."
http://www.economics21.org/commentary/PRRIA-2015-amtrak-reform-subsidies-privatization-2015-03-06

[FlashUNC]

A forumite above scolded us for daring to mention labor. Apparently that's off limits.



"Amtrak’s largest expense is labor, salary, and benefits, which cost over $2 billion in 2014. Maintaining fully-staffed trains on infrequently-traveled routes has contributed to high labor costs, but the pay rate of Amtrak’s employees raise its costs substantially. The average onboard employee made $41.19 an hour on Amtrak in 2012, while railroads that contracted out services to private companies paid their employees $7.75 to $13.00 an hour."
http://www.economics21.org/commentary/PRRIA-2015-amtrak-reform-subsidies-privatization-2015-03-06

I never said any such thing. Don't make some wild abstraction out of my argument.

The engineer should absolutely pay for what he's done. But your posts imply that labor is the sole cause of the ills of the Amtrak system -- from safety to food costs.

Which is absurd on its face. Blaming one group for the woes of Amtrak and passengee rail in this country is resuctivist.

[shovelhd]

I'm not blaming union labor for anything related to this incident. It is a factor of the overall budget, which can affect safety improvements. $21.00/hour to print tickets.

[DreaminJohn]

I'll lay off the political stuff.


Something that no one's mentioned yet: there are times when you want/need the train to go 110 mph. Making the restrictions work only in specific areas is a challenge of technical and logistical areas.

[fuzzalow]

^ No, labor costs have absolutely no effect as a hindrance to safety improvements as a factor in the overall budget. The labor and operating expenses are separate and distinct from the capital budget items such as infrastructure improvements. Safety, speed control and accident warning systems are clearly capital improvements. And as many have already pointed out, Amtrak already operates under a deficit.

Villianize labor is absurd. Populist rage at a succeedingly lowere and lower bar of worker classes driven solely by the false perception that they earn an unfair wage doing trivial, non-taxing labor. This is not conservatism. It is not patriotic. It is blind resentment directed at anyone perceived to be doing better than you.

[ptourkin]

Let's talk some more about how public mass transportation doesn't work:

http://i130.photobucket.com/albums/p250/cecilhayduke/11222588_1097561906926480_7635780531943389950_n.jp g

[Aaron O]

I'll lay off the political stuff.


Something that no one's mentioned yet: there are times when you want/need the train to go 110 mph. Making the restrictions work only in specific areas is a challenge of technical and logistical areas.

I'm not arguing with you when i say this...I just don't understand the challenge...

How is this difficult? I'd think GPS solves this with ease. It sounds like the only real challenge here is setting up a modern switch/control on antiquated equipment.

[makoti]

I'm not arguing with you when i say this...I just don't understand the challenge...

How is this difficult? I'd think GPS solves this with ease. It sounds like the only real challenge here is setting up a modern switch/control on antiquated equipment.

Then your answer is funding. If that's it, then with enough funding it can be done. Either pay to create systems that work with the equipment you have or pay to update that equipment.

[goonster]

I'm not arguing with you when i say this...I just don't understand the challenge...

How is this difficult? I'd think GPS solves this with ease.
I'm not so sure that GPS provides the reliability required of an automated safety interlock system. When my car loses the satellites because of tunnels, trees, or whatever, it is a minor inconvenience. When it fails to stop a speeding train, it is a different sort of problem.

As others have said, this is not a problem of technology, it is a problem of funding and long-term commitment.

In a similar vein, we know perfectly well how to fix all those crumbling bridges, but we have chosen not to . . .

[josephr]

Don't forget the union factor. Less train control by union employees is bad for union employees.

same reason why they still deliver mail on Saturday.

[Aaron O]

I'm not so sure that GPS provides the reliability required of an automated safety interlock system. When my car loses the satellites because of tunnels, trees, or whatever, it is a minor inconvenience. When it fails to stop a speeding train, it is a different sort of problem.

As others have said, this is not a problem of technology, it is a problem of funding and long-term commitment.

In a similar vein, we know perfectly well how to fix all those crumbling bridges, but we have chosen not to . . .

You could definitely be right...I might be overvaluing GPS in the application. My thought pattern was that the GPS automated control was a backup redundant control to the engineer...so if the engineer failed, there's a very strong likelihood that the GPS will work.

[soulspinner]

You could definitely be right...I might be overvaluing GPS in the application. My thought pattern was that the GPS automated control was a backup redundant control to the engineer...so if the engineer failed, there's a very strong likelihood that the GPS will work.

Its not a tecno wonder, the military finds the money for such things.

[goonster]

My thought pattern was that the GPS automated control was a backup redundant control to the engineer...so if the engineer failed, there's a very strong likelihood that the GPS will work.
That kind of thing is trickier than you might think.

I was just reading about Aeroflot 593 (http://en.wikipedia.org/wiki/Aeroflot_Flight_593), where operator input (by a child that should not have been there) counteracted the autopilot long enough to disengage it, and in the ensuing confusion the pilots overrode the autopilot's response, when leaving it to do its thing may have saved the aircraft.

Implementing a "backup redundant control" to a human operator can be extremely complex in practice. While such a system would be much simpler on a train than in aviation, there are all sorts of possible unintended negative consequences.

Some of these issues are considered in this truly excellent article (http://www.vanityfair.com/news/business/2014/10/air-france-flight-447-crash), from which I'd like to pull this maxim coined by a prominent engineer:

"Whenever you solve a problem, you usually create one. You can only hope that the one you created is less critical than the one you eliminated."

[unterhausen]

since the tech isn't there now, it would be a major undertaking to implement it. In a previous life, I used to be responsible for implementing safety modifications on airplanes. It was really difficult and expensive to do that for very simple changes.

[Aaron O]

That kind of thing is trickier than you might think.

I was just reading about Aeroflot 593 (http://en.wikipedia.org/wiki/Aeroflot_Flight_593), where operator input (by a child that should not have been there) counteracted the autopilot long enough to disengage it, and in the ensuing confusion the pilots overrode the autopilot's response, when leaving it to do its thing may have saved the aircraft.

Implementing a "backup redundant control" to a human operator can be extremely complex in practice. While such a system would be much simpler on a train than in aviation, there are all sorts of possible unintended negative consequences.

Some of these issues are considered in this truly excellent article (http://www.vanityfair.com/news/business/2014/10/air-france-flight-447-crash), from which I'd like to pull this maxim coined by a prominent engineer:

"Whenever you solve a problem, you usually create one. You can only hope that the one you created is less critical than the one you eliminated."

This is exactly why i said at the outset that I don't know enough to know what I don't know ;)

[oldpotatoe]

Its not a tecno wonder, the military finds the money for such things.

No GPS guided vehicles ala a train. Navigation and terminal guidance for smart weapons but not speed control or the like. Even drones have a guy flying them.

[Plum Hill]

Look for an article titled Is PTC Going To Work in the November 2014 issue of Trains magazine (check with a local library).

Also: http://www.progressiverailroading.com/ptc/article/Positive-train-control-implementation-chugs-along-despite-challenges--40013 .

[shovelhd]

^ No, labor costs have absolutely no effect as a hindrance to safety improvements as a factor in the overall budget. The labor and operating expenses are separate and distinct from the capital budget items such as infrastructure improvements. Safety, speed control and accident warning systems are clearly capital improvements. And as many have already pointed out, Amtrak already operates under a deficit.

Villianize labor is absurd. Populist rage at a succeedingly lowere and lower bar of worker classes driven solely by the false perception that they earn an unfair wage doing trivial, non-taxing labor. This is not conservatism. It is not patriotic. It is blind resentment directed at anyone perceived to be doing better than you.

You and I subsidize Amtrak. As you say it operates at a deficit. You can use all the GAAP arguments you want. The bottom line is that the bottom line is the bottom line. They can choose to pay ticket button pushers twice what a Target cashier makes, drive up that deficit, and leave less for everything else. That's their prerogative but since I'm paying for it I can give my opinion. Just so you don't think I'm some kind of supply sider, which you probably do anyway, I believe train engineers are subject to many of the same responsibilities as airplane pilots and should be paid similarly.

[zap]

I'm not an engineer, I am a know-nothing that likely just doesn't know what I don't know. This is a bright bunch, and I'm guessing there will be someone who can answer this.

I'm not getting why putting safety features on Amtrak would be complex or expensive. It seems to me that all you have to do is have a gauge that reflects the train's speed (or this can be done via GPS), which they already have. Then all you need is a program of some sort that shows the speed limits along the rail and tracks the train's location (GPS)...and if the speed exceeds the limit by a certain amount, the brake is engaged.

2015 cars all have more complicated features than that. My pohone's GPS can do all of this except automatically engage the brake, but I can't imagine that's especially difficult.

So what am I not getting?

My father has at least one patent in this arena. He and his firm sold and installed several gps based safety systems to local metro systems around the USA. Amtrak-I know my father met with Pelosi, Biden and others at least once but at the time ('09), there was not enough will to allocate funds for this safety program.

[fuzzalow]

Just so you don't think I'm some kind of supply sider, which you probably do anyway, I believe train engineers are subject to many of the same responsibilities as airplane pilots and should be paid similarly.

No shovelhd, I don't think anything about, or draw any conclusions on, anything to do with you political proclivities or inclinations. I only go by what you post. Really. You could extol the virtues of a Marxist/Leninist dialectic in your next post and I'd go with what you wrote. It doesn't matter to me, I just like a good argument! :)

I only go by what anybody posts. Really. I prefer that they don't make a tsk and storm off in a huff like some other guy did when I pointed out he was using patter for ATH when he wasn't talking to people ATH.

[bironi]

I saw the former head of the NTSB this weekend on TV. He said it was a very expensive investment. Congress mandated the change, buy it had taken longer to implement.

[soulspinner]

No GPS guided vehicles ala a train. Navigation and terminal guidance for smart weapons but not speed control or the like. Even drones have a guy flying them.

its not air travel though, the trains path is predetermined, just a gps speed monitor. I know the cost of such things in the weaponry world and it has become cheap. Why for a train its so difficult to integrate.....

[AJosiahK]

financial cost weighted against effectiveness, or really, its too exspensive to ensure accidents like these wont happen in the future.

Had the fortune (good or bad not sure) to meet an amtrak employee. He was dating a friend of mine a while back. some of the stories he had about his personal experiences and the fun they had, really worried me.....

[Plum Hill]

On May 19, the House of Representatives passed HR 2352. The bill extended the Highway Trust Fund authority to finance rail, road, bridge, and other surface transportation from May31 to July 31. Vote was 387-35.(Wow, a whopping two months.)

On May 19, the House defeated a Democratic bid to set aside $750 million in HR 2353 to pay for providing Positive Train Control for passenger trains. Vote was 182-241, basically along party lines.

http://www.miamiherald.com/opinion/editorial-cartoons/jim-morin/imfyp4/picture21007854/alternates/LANDSCAPE_400/colormorin0515

[rounder]

I do not know whether PTC would have prevented the wreck. But I understand that Congress had already approved it and it was being installed along the Eastern AMTRAK line but had not been installed at that curve yet.

I have ridden AMTRAK/MARC daily for almost all of the past 20 years. The AMTRAKs go through the Edgewood station where I board at about 110 if there is no traffic, but goes much slower between Baltimore and Washington because of all the speed limits along the way.

Obviously, the engineer would have known this and was traveling about 105 in a 50 mile zone. PTC might have slowed him down if it was in effect, But why should the system be expected to prevent that type of accident when the engineer was not only driving twice the legal speed limit, but also accelerating at the time.

To me, the problem is not why PTC had not been installed at the time, but why was the engineer driving so fast.

[Plum Hill]

To me, the problem is not why PTC had not been installed at the time, but why was the engineer driving so fast.

Good question, to which there is no answer at this time. The cost cutting by all railroads to eliminate an extra person in the engine could have presented the disaster.
Airlines are required to have two people in the cockpit at all times. Railroads aren't.

[adhumston]

Good question, to which there is no answer at this time. The cost cutting by all railroads to eliminate an extra person in the engine could have presented the disaster.
Airlines are required to have two people in the cockpit at all times. Railroads aren't.

This part is something that is often overlooked. All of the Tier 1 railroads are pushing for single person crews. This IS NOT something the general public wants (whether or not they're aware of it). A 2 man crew could have possibly prevented the Amtrak incident (if in fact it ends up being the Engineers fault).

[Plum Hill]

Darned IPad. I meant 'prevented', not 'presented'.

[fkelly]

Positive train control, two engineers in the cab etc.: all possibly good long term but expensive solutions. Cost benefit versus additional track maintenance, bridge maintenance also not known.

I have hundreds of rides stored on ridewithgps. For 2011 into 2015 I can tell at virtually any point in those rides how fast I was going. I have dozens of segments defined and can compare my times and speeds versus anyone else who has ridden those routes and uploaded their rides. $400 Garmin device and account ($50 per year or so) on rwgps required. Upload of a ride takes maybe a minute. Iphone or Android phone with app loaded will also work.

Have Amtrak buy each engineer and account. Define segments on each route for the ten (or twenty or ... ) worst curves where the speed limit should be lowest. Require each engineer to upload each route after driving it. Or automate the process. Have someone in management monitoring it. Booby prize is for driving the segment above the speed limit. Fine offenders. Fire consistent offenders. I bet you wouldn't have consistent offenders if the engineers knew monitoring was taking place.

Cost probably 1/100th to 1/1000th of the other solutions. No, it wouldn't be positive train control but I'd bet it would be 99% as effective.

[staggerwing]

Probably wouldn't even have to go that far. From what I understand, the locomotives are 'black boxed' just like commercial aircraft, which suggests that, at the very minimum, speed and time are recorded. Wouldn't be surprised if there is something in the mix that also encodes mile marker or track position. It would be a simple matter to have a duplicate data stream encoded to another recording device for automatic daily monitoring.

I imagine the big issue with PTC is one of redundancy, and validation. Putting such a system in place basically says you longer trust a human to make the final decision. If that is the case, the automated system better be 18 levels of infallible.

[fkelly]

I was aware that the trains were black boxed. However, what's recorded in the black box is itself a black box to me. I doubt Amtrak makes that publicly available. Yes, if if the black box that already exists has the pertinent information then monitoring it CONSISTENTLY would be the most direct approach.

All I'm suggesting is that there are ways to solve this problem, to a large degree at least, that don't involve spending millions or take years. There is just no excuse for engineers to be running the trains at 110 mph in a 50 mile zone or for management to let that take place.