Amtrak Roadrailers Going Away

[Anonymous]

QUOTE: Originally posted by M.W. Hemphill

Amtrak #4 LA to Chicago, and #48 Chicago to NYC is scheduled for 65 hours 40 minutes -- with a four-hour connection in Chicago. The once-weekly "Bullet Train" that BNSF tested and UP ran was scheduled for 65 hours, 30 minutes, but it was dropped earlier this year. The current best coast-to-coast intermodal service is at least one day slower than that.

Most intermodal is not particularly time-sensitive; only certain premium-price trains, the principal customers for which are UPS and LTL truckers (which is practically the same thing as UPS). The preponderance of the double-stack trains you see are run on much slower schedules. Intermodal is very price sensitive, however. The difference in price between rail and truck is usually very small -- maybe $50 a box.

I'll admit, I am suprised, and on a number of fronts. On other posts here we've discussed the willingness of shippers to pay a premium to long haul truckers due to the assumption that they are the fastest dock to dock e.g. "when it absolutely positively has to be there yada yada yada....". When one thinks of a "premium" I guess we assume that means a couple of hundred bucks per box. But only $50? And that's the dock to dock comparitive price, right? Not just the rail terminal to terminal difference? I guess I'm taking the comparitive pricing of Fed Ex and UPS package price quotes (not counting overnight and two day which would use air freight, but three day, four day, etc. "ground" schedules) and using those prices as a predictor of trailer/container prices. I would have thought the price spread percentage would have been similar for packages and entire trailers.

Secondly, I would've thought with the added burdens of station stops that Amtrak would be slower than the hotshots, all other things being equal e.g. crew change points, refueling stops, etc. Aren't hotshots allowed to run at passenger speeds? Didn't the Santa Fe used to run 90 mph TOFC through Kansas and Colorado (I seem to remember a TRAINS article regarding the same)? Have the freight railroads really slipped that far on the speed front? Is the market for time sensitive bulk cargo just not there anymore?

[AntonioFP45]

As mentioned above some of Amtrak's long distance trains were being delayed for hours due to the procedures of coupling or uncoupling the roadrailers. By the time connections, inspections, and brake tests were conducted the trains were running behind scheduled. Compound that with the trains running over railroads where freight trains were regularly "prioritized"over Amtrak, and you had a pretty bad report card.

Too bad as the concept could have worked if all the parties involved would have fully cooperated.

[oltmannd]

Here's a few more thoughts:

1. RR suvival is a 3 legged stool. If you drop bulk, intermodal or car load, you don't have enough revenue to support the network. So the trick is to accomodate these three groups with very different service demands on the same mainline network.

2. Things change - what works today will put you out of business tomorrow. If RRs were only in the market niches they were in 1980, they'd all be broke by now. They've had to embrace new markets, such as steamship stack train and new service, such as dedicated finished auto trains and terminals in order to keep from shrinking away. Similarly, if they don't keep looking for more new, good ideas, the traffic base of today will simply be gone in 20 or 30 years. High speed service may be one of those, new, good ideas.

3. Don't foget about frt at pass speeds - RRs think about it all the time. Conail looked pretty hard at an 18-20 hr schedule from NJ to Chicago and 79 mph running was on the table. RRs have been losing out to team drivers on speed - have done well against single drivers. Faster speeds are a way to keep a hand in this market niche. However, maximum track speed often is a very small driver of overall trip speed. It's all the other "stuff" that happens out on the RR that slows the trips. Equipment failure, track&signal failures/slow orders, weather, meets/passes, being held out at terminals, etc, all conspire against fast service. In fact, on Conrail, you needed to keep about 3-1/2 hours slop in the NJ to Chicago train schedules (compared to unopposed run times from simulations) in order to have the train arrive 90-95% on time. Another thing that slows service down is terminal trackage and slow speed connections. It takes many many miles of running at 79 (or even 90) to overcome only a few miles at slow speeds.

4. High speed rail might mean more that just passengers - if we don't blow it. Recent NEC history is a bad precedent! There is some conventional "wisdom" that says "High Speed Pass rail and frt don't mix." Yeah, well, why not! 120 mph Metroliners ran perfectly well with quite a bit of frt on the NEC from 1968 to the late 70s. Then, Conrail took the electrics off - which just happened to be speed control equipped - and replaced them with diesels that were not. Seems to me, that if you do it right, i.e. proper saftey and managment, you can mix quite safely. Of course there are capacity issues, but I don't think there are unsolvable safety issues.

Further, if a network of high speed passenger svc did develop in the US, it would be a shame not to allow frt to move over it "just because." On an incremental basis, the frt would help pay for the investment and operating costs. It could create a new niche market for overnight truckload sevice where no such service exists today. Perhaps the FedExs and UPSs of the world would use it in lanes they now cover with their overnight air parcel service, for example.

[Anonymous]

Thanks, Mark and Don, for the insightful observations.

Some clarification:

1. When I refer to "bulk" I tend to include trailers and containers that have amassed at a terminal and all bound for the same general direction. 100 trailers or 250 containers qualify as a "bulk" load in this context.

2. As for increasing car weights, I will give you a link regarding studies on track maintenance expenditures relative to 264k, 286k, and the proposed 315k standards and how those max weights influence the laws of diminishing returns as applied to increasing track maintenance costs:

<http://www.zetatech.com/CORPQIII44.htm>

It's an interesting approach to quantifying track maintenance costs.

3. I agree with Don that a high speed rail network (125 mph) could result in new markets for time sensitive ground freight/overnight freight. One wonders what the playing field would be like if over the road trucks were still limited to 70mph max while hotshot freight were blowing by at 125 mph!

4. Railroads could do other improvements on the right of way besides increasing weight per axle or improving speed characteristics. Raise the max height allowed for freight cars from 20'2" to 30' and we could start double stacking trailers using wells and racks! David P. Morgan was an advocate of a wider gauge, and if he envisioned something around 10' or so it would be possible to stack two containers or trailers side by side. Of the two, I would think increasing height would be less costly. Food for thought!

[Anonymous]

QUOTE: Originally posted by M.W. Hemphill

Dave, I read the linked document twice. I think I understand it to be a marketing piece for a software sold by Zetatech that predicts allocated track maintenance costs at varying GTMs. But I can't find anything in it about the effects of increasing maximum axle loadings, not even the words 263K or 286K. Can you quote exactly where you found that?

What rationale are you using to qualify TOFC and COFC as bulk? I've never heard anyone in the industry do that before. Is this something new? Who else is using this rationale?

A new high-speed or broad-gauge network would change a lot, agreed. But in the meantime until the government funds that, what do you propose?

Mark, I really hate it when you make me dig back into my files!

I gave you the wrong link, although that study does make mention of the point at hand. The information regarding Heavy Axle Loads (HAL) is from an email correspondence I had with Jim Blaze of ZetaTech and one L. Kauffman in a running argument on shortlines. The study regarding HAL is entitled "Economics of Heavy Axle Loads: Costs and Benefits" authored by Allan M. Zarembski and Jim Blaze. In it, the authors conclude that the law of diminishing returns comes into play when axle loadings are increased from 36 tons (72,000 lbs = 286k car ) to 39 tons (78,000 lbs = 315k car), e.g. the track maintenance costs increase more than the benefits of the increased load factor of the cars, although they do conclude that the benefits of 315k still is better than the older 263k standard. I can email it to you if you'd like.

The HAL subject is something I combine with another assumption, that there is a relative incompatibility of fast freights with drag freights in the final track profile design. Having to construct trackage to handle both time sensitive and cost sensitive commodities ends up eliminating adjustments that would suit one over the other, e.g. the degree of superelevation on curves, a higher incendence of track pounding by the heavier cars that invokes more slow orders, etc. The combined subjects also bring into play another old argument on this forum, that of why the railroads don't go to the three axle truck to accomodate the higher paying load factor while still keeping per axle loadings down to where the Class II and Class III lines can continue to remain viable. If a locomotive with three axles per truck can run over lighter trackage (using radial steering) why can't the same technology be used on freight cars?

The use of the term "bulk" for TOFC and COFC is my own generalization, certainly not anything in the logistics industry. I just use it as an example of the "time sensitive bulk movement" definition (e.g. hauling 100 trailers on a hotshot TOFC consist is a bulk movement of trailers, not a piecemeal movement).

Regarding the last point (high speed trackage, broader gauges, and who would fund such a project) I am serious when I suggest that max clearances could be raised to 30+' to accomodate a sort of two tier TOFC consist, and be done without government help. I am going by the railroads' effort to raise clearances to accomodate double stack containers a while back. The logic of making the necessary improvements to increase the number of paying units per train from 125 to 250 containers is sound, and could be applied to TOFC consists which are still limited to 125 units or so per train. There are corridors out there now which are not encumbered by tunnels or bridge obstructions, where it would only take the raising of a few power lines to accomodate this 30+' height. Since the railroads can charge a premium for the hottest TOFC (at least a higher rate per paying unit than they get for ocean containers), I am assuming it would at least be worth the price of a feasability study by the railroads. Finally, raising vertical clearances would be a lot less expensive than broadening the gauge or building a whole new high speed rail network.

[Overmod]

This thread got interesting in the past few hours.

We ran the numbers intensively back in the mid-'70s for an integral container train between Ivy City and Boston, which essentially would provide 'overnight delivery' for ISO series 1 containers to a bunch of intermediate transfer stops (from which trucks would then make final dock delivery, on a separate schedule). There is no particular reason other than economics why a container train, properly designed and built, can't run at least as fast as a passenger train.

Tilting per se is utterly useless on a freight train. Its purpose is to relieve the inner ears of passengers, not to make the train more stable on curves at high speed -- in fact, a tilt system usually makes the train MORE prone to high-siding. In order to get the benefits of tilt, it's also necessary to perform lateral shift (e.g. at the bolster) to move the roll center physically inboard. And, of course, systems failure at speed will be *anything* but fail-safe...

One of my business mentors is the former CFO for Consolidated Freightways, who pointed out that precise delivery scheduling is MUCH more important than absolute end-to-end speed for almost all classes of practical freight. The volume of 'overnight' traffic in most markets is already saturated at levels that leave FedEx's express division considerably lower in overall CTP than, say, its ground-based services. Think of it this way: an industry normally schedules deliveries of factors via JIT (also known as kanban) scheduling... and has a string of deliveries at intervals. A few extra containers going a bit more slowly isn't going to have substantial impact on profitability -- what's the debt service on a couple of extra days? -- BUT either early or late arrival at the process location has very real, and often very substantial costs associated with it.

This is the place that rail has to address, and quite frankly the existing companies haven't seemed to have done very much that is effective about it... other than trying to concentrate on traffic niches that don't fall in this category.

I don't think there's any particular doubt that track-maintenance costs are nonlinearly much higher at 315K over 263K... and 263K is already too high for high-speed service on existing rail steel in the absence of implemented 'magic wear rate' continuous monitoring. Anybody who thinks differently is invited to look into the metallurgy and physics involved. I, for one, would not much reli***he job of designing high-speed ROW (125mph+) capable of handling routine 263K loadings using existing methods of suspension and damping...

[Anonymous]

Overmod,

You do confirm my own assumptions, e.g. a high speed freight service would have to limit axle loadings to around 55,000 lbs per axle if not even less. That does not necessarily mean each car is limited to 220,000 GW if three axle trucks are utilized. A six axle freight car at 55,000 lbs per axle comes out to 330,000 lbs per car, so the whole increasing-load-weight-to-make-railroads-more-efficient rationale would not be revoked. In that mode, you might be able to have high speed freights and heavy haul freights co-existing on the same line without subjecting one to the other in terms of track profile.

One thing I've noticed when observing passing trains, I always pick up at least three or more instances of the BAM BAM BAM of flat spots on wheels as the train passes, no matter what type of train is passing. I would think that flat spots on wheels of the heavier cars would cause much more track damage than those of the lighter cars, and I doubt the HAL studies that have been done ever take this constant into consideration when determining the cost/benefits of heavier axle loadings.

[daveklepper]

Isn't NS running some freight on the Corridor now? If it is, what a good story for Trains a report on a cab ride would be! But the article should obviously include the economics of the situation, the technical details on the requirements Amtrak places on NS equipment in the service, who the customers are, what the freight is, etc.

I know Guilford has some peddler service north of New York (on the Corridor-Amtrak and on Metro North), and it would be interesting to contrast this freight service with the higher speed freight service south of New York.

[Overmod]

Dave, the flat-wheel problem is one of the greater problems with track integrity, and I agree that you won't find this in the conventional analyses of track wear with increasing axle loads. Part of the reason may be that it's difficult, if not impossible, to quantify the physical effects of "flat wheels" -- which may differ dramatically based on the length of the flat, nature of the 'corners' where the peak transition forces occur, etc.

Where there is no question, however, is that the effects of flat wheels become profoundly magnified with speed increases. At a time when railhead metallurgy is already demonstrably overstressed (pun intended) at current axle loads with current rail steels, it seems strange to me that people are considering higher axle loadings on "existing" track and alignments. I need to read as much as possible of your sources before I can assess their authors' methodology and conclusions, but I suspect they are assuming linear extension of characteristics in areas which will prove to have profoundly nonlinear characteristics. (I'm tempted to say "Hatfield, anybody?")

I am currently investigating cost-effective methods of reducing the problem of wheel-skidding on "conventional" air-braked cars, and of repairing or replacing poorly-worn wheelsets in service (perhaps by using a field-applied temporary material to re-establish profile). I confess to being a bit surprised to find that nobody is using on-wheel lathes to perform regular light re-truing of locomotive wheels (the ACE people specifically mentioned doing this with a Hegenscheidt wheel lathe, 1970s technology)

I wonder whether this thread and the 4-axle thread are beginning to converge. When you go to six-axle cars to 'spread the weight', you're essentially building 'one-and-a-half' four-axle cars, with some theoretical saving of tare weight, but by definition increasing the anticipated vehicle capacity up to (design axle load x 6). Running any part of this capacity 'light', except when expressly booked as a 'balancing' move, is going to qualify as waste. Do we really need the equivalent of a 129' high-cube boxcar when two separate cars will almost always be more flexible?

I think the "best" place for three-axle steering trucks is right where you'd put them: under the articulated joint in stack trains, where their extended side-bearing characteristics would also make your patent applications work better. Running this kind of equipment in unit trains also gives better ability to control where the more-expensive capital equipment is at any given time, and to keep it in actual revenue movement as much of the time as possible.

[Anonymous]

I THINK AMTRAK SHOULD KEEP THE ROADRAILERS AND ANY OTHER MEANS OFHAULING LCL ITEMS IF THE FREIGHT RAILROADS DONT LIKE IT LET EM LUMP IT. THE CRYBABY PRESIDENTS OF THE FRT ROADS MIGHT GET A FEW BUCKS LESS IN THEIR PAY BUT ALL OF THEM ARE OVER PAID AND DONT GIVE A HOOT ABOUT THE PUBLIC OR THEIR EMPLOYEES

[Anonymous]

HEY GUYS THIS WAS ABOUT ROAD RAILERS

[Overmod]

Still is about RoadRailers (and by extension RailRunners)

One of the implicit points that has been made in discussions on these forums is that RoadRailers not only offer operating convenience, but also implicitly limit wheel loading (to a maximum that corresponds approximately to the maximum permissible highway weight per trailer, with some allowance for the rail bogie weight). This is substantially lower than the maximum axle loads in the preceding posts, even if the trailer suspension and rubber tires don't provide any of the weight-bearing and suspension (as they did in some of the earlier RoadRailer variants).

I agree wholeheartedly with tom that Amtrak should make a business niche out of LCL. However, they first have to develop an operational and business model that actually has a positive 'contribution to profitability' -- and then get the political savvy to present that to Congress as an important constituent of Amtrak funding, services, etc.

There is very little doubt in my mind that there's a place for relatively short, rapid RoadRailer trains that also carry a few 'multipurpose' passenger cars, going to all sorts of places that Amtrak currently doesn't serve. There's also little doubt that a single operating entity needs to be responsible for both freight and passenger handling. Where I begin to have doubts is in Amtrak's capability to orchestrate effective "freight" -- to give it the name that so many folks seem to be trying to dance around with terms like "MHC" and "express business" -- in ways that don't compromise its mission to run passenger trains.

One thing I can mention from previous experience is that these trains need to be capable of fairly high peak speed, ideally over a substantial percentage of the route, in order to make up incidental delays and still be at schedule at required intermediate points, as well as to accommodate the incidental stops associated with the 'materials handling'. That's true even if the overall timings recorded for, say, the old Trains Magazines 'speed surveys' show relatively indifferent overall times.

Now, I tend to prefer the term 'slack' (in the operations-research sense) to describe this allowance, rather than Don Oltmann's much more descriptive "slop". One of the points of OR is to quantify the slop, and figure out as exactly as possible how much should be provided, and where -- which is when you can start to call it 'slack' instead. A principal reason being to avoid surprises.

Amtrak's likely profitable 'niche' is going to involve time-critical delivery, rather than least-time delivery (especially since they don't seem to be able to retain the Post Office contract even for priority-mail -- one would think there could be effective political support both for cross-contracting and for appropriate capital subsidy for, say, RailRunners for all those little Post Office independent-contractor folks, but nobody involved seems to have the stones or the savvy to look into this successfully). It's also either going to have to involve service the 'carrier' railroads don't want, or can't successfully handle -- otherwise there can't be a 'win-win' mix, and as a result Amtrak trains will continue to languish... surprise, surprise! ... in sidings while 'competing' freight trains highball along.

Unfortunately, there is no money in Amtrak's budget for smart people to conduct express services planning -- let alone for the required service infrastructure, schedule and dispatch, etc. that are needed to make the thing fly. Since that isn't the case, and since nobody at Amdreck seems to have been able to get an express company, forwarder, etc. to effectively "outsource" this functionality (I won't tell the FedEx part of that story, but someone else might care to share) it's probably a good idea to let it die if, in the eyes of the Government's bean counters, it doesn't 'pay its way' on paper. (THEIR paper...)

Doesn't mean it won't come back, in proper form, as soon as the knowledge and political will appear...

[rich747us]

QUOTE: Originally posted by conrailman

Them Roadrailers was a money pit from the Start when they got them also the BoxCars too. All that money they spend on the Railrailers and BoxCars should have went to buy more Cars? [?]

I have to agree. Amtrak ought to work on improving passenger service. After all, that's what it's there for!

[Anonymous]

QUOTE: Originally posted by M.W. Hemphill

But it gets better. M&E proposed to the Class Is they would be extending new Amtrak trains onto routes currently not served by Amtrak, which would consist of an intermodal train with a rider coach tagging along.

Hmmmmm! An intermodal train with a single passenger coach. I think you've found a way to "privatize" rail passenger operations in the U.S. Just let the Class I's tack a coach onto the end of a double stack or TOFC, tell Amtrak they no longer need to enter the property since they the Class I's are now running the passenger service after a 50 year layoff, and call it good.

In fact, it wouldn't even have to be a full length coach, maybe just a little 30' box with windows for the few that actually utilize passenger service in this country, and maybe it could have a little cupola on top as an abreviated "dome". Think of the money that would be saved by no longer having to use a FRED!