The AirlineSim Technology Demonstrator: A Q&A Thread

English Boards General
21

For the “simulate regional demand instead of per-airport demand” situation:
I think this would be a huge task only to end up with quite similar outcomes - because since AS isn’t a sandbox game, infrastructure is depicted to be as in real life. Think runway capacity, night time bans etc.

What I’d prefer and what would be arguably easier to implement: Have different classes of passengers. I’ve seen that booking classes got mentioned somewhere, but tbh I didn’t dive into it to assess in what context it was mentioned.

But essentially, what I mean with that, is: Leisure travelers have different needs than business travelers. They don’t care for a high frequency, but they like low prices. For business travelers, a frequent network is essential.

At the moment, if I have a weekly demand of 600 pax a week on, let’s say, HAJ-HRG, it makes a lot more sense to fly seven times weekly with an A220-300. In real life, you’d go for a three times weekly A321 connection due to lower production costs.

Also, I’d love to see the cargo simulation remodeled: Maybe create two types of cargo customers. Currently, the AS way of handling cargo is very close to express customers, who usually ship with one of the big integrators (UPS, FedEx, DHL).

But there is another type of cargo customer, who ship’s general cargo and cares a lot less about timely connections, and a lot more about price. If I’d need to adjust the ORS to their needs, I’d reduce the impact of longer connections (something sitting in transit in a warehouse at a cargo hub for 20+ hrs is not unusual). Also, I’d dramatically increase the allowed distance for ground networks to simulate road freight services. Truck shipping costs are nothing compared to air shipping costs, and every day you see cargo being shipped from Munich to Hahn or Liege by truck, simply because this type of customer would not care about the one or two days of delay this takes, they wanna avoid the 20+ days it takes to get from Europe to Asia or vice versa through the Suez Channel.

22

If the city demand (as a whole) is considered as part of the game and also demand is based on population, GDP etc, then the local transportation has to be also reworked. The simple way to put it is that if there is a high-speed railway between the cities pairs, especially city under 1 or 2-hour train ride (or 4 hours driving or cities with ferry crossing etc), then there should be close to no or very low demand to direct traffic (but of course with connection traffic, and as local transport time to the whole trip increases, the demand also increases, e.g, people will have less an issue to drive two hours to catch a 20-hour flight if the price is right or the total transfer time is less due to additional transfer time. i think a lot of people might consider a two-hour drive compared to an additional 30 mins connection plus one hour transfer time).

This is not an issue for some of the counties, e.g., the USA which the general local transport is poor (with some exceptions areas), but if considering countries in east asia where the high-speed trains or local transports are extremely common, it is fairly easy to notice this: there is a very good reason there are close to no flights between Tokyo/Osaka to Nagoya, Peking-Tianjin, HK-Shenzhen/Guangzhou/Macao, Shanghai-Nanking. Taipei/Kaohsiung, because people can travel faster between these cities using trains or local transport network than go through the airport system. This can even apply to the city itself: the airport with convenient transport to the city centre or city limit will be always more attractive than the one without.

in some ways, i would argue that this overhaul is more important than the city demand side because currently a lot of airport slots are taken by unrealistic traffic such as the cities pairs i mentioned earlier, and if you start to use GDP/population density etc, you will only make this issue even worse.

There is also an issue about the point of having city demand as a whole so that players can use other airports in the cities as an alternative way to access city demand: there are a lot of airports in the world that are domestic only: LaGuardia in NY (except if you go through pre-clearance outside US), Itami in Osaka etc. I understand you can have remote immigration facilities as i recall suggested in the past discussion (though i don’t think a lot of these airports or city authorities will ever agree with this practice in mass), but then you have to reflect it at the airport. Also, most of the countries in the world separate domestic and international traffic, so even if the airport is international there is a ratio between domestic and international traffic at the airport. And i don’t think a lot of airports will change their ratio of two in real life that easy. there are airports i see have a flexible terminal structure so that the gate can be used for either purpose, but the number of these gates is small compared to the total gates, and there is also a limit on the passport control side.

Though i am not an expert on this, i don’t think having regional airports have full access to the international market make too much sense, or at least for many places in the world. I really don’t know how the traffic works in Sweden, but i would assume most of the regional airports have flights to other countries but within EU? i think the main issue here is not the Swedish airports need more international demand, it is the Schengen area itself should be more or less considered as domestic (or demand is between international and domestic but closer to the domestic side). There are actually restrictions in real life that are quite opposite from this but allowed in-game, e.g., from my understanding, it is actually illegal for the airlines to sell tickets between a domestic city pair in the USA via a second country (e.g., fly between west coast to est coast via Canada), so this will actually reverse the demand curve. Also, i don’t think most of the countries will feel okay if a second country tries to take away its domestic traffic via an international hub (i literally have this issue on my server: there are a very limited amount of transfer possible airports in Indonesia, and due to the country location and shape, you can literally have domestic pax travel through a Malaysia airport, which i notice when myself also started some routes and notice that demand. i don’t think there will be ever such a demand in real life unless the whole of southeast Asia is like the Schengen zone, and even if that is the case i have some doubts).

Just a side note, i think it would be great if the connection flights are divided between domestic-domestic, domestic-international, international-international, with domestic-international (or international-international if that also requires passport control) having the longest transfer time. Currently, a lot of the airports’ connection times might be too optimistic, especially the ones in US for international connections. I have say as a non-American, i will only book a one-hour connection for international arrival only in a very selected amount of airports because of the border waiting time. You will be very lucky to go through just the border check in an hour for most airports. i think the airports with US pre-clearance literally sell their airport based on this. I think making this kind of change might also help domestic market airlines without banning certain traffics while being more realistic.

I think Martin already have a lot of work to do on the booking side, and I think it will be just a huge hole if demand, especially special demands such as vacation and demographic pairs, also becomes part of the considerations. I think it might be better to get one thing done than have multiple things to complete which might make the project becomes impossible.

just some thoughts. :grinning:

23

Considering you’re thought on direct demand for short distance city pairs I would say the ground connections should be reworked to make sure passengers go by car/train rather than fly. So, overall demand should be high, but demand for flying should be very low (maybe that’s what you meant).

Take Washington-New York corridor for example. New York to Philadelphia is a 1,5 hour drive by car. Flying from LGA to PHL would take approximately 40 minutes. However, if the demand data point in the game would be placed central in Manhattan and Philadelphia the game would need to add 15 minutes on either side for connections and a transfer time of say 40 minutes on either end. That would increase total travel time to about 2,5 hours, 1 hour more then driving. So, unless the cost of flying is a lot lower, people would take the car.

Then take Washington - New York. Driving would take about 4 hours. Flying from LGA to DCA would take about 55 minutes. So, taking the same parameters as from New York - Philadelphia. The complete trip by flying would take 3 hours. 1 hour less than driving. So, people will choose flying over driving, unless flying is very expensive.

So, all you have to do to make these short flights disappear is putting the datapoint in a population centre rather than at the airport itself (which I believe they’re already working on) and making ground connections realistic in terms of duration and costs and allowing passengers to book trips only using ground connections.

As for you’re suggestions of border crossings, transfer times and the effects of clearing customs. Modeling this in game based on real world situations would be breaking the game in my opinion. The game is not to build another American Airlines with an hubs at PHL, JFK, IAD, LAX, ORD, etc. and then just remodelling that real world airline in game, that is only one part of the game of you like to do that. Another part is to try and find alternative solutions, like using a smaller more regional airport as your main transatlantic hub like PVD or try to find that sweet spot on the east coast for flight from the Northeast to the South, even though in real life that airport doesn’t handle more than 1 million passengers per year. Maybe for you Dubai isn’t the big hub between Europe and Asia/Oceania, but Muscat serves that role. The game should allow for those types of choices. So also airports that now don’t see any international flight should be able to handle them.

I do like the idea that international-international, domestic-domestic and domestic-international (or reverse) transfer have different connection times at airports. This might give rise to some interesting flight schedules where in case of wave scheduling domestic flights almost always depart before international flights because passengers transferring to those international flights need more time for their transfer.

24

Yes, but with considering of possible connections after long haul flights.

This will be a very troubling way to model the game because you can have a city that is covered by both airports. Actually, this makes the domestic/international separation more important because most of the domestic airports are closer to the city. Just take NY as an example, if you are a united flyer, you might hate to go to Newark to catch your flights, but you will still do it even though it is further away. and to be fair, what is the definition of the city centre? Manhattan? Brooklyn? I am pretty much sure more people live in Brooklyn. in that sense, you can’t place demand at one point, you need to place the demand in a boundaray.
Also, what about the demand from another city? just use the population density map:

image
image667×526 70.4 KB

The city of Suzhou has no airport, but the population is higher than NYC, and the closer one is in Wuxi and Shanghai. If you do only distance calculation, then you will think people all go to Wuxi. Wrong! people go to the further ones in shanghai because there is a direct train service to the airport while Wuxi is driving only. Not everyone in the world is like Americans that love to drive for no reason… That sounds like the modelling of the whole world local transport network is required for it to work. I am not sure if that’s too much a task…

Not sure how these will break the game. most of the people go through Dubai are connections, so as long as the connections pax is there, you can set up your hub in Muscat. PVD also has a clearance facility, so you can welly do your hub there. The one i want to avoid is something like this:

image
image504×520 105 KB

There are three airports in the Osaka area, The one on the top closer to the city is the RJOO, Itami, which is a domestic-only airport. Most of the international flights go through the one at the bottom, which is RJBB, Kansai international. and that is the main international airport for the city and has a larger international capacity than its domestic. RJBE where Kobe is also mixed-used, but i guess the general capacity will make it less preferable anyway. If you start to use city demand without defining domestic /international only airports, you will have RJOO having more traffic and with a lot of international flights, which is completely wrong compared to how it works in real life. The same story for Tokyo, which RJTT, Haneda, is much closer to the city itself, but the RJAA, Narita, is the main international airport. There are a relatively small number of international flights at RJTT, and the airport is working on increasing it, but you really need to fight to get those international slots. If you don’t consider these, you will make people all concentrated flights at RJTT, and people who come in late in the game have to use RJAA which is a disadvantage in the first place due to distance calculation, which is though true and same in real life.
One impact is there will be a reduction of the number of airports that can be used as an international connection hub, because obviously, you can’t use RJOO to build your international hub as more. but to be fair, most of the countries that have pure domestic hubs already have enough international airports for you to build on, so i don’t see that as an issue. Another impact is it will be harder for you to access domestic traffic in another country, which is probably true in the real world anyway. and that also makes it possible to open up more countries for the players because now they can at least start an airline in the country might still have sufficient domestic demand instead of having the demand taking away by a larger palyer in the nearby country already. I know people might have a preferrable country to start: when i joined the game there are definitely more than realistic amount of players in Germany on my server than it possibly could be in real life, but at least this will help to free some of market up so that the server is not dominated by serval players over the years. I think it is very important for AS to have more attractive older servers than open up newer ones with the same config just because new people think they are not able to start on the older one properly.

actually, this is a good example of the special case in usa as i mentioned. I think you forget about Acela, which the city pairs you mentioned is the only few lines profitable for them because of the demand of not travelling by air. The train is competitive because DC to NY ride is about 3 hours, even though acela is not even a fast train, which will probably still end up the same time if you take a plane. The people who take fights between these cities are not budget travellers, or they are taking the trains. You probably have more people taking these flights as connections, or are less price sensitive.

I think the game has to consider train traffic (not driving traffic) if we go down that route. when relative faster train service is possible, people are not comparing driving to flying, people are comparing train to flying. and usually, you will end up train winning most of the battle if the flight is under 4 hours (with travel through the airport system accounted). There are some exceptions, which is when the train ticket is more expensive than the flights. One example is the Tokyo-Osaka: where the shinkansen is more expensive than flying. In that case, you get demand for flying when you set the price lower than the train price, as the airlines do in real life, but if you set it higher, your demand will just go down like a cliff. actually, this is a very good example of the train preference in real life, because even when the train has a higher price if the total travel time is about the same, people still prefer trains just because it is easier to access and service is generally more predictable inducing other factors like weather etc.

i am not trying to sell train services on AS, especially when i work in aerospace in real life. but i think a lot of issues in the AS currently are caused by the improper modelling of the ground network, especially the train portion.

25

Whether people choose to connect from a long-haul flight to a domestic flight to their destination or take train or car for the last part depends again on travel time and cost. In Europe it’s not uncommon for people to travel by train to the airport, then board an international flight, and the same the other way round. You could easily arrive from North America (or Asia or any other continent) at CDG and take the TGV to Lyon. It’s often faster or just as fast as boarding a domestic flight to LYS and then travel to the city. But that only works well because there is a train station at CDG. Which is very different from for example LHR where there is no high speed rail connection, you would first have to take a train/metro connection to London from where you can take a train to for example Manchester. And as that train isn’t as fast, it is probably faster to take a domestic flight.

Well, this would be a question regarding the new Technology Demonstrator as they are working to making this point or area based demand a part of the new game rather than demand based on the airports themselves.

Modeling all the world transportation networks would be too much work I guess. So, then the question becomes what is more favorable. The current situation where all demand is airport based which causes it’s own set of problems, or a situation where demand is more realistically modeled, but in some situations passengers travel to a slightly different airport then they would in real life.

I doubt that situation with Suzhou, Wuxi and Shanghai is very frequent for the sole reason of ground connection networks. In most cases there would be a train connection between Suzhou and Wuxi and people would still travel mostly to Shanghai Pudong because it just has a lot more connections. Which in game should not be modeled and only occur due to players making more of a hub at PVG.

Well, to me it is part of the fun in the game to make airlines work that might not be possible in reality or that didn’t happen due to circumstances. Like the examples I mentioned before, but any other scenario’s you could conceive like making Osaka Itami the international hub airport.

And if there are more reasonable metrics to limit international flights rather than “it’s not an international airport in reality” I think that would be great. Like LCY can’t be a big intercontinental hub because of it’s short runway and steep approach angle. Or perhaps increasingly strict noise regulations for the airport, or high slot prices or a lack of gates that can handle large aircraft. But I like a game to make as much of thing possible while restricting the choices by introducing “penalties”/downsides.

For example, you can fly transatlantic to LCY rather than LHR of LGW, but to do that you are limited to some small capacity aircraft, which results in lower revenue. But because you offer a faster connection to the city center you might be able to charge more for a ticket. Or the exact opposite where you fly to an airport like BVA rather than CDG saving you money on slot prices and allowing for better flight times as most slots are still available, but in return you have to lower ticket prices because passengers need a reason for the longer travel time as the airport is so far outside Paris.

And if, as a result of those downsides all international flights again concentrate at LHR or CDG, etc. that’s fine and probably means that the modeling in the game is done great.

Well, I completely agree that the ground networks need some proper modeling. Not just trains, but also ferry’s. Passengers that want to get to Bonaire don’t consider flying to Aruba and than take a ground connection. Let alone fly to Venezuela.

26

i don’t think i ever said airports should be punished because they are not used as often as in the real world… but you probably will have an airport that is less favourable if they have a bad ground connection, and i believe that should be represented in AS. Shanghai Hongqiao will not be as much competitive as Pudong if it is without that high-speed train station, just for pure connection standpoint.

Interesting you are putting it that way because the domestic airport is not adding ‘penalties’ to any airport, it’s to reflect a fact that the current AS is not having. If the airport doesn’t have a custom, it is like the current runway length of the airport which is simply not available. Of course, the custom can be added over the years like when RJTT becomes back to more international (it was running with only 2-3 gates for international and mostly to Taiwan and South Korea for a really long time), but so does the runway. Airports with short runways always try to add the runway length for more accessibility. However, it doesn’t change the fact now you can’t land a 330 in LCY, that’s the same airport as you can’t land international flights in RJOO.

As i mentioned earlier in distance-wise, that’s also a good reason they are chosen to become domestic only: the international airport needs to or cannot be built as close to the city centre as the old one, and the old cannot be expanded. A lot of domestic airports actually have been also considered to be closed completely, like RJOO when the new airport is built, but a lot of them was kept open because the local support and economic dependency. If you make domestic becomes international possible, you are technically dismantling the city transportation. Now you are actually putting penalties on the international airport when you switch the demand from airport-based to city-based because most of the international airports will not be able to compete with the domestic-only ones distance-wise. I am more than open to hear if you have any better options than not putting the domestic-only airport as domestic only but for some reason still pushing all the international traffic to where they suppose to because what it sounds like is you need a weighted domestic demand for the city on one side of city, while another weighted international demand on the other side, until finding a city which has two international airports on the two sides of the city, while one domestic-only airport closer to one the international one (wait did i just see NYC)

27

I want to discuss some of the points you all raise up.

Let me preface with two of my own:

  1. Airlinesim was created and exists to provide a realistic airline simulation, or in Martin’s words:
    image
    image1105×524 59.9 KB

If we want to factor in other forms of transit, it changes the element of the game… for this there is already transport simulation games for many methods of transit, and Airlinesim loses its essence. I personally would not like that to happen.

  1. I’m not sure this is the (correct) thread to talk about this stuff…

But now, to your points…

Jan, posts 16/18:

a) Seasonality - of course we don´t want a game we have to be attending to 24/7 to manage, but would you consider making demand for different airports flexible by half-year summer-winter seasons, which could be somewhat manually determined, that airports in XY climate would have a XY% hike in demand during summer or winter.

So if we do twice-yearly shifts, imagine an airline (and these aren’t unrealistic figures, despite their look) that has about 2500 planes spread across a few subs. So to catch this demand change, a player would have to set aside a few days every 6 months to essentially redo 500 plane schedules or something (higher if the model is built on this demand)? Does anyone really want to do this??? I sure don’t…

Additionally, if an airline has traffic rights in say Greece, where do they send their planes in the winter? Does it make sense to ask a player to essentially shut down an airline for 6 months or so? Because I tell you that, there is not enough winter complementary destinations from Greece (and many other hubs) to offset the summer demands (and vv)… so that rules out possible markets for people to play, further aristocratizing servers and corraling the players to battle it out for the “decent” all-year hubs…

Additionally the airlines that do scale back (say your SunExpress) often have other revenue streams within the company to offset the cost of keeping the planes well during the winter. This really can’t be modeled in AS, and the model gets inviable pretty damn quickly.

  • Average annual temperature at an airport (lots of weather data sites out there): This could help determine kind of medditeran climate, which is where summer leisure goes

Well, it’s pretty hot in Kuwait in the summer, but you don’t go there for the beach…

  • Elevation: It´s possible to also get elevation data globally, so it could be used as a definition helper for winter leisure travel

Not every mountain is used for skiing/trekking…

  • Water: the maps that show where the airports are, show the immediate map in the AS even today - if a software could analyze, how much of a picture of a map, that shows let´s say 100km around the airport, is blue (water), it could be an indicator, that the airport is on an island, at a seaside, etc., which are huge demand factors

Yeah, but not every island/seaside is for tourist demand, hardly anyone goes to TLL for water vacation despite being on the gulf there…

What I mean, is that modeling demand that is based on human infrastructure and purpose, modeling it based on raw geography, would do it a disservice.

b) Demand by population in catchement area and not an exact airport - it is possible to insert population “heatmaps” from different sources, recently the game NIMBY rails did it. Would You consider generating demand for given airport based on a combination of population in a defined area and an average local GDP/capita for example? Both accessible datasets. I am mentioning this, because today the demand of different airports is set from the real world traffic, which pushes the virtual players to replicate existing real world networks, while in reality the traffic on the big airports is now given historicaly, not necessarily by such demand. For example the demand in the CGN/DUS/DTM area would be very similar for all three airports originally as of population and GDP, but “by accident” in the real world DUS has the most pax, and CGN is a Cargo hub. But it would be nice if AS players wouldn´t be pushed into doing the same, and instead have the chance to develop concepts on different airports based on a catchment area demand.

Well, the catchment demand area the best way I see it is if we tallied together the demands of the airports (as someone who has seen some demand stuff, it comes per AIRPORT, not per area), if we got all of the airports in an area and aggregated their demands… (and even then we have to settle what is an “area” especially in more ambiguous areas such as in Brazil…), because if we just do raw city areas (or ISO 3166’s) then we can just pair the biggest cities and that’s it… oh wait… After Dropping Chicago - Beijing, American Airlines Considers Ending All Chicago - Asia Flying - View from the Wing - that’s a very good read btw
So is just pairing populous rich cities how airline demand works? No! Is that the way it should be in AS? In my opinion, no… if so like Thomate says below, it becomes a sandbox, and that is not what this game in my opinion is, should, or wants to be… if we want this game to be a realistic situation, the demand (whether city or airport) has to be realistic, and the best quantifiable way to do this is through schedule data / pax numbers per route. Sure, this doesnt allow for creation of demands like LCC airlines do, but can an algorithm such as Airlinesim’s predict the WIZZ-style decisions that are done in a human mind to book to some unknown beach just because it is for 9 quid? Can a computer decide whether a ticket to Durham MME for $9 is worth less than a ticket to Tarragona for $9?
Can an AS-style airline with no ancillaries even break even on this kinda pricing? Should it be what we want? Then some of you complain about “price dumps”… lmaoo

Myquandro, post 19

If I’m looking at travel demand to for example Spain in summer, why is it that there are specific cities/regions that the Dutch go to, and other for the British and others again for the Germans? Why does this clustering happen, is it quantifiable and can it be programmed into the game? And I guess there are many more of these quirks all around the world, because we’re dealing with people that don’t always make logical and easy to understand decisions.

One would think this would be already justified using existing demand data. Why those clusters happen, I frankly don’t know.

For example Australia has high demand from the UK relative to other European countries due to the Political ties and shared history of the countries.
And the same can be seen with many West African countries and France, South Africa and Spain or between various Caribbean islands and several European countries.
But also less obvious is the demand between specifically New York and Israel because of the large Jewish population in New York or direct demand between China and San Francisco for a similar reason.
Looking at industry I would expect strong demand between The Bay Area and Shenzhen and Taiwan. Between Detroit and Southern Germany. Between New York, Hong Kong and London. But where the other examples are more focussed on economy passenger demand, strong relations on the basis of industries would be more focused on business passenger demand.

Are all economy passengers and all business class pax the same and have the same preferences? This is what the booking classes are trying to address.

Thomate, post 20:

At the moment, if I have a weekly demand of 600 pax a week on, let’s say, HAJ-HRG, it makes a lot more sense to fly seven times weekly with an A220-300. In real life, you’d go for a three times weekly A321 connection due to lower production costs

I think this raises again the need for more “old” aircraft and a feasible process for maintenance and retirement of an a/c. And how certain mx providers affect how “long” a plane lasts in service. And when it becomes more viable.

In most long term worlds the oldest plane is as old as the world itself. In Lima, we just got the one year old aircraft. Only in the stagnant ‘phase one’ worlds (< 2010) are any “old” aircraft truly in effect… but then again this also raises the point of how many established network carriers are willing to shift flight plans around and adapt them to new aircraft at the large scale if this retirement came into play.

TWAAir post 22,

This can even apply to the city itself: the airport with convenient transport to the city centre or city limit will be always more attractive than the one without.

How do we model this for every one of the airports? Is it the best use of resources to model the distance between the “center” (which is not obvious in every pattern of development, for example what is the “center” of Paris, or Bogotá?) of an airport vs catchment area (which again is subjective, for example is XCR part of Paris catchment?), for every one of thousands of AS airports? For me, best to use effort elsewhere… like fix unrealistic GN connections such as no ferries between the Canary islands… or EZE <> PDP (only the super rich with their Yachts, otherwise a 15h+ drive? this is “ground network”?)

Just a side note, i think it would be great if the connection flights are divided between domestic-domestic, domestic-international, international-international, with domestic-international (or international-international if that also requires passport control) having the longest transfer time.

Is every international <> international equivalent though?
For example as you say, since the US requires all pax to clear customs before ANY connection (and why many airlines refuse to sell itineraries such as CUN-MIA-FRA via MIA) this takes way longer than a connection that is on a “secure airside” such as in PTY where you clear no security before connecting. Additionally, a connection that is LGW - FRA - VRN will be shorter due to Schengen but is still international for the purposes of AS…

Currently, a lot of the airports’ connection times might be too optimistic, especially the ones in US for international connections.

Yeah… the numbers are pretty arbitrary though (it has to do with the airport size, something that IIRC is also arbitrary). It’s pretty hard to quantify what an “example” transfer looks like without consulting every airport bureau in the world if they even measure such data… and again, prompts the question, is this the best use of our 20 cents a day?

Myquandro, post 23

Another part is to try and find alternative solutions, like using a smaller more regional airport as your main transatlantic hub like PVD or try to find that sweet spot on the east coast for flight from the Northeast to the South, even though in real life that airport doesn’t handle more than 1 million passengers per year. Maybe for you Dubai isn’t the big hub between Europe and Asia/Oceania, but Muscat serves that role. The game should allow for those types of choices. So also airports that now don’t see any international flight should be able to handle them.

I don’t see how it doesn’t. I’ve seen plenty of hubs in ICT, DSM, OMA, HSV, MCT, SHJ, MED, things as such. If your hub is solely reliant on connection data and the geographical and airport constraints are ok, then it can (and is often) done. The issue if it doesn’t work then shifts to the player and how they do it, because it is most certainly possible…

TWAair post 25

Yes, but with considering of possible connections after long haul flights

I wish there was a way to quantify this and isolate this demand from the person who flies due to maybe no will to drive (because it’s a nightmare to navigate the toll systems, perhaps?)

Again, some interesting ideas as to how we model demand… but then we ask how much realism we are willing to sacrifice to have more demand pairs? Is this something we want?

And you aren’t solely constrained to what demand is for an existing airport in real life routes. For example, I run a flight from Madrid to Poznan, Poland. (There’s no OD demand to it or anywhere around from MAD) but it works with little connecting pax. It fills well…

If we’re just looking for new challenges, then simple, just expand the foreign investment list…

28

Well, I think this is the main point. I don’t want the game to push passengers to “where they’re supposed to go”. I don’t have any problem with a situation where JFK is an empty airport in game where only a handful of airlines go to because it’s too far from the city and only chose this because EWR and LGA are congested.

29

What you are saying just implies the slot condition will be much worse on AS and even more unfair competition between old and new players. because you are pretty much running the game by ‘first come first serve’, which makes the server more unsustainable for long term.

sorry, if that’s the goal for the AS changes then i rather the AS be where it is at right now. you clearly don’t see the dominated effect of old players on the old servers. and i can tell you if don’t do it that way, there is a very good chance other players will push you out of the market.

30

I think it’s the opposite. making some airports domestic only would just be more limiting for international flights. Then, when a new player wants to connect to Tokyo and there aren’t any slots available at NRT, then bad luck. But when HND is open to international flights, there is likely some room left at that airport, or when HND is the congested airport there are still slots available at NRT.

And compared to the current situation I don’t think it makes such a difference on competition. Right now airports like JFK, LHR, LAX, ORD and SIN are already congested to the point where at best 10% of slots are available, and you’ve to be lucky if you can find a slot that fits with the schedule and is also available at the other end of the flight. So, players are already moving their flight to alternative airports. Take New York again as the example. JFK and EWR are often full, but have the highest demand. So, new players are forced to connect to LGA or even ISP or SWF. Airports that have much lower demand, so new players can’t compete.

There is the difference that in the current system players that fly to LGA get the direct demand from that airport, so they are actually competing for different passengers. But in the new situation they could compete for the same passengers, but have to find a different strategy, like lower prices. What works better for new players? I don’t know yet. At least looking at the current long term servers it’s clear the current system isn’t working well for new players, so I welcome any change and just see if it is an improvement.

31

To answer your critique

A) seasonality: as I mentioned, seasonality basically applies to leisure passengers. If an airline has 2500 planes (which a few players have now in AS, but actually no one airline on the planet has such fleet), no one is saying the airline has to be oriented on leisure travel. Since the point of the Technology demonstrator is to introduce possibility of variety of business models, this is one factor that counts. Even in the seasonally changing leisure demands, there would still be a ± constant business travel demand, so if a player doesn´t want to change schedules every 6 months, he can concentrate on the high-yield business related demand, which doesn´t experience much seasonality. In practice, that is all You can do today in AS, only one type of passenger. But seasonality would actually allow for more players to find niche markets → someone with big fleet doesn´t want to change schedules every half year? Good, there will be a market for a small player with a small fleet, who for example runs leisure flights between Germany and Spain/Greece/Italy/Morocco in the summer, and in the winter leases out his planes to another player in Canada, who needs them for winter capacity to the Carribeans.

B) Temperature, Elevation, Water:
If you combine these factors with the GDP+population map based calculations, You will actually get about the results that reflect the real world, a few examples.

Let´s say we set up some definitions for the calculations:

  • Summer leisure demand boost for average annual temperatures 18-25 degrees celsius (just an example)
  • Winter leisure demand boost for average elevation (i don´t know) between 1000-4000m above sea level (again, just an example)
  • Summer/Winter leiuse demand boosted airports won´t get demand boost between each other
  • Island demand bonus is not because of tourists, but because there´s no other fast way to travel abroad. See this graph I made of a relation between GDP/capita in European countries and an average annual number of flights/capita in those countries. All the highest numbers are islands - Cyprus, Greece (lots of islands), Iceland, Malta

GDPflights
GDPflights895×550 57.4 KB

Now, a few examples, how the combination of the factors mentioned above will work in practice:

Example 1: Rhodos (RHO): Will get island demand boost all year, will get summer boost because it is in the 18-25 celsius annual temperature range, and the summer boost will be substantial because there is some measurable GDP in Greece. This demand will be realized more from central europe then UK, because a MAN-RHO flight demand gets a bigger distance-penalty then a PRG-RHO flight.

Example 2: Kuwait (KWI), because You mentioned it: Won´t get a seasonal demand boost, because it´s annual temperature is >25 celsius, and even if it did, the market for it from Central europe would be much lower then to Turkey or Greece, because KWI is much further = more substantial distance-penalty.

Example 3: Tallin (TLL), because You mentioned it: Won´t get any seasonality boosts, because it is outside of the defined temperature range.

Example 4: Lukla (LUA): Probably won´t get a winter seasonality boost, because the himalayas around it are higher on average, then the defined range. Even if it did, the GDP in the catchment area is so low, that the eventual seasonal demand boost wouldn´t account for enough for it to be worth flying there much more → The catchment area low GDP is a clear information, that people don´t go skiing there.

Example 5: Innsbruck (INN): Will get a winter seasonality boost, because the elevation in the catchment area matches the defined range. The demand boost will be also substantial, since the catchment area GDP/capita is significant, signaling, that it is actually a place where people go skiing.

C) Area demand:

  • The basic GDP+population layer demand calculation should be based on a presumption of fixed price. That´s the first step. The fixed point can be the average ticket price for a flight of an average lenght. This simple calculation will give You a demand for air travel generated by an area at this fixed price point. And, what more, from there You can calculate basic city pair demands based on this (as a STEP 1), let me give You the example:

AMS-MUC route
Data needed for calculation:

  1. Population and GDP/capita for the catchment areas
  • AMS: population 2 520 000, GDP/capita $81 746
  • MUC: population 5 991 999, GDP/capita $32 532
  1. Average flights/year/capita for the two countries, and thus what amount of GDP/capita “creates” one flight on average:
  • Netherlands: 4,7 annual flights/capita → 1 flight/capita/$17 392 of GDP/capita
  • Germany: 3,0 annual flights/capita-> 1flight/capita/$10 844 of GDP/capita
  1. Global GDP
  • $80 trillion
  1. Route distance: ± 700km
  2. Distance penalty presumption (see in post below, I can only insert one image here):

So let´s calculate:
STEP 1: Total origin/destination pax generated in the two areas (Let´s call this ODPAX):
Population x GDP/capita / 365 days / averageGDPCperflight
AMS: 2 520 000 x 81 746 / 365 / 17 392 = 32 451
MUC: 5 991 999 x 32 532 / 365 / 10 844 = 49 249

STEP 2: Defining a proportion of these pax who will want to fly to the other destination:
Area GDP / Global GDP x ODPAX
AMS: 206 billion / 80 trillion x 32 451 = 84 pax
MUC: 194,9 billion / 80 trillion x 49249 = 120 pax

STEP 3: Applying the distance bonus/penalty:
Combined route pax demand x route penalty (in this case a bonus of about +42%)
204 x 142% = 290 pax

Now, we have a basic p2p demand for the route at a fixed average price (what ever it is) based on the two areas´ GDP and population. Before You mention it:

  • Yes, the number seems to be small, MUC-AMS is a heavily flown route, BUT:
  1. this is a demand on fixed price, the next step is to apply the cost/demand elasticity curve in the system (which can be also pre-defined, and is quite calculatable from real world), so the pax numbers will be higher with lower then base-point prices
  2. Both AMS and MUC are HUBs, lots of traffic on the route are actually transfers, not p2p pax, this will add to the flight (as it indeed does in the real world), if someone would have a hub or interline in AMS or MUC in the game.
  3. I only calculated with the basic MUC and AMS Metro population areas to make an example, if we would calculate with a catchment area of 150-200km, the number would come out bigger.

The basic demand could be calculated for each and every city-pair on the planet based on this, and as next steps You can introduce some manually added demand bonuses or penalties which apply to different circumpstances such as NYC and LON being big financial centers, or a city being a big Oceanic port trading hub, or a country hostile to a specific different country, or a stronger historical connection between some countries (for example the GB commonwealth, or France and the African ex-colonies, etc.) But these are as to numbers factors, that only add to the apparent base, and don´t actually affect that many airports compared to the whole volume of them.

Also Your point of only connecting the big cities - of course not, that´s what transfers are for, and the game does it even today. And real world also - You don´t have for example a full-fare BLL-MRS route, not enought p2p demand for that, instead You have KLM, Lufthansa, Air France flying to MRS and BLL, as well as bunch of other small airports, offering transfer itineraries. And if someone would try a BLL-MRS p2p route, it would have be on a much lower price point to get the volume (so it wouldn´t be profitable, or only with lots of ancillaries as W6 and FR does it).

D) International - Domestic airports:

I think in general there are more international airports than domestic ones. In Europe basically every airport can be used as an international, then for the US pre-clearence system I guess there is a list which can be added to the game manually, as well as a list of airports which would be clearly defined that can not handle international traffic.

E) Concern about using small airports as global hubs:
That won´t actually happen that easily, because even today You have slot limits in the game, and there are naturaly much less slots on smaller airports given their infrastructure, maybe that is something that could be made even more precise (for example to take into account some basic terminal capacities at airports, so it wouldn´t be actually possible to have 60 flights an hours on an airport like BRQ, which has in real life a terminal that can handle max 3-4 flights/hour, so You would have to build Your own terminal to do more (just an idea). Anyway, I wouldn´t worry about making the small airports in big HUBs in AS if the demand calculation changes, if infrastructural limits are set up at least in a way as they are now, the big HUB models will be anyway pushed towards the big airports by nature (which is okay).

I of course didn´t reply to everything, but I mostly adressed the critique of the issues I brought up in the first place. Thanks for Your reaction by the way! :slight_smile:

32

distance bonuspenalty
distance bonuspenalty884×552 30.6 KB

33

actually not really. you are not operating any other transit, but you should be competing with it, or i am not sure how do you even call it a ‘realistic’ airline simulation in the first place because that’s nowhere close to reality. i am not sure how the current ground network is defined, but you could have the current ground transport becomes part of the ORS: if there is a highspeed connection between the cities, and i believe probably the sole important one for long-distance cities pairs, the ors system will include the train with a defined price and travel time. The number of HSR is a pretty finite number, so it is some work, but not an undoable one. Ferry probably is also needed for island traffic, which will have a higher preference for direct traffic.

i will define a ground network as: local transit (bus or trains or underground, i would suggest this to be zero ground connection time, as at the airport), HSR (transit time to the city station), and furry (transit time). I guess for driving you will just do distance, and it should be the least preferred way anyway. I would probably rule out furry for connection traffic because not that many airports have direct ferry access, though i know quite a lot in japan, i would just reduce the complexity. Adding GN only doesn’t work, because players can still dump flights on it due to demand, so you need to have them being part of the equation when pax make the decision on booking the flights. this is a lot of work, but i think you will end up only have a lot of work for the airports serving large cities, and i will say the majority number of the airports in AS are regional.

I think this is very hard and i won’t suggest applying it for intl-intl. i was just trying to state the fact. but you can at least add a standard 30min to the airport for international to domestic connections because an additional 30+ min is better than current, which is 0.

i don’t see it to be applied to AS to be fair. but after so many years of only talking about changes, i rather put the idea there so someone smarter than i am might see an easy way to do it.

Probably unlikely if demand is placed at the city because then you are making HND 10 bar and NRT probably 5 because of the new distance rule. You don’t need to be a rocket scientist to see people will all take a pie at HND first. I can only think of placing both airports as the same demand (disregard distance and consider both of them in city boundary, then this might work as you said, but then you will have a really hard time to start to define how far are other airports that are currently underused by the player, like Ibaraki which also served Tokyo but definitely underused in real life (as HND and NRT is not completely exceeding capacity) and in game. This method will probably be too much work for modelling because now you are mixing the current airport demand system with the distance-wise system. I will be probably even more complex if the pax booking demand, e.g., price vs time-sensitive pax, because now only price-sensitive ones will have a higher chance to go to airports that are further away.

i don’t see the airport being full because more airports are needed; i see it as a result of over-frequent flights. to make the slot available, adding airports close by only delays the issue (for the long term game world). The true solution is so that you should not be able to run an average of the cities per 10/15 mins. I think it is reasonable to add a closer airport so that demand can be accessed, but i don’t think it is ideal to have people forced to use a disadvantaged airport in the first place.

my opinion is you need to make the slots available for the new player, at airports that are equally competitive and with equal demand level per existing players. That means pushing to airports that are distanced away, e.g., ISP and SWF, doesn’t help if the demand is calculated per the distance to the city. The only way i can think of is to enforce domestic-only airports with a frequency restriction so that players can always access the domestic demand easily, and as proven in AS and real-life, especially after COVID, you can see how the domestic market helps airlines to recover and is the easiest one to start. It will help with the current airport-based demand system because you are literally removing any possibility for older players to build an international connection hub there. If the demand is placed per city distance, it only makes it more important to have this policy so that demand is ensured.

34

I don’t see how demand, especially international, would be as sensitive to a little more distance as a 5 to 10 bar difference. Say, you fly from Frankfurt to Tokyo. The shortest distance (FRA-HND) is 9,373km and the longest (FRA-NRT) is 9,394km. Flight times (with 787) is 10h43m to 10h45m. Driving to HND airport from Tokyo city center is 19km, so I would approximate it as 23 minutes (50km/h average) and to NRT is 65km, so 78 minutes. At the other end, from Frankfurt city to airport FRA the airport is 14km away, so 17 minutes. And say there is a 1,5 hour wait at NRT or HND to get through customs, check-in, etc. and 45 minutes at FRA to get off the aircraft and collect baggage.

Now the question becomes, how inclined will a passenger in AS be to save 57 minutes on a total 14h35m trip. If prices are completely equal, most passengers will choose to fly to/from HND rather than NRT, but especially economy passengers are very price sensitive and if the flight to/from NRT is 20 to 30 AS$ cheaper they would likely choose that. And even business travelers are likely to make that trade off.

Now, if the trip was domestic. Say Tokyo to Fukuoka. In this case the flight duration is 1h28m and the drive from airport to city in Fukuoka is 7 minutes. And to check-in, collect bags, etc. I now calculate a total of 1h15m combined. Now the total trip takes 4h08m.

In this case the passenger can save 1 hour of travel time on a 4h8m trip. Now the even economy class leisure travelers will be quick to take a flight from HND rather than NRT. Unless of course the flight from NRT is half the price.

So, the longer the flight the less inclined passengers will be to choose a slightly closer by airport over one further away. And thus I expect airport like HND to quickly become domestic hubs, and international flights depart from NRT. Especially if there is some lower slot price at airports that are further away from the city, which can enable airlines flying from NRT rather than HND to have lower ticket prices and thus be competitive on international routes.

35

I think you just point out the issue: there is no enforcement that you have to increase your ticket at HND or it’s forcing players to use NRT and lower their price. This is an unfair advantage for older players because it means the earlier you start your airline, the higher the possibility for you to be able to run a full-service airline. The later you come in, the chance you are going to an LCC business.

Yes, lower slot price will help, but you can’t prevent older players from running a complete international-domenstic transit hub at HND and taking away all slots. What you are saying is very true for direct traffic, but AS is very much connection dependent simulation, at least at the current stage. My point is you will never have a situation in real life where an airline is taking away all the slots at the airport with all narrow bodies because when there is a slot issue at the airport, the airline will very likely upgrade their planes, which just simply doesn’t happen as often as it should be in AS.

36

For connecting traffic the difference is even smaller. Whether a passenger from Los Angeles to Okinawa flies LAX-NRT-OKA or LAX-HND-OKA, the flight time is as close to the same. So, any difference between the two is based on duration of the connection, which I would assume is the same, and tickets prices which could be a little lower for the NRT connection as slot/landing prices there are lower as I suggested. So, I don’t see why HND would then be the preferred airport for creating a hub.

37

No, it will because new players will all start with direct and local traffic, so they will start at HND, make it large enough and then becomes an international hub further on. I don’t see too many people deciding to start at one airport and then just give it up to build a full connection hub in another airport.

That’s the reason you want to make domestic airport close at the city domestic so you force players to build an international hub at the airport that is further away so that is not taking more limited slots at the airports with better direct demand while they don’t have to use the preferred domestic airport in the first place.

One option is you could make the HND transfer impossible between intl-domestic and with a very high landing cost for international flights. but my fear is older players can still take out the limited slots and running high priced intl flights at the airports, in which most domestic-only airport, not HND but like ITM or even worse like NKM full very fast. NKM is much closer to the city compared to NGO but after NGO opened in real life all intl and even most of domestic flights are moved from there to NGO, so in-game it has only 2 slots compared to NGO which is 7. actually this will becomes a huge issue just when you switch from airport based demand to location based demand

38

Thanks all for the responses.
To respond to some things

TWAAir post 33

actually not really. you are not operating any other transit, but you should be competing with it, or i am not sure how do you even call it a ‘realistic’ airline simulation in the first place because that’s nowhere close to reality

The current demand data is already based on passengers whom chose to fly rather than take other transit. If we want to actively compete against other forms of transport, then the passenger numbers should be increased by the number of people already using trains and cars and ferries and bikes and whatever else in real life.

If we generate a new dataset with no “realistic basis” (such as the one Jan proposes) then sure, we need to take this into account. But the proportion of pax that fly vs other forms will have to be varied per region… in the EU/Japan/the “Global North” transit is much more appealing vs Colombia, where it really is “drive or fly or subject yourself to the ride of your life in some dingy chiva”.

And is the only reason we hate this dataset due to the rigidity of existing schedules? I don’t even see the argument here, people have made hubs in unconventional locations since the dawn of AS and it has worked despite no OD demand. The specific association of demands on a p2p basis continues to be “similar” regardless of if TAM had decided to route all its European through FOR. People still want to go from CWB to LIS - they just will connect in FOR instead of GRU, or BSB, or GIG, or SSA for all I care - and this demand and this dilemma is already, at least in the markets I know, accounted into.
In Limatambo, we have a person is building a neat little hub in Nuremberg to bypass the busy airports in Bavaria, and one whom I’ve seen set up in Lexington (almost) every temp server since I began paying attention to them. A domestic hub in CIA (which I run!) works despite only FR and W6 flying in there, in many cases not domestically. Similarly, Halifax for TATL vs YYJ. Changing the hub and the routes work within this existing demandset. I see no need for a change, unless they keep forecasting this Corona won’t go away… and in a few years the pre-COVID data will become stale.

Jan, post 31

Now, a few examples, how the combination of the factors mentioned above will work in practice:
RHO/KWI/TLL/LUA/INN

I cannot help by pointing out this is a very European model. I play in Europe in Lima, this model makes sense there, but as someone whom is not from there and knows some other markets, I can’t see this working in some other markets outside the developed world. I know you just mentioned Lukla, but you mention it’s really an outlier… what about somewhere like Pasto, Colombia? High, pretty decently hot temperatures even in summer, gets chilly in winter, but no one actually goes skiing there…

maybe that is something that could be made even more precise (for example to take into account some basic terminal capacities at airports, so it wouldn´t be actually possible to have 60 flights an hours on an airport like BRQ, which has in real life a terminal that can handle max 3-4 flights/hour,

Yeah, that’s the issue with basing slots solely on runway length and (HS/LS) connections. I’m not exactly sure of the exact math but I know terminal capacity doesn’t matter shit. It’d be nice as well, if certain capacity was limited to remote stand / gate, as not every gate at an airport will have jetbridges. Sure, some planes won’t need it, but you can most certainly do that, adds a new “realistic” twist.

Brno Airport, to those wondering, looks like this, the yellow is the terminal (with just a few remote stand gates, though it does say capacity of 1k pax/hr) and the pink is the “apron”.
Hugely oversized runway, so slots seem way bigger than they are.

image
image1562×647 65 KB

Back to TWAAir

i don’t see the airport being full because more airports are needed; i see it as a result of over-frequent flights. to make the slot available, adding airports close by only delays the issue (for the long term game world). The true solution is so that you should not be able to run an average of the cities per 10/15 mins. I think it is reasonable to add a closer airport so that demand can be accessed, but i don’t think it is ideal to have people forced to use a disadvantaged airport in the first place.

I think this is a valid complaint, but unless there are some limits to the expansions of the old airlines in certain airports, there is no way to address this. There is a reason those airports fill up - they are filled in the real world (to a certain degree ofc) and they are impossible to get into for new players. Why can’t a startup airline just choose heathrow? It’s full to the brim. The only true way to wedge those is with antitrust stuff but I don’t see this in AS unless it is some sort of “curb” to expansions. I know it’s morally wrong to push the new kid to the sidelines, but it’s what happens in RL.
Plus, in most cases new players should not be in the USA or EU :wink: And I don’t say that out of reserving those in anyway, it’s that they just get so competitive new competition grows slowly and new players get engulfed as soon as low season hits… and why the “new US airlines” (think Breeze which isn’t an LCC or if you want LCC Avelo) don’t operate out of the major airports, same with EU since god knows how many years, and in Brazil too, no new airline has a realistic chance to get into Congonhas… and look what happened when they tried to pry into a market dominated by giants for decades. (First link is from Brazilian media in PT but has some videos (English translation of that article here), other links in English)

So in theory, is it really not realistic (regardless of morality) for new players to have chances at the big airports?

Yes, lower slot price will help, but you can’t prevent older players from running a complete international-domenstic transit hub at HND and taking away all slots. What you are saying is very true for direct traffic, but AS is very much connection dependent simulation, at least at the current stage. My point is you will never have a situation in real life where an airline is taking away all the slots at the airport with all narrow bodies because when there is a slot issue at the airport, the airline will very likely upgrade their planes, which just simply doesn’t happen as often as it should be in AS.

Maybe increasing the max connection time will help here? to something like 12h? It will produce some nastier itineraries, sure, but it would allow for an airline that serves an airport 3xD with A320 to change to 2xD 757/HD767 and still be able to hit 24h’s worth of connections.

TWAAir #37

I don’t see too many people deciding to start at one airport and then just give it up to build a full connection hub in another airport.

Agreed here. At the start, OD demand is what matters. Nearly all players start with small planes and build SH for a reason. In most cases, they tend to wait until certain deps until launching LH and by then the airport is pretty locked (due to comp.) so there tends to be overwhelming SH

Let me give an ex
Say airport XYZ has a strong SH area around and has 10000 slots total.

Airline 1 decides to start with some efficient aircraft (295s for example) and expand like boom
Airline 2 and 3 grow but just a tad bit slower due to using less efficient and harder to fill aircraft (A321ceo for ex)
Airline 4 is not sure what to do, and grows slowly but surely.
Airline 5 decides to use up their $ and buy a 787-8 to launch a LH dest.

Airlines 1 2 and 3 will start with about 75, maybe 100 weekly flights
Airline 4 starts with maybe 50-40
Airline 5 with 7

Then due to natural (exponential) rate of growth, after two months maybe airline 5 grows to about 28 deps, but airline 1 is already at about 450, airlines 2-3 not far behind at say 400 each… Airline 4 has about maybe 150
So then we have 450 + 400 + 400 + 150 + 28 = around 1450 slots (14.5%)
Count in about 50% more slots from foreign/non-hubbed carriers in total (so about 7% more total) so you’re down a quarter of the slots

After 2 months more, airline 5 grows to about 1250, airline 2 adds long haul and grows a tad bit slower at about 750, airline 3 stuck with SH and is about 900, airline 4 at about 350, airline 5 at maybe 125, as the LH exponential growth curve kicks in
1250 + 900 + 750 + 350 + 125 = 3375
Add about 50% more (which sounds high, it is but this is worst case scenario) from nonhub carriers, 3375*150% = half slots gone

After two months more, airline 3 moves to other hub but continues to add deps slowly and breaks 1000 deps, airline 2 sees the LH curve kick in and grows to about 1500, airline 1 grows to 2.5k (2k is oftentimes the point when carriers add LH), airline 4 is hitting upper 500s, airline 5 grows to about 300 deps. At this point airline 6, other domestic carrier in diff major market, splashes a few aircraft and adds 750 deps more… Airline 4 doesn’t last long as low demand enters.
2500 + 1500 + 1000 + 300 + 750 = 6050 slots
Add the 40% of foreign deps (as we take airline 6 into hometown consideration, now) = 8470 (15% slots). And that is only with two airlines whom has LH.

Then it’s a scramble of who gets the rest. In the best case, about 15% slots are long haul, but 85% of slots are still short haul.
Unless we somehow change how airlines grow (which I don’t see happening) then this phenomena keeps happening.

Restricting int’l to certain airports, will mean that either
A. Players will ignore the city airport entirely and operate SH for feeder purpose in the future, from the int’l airport (concentration on one airport, fills faster)
B. People have to build twin hubs (inefficient, expensive)
C. People go back later to the city airport if necessary, after the int’l fills (either to supplement SH routes, or for ex, shift routes with high OD traffic to city airport to open up LH slots in int’l)
D. No Long Haul (airline(s) lose(s) major source of revenue)

None of A-D are great solutions. Add’ly AS does not check what routes are domestic in the “route criteria” when scheduling, so unless this capability is added then this will not work.

39

Agree. I think i am not trying to ask players to compete with any form of transport on any routes, but i think there is a huge issue on SH services: there are a lot of SH flights which are taking a lot of slots from the airport that is completely impossible in real life. i took a quick look at the newer long term server, limatambo, and i assume it already has all the implementation of the slots improvements over years, and i still see many flights between Peking-Tianjin, Shenzhen-HK, Nagoya-Tokyo/Osaka. I think this is an issue that would be better to be solved if somethings is trying to be done with the slot issue. I have no problem when player create connections flights for time-sensitive types of pax, but currently based on the mass of these flights i just don’t see this being the case.

I think it is great to have hubs in impossible locations in real world, which i do the same with my airlines. i don’t think making ultra SH traffic with some alternative transit option will change that though. To be fair i have no problem with the current database in general, i actually prefer some changes on the subjects i mentioned and keep the current database compared to doing the distance based demand system because i think you will only get into more work and more troubles when applying that model.

i know someone will point that out at one point :laughing:

My issues with the current system are these:

  • it is very uncommon in real word to have no competition in one large market. Actually i don’t see AS has an issue on this when i joined in 2013 on kaitak. You have multiple carries in one market and there are like three carries in japan, more than 6 carriers in Europe, several airlines in US etc. I have a competitor in Indonesia. but over time, people leave and now you have one in Japan, soon only two in US, while you still have the same demand compared to before. What you end up with is one huge airline that can take out any new player very fast. so i think we do need a little space for the new players in general.
  • because AS usually ends up with a small amount but large size players in the market, they will spill over flights to all airports with demands in the market. I can spend money and build up huge connection hub in a low demand airport and therefore even reduce the chance for any new players entering into the market. That makes the AS more divert from the reality
  • There are airports that are very congested, but i think you usually still have more than one airline taking away all the slots assigned by the authority. This is not really simulated in the game and what usually happens in AS is you have one carry take out all the limited slots in these airports, which i don’t think will happen in real life.

so i guess the answer is we do need some kind antitrust in game (same as reality), and maybe some additional ones due to the nature of the AS.

Good point. but i think it will make unless you make the connection time have very low sensitivity in booking, which is not really true? I think another issue is players will try to take out most of the slots available by nature to secure the market, so i think you might still need to limit the max slots that can be obtained per player to make this work. I think you might still be able to accomplish this by setting at the high number, e.g., 85-90%. I think there are game worlds with longer connections time? i am not sure how good they work out on slots eventually.

Good point. unless you keep a limit on the percentage of intl flights on intl airports, this will happen. Actually, when you do this you might better have a distance-based demand system instead of the current system we have .

actually i don’t see this as an issue. Most of the airlines operating under the domestic airport policy deal with this issue. So this is very much real.

This will be the reverse of the real-world growth. but i think if you consider the end results, you actually have a system much closer to reality.

i think there are a lot of airlines operating without international long haul in RL, actually probably the majority. I think players will survive with this also, though the airline won’t be as larger as they wanted/dreamed. If you are talking about the domestic long haul, i think the current system has enough domestic demand for this and if everyone is doing domestic SH, they might be forced to do domestic LH anyway because of the reduction in demand due to over competition.

1 Like
40

I wonder if there’s a plan to give players more insight into where are the pax connecting. Right now you can see that a flight has X connecting passengers but you have no idea where are they connecting - thus if it makes sense to, for example, keep certain destinations on the same wave. The same applies to interlining - if I have multiple interlining partners at the same airport, I don’t know which of the interlining agreements makes sense to me and which doesn’t.

I understand it was an issue to store all those data in 2000. But with data storage being much cheaper in 2022, perhaps this could finally be brought in for the next-gen AS?