OvermodI'd submit that there is a further issue, that it's fundamentally impossible to predict the actual 'safety stock' that a given emergency would demand -- particularly as nature alone can deliver something so far out of expected bounds (Diane, Agnes, Sandy, the recent flooding) that no real cost-effective "planned" measures could keep you running. And we have all seen what an 'overoptimized' railroad becomes when even relatively small wrenches are thrown in the works.
Not quite as hard as it might seem. I know at least NS has (or had) disaster plans that assumed a variety of major disruptions. Those plans were based on traffic flow models and train a blocking plans were roughed out. I don't think they took into account crew pool and locomotive fleet planning in any great detail.
A least a couple of times, the plans were useful - most notably during Katrina when all of the traffic over New Orleans had to be diverted.
Of course, back then there was more flexibility in the crew HOS law and in the number of guaranteed starts per month, so there was not such a big deal to keep pools a bit larger and you could keep an extra board churning like mad for a short period. Not so, now.
This sort of stuff can be modeled using discrete simulatons, and such models for crews and power do exist. The problem is that the RRs are so focused on PSR implementaion, that the idea of "safety stock" isn't much of a priority, sadly.
-Don (Random stuff, mostly about trains - what else? http://blerfblog.blogspot.com/)
oltmanndThe biggest danger to PSR that I see is that it will work too well. The crew and locomotive resources will be trimmed to fit the steady state, forgetting that you need some "safety stock" to keep things moving when something goes wrong.
Add to this the problems that occur when "PSR" becomes calculated relative to OR numbers, rather than potential asset maximization. As I suspect is being done in all too many cases even now.
I don't know of any RR that has tried to calculate exactly what that "safety stock" needs to be. It gets discussed from time to time, but nobody ever really does anything to define it or make is a solid part of the plan.
I'd submit that there is a further issue, that it's fundamentally impossible to predict the actual 'safety stock' that a given emergency would demand -- particularly as nature alone can deliver something so far out of expected bounds (Diane, Agnes, Sandy, the recent flooding) that no real cost-effective "planned" measures could keep you running. And we have all seen what an 'overoptimized' railroad becomes when even relatively small wrenches are thrown in the works.
What I'm looking forward to is the effect of "PSR" implementation on the restoration of effective double-track (or multiple track) running, particularly when each main is set up for unidirectional fleeting-type operation with CTC 'passing' capability minimized, essentially used only for unschedulable speed mismatch conflicts. (One immediate consequence being 'optimization' of the existing problems and risks with flat-switching of jobs in yards, now that more trains are coming to and originating in them ... be fun to see how that plays out with crews not being encouraged to board/drop off with the engine in motion or run on ballasted surfaces.)
OvermodI'd be interested to see if Don Oltmann in particular can solve, at least in principle, the problem he brought up last week about the intermodal train that clashes with a nominally faster Amtrak train during the night...
I've been thinking about this and....maybe.
Overall, I think PSR implementation will be bad for Amtrak timekeeping. I think the tendency toward longer, mix freight type trains will result in lower HP/ton and longer "unopposed" run times. Longer trains also take longer to get into and out of sidings, terminals, etc. Longer trains fit in fewer sidings and make overtaking moves more difficult. PSR also values "filling up" the best routes and jettisoning the worst.
However, if you are really thinking about precision scheduling of trains on routes, and attempt to have a train plan that minimized congestion and defines meet and pass locations, you could counteract some of this negative effect.
The biggest danger to PSR that I see is that it will work too well. The crew and locomotive resources will be trimmed to fit the steady state, forgetting that you need some "safety stock" to keep things moving when something goes wrong. I don't know of any RR that has tried to calculate exactly what that "safety stock" needs to be. It gets discussed from time to time, but nobody ever really does anything to define it or make is a solid part of the plan.
JPS1What is the likely impact of Precision Scheduled Railroading on Amtrak’s long-distance trains?
The great possibility -- subject, of course, to Amtrak management (and perhaps some of the line and staff agency people reporting to Congress et al.) -- is that Amtrak trains will be given some 'priority' in the scheduling arrangements and in resolution of emergent situations that results in assured timing quality. This may well be what the recent Amtrak publicity campaign against 'freight delays' is an 'opening shot' to bring about.
I'd be interested to see if Don Oltmann in particular can solve, at least in principle, the problem he brought up last week about the intermodal train that clashes with a nominally faster Amtrak train during the night...
Paul_D_North_Jr Overmod . . . will point even financier-obsessed people at the 'first best' places to expand capability. I beg to differ. The OR-obsessed people aren't interested in expanding capability (or capacity). What they're interested in is the lowest OR - the epitome of 'dumb' railroad - at the expense (literally) of anything else. As a result they're leaving a huge asset grossly underutilized - the track and fixed plant - which could be used to contribute the bottom line with a reasonable marginal net income criteria. Sure, some hump yards can be closed and some second tracks ripped out - but still there are going to be a lot of miles of single track that are underutilized capacity (and no, I'm not advocating more trains just to fill up those tracks because they're there). OR measures only the traffic that's there now, not the traffic that could be had at good rates but is being missed. See Matt Rose (BNSF) recent quotes. The 'dumb' railroad term is appropriate. Some of us will remember "cost per dispatch" from F.H. Howard's story "The Way It Was" in Trains back in the late 1970's, which he concluded justified the following: When a worn brakeshoe was found on a steam locomotive tender, a new one was not procured from the storekeeper - where it then would have been charged to the locomotive's "cost per dispatch" - but instead taken from an old cinder car. The brakeshoes were the same, "but only one [piece of equipment] was subject to cost per dispatch" measurement. Of course, the missing brakeshoe on the cinder car was then replaced with a new one for the next time . . . - PDN.
Overmod . . . will point even financier-obsessed people at the 'first best' places to expand capability.
I beg to differ. The OR-obsessed people aren't interested in expanding capability (or capacity). What they're interested in is the lowest OR - the epitome of 'dumb' railroad - at the expense (literally) of anything else. As a result they're leaving a huge asset grossly underutilized - the track and fixed plant - which could be used to contribute the bottom line with a reasonable marginal net income criteria. Sure, some hump yards can be closed and some second tracks ripped out - but still there are going to be a lot of miles of single track that are underutilized capacity (and no, I'm not advocating more trains just to fill up those tracks because they're there).
OR measures only the traffic that's there now, not the traffic that could be had at good rates but is being missed. See Matt Rose (BNSF) recent quotes.
The 'dumb' railroad term is appropriate. Some of us will remember "cost per dispatch" from F.H. Howard's story "The Way It Was" in Trains back in the late 1970's, which he concluded justified the following: When a worn brakeshoe was found on a steam locomotive tender, a new one was not procured from the storekeeper - where it then would have been charged to the locomotive's "cost per dispatch" - but instead taken from an old cinder car. The brakeshoes were the same, "but only one [piece of equipment] was subject to cost per dispatch" measurement. Of course, the missing brakeshoe on the cinder car was then replaced with a new one for the next time . . .
- PDN.
One thing railroaders as a group have mastered - knowing what aspects of their jobs are being measured and the weights applied to those measurements - they they do whatever gives them the best score - no matter if it is good for the overall operation or no.
The old dictum comes into play - Be careful what you ask for, you may get it!
Never too old to have a happy childhood!
From this link: https://en.wikipedia.org/wiki/DMAIC
DMAIC (an acronym for Define, Measure, Analyze, Improve and Control) (pronounced də-MAY-ick) refers to a data-driven improvement cycle used for improving, optimizing and stabilizing business processes and designs. The DMAIC improvement cycle is the core tool used to drive Six Sigma projects.
When he says 'schedule' he probably means something different from the usual railroad use of the term. I suspect he is setting up something like a PERT-T chart, where every resource is blocked out relative to every other and the maximum number of tasks can be developed; then best use can be made of the remaining 'slack' by in essence finding or thinking up things for the 'idle hands' to do, shutting down equipment or discharging crew when there aren't, and generally arranging the variable costs and the "best use" of capital equipment and plant to suit the volume of the business at hand (and thereby maximize actual rather than "financial" return.
To me this implicitly includes the positive sense of OR, without the usual pratfall distortions that result from the usual cost-cutting-to-reduce-that-index activities on a 'dumb' railroad. It also has implications for finding and developing new business that either fits into or supplants at equal or higher marginal return what you currently have, right up to achievable limits on your plant -- with the added benefit that the DMAIC or whatever cycle you run on your information gathering will point even financier-obsessed people at the 'first best' places to expand capability.
Moyers did turn around IC, and he got SP to the point where it was an attractive merger candidate for UP.
it would be intersting to see which lines are carrying more traffic today as coal declines - the ex-Union Pacific lines or the ex-Southern Pacific-Cotton Belt lines.
EuclidThe later sounds like the minute a car rolls off the hump, it is hustled off to its shipping destination.
Humps? What humps? A primary feature of PSR seems to be getting rid of humps.
EuclidThe former sounds like crews, cars, and locomotives are NOT in perpetual motion. Instead they are waiting for a schedule to begin the motion of transportation.
While you have a point about the cars, it would seem that PSR has a goal to keep on hand just enough crews and motive power to move what has to be moved, presumably on-time.
Terms that have been around for years are "terminal dwell" and "velocity," with the goal to have minimal terminal dwell (humps, especially) and maximim velocity (cars). Those seem to be prime ingredients in the PSR recipe. The shippers may disagree about improved velocity for cars...
I remember discussions in the past involving pre-blocking cars so they didn't have to be switched at every yard they encountered. A group of cars headed for Z from A would stay together all the way through B, C, D, etc.
Euclid ...in order to make transportation pay a higher return.
And this seems to be the primary goal of PSR, particularly in the short term.
It might be interesting to look at the financials of the railroads who have been subjected to PSR in the past over the long term. Were those great returns still around 1, 2, 5, 10 years later, or were the super numbers just a brief flash in the pan?
The multitudinous variables that have been mentioned show up in the database I've been building of trains passing through Deshler, OH. My listing shows almost 110 possible trains right now, of which only about 50 appear with true regularity - all the rest are "as needed," meaning they have to be shoehorned in with all the "regularly scheduled" trains. A good many of those trains (usually unit trains) will cross several divisions in their travels.
It's not unusual to see trains "parked" at Deshler, waiting for access to the North Baltimore IM facility, or for other traffic to clear. And it's also not unusual to have periods of an hour or more with zero traffic through the interlocking. In a perfect world, all trains would move through the area at speed, but clearly, it's not a perfect world.
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...
See my comment at the linked article.
From the article:
In his hands: A 25-page handwritten business plan that, among other things, called for scheduling every movement on the railroad and keeping crews, locomotives, and freight cars in perpetual motion. “It was the beginning of Precision Scheduled Railroading,” Lamphere told the North East Association of Rail Shippers on Thursday. “But we didn’t have a name for it.”
What exactly does this mean? “Scheduling every movement on the railroad” sounds in opposition to “keeping crew, locomotives, and freight cars in perpetual motion.”
The later sounds like the minute a car rolls off the hump, it is hustled off to its shipping destination.
The former sounds like crews, cars, and locomotives are NOT in perpetual motion. Instead they are waiting for a schedule to begin the motion of transportation. Presumably the scheduled departure allows for loads to accumulate to a cost effective train size in order to make transportation pay a higher return.
BALTIMORE — E. Hunter Harrison may have written the book on Precision Scheduled Railroading. But the late chief executive is not the creator of the operating model that’s currently sweeping the industry, says Gilbert Lamphere, the former...
http://trn.trains.com/news/news-wire/2019/04/05-financier-harrison-didnt-create-precision-scheduled-railroading
Brian Schmidt, Editor, Classic Trains magazine
Our community is FREE to join. To participate you must either login or register for an account.