Tropicology: Logging, Lumber, and Furniture

[Copper Maniac]

Now for the experiments including the furniture factory.  In the first two runs, Easy-Does-It (EDI) and Sweatshop (SS) are compared.  Here, the phenomenon observed before -- where lumber mill productivity and resource utilization drops -- rears its head again.  SS seems to offer only a few percent improvement in productivity for the furniture factory.

                    Wage   ManM  Revenue Extra Load  Rate        RSRC=In
CC  HS ---    90,984  7582 328,884  5.9    554.0  0.0731  ------------------
SS  PS ---  108,328  4924 531,030  1.7     355.7  0.0722  299,058=498.4
EDI --- ---     91,388  4154 782,040  0.0     279.3  0.0672  418,980=279.3

                    Wage   ManM  Revenue Extra Load  Rate        RSRC=In            Profit
CC  HS ---    90,888  7574 311,178  0.0    518.6  0.0685  ------------------   214,770
SS  PS ---  108,328  4924 479,170  0.9     320.3  0.0651  282,792=471.3  61,890
SS  --- ---     91,366  4153 808,584  0.0     288.8  0.0695  439,450=293.0  235,010
Treasury96JAN = 1,014,178; Aid 202,000; Other 690,877 Adjust 1,503,055; 991,385

For the SS run I also took a treasury reading at January 1996 and recorded the profit (same for all following experiments).  January 1996 was chosen so that all runs would be finished by then.  I also recorded the foreign aid received each year and the profits/losses shown on each building at the end.  I then ran the equivalent situation but without operating the furniture factory (below).  I am off by $9 on the adjusted treasury, but I'm confident that the books are being kept accurately.  Thus the falling lumber mill productivity when the furniture factory is present has another cause.

                    Wage   ManM  Revenue Extra Load  Rate        RSRC=In            Profit
CC  HS ---    91,488  7624 334,752  0.0    557.9  0.0732  ------------------   237,744
SS  PS ---  108,240  4920 702,255  0.0     468.2  0.0952  285,204=475.3  283,051
Treasury96JAN = 1,014,164; Aid 205,300; Other 703,325 Adjust 1,512,189; 991,394

All profit readings do not include the unsold loads (Extra) sitting in the output queues.

The problem when the lumber mill is operating has to do with the teamsters (anyone surprised?).  I observed teamsters entering the fenced area of the furniture factory and leaving with their loads.  Sometimes they turned around and tried to drop off their loads several times.  The teamsters would then leave without depositing their loads.  They would still have the wheelbarrow graphic for a little while.  This only happened when the furniture factory input queue was at least 15.0.  Occasionally a teamster would succeed in depositing a load at the furniture factory when the input queue exceeded 15.

This was but one end of the problem.  The larger part occurred when teamsters picked up their loads at the lumber mills.  If, at the time the teamster grabbed the load, the furniture factory input queue was at least 15.0, any of several things could happen.  Rarely, the teamster would take the load to the docks.  Rarely, the teamster would head toward the furniture factory.  Most of the time, the teamster would immediately discard the load (the wheelbarrow graphic would never appear).  This is a tale of Tropico teamster theft of titanic proportions!

Aside:  Since the furniture factory is saturated in the above runs, I checked adding the Machining Center (MC), because I had not obtained a productivity figure with it previously.  That result appears below.  The productivity change was 3%.  The hot text claims 10%.  The hot text also claims a price increase for furniture.  As reported previously, the price increase is not implemented (in fact, there is no separate export category for fancy furniture, unlike the separate export category for fine jewelry).  So, I suspect the MC is simply not implemented (except that it will take $8,000 from your treasury and charge your power plant 15 MW).

                    Wage   ManM  Revenue Extra Load  Rate        RSRC=In            Profit
CC  HS ---    90,600  7550 309,150  0.0    515.2  0.0682  ------------------   213,030
SS  PS ---  107,030  4865 516,795  4.4     348.9  0.0717  287,904=479.8  96,101
SS  --- MC    89,980  4090 821,744  0.1    293.6  0.0718  451,666=301.1  236,941

The previously reported results for the timber industry -- specifically, that, if the furniture factories are underloaded, then the productivity loss at the lumber mills does not occur, now makes perfect sense.  If the furniture factory input queue never gets above 15, then there are no losses due to lumber loads being thrown away.  (There can still be losses due to teamsters dying en route.)

It might not seem worth the trouble to build a furniture factory, but, if you examine the results so far, you can see that 2 logging camps will generate a profit of about $230k, and 2 lumber mills will generate an additional profit of about $280k.  Adding one (saturated) furniture factory to 2 logging camps and 2 lumber mills leads to $10k lost.  But if a second furniture factory were added, then the lost lumber loads could be converted into furniture for 2800 of pure profit per load.  That's about 2800 times 150 loads above, or $420k.  Minus labor and maintenance, and minus the about 10k lost from adding the first furniture factory, we can estimate that having 2 furniture factories will lead to about $280k in profit, without even optimizing the labor (restricting jobs at the furniture factories).  This $280k is as good as the lumber mills.

Moreover, if we add the Exhaust Fans (EF) option to the furniture factories, then a 18% reduction in lumber used was previously reported.  For our situation (underloaded furniture factories), this actually means a 22% increase in the amount of furniture exported.  So 809k+420k = 1229k revenue would increase by 0.22*1229k, or about 270k, meaning that a pair of furniture factories could yield about $550k in profit.  That's some serious jack.  Plus, I estimate that one furniture factory with EF option should barely handle the load from one lumber mill with Power Saw option.  The following experiments seek to confirm this.  (Somehow, I shoehorned in a second furniture factory to my already crowded map.)

                    Wage   ManM  Revenue Extra Load  Rate         RSRC=In            Profit
CC  HS ---    90,276  7523  320,028  0.0    533.4  0.0709  ------------------   224,232
SS  PS ---  109,516  4978  675,450  0.0    450.3  0.0905  272,220=453.7  267,954
EDI --- ---   176,044  8002 1,228,136 0.0   438.6  0.0548  675,450=450.3  290,326

                    Wage   ManM  Revenue Extra Load  Rate         RSRC=In            Profit
CC  HS ---    91,020  7585  329,280  0.0    548.8  0.0724  ------------------   232,740
SS  PS ---  109,098  4959  656,265  1.1    438.6  0.0884  282,834=471.4  238,573
EDI EF ---  178,772  8126 1,481,032 0.0   528.9  0.0651  656,265=437.5  559,479

The full improvement possible from the EF option is not yet realized above.  A small teamster problem has reappeared.  This is taken care of in the run below by setting the furniture factory to Sweatshop.  In fact, having multiple furniture factories appears helpful, because the teamster problem only seems to occur if both/all furniture factories have high input queues.

                    Wage   ManM  Revenue Extra Load  Rate         RSRC=In            Profit
CC  HS ---    90,528  7544 308,250  0.0    513.8  0.0681  ------------------   212,202
SS  PS ---  107,052  4866  676,470  3.7    454.7  0.0934  277,956=463.3  266,332
SS  EF ---  178,836  8028 1,538,796 0.7   550.3  0.0685  676,470=451.0  599,174

So, some noticeable profits can be realized with the furniture factories if one is careful to build enough of them.

(continued)

[Copper Maniac]

Now let's consider what happens to the lumber mills and furniture factories when we do not yet have electricity.  The Power Saw option increases lumber mill output by 30% (apparently).  And the Exhaust Fans option decreases the lumber used by the furniture factory by an observed 18%.  So the ratio of lumber mills to logging camps should be approximately 130 to 82 before electricity if it is proper for it to be 1 to 1 after electricity.  I ran another experiment, this time with two lumber mills and one furniture factory, no electricity-requiring options, and trying to reduce the mismatch in productivity as much as possible by setting the lumber mills to EDI and the furniture factory to SS.  We would expect 65/82 of the lumber to remain on the map, and 17/82 thrown away -- actually we expect (from prior runs) about 14/79 thrown away, adjusting for the EDI setting.

                     Wage   ManM  Revenue Extra Load  Rate      RSRC=In             Profit
CC  HS ---    91,032  7586 316,962  1.7    530.0  0.0699  ------------------   220,410
EDI --- ---  110,792  5036 451,650  0.4    301.5  0.0599  227,864=379.8  87,236
SS  --- ---      92,686  4213 834,484  0.2   298.2  0.0708  447,345=298.2  251,295

The lost loads of lumber are slightly higher than expected, but this could be due to teamsters dying while transporting loads.

Summary

1.  With Medium vegetation and Hard economy, the timber industry is rather lucrative.  (Labor costs in these experiments were high, but the general island overhead is ignored.  Net, the profits should be slightly higher, because the wages are the larger disparity.  The layout, however, was more compact than reasonable, so the profits shown, especially the proportions, should be about right.)

2.  A logging camp showed profit of $110k.  A lumber mill prior to electricity showed profit of $95k.  A furniture factory (calculated) prior to electricity could potentially show (net) profit of $205k if built one per two lumber mills -- provided labor at the lumber mills can be suitably restricted (micromanaged).  After electricity, each lumber mill showed profit of $125k.  After electricity, each furniture factory showed profit of $295k if built one per lumber mill.

3.  For the land area consumed, logging camps seem somewhat less profitable than cattle ranches.  (Close to half the space, but with less than half the profit with Medium vegetation.)  However, if one lumber mill and one furniture factory are added per logging camp, then considering the area of these structures and their supporting infrastructure (more workers), the timber industry is more profitable per unit of space -- somewhat less space, and much more profit if the furniture factories are handled properly.  Electricity is a significant factor in making the timber industry superior.

4.  The optimal ratio of logging camps to lumber mills is about 1:1 after electricity, in decently forested areas.  The optimal ratio of lumber mills to furniture factories is about 3:2 before electricity and 1:1 after.  It is better to not build furniture factories at all rather than build substantially fewer than this ratio.  If building furniture factories, but fewer than appropriate, substantially more profit will be shown if the lumber mill labor force is restricted so that all lumber goes to the furniture factories.  When deviating from the ratio, it is much better to build more furniture factories.  The proper ratio depends on how much time the workers spend at their jobs.  This possibly means that, ideally, an 8:5 ratio can be done (prior to electricity), because the ratio of workers is 5:8.

5.  Recommended options for a logging camp consist of the Horticulture Station only, plus Clear Cut of course.  Recommended options for a lumber mill consist of the Power Saw, and optionally the Sawdust Burner if pollution reduction is desired.  (All runs included the Sawdust Burner.)  The Log Debarker is useful if awash in logs (but it is better to build more lumber mills, especially on a profit per space basis).  Industry ratios above assume no Log Debarkers.  Recommended options for the furniture factory consist of the Exhaust Fans only.  The Machining Center option for the furniture factory is broken.

[Copper Maniac]

5.  The Horticulture Station appears to have some effect over the long term.  It's hard to be precise, but it seems to take less than a decade to recover its cost.  In all runs (above, in the thread) the other logging camp option (Tool Shop) did not have any positive effect.

3.  Loggers will chop trees wherever they need to.  You can fill in spaces with buildings if it looks like nothing will grow there for a long time.  If you really don't like the tree layout, for $1000 you can throw in 20 trees.

I figured on, for logging camp:  16 for site, 0 for additional space, 120 for tree cutting area, and 24 for housing and services induced by the 8 uneducated workers, totaling 160.  For a lumber mill:  24 for site, 6 for additional space, and 25 for housing and services induced by the 5 HS educated workers, totaling 55.  For a furniture factory:  16 for the site, 4 for additional space, and 40 for housing and services induced by the 8 HS educated workers, totaling 60.   Total for a combination of one logging camp, one lumber mill, and one furniture factory:  275.

For a cattle ranch, I figured on 323 for the site and grazing area, 0 for additional space, and 6 for the housing and services induced by the 2 uneducated workers, totaling 329.

This assumes a tight-fisted rule with little regard for beauty.  You might add additional spacing/beauty to the business structures.  My lumber map is plenty ugly, as a ranching map would be.

I also assume an orientation toward space-efficient services, such as providing food via fishing.  Trees don't grow in the water.

[Copper Maniac]

The loggers will not confine themselves to a zone.  They will head to your downtown to chop down a tree (if the tree is of the harvestable type).  The 120 number is pure eyeball guess.

In the early maps I did not have roads.  In the later maps I added one road for each logging camp to its nearest lumber mill.  Section of road near the logging camp, I count in the eyeball of the logging area.  Section of road near the lumber mill overlaps with the site.  Section of road in between can be considered to be included in the space for all the housing and services.  My maps are SimCity-packed.

I'm thinking of playing an (almost) all logging map for score to see how good it is.  That should let me know how some things work -- such as expanding by placing services in areas that have been harvested.  Also, electricity is available from the start with the test maps.  Same for building the lumber mills and furniture factories.  It was all Rapido.  The only disadvantage the test maps have are that I let the vegetation be Medium, didn't use education enhancers, didn't use other productivity/revenue enhancers, and I didn't plant any trees.  Planting trees should be a big win.

[Coconut Kid]

There is a lot of solid information in this thread; however, some of the conclusions are probably skewed because some conditions were incorrectly assumed.

From the original Manual (e.g. before P.I. expansion and the Furniture Factory):
One unit of raw resource is converted to one unit of finished manufactured goods.
While recognizing that the paper manual had lots of problems of missprints and errors, it still seems that the quote is not compatible with some of the test findings. I believe that it is reasonable that having variable levels of waste in a wood products production chain is a reasonable feature. Waste is probably not as important an element in the other kinds of production.

In another thread, it was Copper Maniac himself who established that the "throughput" time of factory buildings varies considerably according to whether the workers are actually present or not. The idea that the time it takes to convert input to output does not depend on actual presence of employed workers is a Tropican Urban Legend. So profitability of buildings and their components is highly dependent on maximizing the time workers are in the building and minimizing the time they are doing other things.

On Logging Camps: I opine that Lumberjacks select the next available tree by its maturity and distance from the camp building - distance in 25 tile zones (analogous to construction priority zones). Sweat Shop is more profitable only if you choose to ignore the individual "job happiness" factor. Clear Cut vs Selective Cut is not a question of profitability; it has to do with political repercussions and the negative impact on the beauty overlay. Selective Cut (at least in theory) makes the Logging Camp less offensive to the Environmentalist and Intellectual factions - and to some tourists who might come close to them.

On Lumber Mills: It's too bad the Sawdust Burner was not examined to see if the results are actually visible on the map overlay since other upgrades in the wood products chain affecting pollution were tagged as not visible. I opine the debarker can't be counted as broken until worker actual presence is factored in.

On Furniture Factories: There seems to be a problem with the price increase promised by the Machining Center; but is it promised for a "different product" (analogous to 'fine jewelry') or a differently based increase? Also, there seems to be a lumber load pickup and delivery problem relating to an overfilled input queue. I suggest this needs further observation including the worker present factor before the building or upgrades are dumped.

I suspect Senor Ruina would object to the emphasis I put on the worker present or not factor because the length of the tests averaged out unduely long absences from the job. I don't agree with that. Maximizing the worker's frequency and duration the "Work Mode" as contrasted with the other five normal, 'needs' modes is a significant gameplay problem which requires the development of judgement and skills - it's not simple and averages don't eliminate consistent failures.