Connecting Flight Prices and a More Realistic Game

The problem of connecting flight prices has been brought up a few times in the forum, and I’m sure it’s something most players have thought about from time to time. I know there have already been a few suggestions as to how to make these prices more realistic, but I’d like to make a suggestion of my own that consists of two primary changes.

#1: Automatic Price Ceilings – The ORS already takes into account how connecting affects flight rating. A direct flight has a much higher rating ceiling (100), then a connecting flight (from what I’ve seen 65?). I think it would also make sense if a similar system could apply to pricing. In other words, each player can control the price of individual routes flown by their aircraft. However, when the ORS compiles these routes to make connecting flights, they would apply a price ceiling proportional to the overall flight rating (taking into account things such as price performance of each leg, connecting time, number of stops etc).

For example, in one game world, a flight from CGK to BKK costs a Y passenger AS$ 206. Alternatively, the same flight, connecting through SIN costs AS$ 279. However, the first flight has a rating of 83, and the second has a rating of 38. Let’s say, for example, that in this new system only a max connecting flight rating of 65 allows one to price a flight at 100% of the price of the two legs. In this simplified system, the price of this connecting flight would automatically be priced at $AS 160.

In order to prevent similarly unrealistically low prices I think it might also make sense to induce some sort of price floor, as well as not to have the rating and prices be proportionally linear but something more along the lines of a curve where the largest impact is in the middle ranges.

#2: Connection Caps — Anyone who has read this far has probably thought of the obvious problem with this approach. Taking our previous scenario, let’s say the game world has the demand calculation for CGK before the calculation for SIN. What would happen is that both my CGK-SIN and my SIN-BKK router would be filled with low paying connecting customers, significantly hampering my ability to turn a profit on those routes. What I suggest is allowing each player to set a connecting cap on each route, just like a player can set prices for each route. This means that on the inventory screen, I can cap my CGK-SIN connecting passengers at any amount I want.

Playing out this scenario, let’s assume I have direct demand from CGK-SIN and from SIN-BKK so that I can fill 75% of my capacity on that route. (In order to find this out one would have to do some trial and error). Once I know this I’ll cap my connecting capacity at 25% so that I primarily fill up all of my flights with full-fare paying direct customers and only use connections to fill out my load factor.

Benefits

If for some strange reason, you’ve read this far, I want to lay out a few cool benefits of this change, aside from simply more realistic representation of actual airline prices, without having to resort to a complex system of classes like Airline Sim attempted to do a while back, but never implemented.

#1: Varying Strategies — At this point most experienced players have similar strategies: use a hub and spoke model and peaking to facilitate waves of connections. While it is certainly difficult to do well, this strategy lends itself to certain homogeneity and doesn’t reflect the reality of the world of commercial aviation. Specifically in recent years, the trend has been away from hub and spoke models and toward more point to point offerings. Airline Sim, however, has been notoriously unfriendly to point to point operations. This model might change things.

With the clearly decreasing value of connecting flight, players will have to make serious strategic decisions when choosing to adopt a hub and spoke or point to point approach. The former will likely lead to higher demand, but lower margins, while the latter will have less demand to play with, but higher profits if hitting the right markets.

Inevitably, the best players will find some good mix of the two, more accurately mimicking most real airlines which brings me to benefit #2.

#2: Low Cost Airlines — This to me is perhaps the most exciting possible result of this change in price model. In the real world, low cost airlines such as RyanAir in Europe, or JetBlue in the US, don’t have traditional hubs, and don’t normally attract connecting passengers, unless of course they are self-hubbing (more on that later). Instead, these airlines rely on the strategy that they can attract customers, even in smaller markets, by flying direct flights between destinations at a much lower cost. In essence, they are trying to steal the passengers that might otherwise connect through a mega-hub like Frankfurt or Istanbul, by offering them direct routes that don’t otherwise exist. Alternatively, they will compete with the major airlines on routes they do fly direct, but do so to smaller airports with much lower fees. In both cases, low cost airlines can fly at lower load factors because of the lower costs associated with their flights, and therefore are not as reliant on hubbing as the major airlines.

In AirlineSim this type of airline essentially doesn’t exist because low cost means low rating, means you are low on the ORS and can be easily beaten for demand on your route by competitors. On the other hand, having a connecting flight on a route (which means inherently low ratings), is not as dangerous since one doesn’t need to fill up the whole flight in order to be profitable. A good hub can merely skim the bottom of the ratings from each route and fill up all its flights.

With this new model, there will be a drastic shift in pricing. Pricing routes as they are will no longer be profitable for traditional hub and spoke airlines. By necessity, hub and spoke airlines will need to raise their prices, in order to recoup more from their connecting passengers (this will also most likely necessitate an adjustment in the impact of pricing on the ORS, so that raising prices of individual legs – to a point – does not impact the ORS calculated price proportion in such a way that I actually lose money. Applying a good curve, as mentioned above, also helps deal with this problem). By doing this, they can still operate with high connecting caps (see above), rely less on direct demand, and essentially operate as they were before. What results is a very accurate reflection of the real world: it is very expensive to pay for a direct flight on a major airline, but much more manageable to connect on those same airlines.

With the increase in price by hub and spoke airlines, their direct ratings will necessarily fall. What emerges is that a low cost point to point airline can come in with mediocre scores across the board (cram in passengers, offer few services etc.) and a price point which compares favorably to major airlines (since they’re not catering to connecting passengers anyway).

A possible result is that one would find the ORS would rank flights in the following order for Y passengers: Low-cost point to point, and then either hub and spoke direct, or hub and spoke connecting depending on the ratings and price of each. The radical notion here is that a connecting flight might actually jump over a direct flight due to the nature of each flight’s prices. In other words, the ORS would need to re-calculate the final price performance ratio after applying the proportional price (see above) for a connecting flight, and only then rank the flights available. This would result in only hub and spoke airlines with pristine ratings competing in the demand market with low cost airlines. It would also preserve the game’s emphasis on connections, albeit in a more realistic way.

Almost everything in this section applies to economy passengers only. As in the real world, and as is reflected by the current ORS, business and first class passengers are much more greatly affected by low ratings in other categories, and less affected by price. As such, these passengers would almost always have direct hub and spoke flights first, and low cost airlines last. This would lead to another realistic reflection of most low-cost airlines being economy only. Additionally, as is reflected in the current ORS, the longer the flight, the greater low scores in non-price categories matter. This would also realistically reflect the lack of low-cost long-haul airlines.

Given the previous limitation, and the realities of traffic rights, low-cost airlines would be limited to countries with high domestic (or European) demand, or with high short-haul direct international demand. This would also reflect reality in terms of the home countries of low cost airlines. (A question to explore would be how to recreate airlines like AirAsia which scoot traffic-rights by incorporating subsidiaries in other countries in partner with National companies. I know schemes like this existed in the IPO version of the game, but perhaps even without IPO’s players can set up multinational joint-ventures. A simpler way is just the Domination rules of allowing foreign investment across the board).

Another interesting benefit of low-cost airlines is how they would work in concert with ground networks. In the current ORS, ground-networks significantly reduce the rating of a flight, even if there’s only one actual flight, the ground network segment is considered a flight for the purposes of ratings. In other words, while DMK may be able to draw demand from BKK on a flight to CGK, via ground networks, it will do so at the cost of a big dip in ratings. However, it’s possible that the significant boost given by the lower price of low-cost airlines as compared to hub and spoke airlines, may even allow real-life low cost hubs like DMK, or Hahn, or Bergamo, to compete with the bigger Airports for their own customers.

The interesting thing here, and one that leaves me with a few doubts, is how to scale a low cost airline in AirlineSim with security. I have a few ideas, but none is well-tested. The first is that as with real low-cost airlines, the larger the network the lower they can keep their margins and still grow, and therefore their prices. Even though having a large-hub with many connections provides perhaps an easier path to security as an airline scales, it’s possible that the very price point that a low-cost airline can operate at can be security in and of itself. Another way to boost this (and perhaps just as necessary as change in price modeling to realistically create low-cost airlines) is by charging for landing slots (perhaps maybe above a certain frequency to continue to allow the trial and error nature of research in Airline Sim). This raises the bar for entry for competitors and rewards the current holder of a point-to-point route with some security, while not totally preventing the possibility of growth and change. Perhaps landing slots (which should of course vary by airport as in real life) may also increase in price as the airport becomes more full, further providing security to point to point airlines, while still allowing newcomers to enter the market through creative means.

#3 Interlining — Another real-life phenomenon currently reflected in Airline Sim is interlining, or that one can book a connecting flight in which different legs are operated by different airlines. I think that this change in price model will allow for a more dynamic approach to interlining. Instead of restricting interlining flights to the connecting price ceiling model, interlining companies would be able negotiate to increase or decrease their ORS calculated price up to a certain percentage.

Again using our earlier example, if an airline based in Indonesia and an airline based in Thailand, wanted to interline in order to connect their mutual flights to Singapore, they could decrease their overall cost to undercut and steal demand from an airline already hubbing at Singapore.

Alternatively if two airlines can connect to serve under utilized markets, like in real life, they can continue to maintain high connecting costs. For example, if a passenger were to fly from Ulaanbaatar, Mongolia to Lagos, Nigeria they would still have to pay approximately 15,000 dollars despite making 3 stops. If, however, one booked each leg separately it would come out to less than 4,000 $, meaning interlining flights should be able to raise their prices above 100% of each leg. The idea of booking flights separately brings me to my final point.

#4 Self-Hubbing – In addition to an adjusted interlining model, I believe this change in price model should allow for a new feature known as self-hubbing. Essentially, self-hubbing would recreate connecting flights between interlining airlines as they currently exist in Airline Sim. In other words, self-hubbing would allow passengers with no alternative to fly on different airlines at 100% of each leg’s cost. This is a little complicated, so bear with me.

In this new system, each player would have the option to turn on self-hubbing at any station to which they operate flights. Again using our earlier example, say I operate flights from CGK to SIN, I could turn on self-hubbing at my SIN station.

Turning on self-hubbing would allow my flights at that specific station to connect with any other players’ flights at that specific station, even without interlining. The only catch is that both players must have self-hubbing for that particular station. This can be a coordinated effort, or simply unilateral moves operated in parallel.

Furthering the previous example, if my airline out of CGK turns on self-hubbing for SIN, and another player’s airline out of BKK turns on self-hubbing for SIN, and our flights are within the connecting window of each other, the ORS can connect passengers on our two airlines. These passengers would a) pay full price for both legs and b) bypass any connecting caps as the system would consider them to be direct passengers. This means that there would be (as in real life), no real way of recording whether these passengers are actual direct passengers, or simply the result of self-hubbing. As far as each player is concerned they are direct, but as far as the ORS is concerned, they are connecting. This essentially allows the ORS to function like a flight aggregator such as SkyScanner or Expedia.

The downside to self-hubbing would be the obviously low ratings. In fact, self-hubbing would most likely provide little demand when almost any other alternative exists. It would still help in two very important areas: 1. Start of game – Self-hubbing could facilitate more connections and help players to grow faster despite rudimentary networks. 2. Small markets – A player could develop a fairly sound strategy of flying small market passengers to larger hubs where more routes exist through self-hubbing. This would be a spin on low-cost airlines (and I believe a minor strategy of some existing low-cost airlines), and allow players to squeeze demand out of small markets through point to point flights.

Self hubbing would also help low-cost airlines squeeze just a little bit extra demand out of their markets.

The one question remains in terms of public transport, in order for this concept to work, self-hubbing would need to give a rating which is high enough that compels passengers to travel rather than stay home. Given that self-hubbing aims simply to recreate the current state of connecting flights in the ORS, this shouldn’t be much of a problem, but is still something to consider, especially given the probably-needed adjustment of price impact on ratings as mentioned above.

6 Likes

Hello Sam, and welcome to the forums.

I have had a great read with your points.
I’d like to go over your points, individually. Because I believe that most is all great theoretically, but my biggest thing (and the TLDR) is that I do not see the implementation of any of this with existing ORS code.

#1 - Automatic Price Ceilings
So what I am getting at, is that connecting prices are always cheaper than two O+D flights. This is not always the case, and I’ve done plenty of self connecting to save a few hundred USD. Let me give you an example IRL so you see my point.

Let us think of an example, Las Vegas McCarran - Miami Int’l - San Juan LMM, PR - this pair is realistic due to a lot of Puertorrican seasonal workers in the casinos of Vegas. To set this let’s do March 17 next year. Prices of course as of time of pulling SS.
I know there are countless factors to how a flight is priced, but just to illustrate my point.

On the LASMIA sector, we can see these fares. Settings are meant so fares are comparable, although you know which one has a “better service” (arguable with AA at this point but oh well)

Then on the MIASJU:


AA has some more at 119 (incl the 1057peak flights at 144 and the NK is at 321.

Sure, the connection in MIA is eternal but it counts for the Frontier.
The F9 is actually so long in MIA that it only books a two stop via ORD, at $130.

Then for the connecting one NK shoots up to $641 w/ 20h
And AA is the only clear winner for true potential of conx, possibly due to wave times.
It is $349 for most conx options - the segments at 208 + 119 are less than that.

What I meant to illustrate, is that connections aren’t (and in some cases like you say, when the flights are supposed to rely on OD) encouraged, they are shot up more.
So not the 100 of now, and not always cheaper.

I agree with flexibility of this connecting price, yes - but the way in which it is done should be more flexible and less capped. But then again, to know which connection pairs to stimulate we need to know how do pax connect, which is still tight lipped secrets.
Additionally, if we were to simplify a blanket curve # will not work for every airline…
And what if a pair is not specified for a connection curve? Will it then default to 100% and lose to others who have? Or an airline does not pick it (maybe out of innocence?)
With no tutorial (yet) this falls to exploitation by experienced users / ones who read forums, which can also mean not every player might know (as the game is translated into languages which might not be those spoken on official communication or the actually used forums…)

#2 - Connection Caps - this would really have to wait for a mechanism like booking classes (if that will ever come out).
I like to point people to this, I’m not too sure why, maybe because it’s the only true ‘future thing’ we have to reference. This is LEJ 2016: AirlineSim Convention LEJ 2016 Presentation ENGLISH - YouTube. It’s in german too, if any of you prefer.
This goes over well into what the spirit of the booking class system is/was.

Well, at the point when your airline grows big, who will bother with such micromanagement? If suddenly demand to a certain city pair dies down (say, what might happen to Addis Ababa with the current stuff going on in Ethiopia), and it no longer can accomodate so many OD, or maybe a new entrant enters a market which might then turn oversaturated? This kind of system of adjusting capacities and micro-scale assumes that certain things which are not (and should not) be constant are constant. This is why it’s people’s entire jobs to evaluate pairs for gauge and connection things, and why computers do a lot of it now…
I don’t see how such adjustments to capacities are to be done without the implementation of dynamic cabins and multiple prices/fares per flight.

To the Benefits thing

I don’t really get #1. Sure, AS is a lot of transfer dependent, but so are most of legacy airlines. The LCC thing as you point at later is not what most models in AS try to replicate. So the game is not truly coded for such.

#2
B6 isn’t an LCC, and I hope I’m not the only one that thinks this way.

The AS demand is mostly based on existing pairs between city pairs (with some ‘general’ demand sprinkled around). Therefore, the model of creating spokes, and creating demand is not very feasible. You are chasing demand here in AS (like legacy carriers often do), instead of incentivizing and creating it. This creation, in current and probably capabilities of the AS code, is not really possible, unless we can somehow create advanced demand (like Antalya is a beach destination and cold cities like the beach for holidays). But this means that demand itself has to be majorly redone. And expansion is already planned, to shift demand to cities instead of airports. This implementation when paired with current sources of AS data which are neatly sorted to airport, is a mammoth task. This shift will allow such LCC hubs as you mention to really take off. But for ex as long as the demand is in Cairo International, not a lot will shift to 6th of October unless the demand is taken into the whole Cairo metropolis. Then it comes into question what airports do and do not make up certain metro areas… which means combing through every single AS airport and sorting. A mammoth task.

Then, ancillaries and such have to be a thing to offer such low fares, which is somewhat easier to implement than the other ideas on the post, but it stands as a barrier nevertheless.


As you say with Air Asia, I am a fierce advocate of redoing the foreign investment list.


Economies of scale should be a thing. But it isn’t. So an airline like WN can’t pay less than half for MAXes in AS - so their costs in aircraft acquisition double. Further decreasing margins / disallowing for cheaper fares.
Same with maintenance - loyalty to provider, scale and consistency of schedule are not rewarded whatsoever, and commonality only barely.


Slots should be explored at. I agree. Some limits should be in place and some money should be used. The downside is smaller airlines have no chance to get into the expensive airports, which might discourage certain players who do not understand slot usages IRL, get frustrated, and abandon AS. So some downside to look at, but some exploration of the topic does not hurt and I recommend it.


Interlining should be expanded, particularly among alliances. Alliances at the moment are nothing more than a private forum and a cute 23x23 logo. This needs to change.
Deeper JV-style partnerships should be looked at as a possibility, as maybe the connection between alliance members would trump a connection with only IL agreement.


Loyalty should be looked at too. I think maybe if an airline has served a city pair for a long time, more pax will gravitate to booking their flights.
This will unfortunately discourage newcomers but also incentivize giants to keep running a loss-making route to keep customers happy. This should have impacts on image, as well.


A major issue with #3 Interlining is the cap at 3 segments. This should be expanded if technologically possible. Because not all combos are three segments nor they should be, in my opinion - this only encourages mega hubs like serving every S.Am. 3-bar and up and every European 3 bar and up from LPA or something. as they are able to get a ton more options for three segments including GN…

GN should also be expanded, to include specific methods and reflect accurate transports. Trains should be where trains are, and ferries as well - for ex. there are plenty of famous inter-Canary ferries but no connections in AS - whereas some islands with no links have GN?
This would involve looking at every airport again, but might be a good long term investment.


Completely disagree with your Self-Connect service here. I’m not the most aggressive player but this just opens up floodgates to giving pax to your competition for free.
Plus, which airline economically benefits from this? Not the own airline, due to the staff still needed to run self connects… maybe the partners, but surely it is better to pay a couple thousand dollars a week to ensure you give pax to your trusted partners vs putting them in the wild wild west?
I don’t see how it helps connect smaller spokes.
And I don’t see how this is implemented into ORS without expansion of segment limit.

And finally, remember there are already features in the works, features planned since before 2015, and this game is the twenty-year love affair of one man. And one developer. Surely completely upending his life project and redoing code from the 2000s will take years and probably help.

2 Likes

First of all, really appreciate your response and analysis. One of the things that got me into this game in the first place was the potential for discussions like this, and thinking broadly about both real-world and in-game strategic and logistical questions. On that note, I’d like to respond to a few things, not because I really disagree with anything you said, mostly cause I’d like to continue this discussion cause I find it really interesting.

#1 Automatic Price Ceilings
All your points are well taken and I definitely know that sometimes the prices of individual legs can cost less than the overall connecting prices. I even demonstrated with a much more extreme example in my UBN-LOS route. I guess the question is how to find a balance between something that works for AS and more closely reflects the reality of connection prices being more dynamic.

I do think there is something important to point out. Connecting flights between point A and point B are sometimes (although not always), cheaper than direct flights between the same two points. As you pointed out, they are not always cheaper than the joint cost of direct flights between A and C, and C and B. As of now, in AS a connecting flight from A to B through C, always = AC + CB. In real life there is much more variance.

As you also mentioned, there can be many reasons behind the prices for these flights. Like you said, a connecting flight through a major hub which takes advantage of peaking, can be much cheaper. It’s also true (from what I’ve seen) that in many cases connecting flights are only cheaper than a lower-cost (or even major) airline flying direct on long-haul flights (where LCC’s are sparse). They also help when there’s no direct flight at all.

I did a little more research on the LASSJU route and found some interesting things.

AA prefers to route PAX through its DFW hub in this case, rather than MIA. In fact, the connecting price through DFW is cheaper than the two legs through MIA. The connection through DFW is also a lot cheaper than the sum of its two legs (which comes out to just under 400). I would imagine this is because AA has a much easier time filling its LAS-MIA and MIA-SJU on direct demand (for seemingly obvious reasons). Like you said, there are plenty of people (and computers) whose jobs are to figure out these things, and incorporating this kind of market research into AS just wouldn’t really make sense.

Given all of that, I’d like to make a sweeping generalization. These phenomena essentially all come down to two factors: demand and route network. The oversimplified purpose of an airline – and especially in AS (jumping ahead to your cool take on creating v. chasing demand) is to find demand and move it efficiently along a route network. For most PAX flying from LAS to SJU, they don’t particularly care if they connect through DFW or MIA (the actual difference in flight times is fairly small). AA would rather they connect through DFW, and so that flight is cheaper.

I believe that AS with this change in price model might actually be able to replicate this phenomenon. Applying this real-life example to AS, a player with a hub in DFW and flights to LAS and SJU will probably have less demand on at least the SJU route, than a player with a hub in MIA with the same routes. The MIA-based player will probably get sufficient direct demand on his SJU route not to overly rely on connections. Even if they do use connections, they may be able to keep the price for SJU-MIA high due to the overwhelming demand. As such, the ORS will route PAX through DFW before MIA.

However, this doesn’t really directly solve your problem of connecting flights being occasionally more than the individual legs without route-specific curves. For that I have a suggestion as how to complicate my oversimplified proposal, which brings me to response #2.

#2 Connection Caps
I know you don’t seem to believe in this idea very much and I agree that a more elaborate booking class system would be more realistic, if not overly complicated. In light of the difficulty of implementing such a system, I’d like to offer a revised suggestion which fits somewhere between the crude simplicity of my suggestion and the complicated headache of booking classes.

Let’s say in this new system every route starts with a default connecting cap of 25%. At 25% the proportionality curve works in its default setting. For simplicity’s sake let’s say then that a 40 connecting rating will get an 80% proportional price. In other words, while a direct passenger will pay $100 for that leg, a connecting passenger will only pay $80. At this connecting cap, the flight can fill up with no more than 25% connecting passengers paying discounted fares, but can still fill up with 100% direct demand passengers paying full fare. Meaning this is a one-way cap, it doesn’t cap direct passengers in any way.

Once a player gets going, settles into their routes and starts to scope out the demand they can start to change this connecting cap. In this scenario, unlike in my previous proposal, the connecting cap will not only impact the number of potential connecting passengers, but also perhaps the price. As the player increases connecting capacity on that leg (in order to increase load-factors) the proportional price will fall, in keeping with the laws of supply and demand. Similarly, if the player decreases the connecting capacity, the price will rise. Perhaps even if the cap is under a certain percent, say 5 or 10, the connecting price for this leg may even rise above the direct price for that leg.

What emerges is that connecting routes where each individual leg has relatively lower direct demand will price out lower than the cost of each individual leg. Conversely, connecting routes with higher direct demand will price out closer to, or perhaps in excess of, the price of each individual leg.

It is true that this gives players another element to micromanage, but in my eyes it seems no different than managing prices and maybe it could even be worked nto ASES.

I realize I already wrote a lot, so I’ll try to keep it relatively short in terms of the benefits.

LCC’s
As soon as I posted, I realized I should have used Spirit as an example instead of JetBlue. Totally not trying to create controversy, litigating whether JetBlue is an LCC as opposed to a ULCC, or just not in that category at all is totally not the point of this post, so I apologize.

I agree that city pairing, ancillaries and improvements to ground networks and the foreign investment would be necessary to make this a reality.

I also think that loyalty (both for PAX and vendors) a really cool idea, I’ve seen some other similar (worse) games experiment with the idea.

In terms of PAX, I like the idea of securing loyalty over time, and I also agree it should be heavily dependent on an airline’s image. However, I’m not sure it should be based on airport pair, but airport. For example, if I’m a loyal AA customer in the NY area, I’d prefer to fly any AA flight out of JFK over Delta, and not just JFK-LAX.

In AS, loyalty would then be calculated based on presence at a particular airport, how long the airline has been present and then ratings of those particular flights and image, in general. I also think things like delays and cancellations should factor into an airline’s loyalty value at a particular airport.

By calculating by airport and not route, it also allows for more LCC’s or just general point to point players who can have loyalty at a number of airports, while hub players will be more limited to their hubs.

In terms of aircraft, I also like the idea that a player can sign a delivery contract instead of leases for individual aircraft.

For example (rough numbers), instead of putting down a 3,000,000 deposit on an A320, I could put down 5,000,000 for a delivery contract for 10 A320’s delivered once weekly, or twice weekly, or something like that (maybe even with increasing frequency over time). Perhaps with each delivery I would add a smaller deposit, say 100,000, and I would still pay the same leasing rate. What results is that I tie up less money overall, at the expense of limiting my future buying strategy. In other words, I can grow faster, for cheaper, but must commit to a more homogeneous strategy for this length of time.

I also think this type of deal should be limited to smaller aircraft where one could (more?) reasonably see this type of production schedule. This would also help LCC’s who can sacrifice the need to satisfy a plethora of different markets and therefore require many aircraft (as is needed at a hub), for getting their planes fast and chesp.

IL and SH
I already broke my promise of keeping it short, so I’ll say that I have little to add to your suggestions about interlining, they’re all good.

In terms of self-hubbing the theory is (and correct me if I’m wrong) that airlines won’t need to add more staff to cater to these passengers. This provides them with no extra expense. The ORS does the aggregating, and the passengers are responsible for connecting themselves. This would benefit an airline that doesn’t have traffic rights to a certain route, and might want to connect up with another airline that flies to a joint market, but doesn’t want to pay to interline their whole network. It also helps generate demand for LCC’s. In essence they can work together to raise demand without sacrificing any profit.

Hey guys, thanks a lot for the elaborate discussion :slight_smile:

I’ll post something on the subject towards the end of the week.

4 Likes

I really like the discussion we’re having here, as well.
Some of these points you have made like you said I agree with, and some while I can see the benefits I really don’t see a true path to practicality. We can dream but then again we only live once, to quote that infamous saying.

In terms of the connections, yes. I agree that the example could have been better on my end. Dallas has less OD as you can say. Such routing also opens up the discussion for more hubs to be focused on OD (such as a Miami or a Congonhas) and some more on connections (like a Dallas, a Huntsville to use a common example, or a Brasília). This would mean that connection curves would not have to be universal, but rather a route or hub dependent thing. This would depend on natural integration of bunching flights into “hubs”, something which ASES supports to a smaller degree but of course not present in the Vanilla game. This would mean possibly a declaration such as “Airline has opened a new Hub at X.” which would be a “mega station” of sorts, with some benefit perhaps such as additional benefits in Connection / ORS encouraging conx itineraries through those vs just linking up random flights that match, such as a common example with AA in XNA - plenty of times it offers itineraries like Dallas - XNA - Chicago - somewhere instead of the one stop… Plus some benefits to boot.
These hubs should not be limited to country of HQ as international hubs and focus cities exist, and in AS too - I know one of my alliance mates runs some heavy focus on Tel Aviv despite being Turkish in registration. That would also solve hubs for treatise purposes as well, such as EU or Oceanic.

I agree that the existing sum for connections is insufficient. My proposal is for it to be able to go both ways to both encourage and discourage connection dependant on market and situation.

I support your Loyalty idea. Somehow making market share and loyalty rewarded. This would like I had pointed out decrease competition but would allow for more consistency which would possibly simplify the advanced ORS calculations the other suggested points would need to run properly.

I don’t agree with the loyalty only in hubs idea though. Passengers for example flying to Brunei would probably favor a Malaysian carrier offering many connections one stop via KUL vs a small local fleeting carrier which doesn’t really serve a lot.

As for your aircraft thing, you do not think the existing credit system (even if some aspects are reworked) should be sufficient for this?

I’m still not convinced by self hubbing in terms of practicality, but I acknowledge your stance on the matter as a valid one.

And you can edit posts after publishing for a certain while… :wink:

1 Like

I couldn’t read the end of the thread as I am heading to some IRL work too but I come as a player after a year and many airlines in AS, while studying airline management (including booking and fleet management) IRL.

My main concern coincides with what samwest7172 is poposing. I do not know yet how AS system under the hood works. I have noticed some IRL airlines are just impossible to replicate, and vice versa. I find it odd that LCC can’t make decent margins when IRL they literally are the money makers in the world nowadays.

I have one suggestion : even if say the reputation of the airline is of a lower quality proposal (whateve AS makes of it, the ORS rating, you know that better than me) there SHOULD be a way to propose ancillary revenue. It’s 2021, any company that doesn’t offer

  • extra baggage fare, or not (legacy vs LCC)
  • on board menu (to receive money, instead of spending on the offering which do not exist anymore)
  • it would be nice to be talking about wifi also in the offering
  • priority boarding / lounge / priority security check at the airport / pre-ordered meal / insurance deals
  • local deal with some companies: hotels, rental etc
  • local help from governments: they affect a LOT how an airline survives or not. National airlines get subventions, but private ones can get a contract with the military for standby tickets for the weekend of their soldiers for example.
  • fuel hedging would be nice as a strategical decision. I understand most worlds die after a year but it would be nice to see hedging contracts possible (with or without insurance to not loose too much if you bet bad on the price variation). It would create interactivity with the real world.
  • loyalty to plane brand and all brands overal: when you have 100 B737, you have a good relationship with Boeing in between “you’re stuck with them” and “they want you to order a batch of 50, 100 or 200 to make it look good” and the bigger the order the lower the price of the plane. There should be no need to have so much cash to receive many planes: IRL banks back you up and take a risk with you, you sign a MOE maybe, then later the bank allows the collateral which is debt for you on paper but you don’t have to pay it all per month, planes come when they come.

Just to give an example: my airline is a LFHV carrier (btw this is technically what JetBlue is too: Low Fare High Value) and offers simple pricing, no aggressive price on extra bags at the gate, however we offer wifi and an apparently (i do not use it myself) an industry leading fidelity program which works. We basically offer a mixed strategy, and we’re one of the few airlines which made money in Q3 this yea despite covid. And the reason is: during a flight, we can make 3000 euros of money off alcohol, food, duty free sales. Divided by head that’s sometimes more than 16 euros per seat. In Europe, where fighting for any euro of benefits is the game (most competitive market in the world). We also fly to airports which in AS would not fill a B737 but IRL they make money. Also because the local people have more income than in other areas, and local governments subsidies flights from the capital to the remote airports.

Btw some deeper concepts can be explored: Southwest albeit the most profitable airline for its investment in the world is not “technically” a “low cost” airline: they spend a LOT where typical LCCs cut corners. They don’t cut on salaries in fact they pay more than some majors. The difference: their crew fly the maximum legal time, 900h/year for pilots at least, while majors pilot sometimes fly half of this. That’s whe they cut down cost per flight yet have happy customer which brings a good image which makes the company liked by americans over, say, Spirit which goes the true LCC way, like Ryanair or Wizzair do too including some shady things in the backgound I do not wish to discuss publicly.

In AS, we can only “offer” everything and not “propose” a menu and no ancillary revenue is possible. If I copy my airline, it kinda works because its a LC"HV" so we do fly to nice destinations but some of my planes would be empty, I’d be working with a 10% margin or something like that. What works for now is only, and really only, hub and spoke airlines, which are actually and in reality dying, loosing money unless they merge or receive massive help from their country, reduce their fleet to a few types, and exist in the US or China only (in the middle east, they loose money but the country erases the numbers anyways). Or dying is a big word, but a shift is happening where major airlines are now loosing money because in the long run hubs will still exist but the size they have now is too big and they have way too many employees for operating the same number of flights than new generation airlines do. Self hubbing as described before should work better. Indeed IRL this comes from attractivity (scandinavia/UK → spain for example in europe, central europe to noth africa etc), but having several mini hubs of 5 to 10 aircraft can totally have full loads IRL while most of airlines in AS at the top of a world are still centered in one city in the EU, and a couple in US/China.

Heck, even amazon relies a lot on A to B flights in some areas in the world for their Prime quick delivery program. What they do even on their truck fleet is rely on many hubs (like a LCC) to always have a A to B option on top of having some bigger hubs (souce: I applied to work for them during covid and they explained it all to me)

Indeed I totally understand it’s a lot of work for one man if the coder is still working alone. But people are paying for the website it’s also fair to expect updates in its code and this is my (long) 2 cents from someone in the industry

1 Like

Hello Sedna, nice to see you again :slight_smile: I hope you are still doing well.
I find your valuable perspective interesting and I thank you for your contribution.

I don’t disagree with what you are saying, as with Sam. However I still stand by the skepticism in which this will be implemented to the AS system as is.

The problems with adding LCCs as I see them:

  • How do you have a computer calculate who buys a snack/drink/whatever? Do we set a fixed ratio, probability, etc? If someone is flying on a connecting one, how does one determine what segment the person buys a snack in - and therefore gives the ancillary to?

  • AS is based on fixed demands, and creation of demands on basic premises (eg cheap fares to the beach, even if that beach is the unknown Tivat, but someone prefers to go for 10$ to Tivat vs paying $100 for something like Antalya?) - how does the computer calculate the value of destinations. The game runs on fixed pairs - eg, X # of pax want to go from STR to AYT - not unfortunately X number of pax want a beach holiday starting in STR.

  • This would also have to sort demand and passengers into buckets - not just the basic demographic but also intent of journey and what they value. I’ve talked about this here but essentially sorting demand into certain intents - my basic idea was Leisure/VFR/Work but there are more types and more should be included - for example someone going to hitchhike in Tasmania will not value the same as the family going to Hawaii - but they’d both be “leisure” pax.
    (edit) This dilemma is also discussed from 10:35 onward here
    As demand is acquired in raw numbers (which I can prove as someone who has delved into these demand pools) on the pure OD it is impossible to sample what the intent is though. So the passenger then has to have one set of priorities to correspond with the congruency of the passenger base.
    If someone was heading from Houston IAH to San Francisco for a meeting in Downtown or someone wanting to do tourism in the Bay Area, they are both listed in data as IAH > SFO pax - and to get such data from the real world would prove impossible as in most cases it isn’t sampled (esp domestic), and in the ones that possibly could (Intent of travel, as in customs) - then I doubt any government security agency would be willing to give this data to AS much less the almost 200 members of IATA/ICAO’s security agencies. So that makes it hard. We can make generalizations (such as most pax going to AYT are tourists) but as seen above cases like NYC and SFO are more ambiguous to distinguish.
    Therefore if we only are allowed to have one congruent demand set we must settle for one set of satisfaction criteria. Hence why the ORS is weighted the way it is.

  • I think if we want to get apocalyptic, most of this industry as it is today is going to eventually die or transform. So such argument is hard to quantify

  • Additionally most demand sets would not be “normal” for a few years - even “recovered” countries have changed demand patterns such as a massive increase of demand to Bozeman in the USA and half of the SFO demand. So to predict post-pandemic things we must wait until we either accept the Post dataset as abridged from 2019 or stick with the Pre set until something happens…

  • Pilot workloads and stuff would require a major expansion of the role of the staff, and I personally believe that this should not be where our precious daily cents should be going toward…

  • The loyalty of manufacturer is something I believe is a realistic ask as the implementations for discounts in bulk are already in place in existing codebase. And something I support.

  • We don’t deal really with the function of the terminal - and to the core purpose of the game, this would only really be seen as a boost hidden somewhere in the ORS - sure, a nice thing to implement, but hardly a priority (impo)… unless AirportSim has any plans of making a comeback integrated into AS… (Martin? :rofl:)

  • Baggage - we’d have to mess with predetermined weights of pax to allow for baggage variations - might complicate calculations of MTOW and by proxy reduce capacities or invalidate some flights…

  • As someone who knows what it’s like to serve barren airports here, and knows people who have served arguably worse, “government assistance” would be a miracle to some little tracks…
    An idea: Maybe an Airlinesim Global Development fund is a branch of the AS Bank that offers bonds / payments / discounts based on conditional of serving select airports with X weekly flights - maybe these routes rotates randomly (to not have simulogics have to manually implement this) between certain one and two bar airports per year/6mo/3mo, to incentivize constant new routes… maybe “New bonus for serving (1-bar”)" and this would be advertised on some panel in the Dashboard.
    The selection of the spokes would pool the 1-2 bar airports per continent, and randomly draw only those airports without any Stations (as I imagine it will be easier for the machine to tell than flights actually operating). So it is truly a route with no service, vs a route that is already served and can be viably served without subsidy. And the payment would keep going for 1 or maybe 2 or 3 of the cycles until it is announced that it will stop, then the CEO has to decide whether it can maybe stand on its own (as connection links and such will be done) or not. Airports should have a “cooldown” so they arent selected too close to each other - maybe 4-5 cycles after. And the idea is 3-5 per continent running at a time.
    Of course, this is best paired with an expansion of the FI list to be able to serve more airports, for example if there is one in Bolivia but there is no airline with Bolivian rights, right now that would go unclaimed…
    The payments would probably be doled out with the weekly closing of the airline, as some sort of ‘refund’ on that cost - as that is the way I see it implemented most naturally, though it could also be added like a leasing income at any time into the airline’s timeline
    Payments would follow a full week (measurement is TBD, I propose either a fixed time like Mon0000U or the closing time) of continuous non-cancelled service…

I think that is it for now - but I’ve left you some plenty of ideas to ponder…

Bahaha, not going to happen :smiley:

1 Like

So now that the cat is out of the bag (see the blog or YouTube) I just want to point out how funny a coincidence it is that this discussion popped up just a few days before our announcement. An announcement which we’ve been preparing for two weeks now and the work for which started months ago (when we started working our the project idea and funding application) :slight_smile:

2 Likes

Hahaha, wow! Might be the record for fastest suggestion/discussion to implementation. Impeccable customer service. All I can say is can’t wait to see how things unfold, really excited!

1 Like