Maintenance Costs In Depth

Got a new strategy? Not sure how to do something?
RedKnight
Posts: 139
Joined: Tue Dec 05, 2006 10:40 pm
Location: Atlanta GA USA

Maintenance Costs In Depth

Post by RedKnight » Sat Dec 23, 2006 10:11 pm

*** UPDATE 3/19/09 ***

This long thread has been summarized and more info added, in the Railroads! wiki page on Maintenance. You should see that, for Maintenance info (and add to it, if you like!). The end of this thread now serves for discussion about Maintenance.

Good railroading! - RK

*** END UPDATE ***


Basics

* There are 4.8 maintenance instances a year (25/5.2). Since this doesn't precisely divide into 12 months, you will see maintenances occurring on various months; it won't be a repeating pattern of months, except over long time scales.

* The "Last Year" (LY) Maintenance value (seen in F4) is actually the last 5.2 years worth of maintenance, a.k.a. the total for the last 25 maintenances. These 5.2-year periods are important to how maintenance works, so I'll call them the Maintenance Period (MP).

* Thus, the Lifetime (LT) maintenance cost (F4) is a better estimate of, well, lifetime costs - IF you haven't upgraded your engine. The LT cost shows you maintenance for the route, not the train. You have to start a new route if you want it to be just for that engine.

* As of exactly 5 years, you will have incurred 24 times the Maintenance Cost (MC). Then the next MP kicks in with the 25th maintenance (at 5.2 years). If you have the Lube Patent, $275 is added to your Maintenance cost through that period; if you don't, $550 is added. So a Grasshopper's MC (initially $550) is increased to $1,100 for the second MP, $1,650 for the third, etc.

* If you slow down the game and watch the Profit for a train more than 5 years old, you will see an odd thing. The Profit will appear to go up briefly (even though the train has not made a delivery) and then go back down. What's happening is the maintenance cost incurred 25 instances ago is being dropped, then the new MC being imposed. It's a "first in, first out" function. By noting the prices before (initial value), during, and after (final value) this event, you can see what I mean. The value that got dropped will always be $275 less than the current maintenance cost if you have the Lube Patent, or $550 if you don't. Across this little price jag, an MC that's $275 or $550 less than the current MC got dropped and one that's $275 or $550 more than that previous one got added (the current MC), so the net difference (final value-initial value) is always $1,100 (without Patent) or $550 (with Patent) across the little jag.

* If you have upgraded an old train (with high MCs), those previous, high MCs will be dropping out, until the train is 5 years old (at which point the "proper" ones for the current train make it to the end of that queue). So it can look weird if you don't know what's going on.

* Example: When a Grasshopper is a little over 10 years old, it's LY will say ~$27500 (25*1100), but it's LT will say ~$41,250 (25*550 + 25*1100). LY is showing the last 5.2 years, where the MC is $1,100, but LY covers both MPs.

* The Lube Patent not only reduces the initial MC fee, but also the increase imposed each MP. MC is increased by $550 (w/o Patent) or $275 (w/Patent) each new MP for all engines, regardless of the train's initial MC.

* Trade-In Values use a hyperbolic function. For the percent of original engine cost remaining, use: 1 - (25/(Mo#+25)), where Mo# is the number of months transpired. It falls quickly, then slows; you've lost 50% of value at the two-year point (actually 25 months), but very old trains will retain a few percent of their original cost.

Practical Usage

In order to best utilize this info, the basic question is, "when should I upgrade my trains?"

I re-did the finances another way which I'm sure is better. Train replacement times are a lot more reasonable now ...

As a straw man, the point where they should be upgraded is when the total maintenance cost (the LT value, if you haven't upgraded) equals the cost of a new engine, minus the trade-in value (TIV). This is then averaged across time by dividing by the number of years that have passed. The minimum (across a wide span of years) is the point where you spend the least money per year, given all three costs. It's a very smooth curve; generally any time within 10 years of the stated minimum is fine, except for the very earliest engines.

On to the good stuff...

Without the Lube Patent, optimum replacement times are 10.2 years (GH, Norris), 15.3 years (American to Consolidation), 20.6 years (Pacific to F-Series), and 25.8 years (GS and GP-Series). With the Lube Patent, the GH is 10.2, Norris 15.4 ... anyway, get the attached spreadsheet. :P

While the optimum times might look disquietingly similar (given differing engine and maintenance costs), what does differ for the trains above, is the average cost per year ((New engine - TIV + LT MC) / Years). As stated above, it's a smooth curve around the minimum, and with the later trains, it doesn't make a huge difference if you wait 10, 20, or even 30 years past the optimum. The GP-Series (plus Lube) has an optimum average yearly cost of $22,388 at 36.2 years, but at 72 years, the average yearly cost has only risen to $24,450.

Attached, find an Excel spreadsheet with engine stats, including recommended replacement time. I will post an equation for figuring LT cost at any point in time, when the wiki is up.
Attachments
RR Engine Stats.xls
(21.5 KiB) Downloaded 433 times
Last edited by RedKnight on Sat Mar 14, 2009 9:00 pm, edited 1 time in total.

RedKnight
Posts: 139
Joined: Tue Dec 05, 2006 10:40 pm
Location: Atlanta GA USA

Post by RedKnight » Fri Dec 29, 2006 8:25 am

*** RED ALERT ***

I have been testing ideas at Financier difficulty level a long time, entirely solo (no AIs), and jumped to Robber Baron with 3 AIs for a change. Lo and behold, maintenance fees are definitely higher - and Elijah McCoy's Lubricator Patent has yet to appear (as of 1905, as of this writing).

Quite some time ago in this new Robber Baron game (1880?), I had thought I was way ahead of the AIs in terms of efficiency, when I saw that my Cash barometer was not only not gaining over time - it was actually falling. The F1 key told me the story - ungodly maintenance fees; nothing like what was expected with Financier. (Without Lube Patent.)

I'll work out a full table of maintenance fees and optimal engine replacement time versus difficulty level, at a future time. Until then, suffice to say, at difficulty levels higher than Financier, you should buy new engines earlier than at Financier difficulty level.

Sergo
Posts: 4
Joined: Sun Dec 31, 2006 12:37 am

Post by Sergo » Tue Jan 02, 2007 9:31 pm

I'm looking forward to your next report. I think I've not been replacing fast enough. Thanks for taking the time to unravel all this.

RedKnight
Posts: 139
Joined: Tue Dec 05, 2006 10:40 pm
Location: Atlanta GA USA

Post by RedKnight » Wed Jan 03, 2007 9:26 pm

Thanks, Sergo! I intend to get to the rest, although I have another project going on at the moment unrelated to Railroads!

Unfortunately, I don't have finished games for all difficulty levels. For both U.S. and European engines, for that matter. "Finished" here just means, the game is to the point where the last engine has been introduced.

If anyone has some finished games they can email me, PM me (with the little icon at the bottom of this post; my email will get a notice). Make sure you remember the difficulty correctly. :) (Is there a place to see what difficulty was set to, for a saved game?) It does not have to be any particular scenario; indeed, different ones will let me make sure maintenance is the same across them.

BTW I have updated the above info a touch more (so far) in the Railroads! wiki. When I get to the other difficulties & countries, I'll post it there, but put a notice in this thread.

FWIW one problem I'm facing is simply the number of combinations... there are 34 non-US and 12 US engines, two maintenance levels, and five difficulties, for a total of almost 500 endpoints. Currently my "model" outputs one result at a time, which I cut and paste into e.g. the Excel table attached above. It would be very tedious and highly error prone to do this 500 times, so I'm looking into how to automate outputting the combinations in my spreadsheet.

RedKnight
Posts: 139
Joined: Tue Dec 05, 2006 10:40 pm
Location: Atlanta GA USA

Post by RedKnight » Fri Jan 05, 2007 4:36 am

Ok, I just posted a message in the Railroads! Chat section, asking for volunteers to send savegames. Please check there, anyone who wants to help. (It will let me show the optimal replacement time for YOUR scenario of interest!)

I figured out the problem concerning generating the many possible combinations, thanks to some Excel masters on a CompuServe forum. Now all I need is the data!

RedKnight
Posts: 139
Joined: Tue Dec 05, 2006 10:40 pm
Location: Atlanta GA USA

Post by RedKnight » Mon Jan 08, 2007 3:23 pm

Ok, an updated, complete spreadsheet for all countries, difficulty levels, and lube patent (yes/no) has been posted to the other thread in Chat. I'll also post it to the wiki when it accepts XLSs.

Please see that other thread for any additional posts, since we've got a conversation going there.

RedKnight
Posts: 139
Joined: Tue Dec 05, 2006 10:40 pm
Location: Atlanta GA USA

Post by RedKnight » Sat Jan 20, 2007 2:52 am

I've just done a major revision of the wiki's Railroads! Maintenance page.

This includes a revision of the spreadsheet, with e.g. a new "relaxed replacement time" column, so you can see how much "play" there is in the numbers for optimal engine replacement time.

Also I added a Filter, which makes it easy to only see e.g. U.S. Tycoon engines (or whatever).

Finally, I moved the spreadsheet columns around in a way that I think is more useful to a person consulting it in the game.

Magnu, I didn't thank you openly on the page, but will be happy to, if you like. By the same token - and as is typical for wikis - I also didn't say that I wrote it per se, either. In any event, thanks for your spreadsheet, which made my job so much easier.

BTW if you or any else cares to, I think it would be useful to have Max MPH and Passenger/Freight/Mixed info pulled from the Roundhouse, and added to the spreadsheet. So that it's all right there, for readers. If you or anyone else get a chance to collect that, I'd be grateful. I can send savegames for the countries - although anyone can gather most of the info just by starting scenarios at their latest date. I can get the rest easily.

gforce
Posts: 234
Joined: Thu Nov 09, 2006 3:24 am
Location: Belgium, Brussels

Post by gforce » Tue Mar 20, 2007 6:00 am

It's a pitty that redknight isn't here anymore,
but maybe someone else???
If i take a look at the figures when to replace a train i'm pretty surprised by what redknight is saying... :shock:
I'm replacing my trains somewhere around 10-12 years,
and i really believe this is the best...
let's say you don't want to speed up and you're just replacing by the same loco.
At what time are you replacing your trains in this case???

RedKnight
Posts: 139
Joined: Tue Dec 05, 2006 10:40 pm
Location: Atlanta GA USA

Post by RedKnight » Sat May 05, 2007 5:53 am

hey gforce,

I'm back a short while. Here let me upload all my spreadsheet. Most of it will make no sense but the place the numbers came from are called Maint, M2E, and M2E2. What are you questions?

BTW that was some hella crafty Excel to get those 200 permutations in my pretty spreadsheet output on the wiki. What, no awards?

RK
Attachments
Railroads! Analyses3.xls
(403 KiB) Downloaded 323 times

User avatar
darthdroid
Posts: 368
Joined: Sun Apr 15, 2007 2:42 am

Re:

Post by darthdroid » Fri Feb 27, 2009 5:45 pm

RedKnight wrote:I've just done a major revision of the wiki's Railroads! Maintenance page.

This includes a revision of the spreadsheet, with e.g. a new "relaxed replacement time" column, so you can see how much "play" there is in the numbers for optimal engine replacement time.

Also I added a Filter, which makes it easy to only see e.g. U.S. Tycoon engines (or whatever).

Finally, I moved the spreadsheet columns around in a way that I think is more useful to a person consulting it in the game.

Magnu, I didn't thank you openly on the page, but will be happy to, if you like. By the same token - and as is typical for wikis - I also didn't say that I wrote it per se, either. In any event, thanks for your spreadsheet, which made my job so much easier.

BTW if you or any else cares to, I think it would be useful to have Max MPH and Passenger/Freight/Mixed info pulled from the Roundhouse, and added to the spreadsheet. So that it's all right there, for readers. If you or anyone else get a chance to collect that, I'd be grateful. I can send savegames for the countries - although anyone can gather most of the info just by starting scenarios at their latest date. I can get the rest easily.

Old post, but I just found it, good stuff. Ever gonna tell us the deal on tycoon or robber baron? Shouldn't you be working for an airline or trucking company using decision science software redknight? Sounds up your alley lol.
-Bob the Lunatic

RedKnight
Posts: 139
Joined: Tue Dec 05, 2006 10:40 pm
Location: Atlanta GA USA

Re: Maintenance Costs In Depth

Post by RedKnight » Fri Feb 27, 2009 11:24 pm

Hey darthdroid! How's it going.

You mean Railroad Tycoon, right? I haven't seen a new one since RRT3. Are you still playing it?

I'm not sure what you mean by Robber Baron.

Wow I can't believe I posted that "Railroads Analyses3.xls". Most of it was my own notes while playing around with the game.

I'm a scientist. But if you know of a better paying job with an airline or trucking company, maybe I won't be much longer. Or a railroad company. :)

User avatar
darthdroid
Posts: 368
Joined: Sun Apr 15, 2007 2:42 am

Re: Maintenance Costs In Depth

Post by darthdroid » Sat Feb 28, 2009 3:53 am

RedKnight wrote:Hey darthdroid! How's it going.

You mean Railroad Tycoon, right? I haven't seen a new one since RRT3. Are you still playing it?

I'm not sure what you mean by Robber Baron.

Wow I can't believe I posted that "Railroads Analyses3.xls". Most of it was my own notes while playing around with the game.

I'm a scientist. But if you know of a better paying job with an airline or trucking company, maybe I won't be much longer. Or a railroad company. :)
Hi Red,

No, I meant SMR. Robber and Tycoon as in the highest and 2nd highest difficulty settings for $. You said something about how the maintenance changed when you tried it on highest level ("robber baron"). So that's what I was asking about.

And if you did the job I was talking about (decision scientist for say an airline) you'd be making $400,000 or more.... DeSci seems up your alley, that's all I was saying.
-Bob the Lunatic

RedKnight
Posts: 139
Joined: Tue Dec 05, 2006 10:40 pm
Location: Atlanta GA USA

Re: Maintenance Costs In Depth

Post by RedKnight » Wed Mar 04, 2009 4:13 pm

Hi darth,

Actually I stopped updating this thread when they got the wiki up. For a "final version" of the maintenance issue, see this. There's lots of other good stuff on the wiki. Another thing I'm proud of is the work I did on Resources there. Give it a look!

mikeyc
Posts: 50
Joined: Sun Sep 02, 2007 12:17 am
Location: Australia

Re: Maintenance Costs In Depth

Post by mikeyc » Wed Mar 04, 2009 8:30 pm

RedKnight, I am agreat fan of your work on the Wiki particularly the relationship between speed, various bonuses' and passengers. Your contribution was really clever work and this is a public thank you and a pat on the back.

User avatar
Lowell
Posts: 653
Joined: Sat Mar 24, 2007 2:14 am
Location: Georgia
Contact:

Re: Maintenance Costs In Depth

Post by Lowell » Sat Mar 14, 2009 8:15 pm

You know what's funny, on my single player mods...the trains can run the whole length of the game and never slow down. I also edited the maintance costs on many of my maps as well, reducing the cost in some cases on super sized maps.
So maybe the game treats engines different if the AI is playing :?: As single player has no effect except constant cost of up-keep.

Interesting...

Post Reply