This argument about safety of automatic trains is mostly a twin red herring being used by both sides. Safety is not the issue. Compared to highway deaths, the railroads are not widely perceived as having a safety problem. And most people believe that PTC is the magic bullet for perfect safety.
The real point of automatic trains is to eliminate crew costs, and that includes far more than just employee wages. It also includes the costs of hiring, benefits, liability, overtime, transportation, and lodging. The proper comparison is the “cost of employment,” not just wages.
But cutting these cost will be easily portrayed as corporate greed and hurting families. So it is not a winning position. There is also a tremendous gamble that the investment for automatic trains will not be higher than expected. Rio Tinto exceeded their cost estimate and ended up paying around a billion dollars for their system.
However, improving safety is the most powerful argument to advance anything. If the railroads had data showing that the automatic trains would improve safety, they would win the augment for automatic trains. If the unions had data showing that automatic trains compromised safety, it would kill the prospect of automatic trains forever. The problem is that neither side has the data.
The data is difficult to obtain because it requires proving that the sky would have fallen but for…”
However, there are statistics galore that could be worked into technical analyses and studies that will evaluate the safety matter with automatic trains. I expect a new industry will emerge whereby these various analyses will be sliced and diced, and seasoned with opinions and slants to form literature that will argue whether automatic trains increase or decrease safety.
In the isolated systems of automatic transit systems, some are automatic without operator, some without train operators but with one monitoring safety person who can shut the whole system down in a emergency, and, most frequent, automatic operation with a safety operator on-board that can revert to manual operation if necessary.
Again, these are closed systems, without grade crossings. But they do handle people, not freight. And people can try to hold doors open, have health problems, requires assistance, etc. Freight doesn't have these problems, usually.
York1 zardoz tree68 And that most accidents occur within 25 miles of home Which is why I moved..... tree68 safety advocates were pointing out that most fatal accidents occur at speeds under 45 MPH And to be even safer, quickly accelerate to 70 mph no matter where you are.
zardoz tree68 And that most accidents occur within 25 miles of home Which is why I moved.....
Which is why I moved.....
tree68 safety advocates were pointing out that most fatal accidents occur at speeds under 45 MPH
And to be even safer, quickly accelerate to 70 mph no matter where you are.
A quote from the show Gravity Falls:
"You know studies show that keeping a ladder inside the house is more dangerous than a loaded gun. That's why I own ten guns - in case some maniac tries to sneak in a ladder!"
- Grunkel Stan (created by Alex Hirsch)
It's been fun. But it isn't much fun anymore. Signing off for now.
The opinions expressed here represent my own and not those of my employer, any other railroad, company, or person.t fun any
It blows my mind that corprate America will sacrafice safety to make more profit when it comes to self running trains.
I've worked as an Engineer for 27 years now and can't believe what I've seen come to fruth in that amount of time.
I can recall my late father who worked 42 years on the railroad telling me "I've seen many thing happen on the railroad that old timers said would happen,I can only imagine what you'll see".
Wow how he was right.In the eleven years he's been gone he couldn't have even began to imagine what changes have happened.
But I still fill were some years away in "most" parts of this country as well as Canada for seeing non-occuppied trains roaming the landscape.
To many bugs still in the PTC system,too many cyber threats that can be made , and public as well as industry accepetance .
The threat to the American worker has got to stop somewhere also because we can't live on service related wages alone. Reason for that thought is the amount of hard-core living wages that are disappearing with automation is amazing. I mean can we really handle much more and pay taxes that fund our society ?
Corprate greed has got to be dealt with sooner or later or jobs that have been mainstays of generations of railroaders (I'm 4th) will soon be stories in books .
Collin ,operator of the " Eastern Kentucky & Ohio R.R."
mackb4The threat to the American worker has got to stop somewhere also because we can't live on service related wages alone.
I might point out that the costs of running manned trains are 'service related wages', whereas the considerable capital costs of providing autonomous capability and the infrastructure to support it are not.
Not that I disagree with the point that we're woefully far from making autonomous trains even technically advisable. And I'm involved in the development.
mackb4It blows my mind that corprate America will sacrafice safety to make more profit when it comes to self running trains.
It's been true for at least 150 years, inherent in the ideology. Look at how slow the rails were in adopting various safety appliances, such as Janney couplers, air brakes, etc. often needing government intervention in the pre-deregulation environment.
Overmod any job that does not manufacture or design a product "yes" are considered "service related" but railroad jobs,especially those on class ones's represented by unions ,labors,machinist,electricians,blacksmiths,boiler makers,carmen,conductors and Engineers are union based waged jobs that pay a liveable wage.
Take these jobs away and its less and less "good paying" jobs in the work force with less taxes being paid to the general fund of this country.
Just like social security,it's bound to run out unless we borrow more money from another country.
Zardoz that's a sad part of the mentality of this "new age" profiteers running companies into the ground.
zardoz mackb4 It blows my mind that corprate America will sacrafice safety to make more profit when it comes to self running trains. Lives and equipment are replaceable, and are merely "costs of doing business". As long as forgoing safety ensures more profit, this practice will continue, whether in the railroad industry, the chemical industry, the mining industry......
Lives and equipment are replaceable, and are merely "costs of doing business". As long as forgoing safety ensures more profit, this practice will continue, whether in the railroad industry, the chemical industry, the mining industry......
This is why its hard to explain to some people about the real world of and how it's seen from a railroaders perspective .
Live it and experience it and your mentality would change. I'm right on that I bet'ya
mackb4 This is why its hard to explain to some people about the real world of and how it's seen from a railroaders perspective . Live it and experience it and your mentality would change. I'm right on that I bet'ya
Most of us have a limited perspective of the organization where we work(ed), the industry, and the nation.
A UP engineer that runs a train from San Antonio to Taylor knows more about the route than anyone in the company except fellow engineers, conductors and possibly supervisors. But h/she does not have a top down perspective of the company as a whole, or the industry, etc.
I was not addressing safety on the railroad. I have never worked for a railroad, although I have been a serious observer for decades. I was reacting to the argument put forth by many that executive management in U.S. corporations would sacrifice the safety of its employees to please investors. The idea is simply wrong.
As I mentioned at the top of the page, safety is the issue over which this battle of automated trains will be fought. You make the case that automatic running will pose a safety risk. Others share that view and worry that if something should go wrong and nobody will be on board to react.
The opposing view is that automated trains will be safer because they will not be subject to human fatigue and the mistakes it can cause. But each side of this debate will have to come up with some form of proof to back up their claims.
mackb4 ... railroad jobs,especially those on 'class ones' represented by unions ,labors,machinist,electricians,blacksmiths,boiler makers,carmen,conductors and Engineers are union based jobs that pay a liveable wage.
They are, and it is very true that eliminating them contributes to cutting the heart out of the United States economy as we've come to understand it since exiting the Depression. The question here, though, is more like the discussions about eliminating the 'fireman' position when locomotives no longer depend on a dedicated person providing that job skill.
My own position on this is that the engineer needs to be 'outsourced' to the same sort of facility, perhaps indeed the same facility, that dispatchers are, and supervise the running of nominally 'autonomous' trains via telepresence rather than physically sitting in the seat, being called at wee hours, run around in a van, etc. This in conjunction with someone on the engine (corresponding now to the present 'conductor') who is capable of taking over control of the train in an emergency and perhaps supervise operating it to the nearest effective service or tie-up location, and if necessary perform things like post-UDE inspection or knuckle replacement if that is necessary and a 'wait' for the prospective hi-rail linemen or swooping service drone or whatever is not practical. That preserves most of the current jobs held by trainmen now, and achieves savings not by eliminating employment but by making its provision both easier and less costly.
We'll see people try to cut corners on train service, just as Ed Ellis did with MM&A. We'll see the corners un-cut when that proves not to work, and there are oh! so many places in a PSR-dependent world where it won't work when expected. There are places where a one-man crew may work quite well ... Amtrak and commuter carriers being examples ... but where the added safety from an enabled PTC system or autonomous oversight will add to, rather than replace, those employees.
We have to distinguish between those employees who are necessary and those who we should 'retrain' just as railroads did in the late 1940s when cheap labor and competing sources of 'living wage' destroyed the profitability of nearly any railroad dependent upon steam. Boilermakers and firemen are two examples, but far more important were the ranks of people in various service and support positions, who might have to wait 10 or more years just to advance some small degree, that become unnecessary even at the primitive wages railroads could pay.
The situation is radically different in some instances of PSR imposition, in particular the 'return' of flat switching at so many facilities, where more people are again required to do things right, but no additional hires (or enabling technology of the right kinds) are being provided. Again I'd like to think that sagging overall revenue would serve as a wake-up call for the 'capitalist stakeholders' controlling the ownership, to get them to look at why their assets' "execution" seems to be better but their equivalent of take-home not. There are many things in railroading that aren't amenable to 'automation' of the kind that has changed so many old-line production jobs, and while some of these can be made so with proper AI (much of which hasn't even been framed correctly) I suspect it will be beyond the lifetimes of most of the current hires in the industry that the actual job positions are eventually removed or allowed to wither via attrition.
Of course, you won't hear that from the likes of people touting present-day autonomous train solutions. Unless you ask them the harder questions.
Certain changes like the adopton of ECP brakes, or changing the gage, have to be completed in a very short time. Other changes like automatic trains or dieselization can take as long as necessary. I think railroads will evolve to a state where automatic trains and manual trains can be blended operationally. Along with this transition will come many other smaller changes such as more dedicated trains facilitated by a move to shorter trains, made possible by automation, and made attractive by their greater flexibility. All of this change will be under way for many years. I don't know if it ever reaches a state of completion, or just keeps evolving. But I think it is just begining now and will last for at least 40 more years. With the automatic trains component, there will surely be baby steps in the begining. I think that begins in maybe five years.
First Passenger Flight Directed by Remote ATC Tower
https://www.kongsberg.com/news-and-media/news-archive/2019/successful-first-landing-by-passenger-aircraft-using-remote-towers-in-the-arctic/
Thought this would fit here ...
Overmod Euclid So the self-driving trains can accomplish what monster trains accomplish, but without the monster delays that actually drive up the cost, even as the monster consist under one engineer drives it down. You need to add a detail to the pretty argument to make it work properly. The "logical" conclusion many monster-train proponents will make is that power per ton can be reduced on a long consist with judicious use of distributed power: you must prove that shorter trains moving the same tonnage with the same number of units will produce better fuel economy, power balancing, etc. than the monsters.
Euclid So the self-driving trains can accomplish what monster trains accomplish, but without the monster delays that actually drive up the cost, even as the monster consist under one engineer drives it down.
You need to add a detail to the pretty argument to make it work properly. The "logical" conclusion many monster-train proponents will make is that power per ton can be reduced on a long consist with judicious use of distributed power: you must prove that shorter trains moving the same tonnage with the same number of units will produce better fuel economy, power balancing, etc. than the monsters.
Since the monster won't fit anywhere, it tends to stay moving (slowly). The shorter trains can get held at any siding along the way. Depends upon the dispatcher and other traffic, of course. This is only one of many reasons.
JPS1 . . . I was reacting to the argument put forth by many that executive management in U.S. corporations would sacrifice the safety of its employees to please investors. The idea is simply wrong.
- PDN.
Euclid Overmod Euclid So the self-driving trains can accomplish what monster trains accomplish, but without the monster delays that actually drive up the cost, even as the monster consist under one engineer drives it down. You need to add a detail to the pretty argument to make it work properly. The "logical" conclusion many monster-train proponents will make is that power per ton can be reduced on a long consist with judicious use of distributed power: you must prove that shorter trains moving the same tonnage with the same number of units will produce better fuel economy, power balancing, etc. than the monsters. DO MONSTER TRAINS SAVE FUEL? You raise an interesting question. I have understood that the reason for monster trains is to move more tonnage with less crew cost. Are you saying that another reason is to move more tonnage with less fuel cost? How does that work out? Consider a monster train of 240 cars with distributed power. Compare that to moving the same tonnage in several short trains, each with power only at the head end. Also, assume that the weight of the motive power of the monster train and the several short trains is the same. Assume that both the monster train and multiple short trains with the same tonnage have the same horsepower per ton and the same efficiency of power application to the load. In comparing the monster train to the group of multiple short trains this way, is the fuel usage identical? If fuel usage is not identical, why is that?
I'm not sure it's about less fuel, but using less engines.
Take your 240 car train and chop it into two trains of 120 cars each. For illustration, we'll say the tonnage requires 2 and 1/2 engines for each train. Now you don't have half an engine, so you need three engines for each train. Combine it into one train and you only need 5 engines, not 6.
You can place that 6th engine in storage. Multiply that over the system and maybe you can close some servicing facilities, furlough some mechanical dept people. The 'savings' go beyond just saving on crew costs.
Jeff
I first thought about this in high school when I read a Trains article about Al Perlman at WP. He was advocating a strategy of more, faster, shorter trains at the time, and there were some other railroads (C&NW intermodal?) that were considering the idea at the time. Part of this involved the percentage of time the units assigned to the train had to spend above, say, notch 5 in providing service. (This was the first place I saw the figure that 3 units 'doing the work of 2' might often burn less aggregate fuel for the same tonnage delivered in the same time, and this was in a time-limited model, not the present "JIT" at slower speed paradigm.
Technically we're talking about two kinds of fully autonomous train. One model runs trains in independent comparatively-short consists, probably involving one to three locomotives depending upon the 'granularity of demand' in a particular window of time, but never less than the effective 'train resistance' involved in most economical operation of a given size and type of unit. We'll presume it uses some kind of accurate CBTC so the effective spacing between trains can be minimized ... but note that it can't be reduced to zero, and the interactions between trains proceeding in opposite directions have to be spaced relative to siding density and available length.
The other model would look very much like Don Oltmann's electrified setup for 2040: you have much the same block length and composition as for the small-train model, but each has its own assigned "DPU" and programming, so the train can be thought of as a series of platooned short trains when 'together' but it can be split very quickly and effectively into multiple segments at yards or division points, or recombined into longer trains on disparate lines in a powered version of block switching.
In between points, you can see that the monster train approach requires fewer CTBC minimum intervals, as the 'intervening' spaces are automatically taken up via the platooning (just as would be the case for autonomous trucks) with the draft-gear buff substituting for intertrain interval. You can also see that the aggregate frontal air resistance of the multiple small trains will be higher than the monster; this is a relatively small factor but one which would have real-world implications on fuel consumption.
In multiple operation, several short trains may have to share a given siding, as there is little economy in providing many shorter sidings each with its own switches, starting and stopping, etc.) Each of these involves a certain dwell before the line is 'clear' for opposing traffic, whereas the long trains can arrange to pass each other almost with zero dwell if the head end of one just approaches the switch as the last of the other is clearing, while the opposite-end switch opens and allows the start of acceleration even as the tail of the first clears. Note that the shorter trains have to establish a 'following distance' as they leave the siding, and much of the prospective advantage of short trains over longer ones becomes lost in the dwell time if the advantages of longer sidings were to be retained.
The net effect of a power-balancing move is worse for the short trains, unless train length can be 'balanced' with power moves -- this is somewhat unlikely for reasons associated with why there needed to be a power move in the first place. If full advantage is taken of the 'balanced' power to run in a lower notch to move the shorter train, this may not be a problem and might be a functional advantage in some conditions. But if that economy is not fully realized, the ability of a longer train to accommodate some units 'dead' (considering only a fraction of the dead weight has to be hauled by 'each' of the DPUed engines) may have more value than it would when single engines would require either train derating or an increased risk of stall.
A consideration that has been inadequately discussed, in this thread at least, is reliability: a larger train with distributed power would usually be somewhat tolerant of a unit suffering derating or shutdown during the run. A short-train equivalent will require special attention, as no other 'short train' will be as capable of helping it to the next siding at reduced speed, if at all. One such interruption can easily cascade into the sort of gridlock so quickly encountered 'when dispatchers fail'.
OvermodI first thought about this in high school when I read a Trains article about Al Perlman at WP. He was advocating a strategy of more, faster, shorter trains at the time, and there were some other railroads (C&NW intermodal?) that were considering the idea at the time. Part of this involved the percentage of time the units assigned to the train had to spend above, say, notch 5 in providing service. (This was the first place I saw the figure that 3 units 'doing the work of 2' might often burn less aggregate fuel for the same tonnage delivered in the same time, and this was in a time-limited model, not the present "JIT" at slower speed paradigm.
Before I retired - the Senior Road Foreman of Engines held a required meeting for all Train Dispatchers.
The meeting was to explain Trip Optimizer and Leader operation and why Dispatchers should encourage Engineers to use the tools in compliance with Train Handling Rules.
Secondarily figures were presented that locomotives were more efficient when hauling maximum tonnage in maximum power situations. The figures were presented, I just don't know that I agree with them.
Never too old to have a happy childhood!
Thanks for the input. I understand these points that have been made. There may be many ancillary factors that enter into fuel consumption that are not directly related to the essence of fuel consumed per tonnage transported. As has been mentioned here, ancillary factors include the following:
Length of sidings
Locomotive horsepower being in multiples not matching the tonnage of a short train, and thus added unnecessary surplus power and driving fuel usage higher than necessary.
Shorter trains being less efficient than possible with a given locomotive because it cannot run in full power mode on a short, light train.
Those and other factors may cause monster trains to be more fuel efficient than a series of smaller trains moving the same tonnage. So I am focusing this issue on just that essence of fuel consumed per tonnage transported, and this assumes all other ancillary factors are optimized to not influence fuel consumption in this objective comparison. Therefore, the comparison would not assume just changing over from monster trains to a series of short, light trains in the midst of today’s normal practices.
So, say the comparison is as between the following two approaches:
One train 240 cars long with tonnage and DP matched for optimum fuel efficiency for optimal train performance.
Six trains each 40 cars long, and each pulled by one locomotive with optimum fuel efficiency and optimal train performance.
This comparison assumes running on the same section of track with the average speed of the 240 car train and of the 6 short trains all equal. Also equal would be the consists of the 6 short trains and the corresponding 6X consist of the one long train.
The popular buzz about long trains is that they reduce crew costs and save fuel. Yet, nowhere do I find an explanation of why and how they save fuel. I don’t doubt that they might save fuel in the context of today’s railroad operations. But if they are compared as I propose in the question above, I see no reason why the 240-car train would use less fuel than the series of six 40-car trains. I conclude that fuel consumption for the monster train and six-train set described above would be equal in a straight apples-to-apples comparison.
The operative measure for fuel consumption would be gallons per ton per mile, or something along that line, I would think.
One must also consider that there is probably a certain overhead in fuel usage.
If the 240 car train requires six locomotives, in whatever configuration, then the fuel consumption should equal the six 40 car trains with one locomotive each, assuming the two trains are otherwise equivalent.
Hauling the 240 car train with three locomotives should be more efficient, although the individual fuel usage for the locomotives will be higher.
Hence the gallons per ton per mile.
And, again, this assumes the two trains are equivalent.
In today's world, that would mean six crews instead of one, but we're talking about trains without a crew.
The shorter trains might make the railroad more fluid. Long passing sidings can hold multiple trains. Fleeting would allow the equivalent of a unit train to move over the railroad.
My thoughts.
Larry Resident Microferroequinologist (at least at my house) Everyone goes home; Safety begins with you My Opinion. Standard Disclaimers Apply. No Expiration Date Come ride the rails with me! There's one thing about humility - the moment you think you've got it, you've lost it...
tree68The shorter trains might make the railroad more fluid. Long passing sidings can hold multiple trains. Fleeting would allow the equivalent of a unit train to move over the railroad. My thoughts.
When the B&O Single Track CTC'd the line between Baltimore and Philadelphia (CSX's Philadelphia Sub). The sidings were nominally 9000 feet in length. The 1960's thinking being that with 'normal' freight train size being 4500 feet or less, that double meeting could be easily done. That was 1960's thinking - we are now well into the 21st Century - a 9000 foot train has become a 'mini-train' in the world of PSR. More trains would just create more congestion.
You are correct that the fuel savings have to be measured in fuel used per ton mile - the first generation diesels used a nominal 2 gallons per mile of operation. The current generation of diesels in maximum tonnage service are rated at 5 gallons per mile. Current AC diesels will haul 5 times (or more) tonnage per unit than the first generation diesels were able to over the same territory - the diesels that economically 'killed' steam. Today's AC power hauls significantly more tonnage per unit for all generations upto the present.
From personal experience - The standard CSX Coal train back in the early 90's was 2 Dash-8's and 90 loads at about 10800 tons - between Atlanta & Manchester there are 3 grades - if the weather was anything other than perfectly dry - these trains would stall on one or more of the grades. With the implementation of two unit AC power the trains are now 130 cars and about 18500 tons and the power rarely have stalling issues.
About fuel useage. ----- The BNSF intermodal haulage trains here have all power leading on the front. As well some CSX and some other trains also have all power leading. Now some CSX trains are running DP especially the longer IM trains from JAX and New Orleans. The longer CSX DP trains seem to have much less wheel noise on a curve than those trains with all locos up front.
Is it possible that this wheel noise is an indication of more rolling friction causing a slight increase of fuel consumption compared to some of the DP trains?
tree68The shorter trains might make the railroad more fluid. Long passing sidings can hold multiple trains. Fleeting would allow the equivalent of a unit train to move over the railroad.
Note that there is more to this point. With autonomous trains, at least in theory you can run them 'platooned' either at relatively very short following distance or physically in contact but with no necessity of connecting the brake systems or performing other 'manual labor' to enable the close coordinated operation. That means that the short trains can take up little more space than the hypothetical fully-coupled DPUed-block train, and it becomes possible for one of the trains in the 'middle' to leapfrog those in front if the trains in back are allowed by the signal system to back out of the siding, all of which can be controlled fairly positively.
Now, instead of fleeting (which involves standard separation between trains) you have what can be one solid unit train that may be several miles long, operating past sidings in little more time than one or two separately-fleeted trains would consume, with the usual objection to monster trains removed by the autonomous ability to split into cuts that fit yards 'on the fly' as they get there. So you get the advantages of fleeting when there is a large traffic buildup or mismatch while preserving 'more' of the ability to schedule shorter trains in meets on our '80s-downsized PSR-maintained legacy mains.
Meanwhile, a monster platooned train can also break 'on the fly' into units that fit in sidings, zip into a few, allow an equally-platooned monster train or anything else (say, Amtrak) to pass easily, then sequentially get its sections up to speed 'just in time' to safely reform. This would be a pain to do safely even with well-trained engineers.
There's definitely lower flange force and stringlining tendency in a two-engine DPU consist than 'all engines on the front' -- look at the tendency at the node, then work out in both directions. I see some very interesting experiments into placement of the second engine, too; it need not be on the tail where it used to be (probably because that replaces having to use a FRED) But I don't think it will be any less than the same consist divided into two equal sections each pulled by one locomotive, which is the autonomous alternative. Tests would show rather quickly where, and where not, to expect savings in properly flexible and reliable operation.
I've already discussed some of the sources of lower air resistance from continuous rather than interrupted trains.
In addition, I think that Euclid's proposed train length needs some further discussion: his 40 cars seems more than a little off the practical loading of a single contemporary modern freight locomotive, and I think there's still relatively little point in planning a return to smaller-horsepower GP-size units now that autonomy makes them 'less impractical' to build or run. The short-train consist would be that which 'just loads' a single unit to its economical capacity, and I expect this to be longer for AC units than DC for the reasons BaltACD alluded to for the trains over Sand Patch. It does remain for testing to be done to determine whether shorter 'expediently-dispatched' trains on the original Perlman model would give better fuel saving at 'part throttle' than PSR-arranged 'perfect' ones.
The effect of mismatch in loading is lower for trains with multiple engines, as it is essentially 'split' between the locomotives rather than having to be assigned to just the one.
Note that equipment utilization, which is at least as important as fuel consumption, is worse for the expedient short trains (i.e., you need more locomotives to 'cover' the number of ton-miles moved) but that's not the same discussion as the fuel burned.
I am looking though back issues of Trains to find the article called Fast and Frequent, which describes a new operating principle of the D&RGW. I would like to go back and review that in light of what we are talking about here. I seem to recall that it was about shorter trains, but more of them.
blue streak 1Is it possible that this wheel noise is an indication of more rolling friction causing a slight increase of fuel consumption compared to some of the DP trains?
If you're referring to the squeal on tight curves, I can't see how where the power is would affect that, as it's a function of the solid axles on the cars and the resulting wheel slip on curves.
If the flanges are getting involved, or the wheels are not riding within their "groove," then you may have a point. Think stringlining. On a DP train, some portion of the cars are actually being pushed, vs pulled, so those lateral stresses would not be as pronounced.
Some don't trust automatic trains from a safety perspective. To a limited degree, DP units share some similarities. They are operated remotely by a human. How many accidents have they caused because of throttle or brake remote control problems? With proper PTC, a human in a control room could do the same. Or the train could be totally autonomous.
Our community is FREE to join. To participate you must either login or register for an account.