I had left earlier this year because I had a big airline running and got bored. However, giving it another try on a demo account but not with the usual HUB and spoke but point-2-point connections on routes that exist in real life with sufficient demand but no established routes in the game.
(Air transport -> Air transport measurement - passengers -> Detailed air passenger transport by reporting country and routes)
If you play around with this database you can see real life traffic on selected routes.
For my airline I've chosen routes with an average of at least 50.000 passengers per year since 2010 (since I dont know when AS collected the data) which equals to 136 PAX per day.
All routes had no direct connection besides mine, with a CRJ1000, standard seats, 30% discount from standard price, normal on-board service:
TSF - CRL = average 540 PAX/day real life = 10 bookings in game
TSF - BVA = average 270 PAX/day real life = 10 bookings in game
TSF - BDS = average 160 PAX/day real life = 15 bookings in game
........and this pretty much goes on like this.
Is this because of the current distribution system or due to incorrect data? Is AS traffic based on real life figures (I guess so)?
Lets take TSF - CRL for example. Even with HUBs of other players linking TSF and CRL, should I not get at least 20-40% being the only one with a direct connection?
Confusingly a route that does not exist in real life (TSF-BLQ) showed up with 60 bookings. Why is that?
the "flow" of passengers in AS is different, because of the networks build up by the players. AS only give you a "real world setting" but not the real world itself. And btw, with standard seats, you wont get any passenger. try leisure or leisure plus.
So I can assume that airport X has a certain amount of passengers wanting to travel to airport Y correct? Using my example, in an ideal case 540 PAX per day between TSF and CRL. This should be the baseline.
the "flow" of passengers in AS is different, because of the networks build up by the players.
This is what I dont understand. No matter how many networks there are, there are still 540 pax wanting to go from TSF to CRL or not?
What you are saying is the 540 passengers will rather transfer at an airport than going direct. In my view this contradicts the ORS score system.
with standard seats, you wont get any passenger.
This would be important if I try to compete with a better network but I have direct routing and already a major ORS advantage
There is a video in which Martin is explaining the way demand is adopted from real traffic, answering probably all questions around that topic. You should find the link somewhere in the forum.
answering probably all questions around that topic.
Not really. Assuming we are talking about the same video, Martin talks about absolute and relative traffic data. If I understood it correctly they take the absolute passenger amount of any given airport and distribute this number around the relative data which is made up of real world airline schedules. Since Ryanair operates TSF - CRL we should end up more or less with the numbers I took from "eurstat". He even explicitly mentioned Ryanair as a reference for HHN so I take this should be true for TSF as well.
Do you have the highest (at least a high) score in ORS on this lanes? Maybe there is someone with better connection even though he uses a hub for transfer.
The TSF - CRL route shows traffic of 200k back to at least 2004, so well established. I'd like to play AS again it's just I did HUB and spoke till exhaustion and like something new. Everytime I built my HUBs and they were well running I got bored over and over again.
@ forexlive … new features are coming which will make things more exciting. A connection analyzer, for example. This comes by the end of the year. New performance system will be next, and plans are for fare buckets to be introduced in 2015 as well, hopefully in the first semester of the year. New and exciting are coming…
The AS demand generation system does not and probably never will be accurate down to route level. Real-world schedule data is used, but only in aggregated form and on a far higher than the airport-level. Also, I personally think that it wouldn't be any fun if we simply copied real-world demand 1:1.
The AS demand generation system does not and probably never will be accurate down to route level. Real-world schedule data is used, but only in aggregated form und a far higher than the airport-level.
Thanks for you answer Martin. It just confused me again that there is no point-2-point traffic. At least not on any of the 50 or so routes I tried from various airports.
yes I've seen this thread but he's playing on the latest/newest server. Some of his routes were also my destinations. As soon as airlines have established their network a bit more they will drain his routes.
How should they drain point-to-point demand? Through to the penalty for 2-leg connections within the ORS, this shouldn't be a relevant competition.
I don't think this is true. Let's say there is one point-to-point connection A-B. Now imagine that there are 2 or in contrast 2000 bookable connections A-X-B. Even with the ORS advantage I guess you will feel the competition - you still get the largest single piece of the cake, but the cake is cut into many more pieces. Correct me if I am wrong.
Also, on a new server connections to CRL will benefit from unmet demand to BRU as long as there is not enough supply.
The spill from BRU to CRL should be right. This additional demand is especially threatened by 2-leg demand, as the ground connection is already the first leg.
It’s hard to say how much a A-X-B connection is able to drain from a direct connection. Sure, there might be some share, which is threatened, but I guess if A-B is close to a 99 ORS rating and is serviced with a reasonable frequency, we shouldn’t talk about all or nothing, especially not on short-haul flights. At least that’s my experience, but I might be wrong with this assumption.
I don't think this is true. Let's say there is one point-to-point connection A-B. Now imagine that there are 2 or in contrast 2000 bookable connections A-X-B. Even with the ORS advantage I guess you will feel the competition - you still get the largest single piece of the cake, but the cake is cut into many more pieces. Correct me if I am wrong.
Also, on a new server connections to CRL will benefit from unmet demand to BRU as long as there is not enough supply.
As far as I understand this is what is happening here.
Even if the simulation has an exact copy of demand and 540 PAX per day want to go book between TSF - CRL, there may be say 270 eligible via connections in the ORS before you start traffic on the route so they on average get 2 PAX each.
Now what happens when you start say modestly with 1 flight per day is that each PAX still considers all the other 270 working connections, so even if your route is 5 times more attractive in the weight for the PAX choice your single flights will only get 10 bookings anyways.
So to get bookings you need many flights and many connections, A CRJ1000 with standard seating for a new airline might be too many seats per flight to be able to fill in the competition.
One other thing to note is that demand is not airport based, it is area based. For instance, CDG and ORY are in the same area and so they share traffic. Traffic is allocated to both airports and some to the area in general.
I had this all explained to me in an email from the team because I had traffic data that showed passengers but the route had not nearly as much demand. The demand figures are old. In the US, I believe they are roughly circa 2010. I personally have volunteered to assist with the traffic demand modeling as it is by far one of the biggest realistic features that I believe is missing.
I do not think the demand is area based, I am servicing two same-city airports and I see different demands/flight fillings to the destinations.
Unless AS officially confirms demand is area based (which my empirical observation confirms that it is not) it is irresponsible to say and imply that it is.
It is not, it is airport based but (except Devau) there is public transportation between the airports within an "area". The absolute traffic numbers of all US airport are currently under revision. But this may take some time. So far we have updated 30 regions (28 to come) which will be rolled out with the next data patch. Further details will follow. But due to the economic slow-down in the US in the past years the result may not be satisfying to all regions/airlines ;)