You are all barking up the wrong tree. The problem is not that the costs are not high enough the problem is that there are simply WAY to many passengers,
This is definitely somewhat true.
The issue is, in part, in AE, the way the demand system works.
In the game, it's logical for me to start 1x weekly flights from Auckland to every single Chinese city there is.
Would this fly in real life?
Of course not.
In the real world, the demand is concentrated on a fairly small number of routes, whereas on AE, it's as though the concentrated demand is on all routes.
I'll take an example of Cathay Pacific.
It's European network basically consists of HKG-LHR, CDG, AMS, FCO, MXP, FRA and DME.
In the game, airports in Europe with at least 200Y demand, enough for a daily flight, number about 45. There are countless more with less than this, a European network made up of more than 70 destinations is not at all unviable.
Unless you're Emirates, serving every medium sized airport in a continent simply doesn't happen.
This suggests that, in general,
1. O&D demand should be significantly reduced between city pairs, as the real life demand between two cities includes demand for nearby/connected cities. (Also, just another unrelated suggestion, a GDP + Population + Tourist numbers calculation might produce interesting results, possibly better than just total pax numbers for an airport)
2. Passengers traveling for business should almost always prefer high frequency flights involving connections.
3. Low frequency flights should attract lower yielding leisure travelers.
4. To achieve points 2 & 3, business passengers should be generated each day and choose available flights accordingly, and leisure/family paxes should be generated weekly in most cases.
5. Alliance hubs should play a far bigger part, to encourage more higher frequency services to fewer destinations.
This would help because
-Ridiculous destinations would become unviable,
-Overall demand, and therefore potential for expansion, would be drastically reduced
The differentiation between business and leisure is necessary due to their different frequency demands, and 2x weekly flights to tourist destinations are not unrealistic, e.g. AKL-DPS and many European flights to Havana.
Business pax unable to find a flight on the generation day should perhaps go into the weekly pool, to facilitate 1x weekly flights to places like Kinshasa, where the lower frequency is sensible.
An inherent issue with this, however, is that we would likely need far more worlds, as worlds would quite literally run out of demand...