Hi, I have another question.
Two of the most popular airports at the southern cone of SouthAmerica are SAEZ and SBSP (in this case even more than SBGR). However the estimated demand between SAEZ and SBSP is 0. My question is how this estimated demand is calculated?
a) based on statistics of real life for that time
b) based on passengers that had flown this legs on the game (SAEZ-SBSP it is not flown by any airline)
c) both options
D> None of the above
It would be virtually impossible to get route statistics for every single route in the game to base demand figures off of. It would be millions upon millions of possible routes. Only a relative handful of routes have anything remotely close to actual data to back up their demand in-game.
For the remaining 99.9% of routes, demand is based off a simple algorithm of airport size. Connecting two larger airports will have more demand than connecting a larger and a smaller and a lot more demand than connecting a smaller with another smaller. The further the flight is the more the demand (until some point is reached, where it begins to fall). Historically "domestic" airports will have little demand on international routes and vice versa. It's a bit more complicated than that, but that's how it generally works.
However, it doesn't always work very well. Historically it's been very poor at showing demand within the former USSR and South America -- there are other places, but on a widescale it's relatively nonfunctional in those areas. The system should show demand between SAEZ and SBSP -- I wouldn't necessarily expect it to show what it is in reality, but even within the limitations of the function it should definitely show something as they are both decent-sized airports that both have a fairly large percentage of international shorthaul/longhaul traffic.
There used to be a way of submitting data that would correct these sorts of things. However, with sami's desire to move away from fixed demand figures to "city-based demand" that would generate demand based off fixed economic figures rather than either historical data or an iffy formula, he's said he's not particularly interested in spending more time fixing what we've got now since it's going to be replaced.