General Discussion

Hello All,
I am a senior in high school, doing my senior exit project ...

For my project, I plan to make a computer model in excell, to calculate the expense of running various types of railroad equipment. I plan to use this model to prove the economic superiority of diesel power over steam. Last year, I wrote a 4,000 word senior exit paper on this very topic, so I cannot change.  I can send my paper to whoever wants to read it for kicks...then subsequently tear it apart. I haven't looked at since I turned it in a year ago.

I have only one problem ()...I need lots of data for this. I've raided the library system here in Charlotte, NC. I found some basic data of expenses, but I need data that is much more in-depth. I plan to trek up to the NC railway museum in Spencer one weekend, and see if I can find anything there: apparently there is a museum archive with lots of old Southern Railroad documents.

Does anyone know any good sources where I could find this type of data??? This project is due in over a month, so I've got some time to obtain stuff and work through any problems.

Of course I'll use G-scale models have when I present this product.

Any help would be WONDERFUL.
Thanks in advance!

Can you get access to Interstate Commerce Commission documents? Railroads were required to report annually to the ICC and this would include some cost data for your project. See the PM I sent you.

Well, that's a formidible project. Excel is an excellent modeling tool for this kind of project. The difficult part of fleet comparisons is ... finding comparable fleets. The average steam engine as of 1950 was 27 years old. The average diesel locomotive as of 1960 was about 8 years old. Overall diesel maintenance costs sure looked good. Even today, the average BNSF diesel fleet is about 15 years old. There has never been and probably never will be, in fleet terms, an average age comparable to the average age of steam engines at their heighth. That is a problem for any econometric comparison until you generate an age/cost curve for the respective motive power types. Those curves have been generated and are available. Too, as shops were modernized during the 1950s, that process improved the productivity of maintenance facilities and, while that no doubt benefitted the appearance of diesel fleet maintenance costs, it has to be segregated out as a factor, and that is hard to do without some very careful analytical work.

There are some detailed studies out there done by competent and experienced engineers and subject to peer review. Thomas Thelander did one for Swedish railways in the early 1960s and H.F. Brown did one for British Rail in 1960. Brown's was published by the Journal of the Institution of Mechanical Engineers and is available at their website. I believe Thelander's was also published there, but my copy of his paper is in storage and I don't have it for reference.     

petedabeat wrote:

I have only one problem ()...I need lots of data for this. I've raided the library system here in Charlotte, NC. I found some basic data of expenses, but I need data that is much more in-depth. I plan to trek up to the NC railway museum in Spencer one weekend, and see if I can find anything there: apparently there is a museum archive with lots of old Southern Railroad documents. Does anyone know any good sources where I could find this type of data??? This project is due in over a month, so I've got some time to obtain stuff and work through any problems.

The standard data reference source for these kinds of studies are the Transport Statistics of the United States, although an earlier title was Statistics of the Railways of the United States, published by the Interstate Commerce Commission.

These are generally available only at Federal Depository Libraries. Usually at least one or two public University libraries in each state has that designation. In your state, such libraries are at the University of North Carolina, Asheville, Gardner-Webb University (Dover Memorial Library), Appalachia State University (Carol Grotnes Belk Library), Campbell University (Carrie Rich Memorial Library),  University of North Carolina at Chapel Hill (Kathrine R. Everett Law Library). You should call ahead and ask specifically if they have the Transport Statistics of the United States published by the Interstate Commerce Commission. You indicate you "raided" the libraries in Charlotte, so I assume that the University of North Carolina (J. Murrey Atkins Library) didn't have the ICC reports, even though it is also a Federal Depository Library.

This is a preferred data source because each railroad was required to standardize its reporting to the ICC using the ICC accounting system, and so data is pretty much comparable from one road to another, one year to another.

Unfortunately, for what you are looking for, most of the data is worthless.

And this is because of an accounting anomaly that exists in reporting "maintenance" that needs to be understood before anyone goes off and starts assuming that what is reported as "maintenance" actually reports all maintenance. It doesn't.

The accepted economic service life of a "modern" steam engine was estimated at 30 years, and then they were usually scrapped. The initial purchase price was depreciated over that period of time, and maintenance was ... maintenance.

With the diesel-electric, obtaining comparable maintenance figures is somewhat problematic because the estimated economic service life is only 14 years. Using ICC accounting, any expenditure for maintenance that exceeded 50% of the cost of the locomotive was capitalized, rather than treated as an ordinary expense. So, when a diesel-electric is overhauled, the cost of that overhaul to give it a comparable economic service life to a steam engine doesn't show as a maintenance cost. Just to add a twist, any increase in the cost of the overhaul strictly due to inflation is simply included in the 50% threshold which means that even if the adjusted cost of the overhaul is considerably less than 50%, if the actual cost is over 50%, then the expenditure is capitalized.

In that fashion, the capitalization of such costs capitalizes the effect of inflation as well. The effect on diesel-electric maintenance costs as reported, then, is that substantial costs including inflation, are taken out of the reporting category for maintenance even though they are costs expended directly to keep the individual machine operational.  The significance of that is that ordinary maintenance that continues for a steam engine, for example, reports the entire inflated cost of maintenance as it accrues over the economic service life of the machine.

Succinctly, accounting requirements permit repair costs for the diesel-electric to be substantially capitalized, including effects of inflation on the particular "repair", whereas the nature of the steam engine required reporting of those costs as maintenance at their inflated levels throughout the service life of the machine.

So, while these are purely accounting considerations, what actually shows up under "maintenance costs" for steam and diesel do not reflect, in any meaningful way, a comparison of the costs of maintaining the respective motive power types over comparable periods.

In the hands of a professional engineer experienced in motive power studies, and with access to detailed fleet records and expenditures, it is possible to generate comparable cost figures for steam vs. diesel, but it does require specific access to the accounting records of what was capitalized as opposed to what was reported as a routine maintenance expense. You "can" generate the information from the "Transport Statistics" but you have to have a pretty good handle on statistical analysis.

Cost comparisons, however, have been done in this fashion, and this is the typical result [click to enlarge]:

Regarding other operating costs such as fuel, your hypothesis is on pretty shaky ground, but it will be interesting to see your conclusions.

 

Hmm.  I think I will try the UNC Charlotte library this weekend in addition to the shops at Spencer (UNC library might be better actually).  From what I can tell it is a depository library.  I've been there for projects before, and it has always been quite good.  I had never thought about it for this before now....

Also, thank you very much for the suggestion of ICC records. If they may be inaccurate, I should be able to skew them to fit my purposes!   Thank you MichaelSol for all the information on the ICC accounting records!

I will also search for the other sources mentioned here in my quest for information.

Thank you to all,

Peter

 PS - I'll find a place to host the model online during/after it's constructed, so that you guys can view/play with/comment on it.

You might address a letter to the Norfolk Southern's engineering department and see if they have records that you can use.   Certainly, toward the end of the steam era, the Norfolk and Western had steam locomotives with ages comparable to diesels today and servicing facilities for steam locomotives as modern for the time as facilities for diesels today.  I think this is one way to get a fair comparison.   And despite a tremendous investment in steam, N&W finally did decide to go diesel.

Michel: Do you think if Peter uses maintenance cost data from the steam-diesel transition period that the data might be purer. And what I mean by that is that few if any diesels would have been rebuilt during that time as they were still new.

 

Ed

 

MichaelSol wrote:

petedabeat wrote: I have only one problem ()...I need lots of data for this. I've raided the library system here in Charlotte, NC. I found some basic data of expenses, but I need data that is much more in-depth. I plan to trek up to the NC railway museum in Spencer one weekend, and see if I can find anything there: apparently there is a museum archive with lots of old Southern Railroad documents. Does anyone know any good sources where I could find this type of data??? This project is due in over a month, so I've got some time to obtain stuff and work through any problems. The standard data reference source for these kinds of studies are the Transport Statistics of the United States, although an earlier title was Statistics of the Railways of the United States, published by the Interstate Commerce Commission. These are generally available only at Federal Depository Libraries. Usually at least one or two public University libraries in each state has that designation. In your state, such libraries are at the University of North Carolina, Asheville, Gardner-Webb University (Dover Memorial Library), Appalachia State University (Carol Grotnes Belk Library), Campbell University (Carrie Rich Memorial Library),  University of North Carolina at Chapel Hill (Kathrine R. Everett Law Library). You should call ahead and ask specifically if they have the Transport Statistics of the United States published by the Interstate Commerce Commission. You indicate you "raided" the libraries in Charlotte, so I assume that the University of North Carolina (J. Murrey Atkins Library) didn't have the ICC reports, even though it is also a Federal Depository Library. This is a preferred data source because each railroad was required to standardize its reporting to the ICC using the ICC accounting system, and so data is pretty much comparable from one road to another, one year to another. Unfortunately, for what you are looking for, most of the data is worthless. And this is because of an accounting anomaly that exists in reporting "maintenance" that needs to be understood before anyone goes off and starts assuming that what is reported as "maintenance" actually reports all maintenance. It doesn't. The accepted economic service life of a "modern" steam engine was estimated at 30 years, and then they were usually scrapped. The initial purchase price was depreciated over that period of time, and maintenance was ... maintenance. With the diesel-electric, obtaining comparable maintenance figures is somewhat problematic because the estimated economic service life is only 14 years. Using ICC accounting, any expenditure for maintenance that exceeded 50% of the cost of the locomotive was capitalized, rather than treated as an ordinary expense. So, when a diesel-electric is overhauled, the cost of that overhaul to give it a comparable economic service life to a steam engine doesn't show as a maintenance cost. Just to add a twist, any increase in the cost of the overhaul strictly due to inflation is simply included in the 50% threshold which means that even if the adjusted cost of the overhaul is considerably less than 50%, if the actual cost is over 50%, then the expenditure is capitalized. In that fashion, the capitalization of such costs capitalizes the effect of inflation as well. The effect on diesel-electric maintenance costs as reported, then, is that substantial costs including inflation, are taken out of the reporting category for maintenance even though they are costs expended directly to keep the individual machine operational.  The significance of that is that ordinary maintenance that continues for a steam engine, for example, reports the entire inflated cost of maintenance as it accrues over the economic service life of the machine. Succinctly, accounting requirements permit repair costs for the diesel-electric to be substantially capitalized, including effects of inflation on the particular "repair", whereas the nature of the steam engine required reporting of those costs as maintenance at their inflated levels throughout the service life of the machine. So, while these are purely accounting considerations, what actually shows up under "maintenance costs" for steam and diesel do not reflect, in any meaningful way, a comparison of the costs of maintaining the respective motive power types over comparable periods. In the hands of a professional engineer experienced in motive power studies, and with access to detailed fleet records and expenditures, it is possible to generate comparable cost figures for steam vs. diesel, but it does require specific access to the accounting records of what was capitalized as opposed to what was reported as a routine maintenance expense. You "can" generate the information from the "Transport Statistics" but you have to have a pretty good handle on statistical analysis. Cost comparisons, however, have been done in this fashion, and this is the typical result [click to enlarge]: Regarding other operating costs such as fuel, your hypothesis is on pretty shaky ground, but it will be interesting to see your conclusions.

You may also wish to contact the Association of American Railroads which keeps all sorts of statistics and may already have what you are looking for.

www.aar.org

Also, in reference to the ICC Valuation records. The originals are kept at the national archives (Archive II Building) in College Park, MD. There is a very knowledgable archivist who works with railroads there, who's name is escaping me at the moment who might be of help.

LC 

 

SSW9389 wrote:

Michel: Do you think if Peter uses maintenance cost data from the steam-diesel transition period that the data might be purer. And what I mean by that is that few if any diesels would have been rebuilt during that time as they were still new.

Probably so, but that raises the question about what the diesel maintenance curve looks like "down the road". Too, the maintenance costs of steam are problematic during that period as the fleet was being allowed to age without an eye to sustaining that fleet. And there were still substantial numbers of locomotives built prior to 1915, built with old technologies and old metallurgical knowledge and skills. As you may note from the graph posted above, the differing technologies age quite differently in terms of maintenance requirements -- and the highly complex machine "ages" substantially more rapidly than the simpler technologies of whatever type. That shouldn't be a surprise ... but the rail industry is unique in many of its perceptions.

Even for someone as careful and knowledeable as H.F. Brown at Gibbs & Hill, it was a difficult endeavor to extract meaningful data and, as Brown pointed out, even his evaluation measured, ultimately, the practical results of the last design, a "modern" steam engine, for which there was a large scale application and data -- the Northern -- which supplied the data for comparison to diesel-electric fleets designed and built in the 1950s; and being marketed by the claim that they, in turn, were so much better than the diesel-electric designs coming out of WWII.

They probably were -- so would a steam engine fleet built in the mid-to-late 50s.

But, you go with the data you have, not the data you wish you had.

In this case, any comparison, no matter how carefully done, necessarily compares the two technologies across a generation gap in design and manufacture which necessarily operates automatically in favor of one over the other without offering a meaningful comparison of the technologies themselves.

Brown pointed to the logical fallacy of most studies done at the time, which extracted data in the fashion perhaps suggested by the young man on this thread: but the studies then did not accurately compare diesel with steam. They compared maintenance results of new machines with old machines and discovered that new machines, when new, require less maintenance than old machines, when old. (!)

Just about every purported study done had been done on that basis and that illuminated an additional risk (aside from the fact that part of the maintenance is capitalized and not shown) of just going into the Transport Statistics of the United States and grabbing the maintenance costs of the steam fleet and the maintenance costs of the diesel fleet -- which are nicely separated out in the reports by the way.

Find a library with Trains magazine back through the 40's.  Nearly every issue through the late 50's had articles on operating cost and payback trying to justify steam.

So, this project has begun!   I have begun my spreadsheet (what I have now isn't a model, I know, the model is coming...) and am traveling to spencer tomorrow to look at the NC Transportation Museum archives.  Hopefully I can find some information about individual locomotives, so I can actually start to create a model that will compare a single Diesel vs. a single steam locomotive, and compare costs, proving the Diesel to be more efficient.

 This is a link to the blog I have created for this project.  The spreadsheet is avaliable through a link on the blog:                                                             http://steamvsdiesel.blogspot.com/

Thus far, the blog contains very broad, industry-wide data.   The spreadsheet needs alot of work...

 Also, some of you have mentioned that I use old trains magazines.  I've searched google books and apparently the closest collection of old issues is located at Duke...  I'll go to Spencer tomorrow, and then decide whether I would even need them at all.

MichaelSol wrote:

Cost comparisons, however, have been done in this fashion, and this is the typical result [click to enlarge]:

Just for the sake of completeness and copyright attribution, and in case I am intriqued enough to pursue this information, could you cite the source of this image? 

Thanks... 

Limitedclear wrote:

You may also wish to contact the Association of American Railroads which keeps all sorts of statistics and may already have what you are looking for. www.aar.org Also, in reference to the ICC Valuation records. The originals are kept at the national archives (Archive II Building) in College Park, MD. There is a very knowledgable archivist who works with railroads there, who's name is escaping me at the moment who might be of help. LC

David Pfeiffer

JSGreen wrote:

Just for the sake of completeness and copyright attribution, and in case I am intriqued enough to pursue this information, could you cite the source of this image?  Thanks...

Oops, must have cut off the bottom of the image when I photoshopped it to reduce the image size. The attribution there reads "from H. F. Brown, "Economic Results of Diesel Electric Motive Power on the Railways of the United States of America," Proceedings of the Institution of Mechanical Engineers, 175:5 (1961)."

 

petedabeat wrote:

Hopefully I can find some information about individual locomotives, so I can actually start to create a model that will compare a single Diesel vs. a single steam locomotive, and compare costs, proving the Diesel to be more efficient.

Well, it's nice to see that you are approaching this from a strictly objective perspective!

However, regards "efficiency" I doubt there is anyone who would contest the observation that an internal combustion diesel-electric locomotive is inherently more efficient than an external combustion steam locomotive.