PTC quote

[blownout cylinder]

Randy Stahl

edblysard

Do you have to carry the knuckle uphill, barefoot in the snow?

Of course .. you expect me to wear pumps ?

Hi-heels....

[Murphy Siding]

schlimm

Once again certain folks make a thread ridiculous because they don't like the topic.

     Once again certain folks take life waaay to serious, because other certain folks don't follow their rules.  Lighten up man.

[zugmann]

schlimm

Once again certain folks make a thread ridiculous because they don't like the topic.

That's ridiculous!

[schlimm]

Once again certain folks make a thread ridiculous because they don't like the topic.

[Randy Stahl]

tree68

Randy Stahl

All I'm getting out of this thread is that my next train might get hit by an asteroid while I'm changing a knuckle in a snowbank.

Snowbanks have knuckles?

 

Yes, snow banks have knuckles.. we find them with the snow plows. Makes quite a racket.

[zugmann]

Randy Stahl

Of course .. you expect me to wear pumps ?

Only if you have the legs for it.

[edblysard]

Randy Stahl

edblysard

Do you have to carry the knuckle uphill, barefoot in the snow?

Of course .. you expect me to wear pumps ?

Now I know,  you were a ground pounder/switchman at one time, thats the answer I was expecting! 

[tree68]

Randy Stahl

All I'm getting out of this thread is that my next train might get hit by an asteroid while I'm changing a knuckle in a snowbank.

Snowbanks have knuckles?

[Anonymous]

The fact that the cost exceeds the benefit of this PTC mandate naturally opens the door to criticism.  And since nobody wants criticism, I have to believe that the cost estimate has been made to be as low as possible.  In reality, I expect the cost to be unlimited because the work will go on forever.   The mandate deadline time frame will be subdivided in order to cope with the impossibility of meeting it in one fell swoop. 

Subdividing the timeframe will amount to subdividing the mandate, and this process will continue ad infinitum.  Mandates will have children.  But the key point is that the whole process will be compulsory upon the railroads, leaving them no choice but to pay for it.  That is one big reason why the process will never end.       

[Randy Stahl]

edblysard

Do you have to carry the knuckle uphill, barefoot in the snow?

Of course .. you expect me to wear pumps ?

[edblysard]

Do you have to carry the knuckle uphill, barefoot in the snow?

[Randy Stahl]

All I'm getting out of this thread is that my next train might get hit by an asteroid while I'm changing a knuckle in a snowbank.

[Anonymous]

Bucyrus

Sam1,

I can see how the probable benefit could be predicted by statistical models that project the future based on the past.  But I would say that it would be far more difficult to predict the cost. 

Predicting the cost would be largely a matter of speculation and luck.  The principle of a mandate and deadline injects a wildcard into the normal free market economics.  The fact that it has already been established and accepted that the cost will exceed the benefit is evidence of that wildcard. 

But perhaps another way of looking at the unpredictability of cost is that it naturally results from the undefined endpoint of the objective and scope of the project. 

Predicting cost and benefits, especially for more than a year or two, is dicey.  Most of the corporate planners and regulators that I worked with acknowledged that coming up with an accurate prediction for costs and benefits for more than a year or two is more about playing to the audience than recognizing the constraints associated with accounting and finance predictions.   

[Anonymous]

Sam1,

I can see how the probable benefit could be predicted by statistical models that project the future based on the past.  But I would say that it would be far more difficult to predict the cost. 

Predicting the cost would be largely a matter of speculation and luck.  The principle of a mandate and deadline injects a wildcard into the normal free market economics.  The fact that it has already been established and accepted that the cost will exceed the benefit is evidence of that wildcard. 

But perhaps another way of looking at the unpredictability of cost is that it naturally results from the undefined endpoint of the objective and scope of the project.        

[oltmannd]

oltmannd

From a recent G. Will column:

"

Before returning to Harvard Law School, Cass Sunstein was Barack Obama’s administrator of the Office of Information and Regulatory Affairs, measuring the benefits of regulations against their costs. Testifying to a House subcommittee on Jan. 26, 2011, Sunstein was asked if he could identify an administration regulation whose ―benefits have not justified the cost.‖ He replied:
―There is only one big one that comes to mind. It is called Positive Train Control, and it is a statutory requirement, and the Department of Transportation had to issue it as a matter of law even though the monetizable benefits are lower than the monetizable costs. There aren’t a lot like that.‖
Concerning Sunstein’s sanguine"

So, why didn't the RRs fight against it harder?  I have a couple of thoughts.  What are yours?

One of the Congressmen in the PTC hearing yesterday quoted Will quoting Sunstein.

[tree68]

schlimm

Head on collisions?  Who said that's all we can look at?   Running into the side of another train or the rear of another train also count and could also be prevented.  

Of course, "cornfield meets" are the grandaddy of all collisions.  I suspect that Ed's point is that such catastrophic events are extremely rare, when considering the number of operations that occur on railroads during a given period. 

F'rinstance, just considering the line that runs through my area, there were approximately 3000 train starts over a typical two year period (and maybe more).   That 15 collisions, applied only to this line, means that .5% percent of the trains would have been involved in such a collision, or one in 200.  Raise that number to 30,000 and 15 starts to become a relatively insignificant number.  If you raise that to 3000 starts a day and you have over two million starts - still with just 15 collisions.

That's one incident in 133,333.

To my knowledge, there were no purely rail adverse events on this line during that period, even including derailments.  Pedestrians aren't covered by PTC.

The last major incident on the line - the Fort Drum runaways - would not have been prevented by PTC - there was no locomotive or EOT involved.  Just two "dark" cars.

[Anonymous]

Bucyrus

oltmannd

It is called Positive Train Control, and it is a statutory requirement, and the Department of Transportation had to issue it as a matter of law even though the monetizable benefits are lower than the monetizable costs.

How does one predict what the benefit will be without knowing the cost of the accidents that would have occurred without PTC?

How does one predict what the cost will be when the R&D has not been entirely completed?

Why is it a statutory requirement if the cost exceeds the benefit?

What is the measured quantification of the benefit? 

Predicting costs and benefits is doable, but it is important to understand the constraints. This is especially true for predictions going out more than a couple of years.

The key to predicting costs and benefits is to begin with an appropriate database. Another key is to use the appropriate statistical modeling techniques.  If the database is shallow and/or the statistical techniques are inappropriate, the results will be wrong.  Every statistical technique (method) has inherent weaknesses.  If the researches don't understand them, they are likely to generate the wrong outcome.

Each year for grins I read the President's (OMB) proposed federal budget.  I read all of the summary tables, as well as the support schedules for transportation, at a minimum, and a significant portion of the narrative.. I also read the Social Security and Medicare Trustee's annual reports. These documents contain projections for 10 years or more.  From one year to another the projections, especially those for more than five years, can swing as much as 20 per cent. And that is just over one year.

When someone predicts the costs and benefits of a proposed project as a number, it would only be accurate coincidentally.  An honest researcher projects a range of numbers, together with the parameters baked into his or her statistical model.  

Bottom line?  View all statistical projections (costs and benefits) with a skeptical eye.

[schlimm]

edblysard

From March, 2011 to march 2013, there have been 15 head on collisions, not all involving train on train collisions, of that group only 7 were train on train, 4 due to failure to observe restricted speed, 3 due to a improperly lined switch.

The rest involved Hi rail trucks exceeding track authority, blind shoves on main track, and what the FRA termed “other” motor vehicle exceeding authority, 1 still under investigation.

So we are basically looking at 7 collisions that might have been prevented by PTC.

Head on collisions?  Who said that's all we can look at?   Running into the side of another train or the rear of another train also count and could also be prevented.  

[edblysard]

schlimm

Actually, the phrase in question, from Paul North,  was "Huge asteroids striking the earth and annihilating us all" but I guess that got lost in your sea of sarcasm.  

But the real point (which the railroaders seem to be in denial of), is there are many collisions, many of which are preventable by PTC.  Why the fear?  Thinking this leads to remote or auto pilot trains on the road?   Nowhere else that has advanced train control has moved in that direction.

http://www.ntsb.gov/investigations/2013/scott_city_mo/Chaffee_MO_10-Day_Preliminary%20Report_20130607.pdf

From March, 2011 to march 2013, there have been 15 head on collisions, not all involving train on train collisions, of that group only 7 were train on train, 4 due to failure to observe restricted speed, 3 due to a improperly lined switch.

The rest involved Hi rail trucks exceeding track authority, blind shoves on main track, and what the FRA termed “other” motor vehicle exceeding authority, 1 still under investigation.

So we are basically looking at 7 collisions that might have been prevented by PTC.

7 out of how many thousands of train starts in that same 3 year period.

Bet if you looked, you would find that there are more aircraft running into each other on run way incursions than trains running into other trains.

	Total		Total Year Counts			YTD Counts Jan - Mar		% Change Over Time
	Accs	Pct of Total	2010	2011	2012	2012	2013	2010 to 2012	2011 to 2012	To Mar 2012 2013
-----GRAND TOTAL.......	15	100.0	3	5	6	3	1	100.0	20.0	-66.7
H607 Failure to comply with restricted speed	4	26.7	.	2	2	1	.	.	.	.
H702 Switch improperly lined	3	20.0	1	.	1	1	1	.	.	.
H021 Fail to apply car hnd brks -rr emp	1	6.7	.	1	.	.	.	.	.	.
H221 Automatic block or interlocking signal d	1	6.7	1	.	.	.	.	.	.	.
H306 Shoving movement, absence of man	1	6.7	.	.	1	.	.	.	.	.
H402 Motor car/on-trk rules, fail to comply	1	6.7	.	1	.	.	.	.	.	.
H404 Fail to comply with trn order, etc.	1	6.7	1	.	.	.	.	.	.	.
H499 Other main track authority causes	1	6.7	.	1	.	.	.	.	.	.
M504 Fail by non-rr empto control spd of car	1	6.7	.	.	1	1	.	.	.	.
M505 Cause under investigation	1	6.7	.	.	1	.	.	.	.	.

 

Railroaders are not in denial….we simply work with this stuff every day, and have a good grasp on what will work and what will not.

PTC is one of the not too bright ideas…the money could be better spent and obtain much better results in other areas of our industry.

 

[BaltACD]

blue streak 1

blue streak 1

As others have posted  --  IMHO the NTSB warned the RRs that they were pushing PTC long before the Ca collision.  But they ignored the calls and instead of expanding cab signaling, ATC, & ATS dropped them when the ICC & later the FRA  allowed dropping.   IC is an example.

So they are unfortunately having to pay the price.;

 As a follow up  --  before the Chatworth accident was any RR other han NS planning or actually expanding any of the mentioned upgrades ?  We know NS is putting cab signaling on at least one route and maybe more ?

Any other RRs?  Metrolink has added ATS on all their tracks after Chatsworth.

 

Prior to the Chatsworth incident, CSX was working on what they termed CBTC - Communications Based Train Control.  A real world 'test bed' had been implemented on the Blue Ridge Sub (former Clinchfield) and was involved in day to day testing.

Some of the elements of CBTC remain existant in PTC as it is currently being developed.

[blue streak 1]

blue streak 1

As others have posted  --  IMHO the NTSB warned the RRs that they were pushing PTC long before the Ca collision.  But they ignored the calls and instead of expanding cab signaling, ATC, & ATS dropped them when the ICC & later the FRA  allowed dropping.   IC is an example.

So they are unfortunately having to pay the price.;

 As a follow up  --  before the Chatworth accident was any RR other han NS planning or actually expanding any of the mentioned upgrades ?  We know NS is putting cab signaling on at least one route and maybe more ?

Any other RRs?  Metrolink has added ATS on all their tracks after Chatsworth.

 

[Paul_D_North_Jr]

There's a pretty good and good-sized article about this whole subject - written around the context of SEPTA and the other commuter carriers - in yesterday's (Tuesday, 18 June 2013) Wall Street Journal, starting 'below the fold' on page A-1 and continuing on page A-12 (I believe).  It has a photo of the Bridgeport viaduct/ trestle (apparently in the Norristown area - the article doesn't say whether it's on the High Speed Line [ex=P&W interurban], or the Regional Rail line [ex-Reading RR] ) that SEPTA says it will have to close next month because of a lack of money to both install PTC and maintain existing infrastructure - a bridge, in this case.  The article is titled "Rail Safety and the Value of a Life" by Ted Mann, and an on-line version appears to be 'datelined' as June 17, 2013.  Unfortunately, it appears to be available on-line only to subscribers (which I'm not, either), so for those who are interested I suggest the nearest public library.

- Paul North.   

[BaltACD]

blue streak 1

As others have posted  --  IMHO the NTSB warned the RRs that they were pushing PTC long before the Ca collision.  But they ignored the calls and instead of expanding cab signaling, ATC, & ATS dropped them when the ICC & later the FRA  allowed dropping.   IC is an example.

So they are unfortunately having to pay the price.;

While the NTSB has been pushing things analogus to PTC for years, it is really only within the last decade that the carriers financial situation has warrented there being a mandate.  Prior to 2000, while the carriers showed profits, they were not solid 'bankable' profits as there were concessions made from both the maintenance and investment sides of the business to insure that the stockholders were getting a return. 

In today's financial world, the carriers are earning solid profits while putting record amounts into maintenance and plant enhancement projects.  With that kind of financial footing, the carriers did not possess much of a 'financial defense' against the mandate, and they knew it and thus there was limited complaining about the mandate.

[oltmannd]

schlimm

blue streak 1

As others have posted  --  IMHO the NTSB warned the RRs that they were pushing PTC long before the Ca collision.  But they ignored the calls and instead of expanding cab signaling, ATC, & ATS dropped them when the ICC & later the FRA  allowed dropping.   IC is an example.

So they are unfortunately having to pay the price.;

Shortsightedness at its worst.

Perhaps they stopped listening because the NTSB/FRA had been droning on so long about it, it was falling on deaf ears.

[tree68]

Paul_D_North_Jr

...same points could be said about a requirement to install fire sprinklers in your home, or a business, or a social club, etc.  Suppose the cost would be $11,000 for each $500 of benefits, such as fire insurance premium savings, and/ or the avoided future costs of fire damage.  More scaled up, having to spend $2.2 million for each $100,000 of savings in potential fire damage - when the house itself is worth only $300,000.  How would you - or anyone else - feel about that ?

Oddly, the biggest opponents of the sprinkler requirement are the building trades organizations.  It's been opined that they'd rather see the house burn down (as new construction is wont to do) so they can build a new one...

That would be like the car builders (Greenbrier, et al), locomotive builders, and track construction companies coming out against PTC, because it might mean less business for them.

[schlimm]

blue streak 1

As others have posted  --  IMHO the NTSB warned the RRs that they were pushing PTC long before the Ca collision.  But they ignored the calls and instead of expanding cab signaling, ATC, & ATS dropped them when the ICC & later the FRA  allowed dropping.   IC is an example.

So they are unfortunately having to pay the price.;

Shortsightedness at its worst.

[blue streak 1]

As others have posted  --  IMHO the NTSB warned the RRs that they were pushing PTC long before the Ca collision.  But they ignored the calls and instead of expanding cab signaling, ATC, & ATS dropped them when the ICC & later the FRA  allowed dropping.   IC is an example.

So they are unfortunately having to pay the price.;

[Anonymous]

In thinking about PTC leading to automated operation, I went back and reviewed another thread from last year about crewless running.  It has become one of the prominent references on the topic when you do a Google search.

However, I am puzzled by something.  Why are many of the posts in this thread out of sequential order?  When reading the thread the way it is displayed, comments quoted in some posts are quoted before the comments were actually made.

The dates of the posts clearly show that they are listed out of order in the thread.  As a reference resource, its scrambled sequence makes much of it meaningless. Why have the posts gotten out of sequence? 

Look at it here:

http://cs.trains.com/trn/f/111/t/203077.aspx?sort=DESC&pi332=1

[schlimm]

Found this on the British experience:

The Ladbrooke Grove crash(1999, 31 killed, 520 injured) prompted the roll-out of TPWS acorss the network, which brings trains to a halt if they try and pass a signal at danger (or pass certain point too fast). TPWS doesn’t guarantee the train will stop before the signal – it may stop in the area beyond the signal (UK signals are always some distance in front of the area they protect, so that small over runs pose no safety risk). Every fatal train accident in the UK since 1999 has been caused by track problems rather than signalling issues.

UK: for the period 1991 to 2009, (includes the pre-Ladbrooke period of 9 years)  one fatality for every 10.96 bn pass-km

UK: for 2000-2009, (since TPWS) the figure is one fatality for every 36.66 billion passenger-km.

US: for 1990-2010 one death per 3.4 billion passenger-km.

Two lessons: 1. Highly publicized accidents with a considerable loss of life have prompted government action elsewhere.  2.  Our death rate is very high by comparison.

[Anonymous]

Here is a good article and conversation about the PTC controversy:

http://www.theverge.com/2013/4/15/4226264/positive-train-control-controversy

Quote from the article makes a good point about deaths in train wrecks versus deaths in car accidents:

It’s difficult, however, to argue against a technology that could save lives. A passenger on a derailed train in 2003 recently told the Chicago Tribune, "I think somebody’s life is a lot more important than dollars."

But Banks isn’t so sure.

"Yes, I think the railroads should be as safe as possible," he said. "But it’s insane to force railroads to spend billions in private capital on this project when less than 40 people have died in the last decade in [U.S.] rail collisions while nearly 40,000 people get killed on the highways every year."