Steam Locomotives versus Diesels

[MichaelSol]

QUOTE: Originally posted by up829

Even if taken at face value, Browns study is after the fact while business decisions are based on assumptions about the future, without benefit of a crystal ball.

Can anyone predict what interest rates, inflation, and the price of oil will be 10 years from now? Should a fleet taxi or truck operator invest in hybrid technology, wait for hydrogen, or keep buying conventional gas/diesel? Keep in mind that during Bush I, economic growth was spotty, inflation and interest rates were moderate, we had huge deficits and oil was $30/barrel. Conventional wisdom at the time was that things weren't likely to change anytime soon. During the 90s, we had great growth, low inflation and interest rates, the elimination of the deficit, and $13-$18 oil.

And it is true that "conventional wisdom" is almost always wrong.

And it is also true that a failed management will nearly always claim that the future is impossible to predict when defending their failures, but will gladly demand their bonuses when the future happens to reward their efforts. Failure has no father, but success has a father, a mother, cousins, uncles and aunts, all demanding their bonus.

However, there is nothng accidental about business success. I discussed this a few months ago with one of my former professors, Chuck Snow, Chairman of the Management Department at Penn State. My question was this: "Where does the strategy for business success arise, in business or in the business schools, and what is the role of an academic or a consultant in that process?"

In the context of a major decision like Dieselization, the industry itself will create its success or engender its failure. The utlimate answer surely does come down the decision making process itself, how decisions are arrived at, the quality of the information, and the judgment and experience of the decision makers.

Brown looked at the results of the Dieselization process. Now, notwithstanding "how can anyone predict ...' the question really should be, "how can you know how to make a decision 1) unless you know what goes into the decision in the first place and 2) what are the results that are being sought"?

Evaluating the effectiveness of a decision like Dieselization cannot occur without utlimately knowing what the results are. You can't do that without a study like H.F. Brown's. It doesn't matter whether you "like" the results, or even agree with them. If you don't agree, do your own study. But the point is, any organization, in order to understand its own capabilities, has to know where it goes wrong as well as where it goes right. Call it Monday morning quarterbacking if you will; that's exactly what an efffective, winning team and their coaches do on Monday morning. If they want to win the next game, that is.

To fail to do that is a characteristic of failed enterprises. And further, once an organization begins to rationalize its failures, rather than remedy the decision making process that led to them, you have the classic definition of an organization or an industry in a state of failure.

This kind of a diagnosis is the product of the academic or consultant who is able to take a broad view of many companies and many decisions, and identify the key characteristics that correspond with success, and key characteristics that accompany failure.

Brown's study is essential to that process in viewing the rail industry.

It was not an "obscure" study. It was one of the most important and most controversial studies of its era, perhaps in the history of railroading. Milwaukee Road's L.W. Wylie directed me to Brown and the study. A former Coast Division superintendent and Electrical Engineer, he knew the study well. Milwaukee's motive power expert, H.R. Morgan, could recite it chapter and verse. Every engineer I worked with at Northwest Rail, Walter Gordon (ex-GN). E.E. Van Ness (ex-PRR), Gordon Rogers (see Trains, October, 1963 for his article on the BA&P and its motive power) and a lengthy list of others knew about it. The reason I have the study at all is because we were all talking about it and used it in our work as a primary authoritative reference.

Naturally, it raised the question of the decision making process which led to an enormous investment decision that ultimately generated a negative net rate of return for an entire industry.

Brown identified key areas where the predictions had been right:

1) lower fuel costs
2) lower water costs
3) lower engine house expenses

Brown identifed key areas where the predictions had been wrong:

1) service life of the diesel-electric locomotive.
2) maintenance costs of the diesel-electric locomotive
3) financing costs of the investment
4) the effects of the change in financing method as a result of the shorter economic service life.
5) lack of savings in crew costs
6) significantly higher lubrication costs

None of these had been correctly predicted by the railroad industry.

These were all areas in which the railroad industry had long experience.

What went wrong?

That's the story behind why management should have been all over that decision, because successful companies or successful industries do not rationalize their failures, they study them. They do not write them off as "history". If they did, that would be entirely characteristic of a failed or failing enterprise, and they would soon be history.

The way the rail industry treated the Dieselization debacle, however, has less to do with Dieselization than it did with the way the industry viewed the management process itself. And may go a long ways to explaining the state of the industry over the next few decades.

That process was characteristic of industries that fail.

Best regards, Michael Sol

[solzrules]

HMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM.

[ajmiller]

QUOTE: Originally posted by solzrules

HMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM.

Hey, you forgot an M in HMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM.

[solzrules]

[8][:(][:(][xx(][V][^]

QUOTE: Originally posted by ajmiller

QUOTE: Originally posted by solzrules

HMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM.

Hey, you forgot an M in HMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM.

SORRY. HM.
[:)][:D][8D][:I][}:)]

[Anonymous]

Like Mr. Diehl says - a Monday Morning Quarterback is still a Monday Morning Quarterback, no matter whether his name is Brown or Sol.

ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ

Old Timer

[MichaelSol]

QUOTE: Originally posted by nanaimo73

MichaelSol-
Does Brown say the railroads should have dieselized, but just in a different manner ?
You have posted on this thread that switchers had different economics, and they should have been replaced first. What does Brown say about all of the road steam power that was worn out at the end of WW2 ? (Like CMSP&P's slobber-stacks ?). Does he say they should have been replaced post war with newer steam (like 4-8-4s) or diesels ? Did he have different findings on post-war freight or passenger locomotives ?

He refers to David P. Morgan's Steam's Finest Hour as a useful reference on what modern steam would look like.

He pointed out that 40% of existing Steam power post war was built before 1915. This skewed "averages" considerably against Steam if you used only system numbers for any calculation. He was able to sort according to age class, and also noted that post-1930 steam was highly efficient.

Brown didn't make recommendations. It was simply as the title stated: "results." People could draw whatever conclusions they felt appropriate to their own situations.

What was interesting were the standard power curves, which suggested the thermodynamic basis for the road diesel/steam findings. Above 20 mph, the complex diesel-electric machine lost efficiency, whereas the simpler steam engine gained efficiency, the faster it went above that speed, to a reasonable design limit. That was something inherent in the comparison of a constant horsepower machine against a variable horsepower machine. Steam became very efficient compared to diesel at speeds above 30 mph.

This is probably suggestive as to why Steam-powered roads preferred higher speeds, whereas modern Diesel-electric powered roads prefer average speeds around 20 mph simply from the standpoint of economic efficiency. Of course, that collides with and creates congestion problems due to the slower speeds, suggesting that the congestion problem would be signficantly different today, if railroads were able to take advantage of the economic advantages of higher speed operation with Steam. Now, that would be an interesting study.

Best regards, Michael Sol

[bigfoote]

QUOTE: Originally posted by bigfoote

I like them both.Thats why I model the 50's in N-Scale. I was woundering, how a steam train pulling 12 cars with a diesel helper go the same speed, are they throttled together, or do they use radios?

[Anonymous]

In order to measure whether a sound business decision was made, you also need to know what ALL the assumptions actually were and if they were realistic and achievable at the time the decision was made. Brown's study goes into one scenario, but there may be others. Was management expecting higher traffic levels? More favorable rates from the regulators? More labor savings than they got? Faster schedules resulting in quicker car turnaround and utilization? Any one of these could have made up for the financing costs. Organizations also do things for reasons other than ROI. Consider competitive advantage/disadvantage and its effect on market share. That so many organizations made the same decision suggests something more like d. all of the above.

Drawing another modern day comparison. Was it a bad business decision in the context of the 80s to invest in western oil shale projects? It would seem so viewed from the 90s, but those who hung on and were villified during the 90s are heros now.

[MichaelSol]

QUOTE: Originally posted by up829

In order to measure whether a sound business decision was made, you also need to know what ALL the assumptions actually were and if they were realistic and achievable at the time the decision was made. Brown's study goes into one scenario, but there may be others.

Brown did not study assumptions. He generated a study of results. The discussion period that folllowed his report generated some discussion about assumptions, but, as a carefully focused engineering study, the report itself avoids speculating on any "what-ifs" or "why's". It is probably true that every railroad had different "assumptions" just as each railroad had different needs. The Boston & Maine was not the Union Pacific. The Virginian was not the Santa Fe.

However, when you see that the economic results are negative, if you were a railroad you might take that infomation and go back to your company assumptions and see what they were based on, and how any errors were made in those assumptions. You might be interested in any mistake that cost the company millions, and not only find out why, but who.

I read OldTimer's remarks and TomDiehl's remarks with some sense of wonder. History? The latest annual report is "history." All information is "history".

That is why Law Schools study old cases, and that's why Business Schools teach almost entirely using case studies of "past" performance. Indeed, "history" is the primary instructional tool at the Army War College because it is still the best method of examining the process of how people make good decisions, and why people make mistakes.

Any accountant will tell you, the single best predictor of a company or an industry's future performance is its past performance.

Past performance, of course, is greatly affected by important decisions made, with wide-ranging implications. Dieselization happens to be one of the events in the rail industry for which the results did not nearly measure up to the expectations, indeed, the opposite happened. Like the recent Telecom industry debacle, yes, that does call into question the decision-making process of any number of companies and any number of management teams who all came to the same conclusion at the same time, notwithstanding warnings from both investment analysts and engineers as well.

I gather from your remarks that if railroads had made the wrong assumptions about traffic, cycle time, et.al., that "getting it wrong" on those assumptions might somehow justify 'getting it wrong" on the approach to Dieselization. I'm puzzled, if I'm understanding your point. One mistake does not justify making more mistakes. They're all mistakes. Good companies carefully examine what those are and why, and fix the process. Other companies offer an "oh well, that's history," and go on to make more mistakes.

As I mentioned, that is a key characteristic of companies that fail. Interestingly, judging by the results of the next two decades after Dieselization, there is an argument that the rail industry was one of those industries, for precisely those reasons.

Best regards, Michael Sol

[germanium]

The railroad to heaven will be powered by steam locos, built of imperishable components so there will be no adverse effect on the ROI. The coal used to fuel them will be hand-loaded into hoppers in Hell by those railroad executives who believe in diesels. Well, we can dream !!

[Anonymous]

History is an excellent teaching tool and certainly usefull in figuring out who to blame for legal proceedings, but past performance does not always predict future success and often decisions have to be made based on incomplete or unknown information. The results and an accurate assesment of some aspects may not be possible for years afterward. Those who made the decision and approved it may be long gone and if there's no possibiility to do it over, there's little to be gained by spending a a lot of time looking backwards. CEOs and boards just don't have the time to do much of that.

Business schools also do case studies into something called the Technology Life Cycle, which gets into the risks, rewards, costs, and service life of technology at various points in time. In that context, dieselization and the demise of steam followed a very typical pattern. Early adopters at the leading edge in the 30s took the greatest risk, but solved specific problems i.e. lower cost lightweight passenger trains designed to increase ridership and market share. Early FT adopters used them to solve operational problems with long non-electrified tunnels i.e. Stampede & Moffat. By the post war period when most roads dieselized, the technology was new but proven both by the early users and wartime use by the Navy. Many of the F3's purchased at this time were later upgraded to F9's and continued to run in mainline service well past the date Brown's study was done. Steam was in the obsolete phase of the life cycle and like buying a Pentium 3 notebook or analog cell phone today, had a very short service life.

Some organizations are very sucessful at betting the company on new technology. Boeing has done this repeatedly with the B17, 707, 747, and most recently the 777. Others I won't mention, constantly look over their shoulder at something they shouldn't have done and get run over by something unexpected. Railroads with a few exceptions, tend to be fairly conservative, and as an industry dieselized following the classic business school model, albeit some much better than others. Deploying new technology across a distributed organization is one of the most difficult strategic decisions organizations face. When it doesn't produce the expected results for an entire industry, the reasons are usually complex, similar to engineering disasters which involve multiple component failures in unanticipated ways. Was the engineer who designed the World Trade Center an incompetent fool because he did not forsee a terrorist crashing a fully loaded jumbo jet into it?

[nanaimo73]

Michael,
The Milwaukee Road decided to completely dieselize in 1947, which may may have been influenced by the 1946 coalminers strike. Does Mr. Brown's study mention the effects of this strike on the railroad's dieselization plans ?

By 1947 the Milwaukee Road had purchased ALCOs, Baldwins, Davenports, EMDs, FMs, GEs and Whitcombs. Does the study chastise the railroads for buying locomotives from so many companies ?

[solzrules]

QUOTE: Originally posted by germanium

The railroad to heaven will be powered by steam locos, built of imperishable components so there will be no adverse effect on the ROI. The coal used to fuel them will be hand-loaded into hoppers in Hell by those railroad executives who believe in diesels. Well, we can dream !!

Yes but the Brown study clearly showed that. . . . . . . oh. wait. [:D]

[solzrules]

QUOTE: Originally posted by up829

History is an excellent teaching tool and certainly usefull in figuring out who to blame for legal proceedings, but past performance does not always predict future success and often decisions have to be made based on incomplete or unknown information. The results and an accurate assesment of some aspects may not be possible for years afterward. Those who made the decision and approved it may be long gone and if there's no possibiility to do it over, there's little to be gained by spending a a lot of time looking backwards. CEOs and boards just don't have the time to do much of that.

Business schools also do case studies into something called the Technology Life Cycle, which gets into the risks, rewards, costs, and service life of technology at various points in time. In that context, dieselization and the demise of steam followed a very typical pattern. Early adopters at the leading edge in the 30s took the greatest risk, but solved specific problems i.e. lower cost lightweight passenger trains designed to increase ridership and market share. Early FT adopters used them to solve operational problems with long non-electrified tunnels i.e. Stampede & Moffat. By the post war period when most roads dieselized, the technology was new but proven both by the early users and wartime use by the Navy. Many of the F3's purchased at this time were later upgraded to F9's and continued to run in mainline service well past the date Brown's study was done. Steam was in the obsolete phase of the life cycle and like buying a Pentium 3 notebook or analog cell phone today, had a very short service life.

Some organizations are very sucessful at betting the company on new technology. Boeing has done this repeatedly with the B17, 707, 747, and most recently the 777. Others I won't mention, constantly look over their shoulder at something they shouldn't have done and get run over by something unexpected. Railroads with a few exceptions, tend to be fairly conservative, and as an industry dieselized following the classic business school model, albeit some much better than others. Deploying new technology across a distributed organization is one of the most difficult strategic decisions organizations face. When it doesn't produce the expected results for an entire industry, the reasons are usually complex, similar to engineering disasters which involve multiple component failures in unanticipated ways. Was the engineer who designed the World Trade Center an incompetent fool because he did not forsee a terrorist crashing a fully loaded jumbo jet into it?

Yes, that may be true. You sound like someone with a tad bit of experience in this area. Question: Did business schools discuss "Technology Life Cycle" back in the late 40's or early 50's? Actually, the people who made descisions about the usefulness of diesel as opposed to steam would be higher management. I am guessing that they would have experience of about 20-30 years. I am also guessing that once they completed their business training that they would have not returned to school at a later date (continuing education not being a fad untill the 80's, with some exceptions). This means that they would have been trained in these matter in the 20's or the 30's. Someone with a business degree from the 20's may have a different opinion on risk taking than someone with a degree from the 30's (pre-great depression and post-great depression). Someone also may have a different opinion about technology depending on what decade we are reffering to here. The descision to scrap an entire locomotive fleet one the basis of several manufacturer sponsored studies without any hard data from a previous railroad's experience to me seems like a risky business descision. Would the concept of "Technology Life Cycle" be present in school acadmeics at that time (20's or 30's)? Also, I think that the architect of the WTC was a genious. His plan (and the subsequent engineering) withstood tremendous trama and remained standing long enough for everyone to get out who could.

[MichaelSol]

QUOTE: Originally posted by nanaimo73

Michael,
The Milwaukee Road decided to completely dieselize in 1947, which may may have been influenced by the 1946 coalminers strike. Does Mr. Brown's study mention the effects of this strike on the railroad's dieselization plans ?

No, his was a study of cumulative results of American Class I Railroad experience, since individual company assumptions, fleet distribution (road vs. yard), fleet age, traffic changes, operating characteristcs, etc. among individual roads could yield "artifacts" -- results unique to a particular railroad which, while impacting the results of Dieselization, might not have not reflected the characteristics of the motive power.

When the decision to Dieselize the Milwaukee was made, it was intended to Dieselize the whole road, including the Electrification zones. The Electrification, of course, would have been immune from the effects of the coal strike, and so presumably, since the Electrification was likewise slated to be scrapped, the decision was based on considerations other than the coal strike.

Best regards, Michael Sol

[MichaelSol]

QUOTE: Originally posted by nanaimo73
By 1947 the Milwaukee Road had purchased ALCOs, Baldwins, Davenports, EMDs, FMs, GEs and Whitcombs. Does the study chastise the railroads for buying locomotives from so many companies ?

He didn't look at Milwaukee, except as it appeared the national Class I totals. As I mentioned, his study was not designed to go so far as to look at the impact of individual company policies on the decision, merely to see what the actual economic results of the cumulative decisions were.

Best regards, Michael Sol

[Anonymous]

QUOTE: Originally posted by solzrules

QUOTE: Originally posted by germanium

The railroad to heaven will be powered by steam locos, built of imperishable components so there will be no adverse effect on the ROI. The coal used to fuel them will be hand-loaded into hoppers in Hell by those railroad executives who believe in diesels. Well, we can dream !!

Yes but the Brown study clearly showed that. . . . . . . oh. wait. [:D]

Yes but if you look at the fact that these heavan trains will not be quite profitable to due to the expense of obtaining the coal.......bah!!! [(-D][(-D][(-D]

[AnthonyV]

There are numerous examples of newer technologies displacing existing, well-established technologies in a relatively short time, e.g., transister vs. vacuum tube, turbojet/turbofan vs. propeller, Diesel vs. steam. These events usually signify that the new technology is fundamentally superior to the existing technology. The marketplace typically moves much slower to assume the risk of adopting a new technology with only marginal advantages.

Based on this premise, it would appear that the Diesel is fundamentally superior to the steam locomotive or else it would not have displaced so rapidly steam technology that was in development and use for over 125 years

The argument put forth by both Michael and Brown (Note that I have read Brown's paper.) is that there was no fundamental advantage offered by Diesel technology. Their basic argument is that the Diesel did the same job as steam but with greater complexity and at higher overall cost. Brown's argument is that traffic patterns were changing gradually over time due to the loss of both short-haul freight traffic and passengers, leading to longer and heavier trains. Locomotive technology evolved gradually to meet these needs.

Both Michael and Brown have asserted that multiple-unit capability offers no significant advantages, but complexity. They have also asserted that the Diesel had short economic service lives compared to steam. One advantage that Brown gives to the Diesel is its higher availability, 90 percent to 60 percent.

Based on ICC data, Brown reported that 2.41 Diesel locomotives were required for each steam locomotive due to the lower power of the typical Diesel. Brown estimated that the power of the average Diesel in the 1950's was 1,500 HP. He also estimated that a fleet of 11,800 3,600-HP steam road locomotives was equivalent to the 18,900 1,500-HP Diesel road locomotive fleet in the mid 1950's. This result was derived by adjusting the total available road Diesel horsepower to account for both the higher power per typical steam unit and steam's lower availability.

Brown presented ICC traffic data in terms train miles, gross tons per train, cars per train, revenue ton miles, revenue tons, passenger miles, etc. He presented detailed annual road and yard locomotive inventory data from about 1915 through 1957 in terms of type (steam, diesel, or electric) quantity, and age. He also presented detailed cost data in terms of repairs, fuel, enginemen, facilities, water, and lubricants for road and yard locomotives as filed with the ICC for the same period.

Brown's analysis of ICC data on road Diesel locomotive performance indicates that average fleet mileage for road locomotives in the early 1950's was about 100,000 miles per year, which declined steadily to about 86,000 miles per year in 1957. Note that during this period the Diesel was handling over 85 percent of the train mileage. No such results were presented for the steam locomotive fleet.

Using the data presented in Brown's paper, I performed calculations of steam locomotive fleet mileage and repeated the Diesel locomotive fleet mileage calculations in Brown's paper. The results are discussed below.

Average annual fleet mileage for road steam and Diesel locomotives is shown respectively in Tables 1 and 2 below. Also shown for each year is the average age of each road locomotive fleet and an estimate of the total miles traveled per locomotive, analogous to an odometer reading. The latter was estimated by multiplying the average age for a given year times the average annual mileage for the entire period at the bottom of each table.

Table 1: Steam Road Locomotive Fleet Mileage
Year Average Age Annual Mileage "Odometer Reading"
1915 ........ 15 ........ 26,400 ................ 413,000
1920 ........ 14 ........ 24,444 ................ 386,000
1925 ........ 17 ........ 25,385 ................ 468,000
1930 ........ 18 ........ 19,556 ................ 496,000
1935 ........ 21 ........ 27,054 ................ 579,000
1940 ........ 24 ........ 30,296 ................ 661,000
1945 ........ 26 ........ 36,529 ................ 716,000
1950 ........ 26 ........ 30,740 ................ 719,000
average annual mileage 27,550



Table 2: Diesel Road Locomotive Fleet Mileage
Year Average Age Annual Mileage "Odometer Reading"
1953 ........ 3.5 ........ 101,000 ................ 328,000
1954 ........ 4 ........ 95,000 ................ 375,000
1955 ........ 5 ........ 94,000 ................ 469,000
1956 ........ 6 ........ 92,000 ................ 562,000
1957 ........ 6.6 ........ 86,500 ................ 618,000
average annual mileage 93,700



As shown, the average annual mileage of the steam road locomotive fleet over the period of 1915 through 1950 was about 27,000 miles per year. Average annual fleet mileage was about 25,000 in the 1920's and 1930's, then rose to over 30,000 from 1940 to 1950. The peak was over 36,000 miles per year during WWII.

This is in contrast to the average annual fleet mileage for the road Diesel fleet of about 90,000 miles per year. Note that the Diesel was handling over 85 percent of the train miles by the mid-1950's.

The average steam road fleet mileage (i.e., the "odometer reading") was also relatively low, ranging from about 400,000 miles at 15 years old in the 1920's to about 720,000 miles at 26 years old in 1950. This is in contrast to the average road Diesel accumulating over 600,000 miles by 1957 despite an average age of just 6.6 years.

These results suggest that utilization of the steam locomotive fleet was very poor compared to that of the Diesel. They also suggest that based on average annual mileage, a considerably larger steam locomotives fleet is required to perform the same job as a Diesel fleet. The results also illustrate the point I made in a post a while back: Locomotive economic service lives must be measured in terms of revenue ton-miles (or miles), not years, if the term is to have any meaning.

From 1915 though 1950, average road locomotive mileage was in the range of 27,000 miles per year. It peaked at about 37,000 miles per year during the high-traffic years of WWII. The changes in locomotive utilization during this time period were gradual up to about 1950.

Then a dramatic step-change occurred. Almost overnight, average annual road fleet mileage went from about 30,000 miles per year to 85,000 to 90,000 miles per year.

An argument could be made that this phenomenon is a direct result of purchasing a new road locomotive fleet. It could be argued that this would have happened with a brand-new road steam engine fleet as well.

But would it have?

How would a new steam fleet have looked?

As I mentioned earlier, Brown estimated that a steam road locomotive fleet of 11,800 3,600-HP locomotives was equivalent to the existing 18,900 1,500-HP Diesel road locomotive fleet in the mid-1950's. But is it? Could this fleet satisfy the need to haul heavy mainline traffic as well as work branch-line trains and locals with much less tonnage? How would this fleet be managed? Is there a study that shows how such a fleet could be managed to handle the railroads' needs?

Note that the average annual mileage for Brown's hypothetical steam road fleet would be about 57,000 miles per year in 1957, which is 50 to 100 percent greater than historical levels. Is this a realistic level of improvement?

Or, would more locomotives be needed than Brown estimated? Would a fleet of high-HP steam units be needed for mainline work and a fleet of smaller units be needed to handle the light work? How much would such a fleet cost? What would be required in terms of manpower, repairs, and facilities?

I don't know the answers.

We do know that the Diesel could do both, which is where multiple-unit capability presents its primary advantage. Diesels could be sent out on light-duty branch lines and locals, yet be assembled into a locomotive set to handle high-tonnage trains.

As I said, I don't know the answers to these questions. But like Old Timer and others have said, we do have history to serve as a guide. The steam road fleet was relatively large and utilized poorly for decades. The Diesel road fleet was relatively smaller and was able to be utilized much more effectively, possibly leading to lower overall costs than what is possible with steam, new or old.


Thanks
Anthony V.

[solzrules]

QUOTE: Originally posted by rgroeling

QUOTE: Originally posted by solzrules

QUOTE: Originally posted by germanium

The railroad to heaven will be powered by steam locos, built of imperishable components so there will be no adverse effect on the ROI. The coal used to fuel them will be hand-loaded into hoppers in Hell by those railroad executives who believe in diesels. Well, we can dream !!

Yes but the Brown study clearly showed that. . . . . . . oh. wait. [:D]

Yes but if you look at the fact that these heavan trains will not be quite profitable to due to the expense of obtaining the coal.......bah!!! [(-D][(-D][(-D]

If in your in hell shoveling coal I think the company will get a pretty good labor rate......[}:)][;)][8D]

[germanium]

Can we now draw a veil over these proceedings ? Much erudite reasoning has been advanced, but we are now reaching the "how many angels can dance on the head of a pin ?" stage of fruitlessness.

[solzrules]

Okay, this may be a stupid question, but you can chalk it up to me not having a computer for the last 5 years and not being involved in these conversations (even if I am not now). MichaelSol do you have any actual experience with the rail industry (the Milwaukee Road in particular)? How about AnthonyV? You guys seem to be able to quote stuff that nobody would know unless they were privy to actual company studies.....
I am not asking for details or anything................................[:D]

[solzrules]

QUOTE: Originally posted by germanium

Can we now draw a veil over these proceedings ? Much erudite reasoning has been advanced, but we are now reaching the "how many angels can dance on the head of a pin ?" stage of fruitlessness.

Come on now. These discussions will make a difference some day if the railroads will go back to steam.............We will look like geniouses!!

[MichaelSol]

Anthony, you need to look at what you're saying. You state as follows:

"The argument put forth by both Michael and Brown (Note that I have read Brown's paper.) is that there was no fundamental advantage offered by Diesel technology."

Wow, I think I've said it about 15 times now. In terms of fuel efficiency, and water savings, all due to "technology" there were signficant advantages to Dieselization. Brown says the same thing. From a strictly "operating" standpoint, there was a fundamental advantage regarding at least two specific cost drivers.

Now, you did something odd wth your mileage comparisons. As Brown warned, you cannot compare a 6.6 year old Diesel with a 27 year old steam engine, particularly when 40% of that "steam engine" represents locomotives predating 1915.

The average annual mileage data you present needs to be a little more rigorous. The decline in annual mileage of the Diesel locomotive shows a consistent phenomenon at work. I have no reason to think from looking at the data graphically that it is not linear.

Accordingly, your average annual mileage decline clearly relates to age. If you extend that trend line out to equal what you show for actual results for steam, it paints quite a different picture.

At 26 years, the Diesel fleet would show an annual average mileage of only 15,209 miles. In other words, the fleet would be just about non-functional. The steam fleet as you show it actually averaged 43,240 miles annually at an average age of 26 years, almost triple what a Diesel fleet average would look like.

As Brown points out, 40% of the Steam fleet in 1945 was built prior to 1915. He also points out that Steam constructed after 1930 was of a different quality than steam built prior to that date. Whole new classes of high efficiency, high horsepower units were being put into road service, while older models were relegated to branchline work, and in many cases, simply made inactive although their numbers continued to be counted for statistical purposes (which skews the numbers against Steam).

However, if the 40% of pre-1915 steam is assumed to have been largely functioning at 1915 levels, and "new" Steam represented a different performance level, linear regression shows that post-1930 steam would have been averaging about 43,281 miles annually at the age of 26 years.

As built, "Modern" Steam was an entirely different matter. Milwaukee Road's Alco "Baltic" class, 1934 design, for instance was expected to run 150,000 miles annually under very high speed conditions, and did, and the Hudson class that replaced them did so easily. Milwaukee Road "Northerns" routinely topped 200,000 miles per year.

Further, an annualized "locomotive miles" measure understates that, on a per horsepower basis comparison to modern Road Steam, the Diesel-electrics were operating 800,000 miles annually, combined unit mileage, to haul the equivalent of 200,000 miles with a single unit Northern. If brand new Diesel-electrics were only averaging 100,000 miles per year per unit, then eight such units were realistically necessary to do the work that a Northern was doing.

Then, given the life span of that Northern, 24 such Diesel units, or their overhauled equivalents, were necessary to match the economic productivity of that Northern. Now, the cost of 24 Diesel-electric units compared to the cost of 1 Northern? Is it any wonder that the Diesels required financing where the Northern did not?

In either case, modern Road Steam was at least double the capability of the road Diesel unit, using annualized mileage/economic service life as the basis for comparison.
Averge
Age Diesel Steam
1 108739, 100990
2 104998, 98680
3 101256, 96370
4 97515, 94060
5 93774, 91750
6 90033, 89440
7 86292, 87130
8 82550, 84820
9 78809, 82510
10 75068, 80200
15 56362, 68650
20 37656, 57100
25 18950, 45550
26 15209, 43240
30 244, 34000

You can seen in these figures the reasoning as to why the Diesel-electric was originally thought to have a 20 year economic service life, as its "ability" at that point nearly equalled that of Steam at 30 years. This is exactly what the railroads would have been looking at at the time. As Brown points out, however, the 30 year figure for Steam -- the point at which the economic utility ceased, was not reached by the Diesel-Electric at age 20 years. Instead, it was being reached between 12 and 14 years. The data presented above, then was optimistic.

The data Brown had available, through 1957, represented a Diesel fleet on the average 6.6 years old. The data already showed something was wrong with the assumptions. By the time road Diesels were beginning to reach the end of their economic service lives at age 12, it was obvious. But Dieselization was for all practical purposes complete.

There was no going back, ROI was now such that there was nothing left to finance a complete remake of the fleet once again. Whether it was a mistake or not, the railroads were stuck with it one way or another by the time the operating results became clear. Being "stuck" with a decision is not, as some as have contended, the same thing as historical vindication.

By then, using your annual mileage metric, the figures of modern Diesel vs modern Steam looked like this:
Age .. Diesel ....Steam
1......115410........100990
2......108340.........98680
3......101270.........96370
4.......94200..........94060
5.......87130..........91750
6.......80060..........89440
7.......72990..........87130
8.......65920..........84820
9.......58850..........82510
10.....51780..........80200
11.....44710..........77890
12.....37640..........75580
15 .......................68650
20........................57100
25........................45550
30........................34000

Interestingly, the economic service life of the Diesel-electric represents just about 1 million miles, before replacement or heavy overhaul. That hasn't changed in 50 years. Considering technological advances over that time, that seems odd.

Steam, built prior to 1915, was just as likely to be worn out at 1 million miles or less. Post-1930 Steam, on the other hand, appears from the statistical record to have had a 2 million mile service life, in addition to its higher operating efficiencies at speeds between 20 and 60 mph. It appears that advances in engineering, metallurgy, and other technology were making genuine contributions to the economics of Steam power, and there is no reason to suspect that Steam would not have continued to benefit from those advances. The mystery is why the Diesel has not benefitted in that fashion.

If someone really wanted to raise a ruckus, an interesting study would be as to whether Dieselization, while representing an apparent technological advance, actually represented a technological dead-end that prevented continuing technological development of the mode more susceptible to genuine and continuing technological improvement.

However, taking your analysis to its logical conclusion, you must both look at "young" Steam as well as "old" Diesel, instead of how you handled it, which was to compare "young" Diesel with "old" Steam. A brand new Diesel looked pretty good compared even to new Steam; it looked terrific compared to old Steam. But that just wasn't the whole story. It wasn't even the story at all as it turned out. Road Steam was a better runner when it was 20 years old than a road Diesel was at 10.

And this is more to the point of what Brown was saying. The economic service life of modern Steam power was considerably different than the Diesel-electric locomotive, and while the Diesel-electric locomotive represented technical advantages in regards to fuel and water consumption, notwithstanding its higher availability -- in its early years -- its functional abilities declined at a very rapid rate, compared to Steam power. This had a profound economic impact which overwhelmed its technological advantages.

No doubt, engineering complexity played a significant role in that rapid deterioration.

You did not mention, and I am not sure why since the numbers I generated above are entirely consistent with those curves, the cost per 1000 hp mile curves in Brown's paper which contraindicates what you attempted to put together above, and which plainly reflected the much high maintenance costs on that comparable metric, of the Diesel, as well as the much higher rate of cost growth, which plainly shows as well in my figures above.

Best regards, Michael Sol

[AnthonyV]

Michael:

Maybe I missed something, but isn't it your and Brown's position that the switch to Diesel placed the railroads at a fundamental disadvantage?

As for average mileage, the numbers are very revealing. Locomotive inventory data is presented in terms of numbers and average age. Average annual mileage was very low.

I can think of two extreme cases regarding the mileage and age issue.

One case is the every locomotive traveled the average annual mileage each year. This is obviously an unrealistic case.

The other is that fewer, relatively modern steam locomotives traveled very high mileage (100,000+) each year, and the older ones traveled hardly at all. Wouldn't this present a distorted picture of the average age? For example, isn't this like having a 25 year old car that sits in the driveway and is hardly driven, yet it is included in the average age of the cars in the driveway?

Would not a mileage-weighted average age be more accurate if in fact the newer locomotives were accumulating most of the miles?

The questions that remains in my mind is: What would what would a steam locomotive fleet look like that meets the actual railroad's operating needs? How many locomotives be needed? What would the distribution of horsepower look like?

Looking at it another way, does the steam locomotive fleet in the late 1940's represent what is actually needed in terms of numbers and power distribution?

For what its worth, an example of fleet averages of modern Diesels, I calculated that the BNSF locomotive fleet averaged 90,000 miles per year with an average age of 16 years based on numbers from the following report: "BNSF Railway Company, Class I Annual Report to the Surface Transportation Board for the Year ending December 31, 2001".

As far as Diesel repair costs, I do have some numbers, but I can handle only one aspect of this issue at a time.

Thanks
Anthony V.

[MichaelSol]

QUOTE: Originally posted by AnthonyV

Maybe I missed something, but isn't it your and Brown's position that the switch to Diesel placed the railroads at a fundamental disadvantage?

Well now, you're changing the words. For several important cost drivers, Dieselization offered significant advantages due entirely to the differences in technology. However, on the economic side of things, Dieselization offered an entirely new cost driver -- financing charges -- and a strong tailwind to those financing charges in the form of a substantially reduced economic service life compared to modern Steam.

During the Dieselization process, shop forces could go down dramatically, no doubt. But, by the time the average fleet age began to reach 10, 12, and 14 years, suddenly railroads were confronted with overhaul costs that represented a significant proportion of the original purchase price, plus the need to perform such maintenance on a locomotive by locomotive basis, as essentially custom shop work. Expensive labor and expensive parts.

Brown's study hadn't been able to report on that aspect because the average age of the Diesel fleet was only 6 or 7 years at the time of his study, but his cost curves clearly predicted it and that explains his conclusion that, when those overhaul costs were prorated over the life of the machine, "maintenance" costs exceeded those of the Steam fleet by a substantial margin even though those expenses were, in many cases, capitalized by the railroads rather than reported as repairs.

Best regards, Michael Sol

[MichaelSol]

QUOTE: Originally posted by AnthonyV
For what its worth, an example of fleet averages of modern Diesels, I calculated that the BNSF locomotive fleet averaged 90,000 miles per year with an average age of 16 years based on numbers from the following report: "BNSF Railway Company, Class I Annual Report to the Surface Transportation Board for the Year ending December 31, 2001".

Interesting. Class I railroading is so much different today; the branchline system is by and large gone, a much higher percentage of locomotives operate purely on mainlines, not much stopping for single cars anymore.

However, a 16 year old locomotive is probably about 4.8 years after its major overhaul, and looking at the expected productivity based on the 12 year economic service life estimate set out above, a "4.8 year old" Diesel locomotive would be expected to produce about 90,000 miles per year.

In 1957.

I guess I'm a little surprised to see it in 2006.

Best regards, Michael Sol

[MichaelSol]

QUOTE: Originally posted by AnthonyV
One advantage that Brown gives to the Diesel is its higher availability, 90 percent to 60 percent.

The "availability" measure is just about meaningless for two reasons.

1) For the study period in question, it took at least four diesel units to pull the same tonnage as a Northern. In terms of locomotive miles, that meant a 4 to 1 advantage to steam. Indeed, the diesel fleet would have to show 100,000 locomotive miles to equal the productivity of a Northern showing only 25,000 miles. For a Northern operating 300,000 miles annually, Diesels would be required to operate 1,200,000 locomotive unit miles to move the same freight tonnage.

If the Northern were restricted by its "availability" it could only operate 180,000 miles annually, a loss of 120,000 locomotive miles.

Interestingly, if the four Diesel locomotives were limited solely by their "availability" rating of 90%, they would be able to generate only 1,080,000 locomotive unit miles. A loss due to availability of the identical number of 120,000 locomotive miles lost by the single Steam unit.

The 90% availability looks to have the same ramifications as the 60% availability because of the additive effect of probabilities with each discrete event (locomotive).

However, this is one reason why "locomotive miles" is a tricky and somewhat superficial statistic applied to motive power types of significantly different horsepower. The quality of the Northern locomotive miles are different than the quality of the Diesel locomotive miles.

2) Even today, the 90,000 mile annual fleet average of BNSF locomotives is substantially below what modern Steam was demonstrably capable of in 1950

The Milwaukee Baltic class ran off 150,000 miles per year, but that only represented 7 hours of actual daily work. It was able to sit and relax for the other 17 hours, utilizing only half of its supposed "availability." A modern Diesel with 90% availability is moving only 10 mph at the current average annual mileage. Since average train speed on the BNSF is about 23 mph, that means the average locomotive today is not moving anything about 50% of the time.

Locomotive "availability" was a more or less meaningless figure.

Best regards, Michael Sol

[TomDiehl]

QUOTE: Originally posted by MichaelSol

QUOTE: Originally posted by AnthonyV
One advantage that Brown gives to the Diesel is its higher availability, 90 percent to 60 percent.

The "availability" measure is just about meaningless for two reasons.

1) For the study period in question, it took at least four diesel units to pull the same tonnage as a Northern. In terms of locomotive miles, that meant a 4 to 1 advantage to steam. Indeed, the diesel fleet would have to show 100,000 locomotive miles to equal the productivity of a Northern showing only 25,000 miles. For a Northern operating 300,000 miles annually, Diesels would be required to operate 1,200,000 locomotive unit miles to move the same freight tonnage.

If the Northern were restricted by its "availability" it could only operate 180,000 miles annually, a loss of 120,000 locomotive miles.

Interestingly, if the four Diesel locomotives were limited solely by their "availability" rating of 90%, they would be able to generate only 1,080,000 locomotive unit miles. A loss due to availability of the identical number of 120,000 locomotive miles lost by the single Steam unit.

The 90% availability looks to have the same ramifications as the 60% availability because of the additive effect of probabilities with each discrete event (locomotive).

However, this is one reason why "locomotive miles" is a tricky and somewhat superficial statistic applied to motive power types of significantly different horsepower. The quality of the Northern locomotive miles are different than the quality of the Diesel locomotive miles.

2) Even today, the 90,000 mile annual fleet average of BNSF locomotives is substantially below what modern Steam was demonstrably capable of in 1950

The Milwaukee Baltic class ran off 150,000 miles per year, but that only represented 7 hours of actual daily work. It was able to sit and relax for the other 17 hours, utilizing only half of its supposed "availability." A modern Diesel with 90% availability is moving only 10 mph at the current average annual mileage. Since average train speed on the BNSF is about 23 mph, that means the average locomotive today is not moving anything about 50% of the time.

Locomotive "availability" was a more or less meaningless figure.

Best regards, Michael Sol

Of course this begs the question, "how did they count diesel locomotives?" When the railroads got the first road diesels, they were drawbar connected into semipermanent sets, which, for reasons of taxes and Union contracts (plus others, I'm sure), were counted as a single locomotive. The idea of removing the drawbars and installing standard couplers wasn't brought into the mix until later, when they realized the versatility of being able to mix and match.

[MichaelSol]

QUOTE: Originally posted by TomDiehl
[Of course this begs the question, "how did they count diesel locomotives?" When the railroads got the first road diesels, they were drawbar connected into semipermanent sets, which, for reasons of taxes and Union contracts (plus others, I'm sure), were counted as a single locomotive. The idea of removing the drawbars and installing standard couplers wasn't brought into the mix until later, when they realized the versatility of being able to mix and match.

From an engineering perspective, how they are connected wasn't relevant during the study period. However, because of the practice of GM to call any number of units lashed together in any fashion a "locomotive" for public relations purposes, the Brown study clearly distinguishes the basic "unit" of motive power from the GM terminology.

Best regards, Michael Sol

[Anonymous]

Boy I can't belive this thread has survived to 30 pages, to bad most of it is just drivel.