AirwaySim
Online Airline Management Simulation
Login
Username
Password
 
or login using:
 
My Account
Username:
E-mail:
Edit account
» Achievements
» Logout
Game Credits
Credit balance: 0 Cr
Buy credits
» Credit history
» Credits FAQ

Author Topic: Code sharing (and some solutions)  (Read 902 times)

Offline kimkujala

  • Members
  • Posts: 92
Code sharing (and some solutions)
« on: June 20, 2012, 05:19:20 PM »
Yes... I know that there have been maybe hundreds of topics about this but still. I think I may have solved the problem with code share flights. According to the old topics the problem has been how to deal with connecting flights as the demand and other stuff is so tricky to estimate.

To solve this problem, I have decided to forget this whole "connection flight/transit passenger" thing. This system will only make code share flights available. First of all, I'd like to introduce you an Algerian airline named Air Algeria. Next, I'd like you to meet Air Italy – an Italian airline. Air Algeria is based on DAAG (Algiers) and Air Italy is based on LIRF (Rome – Fiumicino).

So, let's get started. Air Algeria opens a new route between DAAG and LIRF. Later, Air Italy is thinking about to open the same route. The CEO of Air Italy notices that the demand on that route isn't very big and Air Algeria has already much capasity on that route.

Still, Air Italy wants to have flight between LIRF and DAAG. Luckily, there's an option to exercise code share on that route. Air Italy asks from the CEO of Air Algeria if it's possible to make an agreement for code sharing. Now – how this would work.

Air Algeria has a plane flying between DAAG and LIRF with capasity of let's say 150 pax. The plane has about 110 passengers on every flight. There are 40 free seats. Okay, Air Algeria wants to sell these seats to Air Italy. E.g. Air Italy decides to buy 20 seats from Air Algeria. Air Algeria decides the price for the seats. In this example let's say 120 USD / seat. Now Air Italy has a flight between LIRF and DAAG. Air Italy will have a same kind of route window like it has for its own flights. Air Italy has to choose own flight number for that flight. The airline can also order own price for that flight (e.g. 130 USD / seat).

To make this more fair and reasonable, I think the airline that is selling its seats can only charge max. the same price it has for its own customers. Also, the other airline must make the agreement for at least few months. The airline that is actually flying the route takes all responsibilities of the flight – if it's cancelled or delayed it will hit the image of that company. Respectively the airline that has only the code share agreement must market the route by itself.

This creates a win-win situation. Both airlines don't have to fly between same cities with empty planes. The other airline may fill up its plane by selling free seats and the other may have a route to that destination without actually flying there. Plus the economical risks are low for both airlines. Still there's an option to just compete on the same route.

However, Air Italy can't have code sharing for flights from Algiers to other cities. This means that you can only make an agreement for flights to/from your own base. I think this rule is important to keep the system even techincally possible.

Thoughts?

(P.S. To make a little and quick suggestion to this transit passenger problem, maybe the demand could increase on routes when new destinations are available. For example there are 50 passengers/day on first route and once there are more flights from the hub, there could be maybe 80 passengers/day... Still this thing needs more thinking & planning and I don't want to handle it on this topic.)

exchlbg

  • Former member
Re: Code sharing (and some solutions)
« Reply #1 on: June 20, 2012, 10:41:18 PM »
Certainly something like this would be very nice to have, but even your explanation of the basic rules is way more complicated than connecting flights for just your (one ) airline.
Even that is very hard to code, especially when you want to keep game speed,game balance,playability, band width, costs and so on.It is in the pipe anyway, together with a new system of city based demand.
So wishes for new features should keep also the afore mentioned restrictions in mind.
Sami explained them many times before.


Greetings
Christian

Offline TK1244

  • Members
  • Posts: 1245
Re: Code sharing (and some solutions)
« Reply #2 on: June 20, 2012, 11:06:18 PM »
The code share flights can also solve the competition between two airlines from the same alliance (preventing the alliance to fill 200+ % of the demand) on the same route :)

With Sami working on the city based demand feature, I don't think this will be introduced soon, but surely hope to see it in the versions after the city based demand has been introduced!
TK Regional

Offline Jona L.

  • Members
  • Posts: 3361
Re: Code sharing (and some solutions)
« Reply #3 on: June 21, 2012, 12:02:30 AM »
With Sami working on the city based demand feature

;D LOL *rofl*

Best joke of 2012 so far :P

Offline TK1244

  • Members
  • Posts: 1245
Re: Code sharing (and some solutions)
« Reply #4 on: June 21, 2012, 03:37:25 AM »
;D LOL *rofl*

Best joke of 2012 so far :P
Sounds like I'm not up to date... :-[ Wasn't the city based demand feature started developing some time ago...?
TK Regional

Offline Jona L.

  • Members
  • Posts: 3361
Re: Code sharing (and some solutions)
« Reply #5 on: June 21, 2012, 07:33:29 AM »
Sounds like I'm not up to date... :-[ Wasn't the city based demand feature started developing some time ago...?

No offense toward the almighty admin, but:

Sami's workspeed is not maxed out... if he was German, thus efficient, we'd already see AWS 2.3 by now :P , given he is a pilot IRL, I can understand it though.

I am one of the moderators of the -in future to be public- forums of the project. Yet it is on halt...

Offline JumboShrimp

  • Members
  • Posts: 5997
Re: Code sharing (and some solutions)
« Reply #6 on: June 21, 2012, 03:06:23 PM »
No offense toward the almighty admin, but:

Sami's workspeed is not maxed out... if he was German, thus efficient, we'd already see AWS 2.3 by now :P , given he is a pilot IRL, I can understand it though.

I am one of the moderators of the -in future to be public- forums of the project. Yet it is on halt...

I think with software development it is different.  You put a lot of work into it, work efficiently and later realize that you are at a dead end, that you made bad decisions about directions of the project.

So you want to think through how you want to model things upfront.  But this thinking upfront, trying to make good decisions upfront can be very paralysing....
« Last Edit: June 22, 2012, 12:30:21 AM by JumboShrimp »

Offline Jona L.

  • Members
  • Posts: 3361
Re: Code sharing (and some solutions)
« Reply #7 on: June 21, 2012, 10:45:18 PM »
I think with software development it is be different.  You put a lot of work into it, work efficiently and later realize that you are at a dead end, that you made bad decisions about directions of the project.

So you want to think through how you want to model things upfront.  But this thinking upfront, trying to make good decisions upfront can be very paralysing....

Obviously someone took me too serious here :-[

Isn't it the all time joke of AWS, that development proceeds slowly? At least I remember so from earlier years of the sim...

Glob-Al

  • Former member
Re: Code sharing (and some solutions)
« Reply #8 on: June 25, 2012, 02:43:39 PM »
Like the OP, I have been thinking a bit about codesharing recently - thinking which has been speeded up by the hints in this thread that city based demand may still be a long way away. Now I agree that for codesharing to serve its real life purpose, we need a connections system in place first. But I think a form of codesharing brought in to AWS as it stands now could still be worthwhile because it could serve two very different purposes:
(1) as a quasi-replacement for ABCBA routing.
(2) as a hedge mechanism to support new startup airlines competing against big incumbents (which in turn means that the demand model can be tweaked so that price becomes more significant than frequency but without big airlines being able to kill of small competitors by starting a price war)

Now this may fail the "real life test", but I don't think it would be a big success in any "playability test". Here's how I'd see it working:

- An airline can put up for sale a block of 20% of its seats on any given flight. They can advertise these to a specific airline, to airlines in their alliance, or to all airlines. They will specify a percentage of the default ticket price (within a range - say 60-80% - to avoid abuse) that they would charge an airline for purchasing these seats. If it succeeds in selling one block of 20%, an airline can then list a second block of 20%, but they would have to be bought by a different airline, and after two blocks that's it - so you will always have to try to fill 60% of your seats yourself.

- An airline can buy one of these blocks of seats if it flies from one of its own bases in to the airport that the proposed codeshare flight departs from, and provided purchasing the block does not mean it would have more codeshare seats going out of that airport in a week than it flies passengers in to it (i.e. if you fly 500 passengers a week to airport x, the max number of codeshare seats you can buy out of that airport x is 500 per week). When an airline makes a purchase, it commits itself to buying 20% of the seats on every single instance of that flight for a year, and at the % of the default ticket price (i.e. 60-80%) agreed in the contract. It also has to pay a fee (perhaps in staffing costs). It then sets a flight number for this flight and sells the tickets on at whatever price it wants - acting almost as a travel agent if you like.

In other words the selling airline is getting a guaranteed sale of 20% of its seats, but at a discounted price compared to what it would be able to get if it successfully sold the seats itself and at full price. In essence it has taken a hedge, and the buying airline has taken a risk. In that sense, it's quite simple, but the beauty comes in what it could mean for gameplay. Take an example:

I am Airline A, part of the Green Alliance and based at LHR. Our competitors the Red Alliance have a player (Airline B) who is bossing the market at JFK and starting to get very big, and scoring lots of points for Red Alliance. Airline B's only competitor is Airline C, a small startup that Airline B is trying to defeat by dropping his prices on any route Airline C flies. Currently demand LHR - JFK is 1000 per day, and I offer one flight with 200 seats (Airline B offers 900 seats and has the majority market share). Meanwhile demand JFK - MSP is 600 per day and demand JFK - MIA is also 600 per day. Both Airline B and Airline C fly four flights a day to each of these destinations, with 150 seats per plane. But because of lower prices Airline B is taking 65% of the market. Airline C therefore decides to offer a block of 20% of the seats (i.e. 30 seats) on each of its flights to these two destinations as a codeshare, charging 75% of the default ticket price. I buy the rights to codeshare on each of the four JFK - MSP flights and price my seats aggressively to get a chunk of the market share and cause Airline B some headaches. I also do likewise for two of the JFK-MIA flights but at that point I can't buy any more codeshares because I now have 30 seats on 6 flights = 180 seats, and buying another block of 30 would take me over 200 - the number of seats I currently offer with my own airline flying in to JFK each day. If I want to buy these extra codeshares I'm going to need to man up and offer more seats LHR - JFK in direct competition to Airline B. Or perhaps I will see if I can get my Green Alliance partners based at CDG and AMS to buy them...

So, who benefits from this scenario? Well:
- I, Airline A, benefit because I can now find a way to challenge my global rival Airline B, even though I can only go head to head with him on one (or a handful if we both have multiple bases) of routes. I can compound this by providing some financial stability to a startup who wants to compete with him on other routes. Meanwhile I have also benefitted from adding extra destinations served, which I believe is one of the things that can score points for my alliance. (and if not should be  :P )
- Airline B clearly does not benefit, as his monopoly is being challenged. BUT, many players have said that running a big airline with no competition can get a bit boring, and this would certainly force him to keep his attention up and/or to work closely with his alliance. And of course he could try to do the same thing back to me - in fact, as he's flying 900 seats a day in to LHR he could buy quite a lot of codeshares from there if he wanted. Meanwhile, his own direct losses from my involvement are capped because of the number of seats I offer flying in to his base - probably he has loads of other routes he can still make money on.
- Airline C is a big beneficiary as they are able to hedge their risk, and establish a foothold in the gameworld at a big base. It's still going to be tough competing with Airline B, but if he can sell lots of codeshares he helps protect himself from a price war.
- The Green Alliance can be encouraged to work together by all building up a portfolio of codeshares out of JFK to challenge Airline B - good for alliance teamwork. Maybe in time we could even seek to bring Airline C on board.
- The Red Alliance can be encouraged to work together to strike back, or perhaps to help their team mate Airline B if he does run in to trouble by buying codeshares on his flights or offering him cheap ones on theirs.
- Those who hate competition based solely on frequency benefit because this will allow price to become a more important factor in demand without killing mid-game startups.
- The other airlines in the world benefit because potentially lower profit margins for big airlines (from more competition on price) means they may be slower to open more bases and unable to control as many slots in production lines.

Now, I wouldn't pretend this is a perfect system. It would definitely take some coding (although nowhere near the scale of city based demand and/or passenger connections I wouldn't have thought), and it might fail some people's "reality test" - to be clear, there is no sense in which my passengers on LHR-JFK are actually connecting through to these flights (except in my imagination!). But I think it would add a lot to gameworld playability IMHO.

Thoughts?

Glob-Al

Glob-Al

  • Former member
Re: Code sharing (and some solutions)
« Reply #9 on: June 25, 2012, 02:44:55 PM »
As a couple of footnotes to this (I thought the original post was long enough so am putting these here!)

- I mentioned limiting the price range at which codeshare seats could be sold to 60-80% of default ticket prices to avoid abuse. But an interesting tweak to this would be if for routes with daily demand of less than 50, the upper end of this limit was relaxed to say 130%. In that way a big airline could overpay for seats on a little regional connection flight, acting as a loss leader in order to expand their network (does number of destinations served count towards CI - if not, maybe it should?) - whilst effectively subsidising a small regional carrier and perhaps making this sort of airline more economically viable in game.

- Another interesting twist that I'm struggling to get my head around if it could work, would be if for every codeshare flight established on a route, route demand increased by a small amount - at the expense of another route departing from the same airport but that currently had no-one flying it. This would be a very crude simulation for passengers choosing to take an indirect flight as there wasn't a direct one. But it might well be more trouble than it was worth!


exchlbg

  • Former member
Re: Code sharing (and some solutions)
« Reply #10 on: June 25, 2012, 06:01:13 PM »
Implementing this you would need to recalculate all demands, because given demand already calculates connecting/codeshare  passengers.
To make sense of connecting or codeshare flights city-based demand has to be implemented first, otherwise numbers of passengers would rise extremely unrealistically.

Glob-Al

  • Former member
Re: Code sharing (and some solutions)
« Reply #11 on: June 25, 2012, 11:31:45 PM »
Implementing this you would need to recalculate all demands, because given demand already calculates connecting/codeshare  passengers.
To make sense of connecting or codeshare flights city-based demand has to be implemented first, otherwise numbers of passengers would rise extremely unrealistically.

Hi exchlbg,

I'm not sure if your comment was in response to my post or not, but just to say the point of my idea is that you would actually not need to recalculate them. My approach would mean code sharing would just be a way of competing on other routes without opening a new base, rather than modelling connecting passenger demand. I agree it would be good to do this one day but it will be a hugely complex task - something like my suggestion would help make the game fun to play in the meantime.

Cheers,
Glob-Al

exchlbg

  • Former member
Re: Code sharing (and some solutions)
« Reply #12 on: June 26, 2012, 12:55:26 AM »
Ok, maybe I got your point not right by the first reading.But nevertheless I can´t see any point of buying another airlines seats, that this airline is forcasting not selling by itself.
And talking of coding... do you really think implementing ditribution of seats,negotiating processes, calculating rule violations, and many more I can´t think of yet could be done just like this? It would take a serious amount of time coding,testing, balancing that is already needed for a realistic distribution system, implementing cargo, and realizing connecting traffic, first for your airline, later maybe inbetween airlines/alliances.
I think codesharing would just make no sense without connecting passengers, only this way you can successfully act as a feeder airline.

Greets Christian

 

WARNING! This website is not compatible with the old version of Internet Explorer you are using.

If you are using the latest version please turn OFF the compatibility mode.