Attractions: What influences generated pax?

[cy087]

(Ver 112.2 r6390 PAKset 128 official)

This has got me absolutely flummoxed!  At one edge of the map I have a cross-country ski resort and at another edge there is an oasis.
The cross-country ski resort is producing around 37 passengers/month at the local bus stop, the oasis on the other hand occasionally (3 months per year) produces 2 or 3 passengers a month.  According to the info popups the Passenger levels are:
Ski resort 600
Oasis 100

They are both "about" the same distance from the nearest town, they are both attached to the same level of stop, the stops use the same type of vehicle, the towns populations are 9499 for the ski resort and 7139 for the oasis, they are both huge blobs on the map etc etc.  Other "in-town" attractions like the Elementary school (Passenger level=135) are producing around 30 pax per month.

I have been looking for hours for something that would explain what is going on here to no avail. I give up! Just what, what, does influence the number of passengers generated by an attraction?  (Purple frustration level!  )


Cy

[kierongreen]

Attractions don't produce passengers as such - they generate as many as arrive from what I remember. So if you don't have a decent service serving the ski resort then not many passengers will want to make the return journey. This could also happen if you've got a bottleneck on your network somewhere....

[Ters]

As far as I know, passenger level and total population that has access to the attraction is all that matters. It's possible that the elementary school is similar to the oasis and that the remaining passengers on that stop come from other surrounding buildings.

I don't think the type of stops or vehicles matters for passenger generation, not until passenger generation stops because the stop overflows. Total population with access to the attraction depends on how many transfers they have to make (there's a limit somewhere), and a range that increases over time (more people travel further more often in 2000 than 1900). Passengers are also generated in pairs, one traveling from A to B and another traveling from B to A. It's possible that passengers are primarily generated at their homes, so that the passengers appearing at attractions heading home are only the reflection of those going the other way. If those wanting to go to the attraction can't get there, no one will leave the attraction either.

I might be wrong, though.

[prissi]

Assuming both are no city attractions: Did you cover all tiles of the oasis with your stop? And again, as mentioned, are there enough possible connections ot the oasis. What about no route passengers, does they scale as expected?

[cy087]

Thanks for you interest.  Here's what I can glean so far:

Factor	Ski resort	Oasis
Avg* Monthly Arrivals	80	5
Avg* Monthly departures	70	5
Stop coverage	About 50%	100%
Stop capacity	32	32
Traffic bottlenecks	Yes, but pax are still generated wanting to go to the resort	No, all feeder routes are well performed
Service levels	1 x 38 seater minibus, from nearest city interchange stop to resort	1 x 38 seater minibus, ditto
Line complexity	No intermediate stops on route	Ditto
Trip distance	41 tiles	28 tiles
Proximity to other pax generators	There is a solar collector farm about 6 tiles distant (radially) but it has it's own bus stop. The solar farm is on the same road!	There is a iron mine about 8 tiles distant, not connected by the same roads.
No route pax	Zero at resort stop	Zero at resort stop

_{* Monthly averages are just from eyeballing the stop charts}

So the only big differences are:

• the stop coverage, but I would have expected that the ski resort would be more affected. 
• the proximity of a factory on the same roadway.

Any thoughts/comments appreciated.

It's possible that passengers are primarily generated at their homes, so that the passengers appearing at attractions heading home are only the reflection of those going the other way.

Interesting.  I have a recollection that pax were generated travelling "from" the ski resort before there were any trying to go "to" it.  I will have to build a model to check this.


cheers
CY

[kierongreen]

The problem is that even if passengers are being generated to go to the resort, if they don't arrive there won't be any return journeys. To be sure this isn't the problem you'll have to increase network capacity until there are no waiting passengers across your network.

[cy087]

OK, maybe I have to rephrase the question to:

"What do I need to do to convince my Simuls that they want to go to the oasis?"

The number of pax going to is low as well, the arrivals reflect the number of pax at the town station i.e. there are no pax waiting to go from town to the oasis, they go as soon as the bus arrives at the town stop.

On the other hand, I can't keep up with the number at the other town who want to go to the ski resort, i.e. there are always some waiting to go there at the town end of the line.

The thing I am trying to work out is why there is such a large disparity between the pax.  Again, if anything the "attractiveness" of the ski resort is lessened by the fact that they cant get there, whereas no-one wants to go to the Oasis?



CY

[Ters]

One thing to take note of, if you haven't already, is that the numbers aren't as skewed as they first appear to be. If the ski resort has level 600, and the oasis level 100, then 70 passengers to the ski resport would equal 11 or 12 to the oasis (I think). As such, the number of passengers to the oasis is just about half of the expected, not less than a tenth. (Though the fact that only half the ski resort is covered might mean it's a quarter.)

There are maps in the dialog for each town showing where passengers from that town want to travel to. Try using them to see which town has lots of citizens wanting to go to the oasis, but who find no route. See the help text for how to read those maps.

[cy087]

One thing to take note of, if you haven't already, is that the numbers aren't as skewed as they first appear to be. If the ski resort has level 600, and the oasis level 100, then 70 passengers to the ski resport would equal 11 or 12 to the oasis (I think). As such, the number of passengers to the oasis is just about half of the expected, not less than a tenth. (Though the fact that only half the ski resort is covered might mean it's a quarter.)

Ah! Thanks, I hadn't considered that ratio.  Your mention of 11 or 12 makes me wonder if the "Passenger Level" displayed for the attraction is a "yearly" figure?  No, that cannot be. It would mean that most city houses would generate about 1 pax per year or two.  So I am unclear on how to interpret that figure.

There are maps in the dialog for each town showing where passengers from that town want to travel to. Try using them to see which town has lots of citizens wanting to go to the oasis, but who find no route. See the help text for how to read those maps.

The minimaps are just as I expected.  There are 4 towns in this game. On each one of them there are some yellow dots and lots of orange dots for the ski resort (in both months) but no dots for the oasis.  No-one wants to go there 

One final question before I close the line down and save poor "Hinterland Tours" from going broke, would the fact that the oasis is built on the wrong climate ground have anything to do with it? (temprate rather than desert).

CY

[Combuijs]

The problem is that even if passengers are being generated to go to the resort, if they don't arrive there won't be any return journeys.

Not true, when passengers are generated for a journey, the return journey is immediately generated as well. So both start at about the same time.

If the ski resort has level 600, and the oasis level 100, then 70 passengers to the ski resport would equal 11 or 12 to the oasis (I think).

As far a I know the level is not linear with the number of passengers generated, e.g. an attraction of level 600 does not necessary generate 6 times as much as passengers as an attraction of level 100. How they actually compare I do not know. It is however the basic mistake in cy087's reasoning.

One final question before I close the line down and save poor "Hinterland Tours" from going broke, would the fact that the oasis is built on the wrong climate ground have anything to do with it? (temprate rather than desert).

Climate nor other factories/attractions in the neighborhood influence the number of passengers generated. Level and distance are the only important things. (Distance plays a role, as in early years more local traffic is generated, and in later years more global traffic is generated. I don't think the difference in this case is enough to account for the difference in passenger generation.)

[Ters]

Ah! Thanks, I hadn't considered that ratio.  Your mention of 11 or 12 makes me wonder if the "Passenger Level" displayed for the attraction is a "yearly" figure?  No, that cannot be. It would mean that most city houses would generate about 1 pax per year or two.  So I am unclear on how to interpret that figure.

As far a I know the level is not linear with the number of passengers generated, e.g. an attraction of level 600 does not necessary generate 6 times as much as passengers as an attraction of level 100. How they actually compare I do not know. It is however the basic mistake in cy087's reasoning.

The help text describes the number as "relative popularity". I understand that to mean that 600 is 6 times more popular than 100. It's different from the number stored in the pak files, which I think is the non-linear one. The number is however not the number of visitors, just a component in some bigger formula, which also includes total world population and how far passengers are willing to travel at the time.

[cy087]

This is now becoming clearer.

(First off, I have to admit that I had some mental roadblocks and presumptions that made this a slow and painful learning experience.  So thank you all for your perseverance.)

   
• Attractions don't generate pax: I understand that now but am was unsure about how that relates to the monthly arrival/departure rates for a stop serving a "successful" attraction (and no other pax generating object). If the "passenger pairs" theory were correct then it would suggest that, if transport capacity is sufficiently high enough then the number of pax leaving the attraction in a given month should never exceed the arrivals and should, and in fact should be equal to the arrivals.   I rigged up a game to check this and it is correct (with very small variances due to vehicles arriving/departing very close to month ends).
• Distance from the town and game year affects pax numbers: I am almost clear on this.  I have read the simuconf.tab comments and can see the effect of the locality_factor and how in the game it will limit the number of pax wanting to go to the attraction early in the game.  The exponential way in which this config parameter is applied is interesting and I have given myself some homework to do creating games with different locality_factor profiles to see what happens.
• Other simuconf parameters: From reading the comments it is apparent that factory_worker_percentage and
  tourist_percentage could affect pax numbers to land attractions.  Why it isn't totally clear is this. If the values for these two parameters are 33 & 16 respectively, i.e. 49%, it leaves 51% for intercity/intra-city travel.  I presumed that land attractions are like factories outside the town boundaries where the pax come from the 33% factory_worker_percentage, so the tourist_percentage (16%) is spread across attractions in the town and ones reacheable outside the town.  Is this true or do land attractions get their pax from the 51% inter/intra-city pool?  Further, which pool is used for monuments?
• James Pett's document: I stumbled across this only after starting this thread.  It makes things a lot clearer!
• Passenger level is really "relative popularity": Now that was the real thing!  Sorry it took several tries to get that through my thick head! 

Thanks to all again.
CY

[prissi]

The arrival rates are not neccessarily equal. Passenger can get lost. Also due to growth passengers arrival will be usually higher than departure. In care of discarding passengers when routes are overcrowded and different routes for foward and backward, also asymmetric number could happen. This is very tough to test. But, be assured it is programmed that way ... (Moreover, more mail is generate from attractions than going towards them intentionally.)

Distance from the town of origin is the important point. Or, in case of absolute numbers, the distances from the biggest cities of the map.

Land and city attractions from those 16% (albeit this number is affected by the supply of workers to factories). Monuments are normal city buildings in this regard, and thus very attractive targets.

In experimental passenger routing (and hence) passenger generation may produce different results.

[Tazze]

I just post what I know.
the places that can makes pax are three: city hall,city buildings (res,com,int),monument.

Attraction and industry can't makes pax,and so we must connect varies route.

By the way, monument included in the 51%:
the 51%=city hall+city buildings+monument+HQ

[cy087]

Thanks  prissi & Tazze, that clears the pool percentages up.

Attraction and industry can't makes pax,and so we must connect varies route.

(my emphasis)

Which just begs the question, what are the numbers plotted in the Factory Chart on the Production/Boost tab "Passengers: Generated" and "Mail: Generated"?

CY

[Ters]

It has something to do with how production is boosted by passengers (workers) being transported to (and from?) the factory. Similarly for mail (orders i presume).

[Tazze]

Which just begs the question, what are the numbers plotted in the Factory Chart on the Production/Boost tab "Passengers: Generated" and "Mail: Generated"?

It has something to do with how production is boosted by passengers (workers) being transported to (and from?) the factory. Similarly for mail (orders i presume).

@ Ters: It's alright!

@ cy087:
Though You might think that workers has stayed in the factory from starting, This designate The number and area of workers.


By the way,The destinations ratio is "worker:tourist:lest = 33：17:50"
(the source is Simutrans Wiki)
The percentage of pax who goes to an Attraction is calculable with The destinations ratio.

formula: (0.17*passenger level)/total passenger level in the map

for example, they are only a oasis and a Ski resort. and then I want to know The percentage of pax who goes to a oasis:
(0.17*100)/(100+600)=0.02428571....=(2%)

Try also (cation:this site is very very old...)

[prissi]

Ters as well as me know the code: Have a look at simcity.cc, step_passagiere bze get_passagierziel()

The ratios can be set freely in the configuration, an the number of workers is just a maximum number. If there are no factories connected to a town, then the ratio is obviously different.