AirwaySim

I have been working on a full rewrite of the passenger demand distribution system lately. The old code is based on early days of AWS and has been updated during the way with various bits here and there, and a bit of a cleanup was needed.

There are no major new features but this update will change some aspects of the demand distribution (more specs & details later!) and fix also a couple of bugs (for example a memory leak that has caused problems in MT#6 a few times). This update will be available on any new game worlds. It will still be called v.1.3 for practical reasons, instead of a new version number (to which it very well would be entitled for).

Since this update is a core-level change, testing is needed. Instructions are as follows (see this post later again for further info).

* Cost to join is 2 credits (flat fee). It will be refunded when the game is complete. Requirement for refund is that you are still a member of the game world when it closes down.

* The world has 10 minute days, $100 million start cash. Please get as many planes and routes as you can. Certain features like bases and alliances are disabled.

* Coding is still in progress, and some of the features are missing (see the end of this post).


* Please test especially the following:

   - passenger demand numbers (route planning). Are some routes that should have demand missing the pax demand figures (0 demand)?  Or otherwise any gross errors on the figures?

   - once the time gets started, please test the passenger distribution in a normal way first (to see that everything is logical), and after that please try to do something silly, like adding way too many flights (oversupply) or way too much frequency compared to what is reasonable, or play with the prices too.

   - test separately also the effect of ticket prices (up / down from default).

   - (Note; there is no randomization in the route sales figures; for easier testing - final version will have daily random fluctuation)

   - (Note2; aircraft purchasing limits for first turn and weekly limits are removed for testing purposes)


* Please report bugs on the bugs forum, and please notify clearly that it's about the test scenario. If it's about the pax distribution, don't touch any route data (like prices or flights) before it's investigated.

* Please post all other comments to this thread. Game does not have own forums (or alliance or press release functions).

* Disclaimer: This is a test. Game may be shut down at any point, and it will NOT run for the entire duration.



Feature / changes list of the airport-to-airport demand system:

- This feature determines how many passengers wish to travel from airport A to B.

- The system has been technically modified so that the demand values are calculated and stored differently than before. This is needed for the possible next new demand system (city demand). The actual calculation of the demand figures is exactly the same as in current games though.

- New feature for the system is that any demand changes are gradual. The demand values (airport to airport) are updated on certain intervals and if the values change, there is always a maximum change percentage per game month. (Please note that this may cause issues to airports where huge demand changes occur overnight, like fall of CCCP; untested so far)


Feature / changes list of the new demand distribution system:

- This feature allocates the demand to available flights, based on the data of the route and flights.

- The background process doing all this magic is some 30% faster than before, and a memory issue has been fixed that has caused issues in previous game worlds (time freeze).

- New system also features better debugging and testing abilities for me to make test runs of certain route combinations without the need to actually create the airlines and routes in game interface.

- Frequency limits for routes have been adjusted. The system includes now two separate checker mechanisms for this; another factors the plane size vs. route distance, and other factor is route frequency vs. route demand. In short, players are expected to use the appropriate sized aircraft equipment for routes (usually meaning a bit larger planes than in the past game worlds). Information to user interface about these will be built later.

- Minimum intervals for flights are also adjusted (= the time needed between two flights on same route by same airline). Again, if too close, they are counted (nearly) as a single route. 60 minutes is always recommended, but goes down linearily on busy routes.

- Technical stopovers carry about 20% penalty vs. non-techstop flights (each stopover adds another ~20% penalty). (increased from previous)

- Flight delay / cancellation and route oversupply calculations have also been technically modified, and made much simpler than in the past. (allows to develop the delay/cancellation module more easily in the future)

- RI plays a larger role now than in the past. It still grows like before from 0 to 100 and stays there, but effects of low RI have been increased. (RI will, perhaps, become a fluctuating number in the future instead of static once reaching 100). It is advisable to start new route with smaller equipment and less frequency, rather than blast the full demand right away.

- As before, the route passenger demand increases if there is a lot of seat supply, or low average prices offered by airlines on the route. And like before if there is no supply in let's say C class the pax demand will move to Y/F class if there are available seats there. (nothing new here, just adjusted from current systems). Passengers will also choose not to fly if the airlines offer too high prices or otherwise a poor product (= demand is not static)

- Price effect is more noticeable and makes greater impact to route sales. It is also more safe to increase prices over the suggested default level since there is not so steep "cliff" anymore on the pax travel desire with increased prices as before.

- Of the ~12 factors affecting passenger distribution the following are currently modelled in this test version: Prices, flight frequency, techstops, aircraft size.

- The following are not yet modelled, but will be added before the final release version of the system: dep/arrival times, route flight duration, aircraft type/condition bonus, company image effect, alliance effect, onboard service/seat quality, and few other minor factors.

Perhaps open the gates and flood the used market so we can pick up some planes?

So I jumped right in and decided to jump right to the frequency oversupply/frequency problem.  The clock hasn't started yet, but once all my aircraft are delivered I will be flying 38x daily from O'Hare to LaGuardia with 737s (5100 total seats avail daily).  I thought I was being slick and overloading on frequency and capacity, which according to the game I should be.  However, I was curious as to the actual frequency in real life as the demand is ~4000 pax/day.  So I did some research and in the real world there are 35 daily flights!

Here is the breakdown of aircraft used:
14x MD83
11x E170
6x A320
2x B752
1x A319
1x B739

So in the real world, the largest aircraft is a ~190 seat 757-200 (or B739) and the smallest is a ~75 seat E170.  The average seats sold per flight, based on 4000 pax/daily, is 114 pax/plane.  My average seats per flight is 136 which means I should have a ~84% load factor assuming no competition.

You stated that we should test the pax distribution system in a normal way first and then try to do something silly with oversupply/frequency compared to what is reasonable.  At this point, I'm kind of confused as to whether what I did is "the normal way" or "the silly way" since it mirrors the real world.  Can you further elaborate as to how the new algorithm is supposed to work so we can attempt to "break" it?

Nothing left in the used market.... so.. placed an order for a bunch of new jets...

Yeah same here. I think it'd work out better if you flood the markets with a few decent domestic aircraft (A320 family /B737 family). Otherwise most of this test is going to be spent waiting for new aircraft deliveries.

Same here... nothing left in used market... just ordered 10 B737... but have to wait 3 months (game time) before first delivery... perhaps move the new jet delivery much faster for the first few days so we can get as many jets as we can with that $100M?

Quote from: tacsniper on July 04, 2012, 04:18:20 AM
Same here... nothing left in used market... just ordered 10 B737... but have to wait 3 months (game time) before first delivery... perhaps move the new jet delivery much faster for the first few days so we can get as many jets as we can with that $100M?

Agreed, just put 6 A320 series aircraft on order.

The test world will move quick enough with the 10 minute game days that i dont think there is any need to flood the market or speed up new deliveries

In the other quick beta a couple of months ago it didn't take me long to get a good fleet going.

Quote from: LemonButt on July 04, 2012, 12:57:54 AM
flying 38x daily from O'Hare to LaGuardia with 737s (5100 total seats avail daily).  I thought I was being slick and overloading on frequency and capacity, which according to the game I should be.  However, I was curious as to the actual frequency in real life as the demand is ~4000 pax/day.  So I did some research and in the real world there are 35 daily flights!

So in the real world, the largest aircraft is a ~190 seat 757-200 (or B739) and the smallest is a ~75 seat E170.  The average seats sold per flight, based on 4000 pax/daily, is 114 pax/plane.  My average seats per flight is 136 which means I should have a ~84% load factor assuming no competition.

For this route, I bet the 35x daily flights are not supplied by a single airline. That's the main point here...

I pre-checked ORD-LGA, and with present settings the "minimum optimal" plane size for that route (demand 3300pax/day) is a 83-seater (below that you will get a penalty for passengers since they don't like to fly any smaller plane for that distance). And another factor; after 11 daily frequencies (which is considered the "maxmimum optimal frequency" for this route) you will see noticeable drop too.


Used market is also refreshing normally now, that setting was off...

Happy to try the Beta but struggling with the flu at the moment  :'(

I've joined also and loaded leased eight 737-300/400 aircraft. Initially I've loaded 16 daily flights from Oslo to Bergen with 737-400s which meets demand quite well in game. I've also overloaded the route between Oslo and Kristiansand by around 50% and underspplied Oslo to Bodo to see how the routes perform in comparison.

Similar to another player above, I have compared the AWS demand figures with the real world schedule (this is particularly easy for Norway using the avinor website).

In September 2012 (I ignored peak summer season as frequency reduces greatly over summer as its a business route mainly) there are 28 daily flights on a 737 (Norwegian and SAS) between Oslo and Bergen, plus 3 Dash 8 flights (these are multi-stop flights however). Even though this game is 2004 vs today, it seems like true demand is much greater than the AWS model suggests. The same can be said for other domestic routes from Oslo:

September 3rd 2012 (Mon)
Stavanger 28x daily 737
Trondheim 30x daily 737
Bodo 11x daily 737
Tromso 14x daily 737
Kristiansand 9x daily 737

The list goes on... The domestic market is Norway is huge.

I'd be happy to put together a full list of up to date stats for the whole of Norway if it would be helpful/used.

Quote from: sami on July 04, 2012, 11:08:00 AM
For this route, I bet the 35x daily flights are not supplied by a single airline. That's the main point here...

I pre-checked ORD-LGA, and with present settings the "minimum optimal" plane size for that route (demand 3300pax/day) is a 83-seater (below that you will get a penalty for passengers since they don't like to fly any smaller plane for that distance). And another factor; after 11 daily frequencies (which is considered the "maxmimum optimal frequency" for this route) you will see noticeable drop too.


Used market is also refreshing normally now, that setting was off...

This opens up a huge can of worms:

1. 35x daily are supplied by several airlines and not just one, but the only reason this is true is because of real hubs in ORD and LGA creating this need.  Using AWS's current static demand model, this is not taken into consideration.  This means if you have no competition, you would be forced to fly 11x daily with 300+ seat aircraft to meet demand.  Once the competition shows up, you'll be forced to switch to 150 seaters flying 11x daily to avoid going broke.  This doesn't make any sense.  The obvious way to counter this in my mind would be revenue management where you could fly 11x daily with 150 seaters and overwhelming demand where you can make just as much money as a 300 seater.

2. Is the minimum optimal plane size handled as discrete or continuous data?  I went back to verify and that E170 is actually an E175.  An E175 can have 78 seats in 2-class config or 88 seats in a 1-class config.  Does configuring for less seats mean you get a penalty (minimum is 83 seats) or how is this data handled?

3.  What is the minimum optimal plane size calculated using?  Just flight distance or does demand factor in?  A CRJ200 only has 50 seats, but there are many flights IRL that are considerably longer than the ORD-LGA flight (635nm): http://flightaware.com/live/aircrafttype/CRJ2?;offset=0;sort=DESC;order=filed_ete  The longest flight with a CRJ200 in the air as I type this is from Des Moines to Phoenix, which is a 1200nm flight with a 50 seater!  You said pax don't want to fly a smaller plane for "that distance".  With a jet, whether it be a CRJ or E-jet, the flight is only 1:45.  I guess my big concern here is that it's already very difficult to run a regional airline, so penalizing them for flying CRJs etc long distances is counterproductive, especially since the number of potential flights for a regional is very limited outside of the US/Europe where things are close together.

4.  My understanding of the frequency issue are the transatlantic 757 flights and other "long range narrow body aircraft", which is why people want ETOPs etc. to combat abuse.  I think flights under 500nm in length and over 2000nm should be the only ones subject to ideal frequency limits (or some other arbitrary limit).  The frequency problem is not really in issue in the 500-2000nm range IMO.  The under 500nm routes with heavy demand, such as Taipei to Hong Kong are subject to abuse as well as the trans-atlantic routes.  I initially thought the upper limit should be 2500nm, but then I remembered I used to fly 24x daily with MD95 between SFO and JFK, which is part of the abuse we're trying to end :)  Other restrictions could be based on domestic/international flights (for US at least) and jet/prop aircraft.  I can fly a Dash-8 from Chicago to Los Angeles since the range is 1510nm, but it would take 5.5 hours since it is a prop, but making that flight with a CRJ200LR will get there in 4.5 hours.  However, the Dash-8 would be preferred because it is a "bigger" plane based on number of seats and even though the CRJ would be preferred by passengers IRL, the Dash-8 will receive preference in AWS.

5. Once city-based demand is implemented, routes like ORD-LGA will likely get bigger--not smaller, since all the demand from LGA/EWR/JFK would be/should be lumped in together as will the demand from ORD/MDW.  This is probably going to be around 10,000 pax/day I'm guessing and 11x daily isn't going to cut it.

6. Can you provide a sample matrix on minimum plane size, maximum frequency, and how demand plays into this to give us an idea of how it works?

- The first message of the thread has been updated; contains information about what's changed.

- Time starts to tick now.

- If you are wondering why some route has certain sales, please post a question here, with the appropriate route details. Do note that RI effect is quite high at the beginning - 0 route image combined with low demand and multiple daily flights is not profitable, but once RI reaches 100 you will attain full sales.

- Note (from the first message) that parts of the values affecting demand distribution are still missing (like effect of route dep / arr times). These will be added later on.

Hi sami, I am based out of Baku (UBBB) and as the first day changed the first l/f results came through. They seem far too low in my opinion, I operate one flight a day to each of four destinations from Baku (one of which detailed in the attachment) and even though the aircraft nearly/only just covers the advertised requirement, I am only getting around 5% l/f.
Regards, Gianni

Quote from: Gianni on July 04, 2012, 04:30:59 PM
Hi sami, I am based out of Baku (UBBB) and as the first day changed the first l/f results came through. They seem far too low in my opinion, I operate one flight a day to each of four destinations from Baku (one of which detailed in the attachment) and even though the aircraft nearly/only just covers the advertised requirement, I am only getting around 5% l/f.
Regards, Gianni

Having the same problem. My Miami-Aruba route has gotten the same LF for 4 days straight, (10.8%). Also, my routes that I fly once a day are getting LFs of around 15%.

Quote from: ARASKA on July 04, 2012, 04:41:19 PM
Having the same problem. My Miami-Aruba route has gotten the same LF for 4 days straight, (10.8%). Also, my routes that I fly once a day are getting LFs of around 15%.

Agreed, every flight is hardly pulling any load factor at all and also isn't showing up in the income statement as it should. Two of my flight are showing prose at 70% LF but only after advertising from the beginning.

Quote from: ARASKA on July 04, 2012, 04:41:19 PM
Having the same problem. My Miami-Aruba route has gotten the same LF for 4 days straight, (10.8%). Also, my routes that I fly once a day are getting LFs of around 15%.

In sami's updated first post today:

' - RI plays a larger role now than in the past. It still grows like before from 0 to 100 and stays there, but effects of low RI have been increased. (RI will, perhaps, become a fluctuating number in the future instead of static once reaching 100). It is advisable to start new route with smaller equipment and less frequency, rather than blast the full demand right away.'

I would think that this more pronounced RI-related effect is being rather severe in the early stage?

Quote from: brique on July 04, 2012, 04:46:39 PM
In sami's updated first post today:

' - RI plays a larger role now than in the past. It still grows like before from 0 to 100 and stays there, but effects of low RI have been increased. (RI will, perhaps, become a fluctuating number in the future instead of static once reaching 100). It is advisable to start new route with smaller equipment and less frequency, rather than blast the full demand right away.'

I would think that this more pronounced RI-related effect is being rather severe in the early stage?

Then how come two airlines have 100% LFs since the first day?

exactly what I was going to ask.

Quote from: sami on July 03, 2012, 10:29:48 PM
   - (Note2; aircraft purchasing limits for first turn and weekly limits are removed for testing purposes)

If the purchasing limits are removed, how can I get this message:

At least other of the two is flying on intra-Japan routes with 10000 daily pax demand. If you reach still 20% of those people (they "hear" of your route), you still have a whole lot of pax to fill your F100. But if you have a 200 pax route and still reach the same 20% of the people when RI is 0...

But anyway, the sales of Gianni's route seem a bit too low, I am looking to this.

Also another thing I noticed was that my RI has not gone up at all on any of my flights.

Quote from: ARASKA on July 04, 2012, 04:47:54 PM
Then how come two airlines have 100% LFs since the first day?

No idea : just thought that the passage quoted may offer some explanation for ultra-low L's on new routes, if its not consistent, then Sami may be able to offer a better one.

additon : Ah, he has....

I have gotten exactly the same LFs since day one... I don't call that realistic

Quote from: libertyairlines on July 04, 2012, 04:51:32 PM
Also another thing I noticed was that my RI has not gone up at all on any of my flights.

It takes 6 months or so to reach 100 from zero, so it does nothing in a game week really.

Quote from: ARASKA on July 04, 2012, 04:53:40 PM
I have gotten exactly the same LFs since day one... I don't call that realistic

Read the instructions properly first please. All randomization is OFF.




Edit; Also, server time is stopped while I check these things + the bug report on income sheet.

Should I be concerned that there was no income from ticket sales even though on my Manage Routes it shows that I have tickets that are selling?

Quote from: sami on July 03, 2012, 10:29:48 PM

- Technical stopovers carry about 20% penalty vs. non-techstop flights (each stopover adds another ~20% penalty). (increased from previous)

While this is a good idea in MT scenarios, will airlines face this penalty in JA and DOTM scenarios where there is no equipment available to serve routes non-stop? There is not an aircraft that can fly over 4500nm until the DC-8-62/3 in the late 60's, and no aircraft capable of non stop flights over 5250nm until the 762ER in ~1984. If there is a 20% penalty on a DC-8 in 1961, for example, many routes will become un-servable in earlier scenarios.

Don, I think the key words in what Sami said there are "versus non stop." As I read it, if all airlines on a route are tech-stopping, the passengers will live with it. But if one airline is tech-stopping and the other is not, they'll prefer the direct flight by quite a wide margin. Hopefully that's the case anyway.

Quote from: --miki-- on July 04, 2012, 04:48:49 PM
If the purchasing limits are removed, how can I get this message:

2 per day limit still active, but time is stuck :P

Quote from: sami on July 04, 2012, 04:54:52 PM
server time is stopped while I check these things + the bug report on income sheet.

Time is running again, incomesheet bug is fixed, and so is the route image issue (route image reduction was done twice).

Since there was a modification to the route calculations the ROUTE SALES HISTORY DATA IS CLEARED FOR EVERYONE, so that "old" results do not mess up..

Quote from: Glob-Al on July 04, 2012, 05:11:54 PM
Don, I think the key words in what Sami said there are "versus non stop." As I read it, if all airlines on a route are tech-stopping, the passengers will live with it. But if one airline is tech-stopping and the other is not, they'll prefer the direct flight by quite a wide margin. Hopefully that's the case anyway.

Yes, this is what I am after. But if all flights are with a stopover, some pax may choose not to fly (unless you compensate by lower prices for example). (= same story if the flight would be middle of the night, or poor CI .. etc)



Edit: MONEY has been also reset to everyone, so that nobody happens to go bankrupt ;)


TIP: When your RI is low, lower the prices to gain more pax!

Quote from: sami on July 04, 2012, 11:08:00 AM
For this route, I bet the 35x daily flights are not supplied by a single airline. That's the main point here...

I pre-checked ORD-LGA, and with present settings the "minimum optimal" plane size for that route (demand 3300pax/day) is a 83-seater (below that you will get a penalty for passengers since they don't like to fly any smaller plane for that distance). And another factor; after 11 daily frequencies (which is considered the "maxmimum optimal frequency" for this route) you will see noticeable drop too.


Used market is also refreshing normally now, that setting was off...

First, LF on my one operating route (AMS-LHR) looks good. 504 seats offered on the 200nm route on a 126 seat A319 4x daily. LF is 85%. Just added an ATR-42 on some sub 100nm routes, will see how those fare later today.

Now, regarding the statement that there will be a "penalty" on a plane with less than 83 seats. What kind of penalty? Similar to the route overlap function where LF's plummet, or the "frequency" penalty that has been discussed where, say, a 76 seat EMB-175 would count as .915 frequencies but all seats would still sell assuming no competition?

Just trying to wrap my head around this. I am really trying to figure out how an airline will be able to operate penalty free with 3 fleet types.  :-\

Don

Hi sami, thanks for that, now achieving more along the lines of what I'd expected (30%-50%)
Regards, Gianni

Quote from: sami on July 04, 2012, 05:16:39 PM
Time is running again, incomesheet bug is fixed, and so is the route image issue (route image reduction was done twice).

Since there was a modification to the route calculations the ROUTE SALES HISTORY DATA IS CLEARED FOR EVERYONE, so that "old" results do not mess up..

Thanks.

My LFs were 22.5% before the change but they're in the 70%-80% range now.

Feb 8 2005 I had $2655K cash.

Feb 9 2005 I have $100M cash.

Was this a gift or is something wrong?

Please try the effects of lowering or changing prices too.  From small increments (1-5%) to larger ones (>50%).

Quote from: sami on July 04, 2012, 05:16:39 PM
Edit: MONEY has been also reset to everyone, so that nobody happens to go bankrupt ;)

Not flying any route yet (a/c in configging) my money just jumped back to 100M.... is that supposed to be like that?

Also the Used market limit is still up :/

Also my credit rating just jumped down to D :?

Bought used planes at 60% condition and had them fly for over a week. The delay page shows massive delays (1h 50 min!), but no cancellations. Doesn't make sense, somehow, since I scheduled them 115min apart. Some routes were flown less than others although I got both at the start day, so I guess there were some cancellations after all (CYVR - CYYC routes).

I have more junk incoming, I will leave them at their low percentages overnight so that you can play with the code, if you like and use my airline as testbed.

Edit 20:06: CYVR - CYPW route does not show pax limitation in graph (it shows on route edit)

Edit 20:23: If someone wants to test the frequency issues, fly against me from CYYC or KSEA (have both routes fully supplied with A300s - especially the latter would be kinda funny IRL)

I've got something odd.  On RJAA-VTBD I'm actually managing higher load factors in C class than in Y class in spite of my company image of 2 and a route image of 4.  I thought it was supposed to be harder to attract those C class passengers with values so low.

Edit: only in one direction though...maybe it's just a daily fluctuation since you cleared the history?

IMHO I'm getting weird oversupply warnings. For example BIKF-BIAR - it shows between 60-80 pax/day and I supply 116 (2 flights). Two warnings already. Are real pax figures really below 58?

Quote from: AndiD on July 04, 2012, 05:53:12 PM
Edit 20:06: CYVR - CYPW route does not show pax limitation in graph (it shows on route edit)

Payload is limited only CYPW-CYVR.

Quote from: ucfknightryan on July 04, 2012, 05:54:22 PM
I've got something odd.  On RJAA-VTBD I'm actually managing higher load factors in C class than in Y class

The system does not calculate load factors. It calculates how many seats you actually sell, and LF is a statistical value based on that.

But anyway, the C/F class RI effect has been increased.

Quote from: chiveicrook on July 04, 2012, 06:11:52 PM
IMHO I'm getting weird oversupply warnings. For example BIKF-BIAR - it shows between 60-80 pax/day and I supply 116 (2 flights). Two warnings already. Are real pax figures really below 58?

Checked, BIKF-BIAR true demand at a certain date is 69.3pax/day. You have 116 seats. => should not be a problem.

And indeed there was a bug, it thought demand was 55 seats ..

My flights are scheduled with 1% chance of delay but my on time departures are 20%. It looks like no airline is above 75%.

https://www.airwaysim.com/game/Info/Airline/27/#AirlineInfo

Quote from: ucfknightryan on July 04, 2012, 05:54:22 PM
I've got something odd.  On RJAA-VTBD I'm actually managing higher load factors in C class than in Y class in spite of my company image of 2 and a route image of 4.  I thought it was supposed to be harder to attract those C class passengers with values so low.

Edit: only in one direction though...maybe it's just a daily fluctuation since you cleared the history?

@ EDIT: no daily fluctuation as Sami said in first post

@ rest: It is just harder/impossible to fly Magic carpets with more C than  SEATS available, You can still have highe C/F LFs than Y, if you have a "normal" configuration ;)

Route cancellations did not register to route history. This is fixed now.

Quote from: jetwestinc on July 04, 2012, 05:18:00 PM
Now, regarding the statement that there will be a "penalty" on a plane with less than 83 seats. What kind of penalty? Similar to the route overlap function where LF's plummet, or the "frequency" penalty that has been discussed where, say, a 76 seat EMB-175 would count as .915 frequencies but all seats would still sell assuming no competition?

I am also wondering about this.  Let's define concepts first:

- Frequency benefit is:  Capacity is not treated as equal.  Capacity in 1/2 size aircraft is worth 2x (will have 2x LF) on certain routes

- No frequency benefit:  All capacity is equal, regardless of the aircraft size.  LF is equal regardless of aircraft size

- Frequency penalty (small aircraft penalty):  This is new one. if really implemented, it would mean that capacity of in aircraft that is too small would be penalized.  Meaning capacity in aircraft that is too small will have lower LF than capacity in larger aircraft.

Quote from: sami on July 03, 2012, 10:29:48 PM
- Technical stopovers carry about 20% penalty vs. non-techstop flights (each stopover adds another ~20% penalty). (increased from previous)

- The following are not yet modelled, but will be added before the final release version of the system: dep/arrival times, route flight duration, aircraft type/condition bonus, company image effect, alliance effect, onboard service/seat quality, and few other minor factors.

The extra 20% penalty seems a bit harsh if the flight duration is going to be implemented.  Why not just make flight duration significant enough so that the extra delay of tech stop makes it less desirable vs. faster competition, but would not add penalty if flying the route alone?  There is already a dis-incentive to tech stop flghts to gain frequency advantage.

There may be low demand routes (4000nm ~130 pax in MT worlds) that can now be served by tech stopped A320/737, at above break even.  With 20% (LF?) penalty, a bunch of marginal routes will not be viable.

Or, as Don mentioned, in JA worlds, a lot of routes will no longer be viable with the penalty, since there is no aircraft in existance capable of flying the route.

BTW, I really love the lack of randomization in pax distribution.

I would go even a step further: No randomization on route demand screen...

Also, I would disable randomization of route demand screen in live games on routes where an airline reaches cerain RI threshold (RI of 90 or 100).  There is no one who knows more about the demand of a route than airline flying it daily...

Quote from: sami on July 03, 2012, 10:29:48 PM
- RI plays a larger role now than in the past. It still grows like before from 0 to 100 and stays there, but effects of low RI have been increased. (RI will, perhaps, become a fluctuating number in the future instead of static once reaching 100). It is advisable to start new route with smaller equipment and less frequency, rather than blast the full demand right away.

Wow this one sure is noticeable.  Same equipment flying 2x daily RJAA-VVTS at ~100% demand gives lfs of ~35% whereas same type of aircraft flying RJAA-EGLL 1x daily only 4 days a week at ~7% of demand gives 80%+ LFs.  RI is 11 for VVTS and 4 for RJAA

Also, satisfying ~75% of demand on RJAA-VDSR with the same equipment 1x daily provides no noticeable improvement in lfs compared to 100% demand on VVTS. 

Quote from: sami on July 03, 2012, 10:29:48 PM
- Frequency limits for routes have been adjusted. The system includes now two separate checker mechanisms for this; another factors the plane size vs. route distance, and other factor is route frequency vs. route demand. In short, players are expected to use the appropriate sized aircraft equipment for routes (usually meaning a bit larger planes than in the past game worlds). Information to user interface about these will be built later.

Quote from: sami on July 03, 2012, 10:29:48 PM
- Technical stopovers carry about 20% penalty vs. non-techstop flights (each stopover adds another ~20% penalty). (increased from previous)

You mean tech-stopping 727s across oceans to instantly kill the widebodies will no longer work?

Woohoo!

No idea if this is an indication of a problem, but my KSFO-KLAX flight on normal day times (6-24) don't get as much passengers as the one in the night (00.30 and back on 4 in the morning) ... it's just a little bit more pax than on day flights, but I'd expect a lot less? Am I wrong? ... I added more capacity with different plane sizes to investigate this (and other things) further.

Flight dep/arr times do not (yet) matter at all. Please read the first post.

And a little question about the used aircraft page... can't you store the settings that were used the last time and directly restore them when I return to this page? When I click on an aircraft and then press the back-button I always have to redo the whole filter settings... ok, I could open the linke in a different window, but sometimes I don't think about that when searching an aircraft.

Pax allocation issue:  Classes

Day 1: 1x A332 on CDG-BOS ~3000nm

Supply:
Y: 234
C: 16
F: 8

Allocated pax:  
Y: 116
C: 13
F: 2

Day 2, Added 1x753 + 1x752

Total Supply: 332 + 753 + distance limited 752
Y: 234 + 222 + 162 = 618
C: 16 + 15 + 15 = 46
F: 8 + 0 + 0 = 8

Allocated pax: 332 + 753 + distance limited 752
Y: 46 + 46 + 38 = 130 (up 14, ok I guess)
C: 3 + 3 + 3 = 9 (down 4, probably not ok)
F: 0 + 0 + 0 = 0 (down 3,  not ok)

LFs on Day 2, just looking at Y:
332: 19.7%
753: 20.7%
752: 23.5%

I think this may be a problem.  If the ideal aicraft is ~250 pax aircraft, if would be roughly the 753.  Therefore, slightly lower LF on A332 is expected (753 seats count for more seats than A332 seats on route this distance).

What does not make sense is 752.  It should have the LF equal to 753 (with no frequency, or smaller aircraft size benefit on this route).  Or, it should have lower LF (if there is an actual penalty for aircraft too small).

Quote from: JumboShrimp on July 04, 2012, 08:03:58 PM
Pax allocation issue:  Classes

Probably related to low RI.

Changed RI of that route to 100 to test. (probably will be 100% LF on all flights though since the demand is not met?)

Quote from: EsquireFlyer on July 04, 2012, 07:48:38 PM
You mean tech-stopping 727s across oceans to instantly kill the widebodies will no longer work?

Woohoo!

Tech stopped narrowbodies should not be able to kill non-stop widebodies for completely different reasons:
- narrowbody smaller than ideal aircraft for the route
- generally slower narrowbodies + tech stop time result in greater duration of flight.  Once coded, that should give advantage to non-stop widebodies.

Given this, additional 20% penalty for tech stop is unnecessary, IMO

Quote from: sami on July 04, 2012, 08:00:08 PM
Flight dep/arr times do not (yet) matter at all. Please read the first post.

Sorry, didn't see that... the rest seems to be good so far... I'm getting 85 pax with every plane, no mather if M90 or 763 ... that's much bether now.

Quote from: JumboShrimp on July 04, 2012, 08:09:24 PM
Tech stopped narrowbodies should not be able to kill non-stop widebodies for completely different reasons:
- narrowbody smaller than ideal aircraft for the route
- generally slower narrowbodies + tech stop time result in greater duration of flight.  Once coded, that should give advantage to non-stop widebodies.

Given this, additional 20% penalty for tech stop is unnecessary, IMO

Some narrowbodies (e.g. 727) are very fast. So the second point would apply to 737s for example, but not 727s.
IMO the additional 20% penalty for tech stop is appropriate because it is realistic--in real life, pax do not like techstops, especially when there is a nonstop available.

Maybe the 20% penalty should only apply if a competitor is flying nonstop, and not when the techstop is the only option? But there should definitely be a techstop penalty when a competitor is flying nonstop. The techstop player should have to either lower prices significantly, or take a LF hit.

Quote from: sami on July 04, 2012, 08:06:23 PM
Changed RI of that route to 100 to test. (probably will be 100% LF on all flights though since the demand is not met?)

Fast-forwarded the time a bit (due to B checks) and now the RI is back to 3.5 where it was.

If you look on the seats sold (extended route data) for last 2 days, seems logical. (?)


--
I'm off now; any further changes/fixes tomorrow evening.

Quote from: sami on July 04, 2012, 08:06:23 PM
Probably related to low RI.

Changed RI of that route to 100 to test. (probably will be 100% LF on all flights though since the demand is not met?)

I am guessing that a single calculation is performed for pax allocation rather than 3 class independent.  That and some rounding probably resulted in the 3 F pax disappearing when additional equipment was added on the route.  I don't think it is a problem.

Thanks for changeing the RI to 100.  I will re-test once I get some aircraft delivered that will allow me to oversupply the route a little more.  Right now, I am only at about 115% of demand...

I am still seing LF that is too high on the 752, BTW:
332: 65.8%
753: 69.4%
752: 79.6%

In general, 752 should not be having higher LF than 753.  Possible reasons:
- the system is not looking at the effective capacity.  My 752 is distance limited to lower Y limit.  I will substitute another 752 once it arrives and retest.
- the system is looking at max HD capacity to figure out ideal aircraft for the route length.  If that's the case, I think the formula could use some tweaking to shift the minimum aircraft up by 10 to 25%.

I was in the last test game world which must have just ended, but dont think i got the refund that it said there would be last time and says there will be this time. >:(

Quote from: bob-b on July 04, 2012, 09:03:01 PM
I was in the last test game world which must have just ended, but dont think i got the refund that it said there would be last time and says there will be this time. >:(

Test games provide way more than 2 credits worth of fun.

Quote from: EsquireFlyer on July 04, 2012, 08:22:08 PM
Some narrowbodies (e.g. 727) are very fast. So the second point would apply to 737s for example, but not 727s.
IMO the additional 20% penalty for tech stop is appropriate because it is realistic--in real life, pax do not like techstops, especially when there is a nonstop available.

Maybe the 20% penalty should only apply if a competitor is flying nonstop, and not when the techstop is the only option? But there should definitely be a techstop penalty when a competitor is flying nonstop. The techstop player should have to either lower prices significantly, or take a LF hit.

That's one of the reasons why I don't think there should be a penalty for tech stop, and things should only be handled by travel time.  45 minutes tech stop of 727 (ir 737.A320) adds 10 to 15% to route travel time vs. non-stop narrowbody (on a transatlantic route).  Sami could factor this travel time difference whichever way he wants to.  He could make it significant enough to amount to 20% hit on LF...

The passenger allocation based on time will happan only vs. other competitors.  It would not be an across the board penalty.  So if there is no competition, the speed of flight should make no difference on LFs (whether it is just a slower aircraft or tech stopped aircraft).

I set up some routes to test differences from small price changes.  On RJAA to RKSI I've got Dash8s flying at default, default + 5% and default -5%.  Both the routes above default pricing and the routes below default pricing are attracting less passengers than the route at default pricing, which seems odd.

Also, how seriously should I be being penalized for flying Dash8s on a route with ~2000 demand?  I don't seem to be being hit very hard for it at the moment.

More substantial price changes, however, seem to be working more as you would expect.

RAJJ-EGLL is seeing at lest 20% less passengers at least with a 25% price hike and at least ~20% more with a 25% price cut.

Edit: of course that route might not be there for all that long as my airline appears to be on the fast track to bankruptcy.  I should have waited until the used market was set to refresh as 747SPs are turning out to have been a really bad idea  :laugh:

Is the appropriate aircraft size determined by flight length and demand or just demand?

For example, let's say demand is 2000 on IAH-ATL and 2000 on IAH-LHR. If the A330 is appropriate for IAH-LHR, must we fly A330s on IAH-ATL as well or would an A320/A321/737/etc be appropriate?

In real life you wouldn't usually see an A330 or any widebody on such a short domestic route.

Quote from: BryanIAH on July 04, 2012, 09:37:15 PM
Is the appropriate aircraft size determined by flight length and demand or just demand?

For example, let's say demand is 2000 on IAH-ATL and 2000 on IAH-LHR. If the A330 is appropriate for IAH-LHR, must we fly A330s on IAH-ATL as well or would an A320/A321/737/etc be appropriate?

In real life you wouldn't usually see an A330 or any widebody on such a short domestic route.

There is an appropriate aircraft for a given distance.  A330 may be a it for IAH-LHR.

As far as IAH-ATL, the appropriate aircraft distance-wise, it would be a much smaller aircraft than A330.  But Sami as adding appropriate (or optimal) frequency for the route.  That is a bit of a mystery right now, very little info on that is available and how the mechanism would work.

Quote from: bob-b on July 04, 2012, 09:03:01 PM
I was in the last test game world which must have just ended, but dont think i got the refund that it said there would be last time and says there will be this time. >:(

Really? for the sake of $1?

Hi sami, could you possibly check my delays, I'm currently at around 35% on time, but have settings all exactly the same as other games where it is 95%. It says it is primarily down to weather & technical.
Regards, Gianni

I think as a good rule of thumb that seems to be showing up in my airline is that having aircraft on say 350 pax routes that allow 2 frequencies per day and with a little demand left over seems to be working for me. You absolutely don't want to put on a 160-185 route a 160 pax 737 or A320, it seems that it is much better to say take two E-170/175's on the route or similar pax capabilities. Or alternatively you could make drastic reductions in tickets prices, default gives way low LF and I find up to 40% worst case for short haul from default pricing to give good profit and LF, long haul still sorting out but it is looking potentially over 50% reduction in ticket prices at the beginning of route with low RI. Although you find that will 300pax route B772 seems too big, it seems that A332 or B762/763 seem to be better options early on with this. Long haul loss are driving my airline into the ground although I'm trying to find the profitable spot. You should be able to increase them as the RI goes up.

Basically (I think) you don't want biggest is best and supply the demand, you actually want to supply frequency in flights aswell, this it seems is an very important factor in LF especially early on it seems. The update seems to be encouraging players to undersupply the demand which I think will be better for competion between airlines or spend fortunes on getting RI up high for the start of the route.

Anybody got any other additions?!

I really like the changes to the workings of the RI & LFs

It seems more "normal" to have a brand new airline slowly build up its LF instead of jumping on a new route and getting up into the 70/80s

Out of interest, does anyone else have a big spike in passenger numbers on the 25/02?

Quote from: markj23 on July 05, 2012, 12:22:11 AM
I really like the changes to the workings of the RI & LFs

It seems more "normal" to have a brand new airline slowly build up its LF instead of jumping on a new route and getting up into the 70/80s

Out of interest, does anyone else have a big spike in passenger numbers on the 25/02?

I did not notice one, but when I see one, it is when one of several flights to the same destination is not flying (because of cancellation or a check, other flights go up proportionally...

Quote from: JumboShrimp on July 05, 2012, 12:25:35 AM
I did not notice one, but when I see one, it is when one of several flights to the same destination is not flying (because of cancellation or a check, other flights go up proportionally...

Ok im silly - i missed a flight when i was checking and can see the cancellation of an earlier flight now. I had seen small bumps before but this was a pretty big one

Quote from: markj23 on July 05, 2012, 12:22:11 AM
I really like the changes to the workings of the RI & LFs

It seems more "normal" to have a brand new airline slowly build up its LF instead of jumping on a new route and getting up into the 70/80s

Out of interest, does anyone else have a big spike in passenger numbers on the 25/02?

On this, it certainly seems like RI is acting almost like a proxy for the percentage of the number of people wanting to fly the route who know about your service. For me:

On my PVG > HKG route (where I'm supplying ~50% of demand) RI is 28 and LFs are around 72%
On my PVG > DUB route (where I'm supplying almost precisely 100% of demand) RI is 28 and LFs are around 40%

I'm going to experiment now with what a deep price cut does to this.

Quote from: Glob-Al on July 05, 2012, 12:50:44 AM
On this, it certainly seems like RI is acting almost like a proxy for the percentage of the number of people wanting to fly the route who know about your service. For me:

On my PVG > HKG route (where I'm supplying ~50% of demand) RI is 28 and LFs are around 72%
On my PVG > DUB route (where I'm supplying almost precisely 100% of demand) RI is 28 and LFs are around 40%

I'm going to experiment now with what a deep price cut does to this.

I am seeing the same results.  I think I like this change as well...

Quote from: chiveicrook on July 04, 2012, 06:11:52 PM
IMHO I'm getting weird oversupply warnings. For example BIKF-BIAR - it shows between 60-80 pax/day and I supply 116 (2 flights). Two warnings already. Are real pax figures really below 58?

check your return flight demand, I've noticed it can be a bit less, maybe enough to trigger the warning

I'm running a mixed fleet (B733 and A310) on the CYVR-KSEA route (short hop, single supplier, 4 flights/day, demand is 60% of my supply). Interestingly, the pax are equally distributed across the four routes, which in turn leads to load factors of 60% (733) and 30% (A310). Again, this would benefit smaller planes over larger ones, wouldn't it?

On the other hand I get immediate 97%+ load (20 C pax!) being the sole supplier on the CYVR-CYYZ route (280 supply/2800 demand), even with a single-digit (!) CI and RI.

Working as intended?

I wish this wasn't a bug, but it has to be ;D

Quote from: bob-b on July 04, 2012, 09:03:01 PM
I was in the last test game world which must have just ended, but dont think i got the refund that it said there would be last time and says there will be this time.

You did if you were a member of that world still when it ended.

Just started flying some long haul routes with A340-300x. For those routes which were not being flown (eg. KJFK - VHHH)- LF is straight away up in to 80-90s. However where competition is already flying the route, (eg. KJFK - LFPG) LFs are in their 20s. My CI is 12. Currently only flying each route 4 days a week.

Hi,

LF are acting strangly.

(https://lh6.googleusercontent.com/-c1DDb2AiMT8/T_Viq-YNbeI/AAAAAAAAAKw/agkjYJXZNTA/s800/NTE-ORY%2520route.png)
LF figure

You can see that suddenly the LF in Y/CL dropped down, then the business increased. There is no competition on the route, only 2 flights a day with 738 (5/180). The only thing I changed on may 10th is the ticket price.

(https://lh3.googleusercontent.com/-rDxHmIRClj0/T_VkbO3lydI/AAAAAAAAALM/_2t8LhS3lVM/s912/NTE-ORY%2520route%2520demand.png)
Pax demand

Is it a normal situation or bug ?

regards

A little more cash would be much appreciated now :)  ;)

Hi Sami,

I have 12 A321s with 3 flights per each going between Haneda and Sapporo, equaling 36 flights a day. When I go to schedule the 37th, I get the below warning. I thought the limit was higher than 40 flights a day but anyway, can't schedule the 37th.

The company and route image are important but reducing prices drastically works as well. I think it's also based on demand. Flying 130 seats on a 140 demand route will take longer than flying 100 seats on a 10.000 demand route. I think.


Sami one thing I realized. Is the time frequency limit removed? I am flying out of Copenhagen to Goteborg. Two of my flights depart 30 minutes after eachother. One is flown with an A319 and the other with an Embraer 170. The total demand is 790 pax per day.

In total I fly 8 times a day but these two routes depart 30 minutes after eachother. Still the loadfactors are 80% and 90%. I do remember reading something in the previous version that penalties occurred if flights leave within an hour of the other one.

The "flights too close to each others" penalty is already active.

However if you have a route with low RI and huge demand, you still may be able to get nearly full planes despite of the flights being too close.

Quote from: michael on July 05, 2012, 11:55:36 AM
I have 12 A321s with 3 flights per each going between Haneda and Sapporo, equaling 36 flights a day. When I go to schedule the 37th, I get the below warning. I thought the limit was higher than 40 flights a day but anyway, can't schedule the 37th.

Have to check


Also. MONEY HAS BEEN RESET TO $100MIL AGAIN FOR EVERYONE.


Edit: Otherwise all OK for everyone? Other feedback and comments so far?

With everyone having many millions of money and several top ups, will it be feesible to start a game with only a few million and get good enough LFs to get enough profits to survive and grow. And I mean the average player too, not just those having played the beta, as they might not know the ideal aircraft/size for routes.

Quote from: alexgv1 on July 05, 2012, 01:08:23 PM
With everyone having many millions of money and several top ups, will it be feesible to start a game with only a few million and get good enough LFs to get enough profits to survive and grow. And I mean the average player too, not just those having played the beta, as they might not know the ideal aircraft/size for routes.

That's a good point... It's relatively easy with millions in the bank and plenty of aircraft on the used market. The start if a real game is not so easy. It may however, be a good thing as it may force players to be consider more carefully which aircraft to lease rather than grabbing any aircraft they can get their hands on. This will of course probably mean that players will be chasing the popular types even more though....

Quote from: ucfknightryan on July 04, 2012, 09:29:22 PM
I set up some routes to test differences from small price changes.  On RJAA to RKSI I've got Dash8s flying at default, default + 5% and default -5%.  Both the routes above default pricing and the routes below default pricing are attracting less passengers than the route at default pricing, which seems odd.

Also, how seriously should I be being penalized for flying Dash8s on a route with ~2000 demand?  I don't seem to be being hit very hard for it at the moment.

I started flights on the same route, but with MD-11 widebody planes (capacity is 1060 pax/day). Would love to compare our numbers on that route when you have a similar number of capacity on that route ;)

These a finances for B772 in my fleet, how can I be making 2.2 million this week when my takings every week are about 600,000! Bug I think?!

Financial overview
Values show the actual profit/loss of the aircraft including all fees and payments.
Profit yesterday
97 947 USD
Cumulative profit this week
2 232 025 USD



Cumulative profit last week
-272 614 USD
Financial overview, weekly estimate
The sums are an estimate of next week's incomes based on previous day.
Sold tickets
609 852 USD
Line maintenance (A+B)
-19 711 USD
Insurance
-80 548 USD
Fuel cost
-237 126 USD
Route fees (1)
-115 617 USD
Weekly leasing cost
-403 977 USD
Total
-247 127 USD

EDIT: This is also happening with other aircraft like 762's 1.5 million cumulative profit with only 500,000 in ticket sales. Am I missing something here?!

In terms when we start with a few planes, things will be a lot easier, growth will be slower and once you've realised how much you need to drop prices a good short haul aircraft fleet will be good early game, I have to say going long haul quickly may not be as easy as in the past. You also need to make sure you undersupply the route and get more than one Frequency in there it seems (or at least allow space for two frequencies). My short haul are very good but my long haul flights are a pig to get right, I think when CI comes into it in the real game you may find opening up long haul routes easier after a few months when CI is higher and you are making good money from your short haul flights.

I think rule of thumb for plane choice:
Anything below 300 pax demand a plane that can supply two freq per day approx 110-150 pax if short haul or 220 to about 260 or so with long haul.
400-500 pax demand for long haul Approx 300 pax carrying capacity for long haul or 180 or so for short
I'd say anything above 500 pax demand the biggest things you can reasonably get on the route (757's for short haul 380 for long haul for example).

This is what seems to be getting the easiest and best results for me. Got A333 working very well and B772 working well on this rule. B747's need plenty of pax demand to work it seems. I.E. At least and preferably over 500 pax. Still need to lower prices though.

It seems as if it is going to be much harder to run a regional airline or even start an airline with the new LF system that depends on RI so much. let's say you wanted to start out of Ekaterinburg for example. There are lots of routes that you could put 737's on to Europe like London and Madrid but the demand is only 130 Pax. per day. Previously, it would be feasible to start an airline like this bit with the new system, the LFs would be too low to run a profit. The same could be true for restarting in the middle of the game world. If you are trying to fly all the smaller demand, not served routes before you attack the your competitors, you won't be able to turn a profit doing this. Could this be tweaked?

You downsize aircraft, you would look for E-jets and F-70 possibly F-100 etc. You just have to adjust what aircraft you would choose on the route, possibly use Turboprops aswell. You may only fly a 130 pax route once a day with 70-80 pax aircraft but this works whereas using 125 pax aircraft on a 130 route doesn't work (as easily if at all) due to the need of frequency, your optimal aircraft for a route is smaller than the demand on that route it seems. You may have a more varied fleet with E-Jets, ATR's and 737's/320's for higher demand routes and possibly the odd long haul aircraft if you have those routes available to your airline. Trust me, smaller aircraft on leases are working very well for me, BAe146's making quite a lot for me and working well.

Basically don't always think in 737's!!!

But part of the aim here is to make this more realistic. I don't think Turboprops or F70s from Ekaterinburg to Western Europe really ticks that box.

So I share araska's concern that it will be particularly hard for airlines starting up later in the game. Otherwise though I really like the new system, so I hope this issue could be solved with a small tweak. Perhaps a one-time boost to RI (not applicable to day one) if you open up a route that no one else currently flies - encourage people to take on the empty routes and be able to grow more quickly on them?

I like the new system. Most airlines in the Real World can't just start off with A320s and expect full planes unless they have really low prices. Otherwise, no one will know who they are. In this game, it makes sense for airlines to start off with small aircraft while they build their CI and RI.

Anyway, Sami; I'm not sure if this is how its supposed to work but, I have a DEN-SLC route that has a RI of 70 and a LF of 80% while my DEN-LAS route has a RI of 60 and I'm getting a LF near 95%. Is this a bug?

I'm flying both routes with the A320 Family. SLC gets flown 4x a day and LAS 9x a day. I meet the demand and leave just a little left over.

Oh and apparently I'm attracting 'C' Customers a lot easier than 'Y.' 100% on 'C' and 76% on 'Y'. Both prices were decreased about 15% at their opening.

Quote from: AAL558 on July 05, 2012, 03:14:34 PM
I like the new system. Most airlines in the Real World can't just start off with A320s and expect full planes unless they have really low prices. Otherwise, no one will know who they are. In this game, it makes sense for airlines to start off with small aircraft while they build their CI and RI.

Anyway, Sami; I'm not sure if this is how its supposed to work but, I have a DEN-SLC route that has a RI of 70 and a LF of 80% while my DEN-LAS route has a RI of 60 and I'm getting a LF near 95%. Is this a bug?

I'm not asking for full planes, I would just like to have 60% LFs instead of 25% LFs.

Regards, Sean

I think that it would be interesting to start an airline in these conditions with the normal amount of money, but I think it will drastically help stop 70% of airlines succeed straight away, which just isn't realistic. I think the increased dependence of l/f on ci and ri vastly improve the reality of the game. I did however notice that when I decreased prices by 30%, there wasn't particularly an increase in l/fs, I will however try it again in a few months with improved images.
Gianni

Quote from: sami on July 05, 2012, 12:20:09 PM
Edit: Otherwise all OK for everyone? Other feedback and comments so far?

I would like to test where and how the "higher than optimal" frequency kicks in.  I am guessing that it has to be tested by 2 players...

Any example of what should be considered higher than optimal frequency?

Quote from: AAL558 on July 05, 2012, 03:14:34 PM
I like the new system. Most airlines in the Real World can't just start off with A320s and expect full planes unless they have really low prices. Otherwise, no one will know who they are. In this game, it makes sense for airlines to start off with small aircraft while they build their CI and RI.

Anyway, Sami; I'm not sure if this is how its supposed to work but, I have a DEN-SLC route that has a RI of 70 and a LF of 80% while my DEN-LAS route has a RI of 60 and I'm getting a LF near 95%. Is this a bug?

I'm flying both routes with the A320 Family. SLC gets flown 4x a day and LAS 9x a day. I meet the demand and leave just a little left over.

Oh and apparently I'm attracting 'C' Customers a lot easier than 'Y.' 100% on 'C' and 76% on 'Y'. Both prices were decreased about 15% at their opening.

When RI is high enough (and yours is) C pax should have no problem flying your airline.  If there are very few C class seats available, high C demand would be appropriate...

Here is an example of what I assume is a route working correctly:
https://www.airwaysim.com/game/Routes/Planning/X/KORD/KMSN/?TB_iframe=true&width=800

It looks like demand is split based purely on seats avail versus filling up planes based on frequency.  Please confirm this is how a route should be working when you can sami.

Quote from: ARASKA on July 05, 2012, 03:16:04 PM
I'm not asking for full planes, I would just like to have 60% LFs instead of 25% LFs.

Regards, Sean

It depends on the route supply and demand.  Some start with higher than 25%.  If you supply small percentage of demand at the beginning, you will have higher initial LF.

Quote from: LemonButt on July 05, 2012, 03:29:26 PM
Here is an example of what I assume is a route working correctly:
https://www.airwaysim.com/game/Routes/Planning/X/KORD/KMSN/?TB_iframe=true&width=800

It looks like demand is split based purely on seats avail versus filling up planes based on frequency.  Please confirm this is how a route should be working when you can sami.

Interesting.  I can see the supply, but I can't see the RIs if you and your competitor.  Under the old rules, everything else being equal (RI, CI etc), the 735 would be slaughtered.  So I am guessing that the "higher than optimal" frequency thing is kicking in....

Quote from: AAL558 on July 05, 2012, 03:14:34 PM
Anyway, Sami; I'm not sure if this is how its supposed to work but, I have a DEN-SLC route that has a RI of 70 and a LF of 80% while my DEN-LAS route has a RI of 60 and I'm getting a LF near 95%. Is this a bug?

This depends entirely on the demand of the route.

Route Image is a factor of people who know your route, and with RI 0 this is about 25-35% of the potential demand.

If you have a demand of 100 pax daily, you cannot get more than ~30-50 pax with RI 0 (depending on other factors). But if you have a route that has 5000 pax/daily demand, you can get rather full planes right from the start. At least now when there is no competition yet.

Quote from: LemonButt on July 05, 2012, 03:29:26 PM
Here is an example of what I assume is a route working correctly:
https://www.airwaysim.com/game/Routes/Planning/X/KORD/KMSN/?TB_iframe=true&width=800

Does not look too bad to me?  You have 11x daily with C208 with is clearly "too much" for that route.



Also, for any new actual game worlds the start money will be increased. This allows the airlines to run some losses while they build RI/CI.

I still see the strange behaviour with type-pax-distribution... I fly SFO-EWR at the same time with different types (ok, almost same time) ... but what I can see is, that larger aircraft get more pax... e.g. 320-> 55, 321-> 66 ... M83,M88,M90 -> all 30 ... on the other side I get the exact same amount for the M11 and 763 what I think is good... did you add a "bigger plane"-preference for the passengers?

I haven't run into anything odd yet. It seems to be working fine for me.

Quote from: meiru on July 05, 2012, 04:07:49 PM
I still see the strange behaviour with type-pax-distribution... I fly SFO-EWR at the same time with different types (ok, almost same time) ... but what I can see is, that larger aircraft get more pax... e.g. 320-> 55, 321-> 66 ... M83,M88,M90 -> all 30 ... on the other side I get the exact same amount for the M11 and 763 what I think is good... did you add a "bigger plane"-preference for the passengers?

Looking at your route, I am assuming that your RI is still on the low side.  With that, the system seems to work as it should.

I am guessing that when your RI gets to 70-100 range, your smaller aircraft will get full first, and than even the larger aircraft, since you are supply roughly matches demand...

Quote from: JumboShrimp on July 05, 2012, 04:32:17 PM
the system seems to work as it should.

Can you explain me, why I should get less passengers when flying an 320 instead of an 321? Or even less with the M90 instead of 320 ??

RI is at 60 at the moment... but this is a bit against the airplane-family concept... 77 on 321 -> my thought: ok, then I could use an 320 -> result: I only get 59...

Extending the lease -> why can the lease price go up, when I extend the lease? let's say, I want to extend it by 1 month... why do I then have to pay 807k per month instead of 540k before? I mean, it's longer... normally the price is lower then...

Quote from: meiru on July 05, 2012, 05:29:54 PM
Can you explain me, why I should get less passengers when flying an 320 instead of an 321? Or even less with the M90 instead of 320 ??

RI is at 60 at the moment... but this is a bit against the airplane-family concept... 77 on 321 -> my thought: ok, then I could use an 320 -> result: I only get 59...

Larger A321 size aircraft is more appropriate aircraft for the route than A320.  Since A320 is smaller than appropriate, it gains no frequency advantage on A321.

When aircraft has no frequency advantage, passengers are allocated purely by capacity. 

Quote from: meiru on July 05, 2012, 05:50:52 PM
Extending the lease -> why can the lease price go up, when I extend the lease? let's say, I want to extend it by 1 month... why do I then have to pay 807k per month instead of 540k before? I mean, it's longer... normally the price is lower then...

This is not new. If you renew or take a new lease for a short term, then the lease cost will be higher. Extending the lease for a long term will usually result in lower fees.

Quote from: NorgeFly on July 05, 2012, 06:06:27 PM
This is not new. If you renew or take a new lease for a short term, then the lease cost will be higher. Extending the lease for a long term will usually result in lower fees.

I'm talking about "extending" the lease... so, it's still running for 7 years and I want to extend it to 7 years and 3 months -> result: much higher price than before

Quote from: meiru on July 05, 2012, 06:27:48 PM
I'm talking about "extending" the lease... so, it's still running for 7 years and I want to extend it to 7 years and 3 months -> result: much higher price than before

Ok I see your point, that is a bit weird. But why would you want to make such a small adjustment so far in advance?

Quote from: JumboShrimp on July 05, 2012, 05:59:32 PM
Larger A321 size aircraft is more appropriate aircraft for the route than A320.  Since A320 is smaller than appropriate, it gains no frequency advantage on A321.

When aircraft has no frequency advantage, passengers are allocated purely by capacity. 

ok, but... you know... treat A320, M90 and 737 differently on these routes (and they are flown with these planes in reality) gives some aircrafts an advantage over others. I know, it's good to do that generally, but the types that are direct competitors shouldn't get different penalty values.

Quote from: NorgeFly on July 05, 2012, 06:30:47 PM
Ok I see your point, that is a bit weird. But why would you want to make such a small adjustment so far in advance?

I use this when I buy new jets... then I know when the arrive and I ajust the leasings of the older ones... (or because I don't want to pay for d-checks  ;D but fly the jet until this happens)

Quote from: meiru on July 05, 2012, 06:33:32 PM
I use this when I buy new jets... then I know when the arrive and I ajust the leasings of the older ones... (or because I don't want to pay for d-checks  ;D but fly the jet until this happens)

Ok I understand. I do it slightly differently, I set the leases to manual renew and then before the expire extend them by a few months if I need to. That way the high fees are only paid for a short while.

But we're drifting off thread topic now  ::)

Quote from: meiru on July 05, 2012, 06:31:12 PM
ok, but... you know... treat A320, M90 and 737 differently on these routes (and they are flown with these planes in reality) gives some aircrafts an advantage over others. I know, it's good to do that generally, but the types that are direct competitors shouldn't get different penalty values.

First, your SFO-EWR route has certain length and calls for certain kind of aircraft.  Shorter route would call for different (smaller) aircraft.

As far as penalties, there are no penalties as far as I can tell.  Only varying degrees of frequency bonus (or no bonus).

Someone should check into the HNL Routes. Seems to be an Issue there with Super Large amounts a PAX
|

https://www.airwaysim.com/game/Routes/Planning/PHNL/PHKO/
https://www.airwaysim.com/game/Routes/Planning/PHNL/PHOG/

Quote from: JumboShrimp on July 05, 2012, 06:59:49 PM
First, your SFO-EWR route has certain length and calls for certain kind of aircraft.  Shorter route would call for different (smaller) aircraft.

As far as penalties, there are no penalties as far as I can tell.  Only varying degrees of frequency bonus (or no bonus).

absolutly correct! ... but A320, M90 and 737 (the 700) are too close to get different results... that's what I say !

Quote from: meiru on July 05, 2012, 08:41:21 PM
absolutly correct! ... but A320, M90 and 737 (the 700) are too close to get different results... that's what I say !

You have a bunch of them departing at the same time, which is going to screw things up....
https://www.airwaysim.com/game/Routes/Planning/KSFO/KEWR

Quote from: JumboShrimp on July 05, 2012, 08:48:57 PM
You have a bunch of them departing at the same time, which is going to screw things up....
https://www.airwaysim.com/game/Routes/Planning/KSFO/KEWR

Not at all... if I do this with other types, they get the same amount of passengers. E.g. ATR-72 and MD90 on shorter routes... both got exactely the same amount of passengers.

And by the way... I do this to eliminate all other factors except the aircraft type. Like this I can check if the type is making any difference and it does... of course, that's good... but it's too much I'd say.

Quote from: meiru on July 05, 2012, 10:08:48 PM
Not at all... if I do this with other types, they get the same amount of passengers. E.g. ATR-72 and MD90 on shorter routes... both got exactely the same amount of passengers.

And by the way... I do this to eliminate all other factors except the aircraft type. Like this I can check if the type is making any difference and it does... of course, that's good... but it's too much I'd say.

But the time of day check is still disabled, as far as I know.  Spacing between the flights check is enabled.

My experience with inadequate spacing between flights is that all bets are off, the results will be screwy.

I'm not sure the tech-stop penalty is high enough / working as it should. On PVG > DUB I have a 772 and a tech-stopping 737. The 772 is claiming an average of 137 passengers, the 737 gets 107 (for a nice 85%+ load factor). That implies to me (although obviously I can't test this by myself) that if one airline flew a 772 and another 2x 737 with tech stops, the airline flying the 772 would get killed.

Speaking of which, I'm a bit concerned that a lot of the things that we want to test here require competition on routes to see how they work properly. But with only 75 airlines in the whole world I'm not sure how much of that we're getting?? If anyone wants to shift over to PVG and run some experiments happy to work with you to see how it turns out.

Quote from: sami on July 05, 2012, 04:07:39 PM
Also, for any new actual game worlds the start money will be increased. This allows the airlines to run some losses while they build RI/CI.

This won't necessarily fix my problem with the new system...     

Quote from: Glob-Al on July 05, 2012, 11:42:35 PM
I'm not sure the tech-stop penalty is high enough / working as it should. On PVG > DUB I have a 772 and a tech-stopping 737. The 772 is claiming an average of 137 passengers, the 737 gets 107 (for a nice 85%+ load factor). That implies to me (although obviously I can't test this by myself) that if one airline flew a 772 and another 2x 737 with tech stops, the airline flying the 772 would get killed.

There is a penalty for tech stop.  See here A321 vs. A321 with Tech stop that goes along with this post:
https://www.airwaysim.com/forum/index.php/topic,41433.msg223922.html#msg223922

And the penalty is not fully baked in.  It is going increase when flight duration is taken into account.

I think the 737 is getting way too many pax here regardless of the tech stop.  It would be way too many even if 737 was able to fly the route non-stop.

BTW, your test gives you enough info.  You can probably yourself add another 737 without exceeding the oversupply limit.  Testing it yourself removes all the other variables (different RI, different CI) it 2 players are testing.

Quote from: Glob-Al on July 05, 2012, 11:42:35 PM
Speaking of which, I'm a bit concerned that a lot of the things that we want to test here require competition on routes to see how they work properly. But with only 75 airlines in the whole world I'm not sure how much of that we're getting?? If anyone wants to shift over to PVG and run some experiments happy to work with you to see how it turns out.

You can also work with someone at a different airport, where 2 people are flying route between 2 HQs.

Is seat comfort significantly more important in the beta compared to previous versions?

I know the first post says this isn't yet modeled, but will it be changed from MT6/DOTM4/JA6/etc?

ConnectCanada Update

7 day scheduling is complete on 25+ A320/A321s and 14+ A330s. Some of my flight times are unrealistic (departures/arrivals in the middle of the night) since passenger preference for departure/arrival time isn't active yet.

Here's my beta plan going forward:

0-1500nm: A320-200 (10C 138Y 148 total)
1500-2920nm: A321-200 (15 C 168Y 183 total)
2920-5460nm: A330-300 (Factory default)
5460-7210nm: A340-600 (Factory default)
7210-8620nm: A340-500 (Factory default)

YYZ-MEL and YYZ-PER are the only major routes >8620nm but demand is only 150 and 90 pax per day.

Some A320s and A321s might be switched around but I haven't reached that point yet. I also might put an A321 on YYZ-LHR (limited to 144 pax) but I don't have enough aircraft to test that at the moment.

I should be able to fill all demand out of YYZ if the beta runs through 2015.

I'm struggling a bit to understand exactly what is going on from a pax distribution perspective. I've read the change logs over and over and they seem to make sense, but I'm find it to be far less forgiving than AWS has been traditionally.  Even with a higher amount of start money, I suspect there will be huge waves of bankruptcy as people figure out all of the changes that have happened the hard way once this is implemented in a full scale game world.

Here's an example that I don't think makes a ton of sense - LAX -> IAH.  Demand is ~2000 and I'm supplying ~1000 seats at 6x daily (30 minute intervals) on 3x A321's.  Two of them are configured 35/138 and the other is at the stock 192Y.  On the C/Y configed planes, I'm selling 81Y seats (for a 55-60% load) versus the all Y plane selling 56 seats (for a 30% load).  RI isn't very high yet (currently at 11), but it is sticking out like a sore thumb - my understanding is that I should be seeing similar loads across all planes of the same type, but it seems that actual seating configuration is having a rather drastic effect.

G4035-G4046 flight numbers. Is this expected?

Quote from: schro on July 06, 2012, 02:00:56 AM
but it seems that actual seating configuration is having a rather drastic effect.

apparently seating quality is not modeled yet on this server test... So, I'm having this confusion too, since all my HD flights are having less pax than the STD ones, and I can't figure exactly why...

Quote from: schro on July 06, 2012, 02:00:56 AM
Here's an example that I don't think makes a ton of sense - LAX -> IAH.  Demand is ~2000 and I'm supplying ~1000 seats at 6x daily (30 minute intervals) on 3x A321's.

Sami mentioned somewhare that he may be changing the minimum spacing of flights.  While 30 minutes was definitely safe in MT6 for 2000 pax route, who knows if it is still safe.

I would change flights to have 1 hour testing and see if it makes any difference.  Or change at least one flight so that it is not within 1 hour of another flight....

Quote from: JumboShrimp on July 06, 2012, 02:34:00 AM
Sami mentioned somewhare that he may be changing the minimum spacing of flights.  While 30 minutes was definitely safe in MT6 for 2000 pax route, who knows if it is still safe.

I would change flights to have 1 hour testing and see if it makes any difference.  Or change at least one flight so that it is not within 1 hour of another flight....

Well, he did ask not to monkey with it until thigns like this could be investigated, but its not behaving like a too close interval trap - i.e.

0500 - 81Y sold
0530 - 56Y sold
0600 - 81Y sold

If the spacing was too close, all 3 flights would be nuked - the 81Y sold seems to be consistent with what I've seen on other flights that I've done.

I'm also not liking that routes that used to be able to be used as "fillers" on a schedule aren't working well either. For example, dropping a 192 seat A321 into a 130 demand city that's less than 300 miles away results in about 50 seats getting sold. These additional nuances seems like it is taking scheduling from a livable pain to one that I don't have time to think through and implement in a successful  manner.

I'm having this problem with HD vs STD...

I have these 4 flights...

In these 2, it's all working good... same pax for the HD and STD configurations.
(https://www.airwaysim.com/forum/proxy.php?request=http%3A%2F%2Fi47.tinypic.com%2F5tz90z.png&hash=e7255c4f98ea432097b8fef256b42b61da349983)


But, in these 2 other flights I can't figure out what's going on (and why's the difference)... Also, notice the difference between the to and from flights... that should not happen...
(https://www.airwaysim.com/forum/proxy.php?request=http%3A%2F%2Fi48.tinypic.com%2F2z565pk.png&hash=20099a8296b3411149fb0b6408b254635e0e7326)

(470Y 20C for upper flights)
(628Y for bottom flights)


EDIT: the relation is good for the last 2 flights... the problem seems to be that trip with lower LF -TM564- (can't figure why)

Quote from: schro on July 06, 2012, 02:39:14 AM
Well, he did ask not to monkey with it until thigns like this could be investigated, but its not behaving like a too close interval trap - i.e.

0500 - 81Y sold
0530 - 56Y sold
0600 - 81Y sold

If the spacing was too close, all 3 flights would be nuked - the 81Y sold seems to be consistent with what I've seen on other flights that I've done.

I'm also not liking that routes that used to be able to be used as "fillers" on a schedule aren't working well either. For example, dropping a 192 seat A321 into a 130 demand city that's less than 300 miles away results in about 50 seats getting sold. These additional nuances seems like it is taking scheduling from a livable pain to one that I don't have time to think through and implement in a successful  manner.

These are only start-up issues, not long term issues.  When starting up, supplying small percentage of demand will result in higher LF.  192 pax aircraft on 130 pax route is definitely not a good idea when you start.

When you airline is up and running (and solidly profitable), these issues of low LF on low RI routes will be just noise...

Look for Sami's post on low RI:
"Route Image is a factor of people who know your route, and with RI 0 this is about 25-35% of the potential demand."

So on 2000 pax route, your potential market starts between 500 and 700 pax...

Some (potentially) weird issues on my CYVR-CYZF routes:

100% of C seats filled on the way to YZF and 0% C seat demand (and 0% seats filled) on the return route.

Quote from: AndiD on July 06, 2012, 03:59:42 AM
100% of C seats filled on the way to YZF and 0% C seat demand (and 0% seats filled) on the return route.

Did not quite understand this? Is there C demand both ways?



And for info to everyone; there will be some minor changes to the systems later tonight.

Quote from: schro on July 06, 2012, 02:39:14 AM
Well, he did ask not to monkey with it until thigns like this could be investigated, but its not behaving like a too close interval trap - i.e.

0500 - 81Y sold
0530 - 56Y sold
0600 - 81Y sold

If the spacing was too close, all 3 flights would be nuked - the 81Y sold seems to be consistent with what I've seen on other flights that I've done.

This does not appear to work properly, I have seen same with my ATR airline on one route. It reduces only the other route of the two overlapping.

Quote from: sami on July 06, 2012, 04:14:20 AM
Did not quite understand this? Is there C demand both ways?

I have a similar problem...

35.4% in the way RJFF and 69.6% in the return flight...

Quote from: sami on July 06, 2012, 04:14:20 AM
Did not quite understand this? Is there C demand both ways?

No, only on the route from the larger airport to the smaller one.

Quote from: JumboShrimp on July 05, 2012, 08:48:57 PM
You have a bunch of them departing at the same time, which is going to screw things up....
https://www.airwaysim.com/game/Routes/Planning/KSFO/KEWR

ok, statistics again... I fly all planes at the same time (+- 5 mins because I don't have enough slots) ... that's what I get:

M80/90 -> 37  (all get exactly the same amount!)
744, 772, 343 -> 46 (all get exactly the same amount!)
737 -> 54

... sorry, but it can't be correct like that...

Suddenly no demand between ZBAA - VTBD (Beijing to Bangkok)
Don't know if this is a bug or not.

T

No Demand or just no seats sold?

Quote from: meiru on July 06, 2012, 08:16:46 AM
ok, statistics again... I fly all planes at the same time (+- 5 mins because I don't have enough slots) ... that's what I get:

M80/90 -> 37  (all get exactly the same amount!)
744, 772, 343 -> 46 (all get exactly the same amount!)
737 -> 54

... sorry, but it can't be correct like that...

This is not a valid test at all in the first place as you have overlapping flights.

And as I did post already earlier today, the overlapping thingy is not working properly now.

Quote from: AndiD on July 06, 2012, 06:51:10 AM
No, only on the route from the larger airport to the smaller one.

That's how it is then. Demand is not the same on all routes... (should be in the same ballpark but differences are possible)

Used market coughed up 9 new aircraft in the last day - any chance to increase that with those sitting in storage? ;)

Quote from: michael on July 06, 2012, 08:31:06 AM
Used market coughed up 9 new aircraft in the last day - any chance to increase that with those sitting in storage? ;)

I guess sami wants us to use other a/c types as well :P

I don't know if this is thanks to the new system, or thanks to my competitors not supplying C/F seats on the route, but:

My gap filler flight LHR-CDG flying once daily (vs. >15 daily flights by either of my competitors) has a 100% LF in C-class :) Y is, despite massively lowered fares crap, and F is not available within Europe anymore apparently :/

cheers,
Jona L.

Quote from: sami on July 06, 2012, 08:27:09 AM
No Demand or just no seats sold?

No demand. Even got an over supply warning.

Thomas

Got some pretty amazing LFs going on! All these flights have no competition and Im currently under supplying. Have just increased prices by 20% to see effect

Well i've got a route of 350,000 a day on 772 to KMIA and the aircraft isn't making any profit even though it predicts in should be making around 500,000 per week. It is actually earning more per day by over 100,000 than my most profitable 772 which is raking it in. Got the same problem with some of my 762's, they are predicted to make profits but are not making them! Just seems that this isn't making sense! Can you take a look please!

I agree with Pukeko.

In previous games I never had 100% loadfactors and now I have, even with default prices and most important.... oversupply.

One flight is on 100% while the other one is 74%

Although this makes sense on itself I guess I'm baffled that the small plane is obviously the most popular as that is sold out and the airbus is not.

Here are some screenshots...

(https://www.airwaysim.com/forum/proxy.php?request=http%3A%2F%2Fi145.photobucket.com%2Falbums%2Fr240%2FYulesRo%2FBillundDemand.jpg&hash=488c70495c867547061f96bfa861cdcc3d7865b7)

(https://www.airwaysim.com/forum/proxy.php?request=http%3A%2F%2Fi145.photobucket.com%2Falbums%2Fr240%2FYulesRo%2FBillundLoadF.jpg&hash=cba04e6843a695038f79992b8a0a2b9516f59f0f)

Can somebody explain why people rather fly a small regional jet halfway the day over a comfortable baby bus early in the morning.

I bet that if you look at the actual number of seats sold, the A320 sells more, or as many as the E170.

= stop staring at the loadfactor people....  :P. You cannot judge anything from the route sales by looking at load factor, it tells nothing by itself.


And, checked, and correct. The small plane is NOT more popular. Airbus has sold more seats (partly mainly because the embraer is full, but they are both fully suitable planes to this route and should get exactly same sales if they would have capacity available)

And also. Read the initial post's information (too many times now when I have to remind of this ...  :-\). All randomization is off, and so is the flight time of day checking. (means that you will get constant 100% lf, which in reality indeed very rarely happens all the time, and flight dep time does not matter ..yet)

I just don't get how an airplane that shows 100% in every flight can average 75% LFs... 

Okay sure... I understand that this doesn't test functionality between plane types.... I am just asking how that math adds up.

Quote from: sami on July 06, 2012, 12:07:40 PM
I bet that if you look at the actual number of seats sold, the A320 sells more, or as many as the E170.

= stop staring at the loadfactor people....  :P. You cannot judge anything from the route sales by looking at load factor, it tells nothing by itself.


And, checked, and correct. The small plane is NOT more popular. Airbus has sold more seats (partly mainly because the embraer is full, but they are both fully suitable planes to this route and should get exactly same sales if they would have capacity available)

And also. Read the initial post's information (too many times now when I have to remind of this ...  :-\). All randomization is off, and so is the flight time of day checking. (means that you will get constant 100% lf, which in reality indeed very rarely happens all the time, and flight dep time does not matter ..yet)

I didn't say that anything was wrong with it :) I'm fully confident of your capabilities and my own incapabilities Sami :D

Was just wondering why the Airbus is empty and the Embraer is fully loaded.

Quote from: swiftus27 on July 06, 2012, 12:25:51 PM
I just don't get how an airplane that shows 100% in every flight can average 75% LFs... 

Umm.. Huh?

Quote from: sami on July 06, 2012, 04:24:58 AM
This does not appear to work properly, I have seen same with my ATR airline on one route. It reduces only the other route of the two overlapping.

Ok. I've changed the interval to 1 hour to see how that impacts it.

Quote from: sami on July 06, 2012, 08:28:18 AM
This is not a valid test at all in the first place as you have overlapping flights.

And as I did post already earlier today, the overlapping thingy is not working properly now.

... well, I don't comment the first one, but you're right. The overlapping-problem is messing up everything. If I distribute the flights over the day uniformly it works as expected.

Some changes have been made:

- overlapping flights checker is disabled until the issue is fixed.

- the math behind the "passenger's desired aircraft size for a route" has been adjusted a bit. It takes now into account long domestic flights (like intra-USA) and allows smaller equipment there. On the other hand intercontl. flight requirements have been increased (= penalty for 737 sized jets over atlantic or such). Also calculation is changed so that pax see the whole fleet of planes as similar, in other words the A318 and A321 would look the same to them.


MONEY IS ALSO RESET.

Quote from: Pukeko Airways on July 06, 2012, 10:33:44 AM
Got some pretty amazing LFs going on! All these flights have no competition and Im currently under supplying. Have just increased prices by 20% to see effect

Still getting 100% on these flights with the 20% increase. I guess this makes sense, if there is a big demand and the route is undersupplied by a monopoly (as well as no randomisation as Sami sayss), passengers will be willing to pay more. Going to increase prices by 20% again... and again and again to see the breaking point.

Yes, that's the whole point really. If there is undersupply and you have RI100 ("everyone" knows you fly), you can increase the prices.. But increasing prices makes people more unwilling to fly with you, and you may hike them up until you reach the sweet spot where you have planes full enough and get max profit still. With oversupply it's opposite then.

But this may not be so easy then anymore to see or use when all the other factors are added in.


I am also considering of making the demand grow gradually after game start, like slots. So there would not be tons of unfilled demand to begin with. But may be a tad difficult.

Quote from: sami on July 06, 2012, 03:25:36 PM
I am also considering of making the demand grow gradually after game start, like slots. So there would not be tons of unfilled demand to begin with. But may be a tad difficult.

That would make getting good enough LF's to profit from a small airport even harder.

Quote from: ARASKA on July 06, 2012, 03:29:57 PM
That would make getting good enough LF's to profit from a small airport even harder.

Obviously I would not slash 80% off a route that has 70pax demand. Wouldn't make any sense.. But makes more sense on the big routes.

Sami I've just been checking the demand from Moscow Sheremetyevo. I know most domestic flights are from Domodedovo but the 80 and 70 passengers to cities as Sochi and Perm and Tyumen seems a bit low.

Got no real information about this but I know Aeroflot flies those routes daily with at least a 737 or A320.

Just saying it, as I don't know how to get to the correct numbers, sorry.

My a/c screen:  I looked at one of my planes that showed an avg lf of 74%.

Clicking on that plane, it showed 100% lfs on every route. 

How is plane avg 74% lfs when every route is 100%

I don't like this penalty idea that you use to solve the frequency abuse problem... now I don't get pax on SFO-FLL (M90 and 320) but I saw, that it's flown with 319's in real world. So... I think, making the "overlap-area" wider for longer flights makes it more interesting to use larger aircrafts on those flights, because it simply is less expensive to operate... (in my solution you could use an other curve type or an other mixture of types).

Solving problems of unrealism with other non realistic ideas is like trying to push a light switch with a 4m bamboo cane. It's easier to try to move closer to the switch than trying to "improve" your tool.  ;D

Quote from: meiru on July 06, 2012, 05:16:04 PM
now I don't get pax on SFO-FLL (M90 and 320) but I saw, that it's flown with 319's in real world.

Your route image on that route is practically ZERO and you are also oversupplying the demand. That's the problem.  (= try with a single flight first)

Even when the RI is full, you cannot expect full planes due to the oversupply (305 supplied, demand is 180)

So. No thank you.

Quote from: sami on July 06, 2012, 03:18:55 PM
- the math behind the "passenger's desired aircraft size for a route" has been adjusted a bit. It takes now into account long domestic flights (like intra-USA) and allows smaller equipment there. On the other hand intercontl. flight requirements have been increased (= penalty for 737 sized jets over atlantic or such). Also calculation is changed so that pax see the whole fleet of planes as similar, in other words the A318 and A321 would look the same to them.

I am not sure if there needs to be distinction between intra USA and the rest of the world.  2 of the 9/11 aircraft were 767, 2 were 757.  Both larger than 737.  757 is still common on trans-continental flights.

I think it would be better if the code did not get convoluted with exceptions, and this sounds like one of them that may not even be necessary.  I think all we need is a fighting chance for the widebodies, so that they don't automatically get slaughtered by A320/737.  The way it can happen is when on longer distances, the smaller and smaller aircraft would not get higher and higher LFs...

Let me re-phrase. A "domestic longhaul" factor has been added so that you can also potentially use the 320/737/md-90 series there. But they are still considered "a bit too small" for those routes, and pax do prefer the bigger planes there too but just less than in intl routes.

ok, forget SFO FLL... the key point is -> your code will be a collection of thousands of special cases and it will be too complicated to work

No it is not, and will not be. End of that story, ok?

For example the long domestic rule is just one line of code, simple as "pts * 0.8" or that way. So thank you, I am very well capable of knowing how complicated it is or not.

Quote from: sami on July 06, 2012, 05:44:55 PM
No it is not, and will not be. End of that story, ok?

For example the long domestic rule is just one line of code, simple as "pts * 0.8" or that way. So thank you, I am very well capable of knowing how complicated it is or not.

Sami,

I admit I have not had as much time as I would like to build in the test world, but this question is still lingering. The "frequency bonus" (.8 above, for example)...does that mean that best case scenario I can only carry 80% of the potential passengers that would have flown on that route since the plane is "too small"? In other words, will that mean the flight is, in essence, artificially restricted to an 80% load?

Thanks!

Don

Quote from: jetwestinc on July 06, 2012, 05:50:38 PM
I admit I have not had as much time as I would like to build in the test world, but this question is still lingering. The "frequency bonus" (.8 above, for example)...does that mean that best case scenario I can only carry 80% of the potential passengers that would have flown on that route since the plane is "too small"? In other words, will that mean the flight is, in essence, artificially restricted to an 80% load?

That was just an example to show simple that part of the code is. Firstly, there is no separate "frequency bonus" even coded, or never has been. It's just a byproduct of the process and the updates made are ways to harness it a bit.

The 0.8 I mentioned above is just a factor in one variable, not relative to the entire demand in any way.

In other words with proper values (prices, RI, CI ..etc. etc, as usual) you can get the 100% demand to you, or even 130% of the demand if you are offering low prices or such (as mentioned the demand is not static and can grow there).

Quote from: sami on July 06, 2012, 05:35:20 PM
... and pax do prefer the bigger planes there too but just less than in intl routes.

Passengers select flights according to some rules... they don't like high prices and if they can chose, the take flights closer to the prefered time when the want to fly... now, with larger aircrafts you can provide smaller prices. That's why larger aircrafts work on those routes... not because of the fact, that passengers "like" bigger planes more... let's talk about the 757. She has the same fuselage like the 727 and the 737 had! So, passengers don't even "feel" the difference between a 737-200 and a 757-200 ... so, it can't be a criterion when selecting a flight... it's just an economical calculation.

So, when the departure-time-match get's more irrelevant and the price more relevant on those routes... you can't beat the large planes with high frequency and small ones.

Quote from: meiru on July 06, 2012, 05:58:41 PM
Passengers select flights according to some rules... they don't like high prices and if they can chose, the take flights closer to the prefered time when the want to fly...

Correct basically, but already out of the scope of this topic.

But since people "do it wrong" by using too small planes on too long routes, to simplify, there is a need to change it. In other words to make it more realistic, by for example discouraging people to fly A321 across Atlantic. Simple.

(One factor is the missing cargo too (among a few other things).)


But I guess it's better for me just to announce the updates, and won't discuss any technical matters here (and just perhaps publish guidelines in the manual), since most of my time goes into this talk now. So just saying: it will be done, it has been already done, and all what's left is some tuning in this part. After that the remaining variables affecting will be added, and fixing of the rest of the issues noticed and it's good to go.



edit; Also, bumped your FLL route's RI to 100, so you can see it "properly" now without the RI effect - should be all good.

edit2; checked it now. Demand for Wed is 176, you sold 90+90 => 180 seats => all demand + few extra sold. => Works just fine.

Quote from: sami on July 06, 2012, 06:02:46 PM
Correct basically, but already out of the scope of this topic.

But since people "do it wrong" by using too small planes on too long routes, to simplify, there is a need to change it. In other words to make it more realistic, by for example discouraging people to fly A321 across Atlantic. Simple.

I think, if the price (and maybe quality -> smaller plane seats should have lower quality than those on larger aircrafts) gets more weight, it will be solved automatically... at least I could imagine, since then the small aircrafts aren't the optimal ones anymore, because you can't offer so small prices.

Doing it "wrong" should simply cost you too much... but not generally mean that passengers don't like it :-) ... if you offer me a F flight from Europe to NY for just 10$ I would also take that... but it's hard to run an airline profitable like that.

I wasn't fast enough to get in on the beta, but I've been following the discussion. Is the "widebodies are better longhaul" calculation based on demand? Meaning, a 767 > 737 on BOS-LAX but not necessarily on BOS-EUG (or any transcon route with demand around 150-200) since the 767 is way too much airplane for that route...

There may be some changes to pax demand figures over the next game months; found a bug there.

Quote from: tvdan1043 on July 06, 2012, 06:35:44 PM
I wasn't fast enough to get in on the beta, but I've been following the discussion. Is the "widebodies are better longhaul" calculation based on demand? Meaning, a 767 > 737 on BOS-LAX but not necessarily on BOS-EUG (or any transcon route with demand around 150-200) since the 767 is way too much airplane for that route...

There really aren't many nations that have domestic flights of 2000nm or more. Just sayin... you're right though.  I fly from CLE to SFO on a 737...  to HOU on a 737 to LAS on a 737...

Quote from: swiftus27 on July 06, 2012, 05:14:07 PM
My a/c screen:  I looked at one of my planes that showed an avg lf of 74%. Clicking on that plane, it showed 100% lfs on every route. How is plane avg 74% lfs when every route is 100%

Screenshots needed, or at least some info on which plane this is .. etc...

Quote from: swiftus27 on July 06, 2012, 07:16:06 PM
There really aren't many nations that have domestic flights of 2000nm or more. Just sayin... you're right though.  I fly from CLE to SFO on a 737...  to HOU on a 737 to LAS on a 737...

And SJU, CUN, SJD on a 320, SAN (occasionally), etc. Another country with long domestic is China, some flights topping 1700nm. If small planes like the E-Jets are going to get penalized these routes will suffer (although your explanation above on how the pax allocation works seems to indicate these routes will be fine)

Don

Quote from: jetwestinc on July 06, 2012, 07:28:34 PM
And SJU, CUN, SJD on a 320, SAN (occasionally), etc. Another country with long domestic is China, some flights topping 1700nm. If small planes like the E-Jets are going to get penalized these routes will suffer (although your explanation above on how the pax allocation works seems to indicate these routes will be fine)

Don

if two flights were identical times and prices, people would always gravitate to the airline they know/trust first and the size of the plane second. 

Quote from: sami on July 05, 2012, 05:03:31 AM
You did if you were a member of that world still when it ended.

Uh, no, he probably didn't. At least looking at my credit history, the refund was never applied and I did play to the end (unless it was restarted at some point and I didn't notice, so the airline bk-ed). Then again, I figured that the two free weeks I got in JA6 (alongside everyone else) and the ability to look up future planes in the Test world and make plans accordingly did go some way to make up for this.

Seems that narrowbody vs widebody problem is still not solved  :-\
(https://www.airwaysim.com/forum/proxy.php?request=http%3A%2F%2Fi.imgur.com%2Fpnq3G.jpg&hash=9738339d9bb354bd68fbcd616bdedbc7847476b0)

Quote from: Pukeko Airways on July 06, 2012, 03:20:11 PM
Still getting 100% on these flights with the 20% increase. I guess this makes sense, if there is a big demand and the route is undersupplied by a monopoly (as well as no randomisation as Sami sayss), passengers will be willing to pay more. Going to increase prices by 20% again... and again and again to see the breaking point.

Still sitting at 100%, although flights with competition are suffering a bit. Time to increase by another 20%!!!

Quote from: l33ch86 on July 06, 2012, 08:13:16 PM
Seems that narrowbody vs widebody problem is still not solved  :-\
(https://www.airwaysim.com/forum/proxy.php?request=http%3A%2F%2Fi.imgur.com%2Fpnq3G.jpg&hash=9738339d9bb354bd68fbcd616bdedbc7847476b0)

I don't know the level of oversupply of that route, but it looks like a big improvement over MT6 for this set of flights.  In MT6, if each airline supplied 100% of demand, the A321 airline would have > 66% if tge market.  So it is not as bad, but still quite bad...

Quote from: l33ch86 on July 06, 2012, 08:13:16 PM
Seems that narrowbody vs widebody problem is still not solved  :-\
(https://www.airwaysim.com/forum/proxy.php?request=http%3A%2F%2Fi.imgur.com%2Fpnq3G.jpg&hash=9738339d9bb354bd68fbcd616bdedbc7847476b0)

At least you get 40% Market Share, in former game engine it would have been in the <30% area.

Have you checked the CI of both, yours and his company? maybe he gets a boost through that ;) Or he offers a lower price. As per what sami said, that impact should be bigger than in former games. I'd like to test that somehow. If anyone wants to try on a LH route to/from LHR we can discuss price settings through PMs ;)

My thoughts about frequency rape and solving it AWS-way...

1. In AWS crew requirements are a bit unrealistic, salaries for Very Large airplane pilots are about 50% higher than for Large AC (that's realistic I guess), but for each Very Large AC there needs to be about 85% more sets of pilots than for each Large AC (based on my data from MT6, each AC has 2 pilot crew). That means that flying wide-bodies will get you a really dramatic pilot + all the overhead staff salary increase (this means that Very Large AC also need more sets of cabin crew). If this is sorted out in a more realistic manner then Very Large AC = wide-body will get a bit more competitive. One option is to increase pilot-set requirements for smaller AC, other option is to calculate crew requirements realistically based on scheduled flights (if I understand correctly currently it's done so that each AC class has it's static crew requirement which is absolutely not dependent on if it flies 1 or 18 hours per day).

2. 05-00 flight window and other time-related stuff. Who the hell wants to fly at 05 in the morning anyway??? IRL you need to be at the airport like 1:30 before the flight and add at least 30 mins for commute and 30 mins for waking up. That means you'll be looking at waking up at 02:30. This flight window needs to be decreased somewhat. Currently it seems to be that demand is quite evenly spread out between 05-00, but it should be more precisely concentrated on rush hours. That way people could schedule bigger airplanes for rush hours and smaller AC for other times. This will also favor bigger aircraft.

3. Does average Joe really care about what AC is he flying on? I think that most of the people do not check before buying the tickets what AC will they fly on. Even I as aviation enthusiast will check the AC AFTER I have already bought the tickets. And I'm pretty sure that average Joe will not even do that. At best they will know the AC type when they board. I think that after the WOW-factor has gone (which is after couple of flights) even aviation enthusiast does not really care if AC is new E195 or B777. After all - with smaller AC you get less waiting for boarding and de-boarding. Time and price are what matter, not AC size.
If I understand correctly your current new formula favors A321 over A318, why is that? Why should A321 sell more seats than A318 when there is oversupply and there are not enough pax to fill the A318(-s)? I think that A318 and A319 (and 736 and 737) are already in disadvantage (see their production number in MT6), why make them even worse?

4. Penalty for tech-stopping when there are non-tech-stop flights available - this is really good and realistic!

5. Slot fees. Maybe in addition to cost of acquiring slots there should be also a fee for using it daily. I know that there are route fees but AFAIK those are based mostly on the weight of AC. Maybe you could change it so that there is a flat fee for using the slot and it would not depend on the weight or class of airplane (if the fee already exists then make it bigger) and also a portion of fee that is based on weight or class (like probably now). Also a small step to favor bigger airplanes.

To sum it up - I think you need to create economic factors for players to prefer bigger planes on certain routes not some silly formulas like "if route has demand X and length Y then ideal plane size would be Z" and if it's smaller or bigger then it gets penalized...

Probably not the best thread for the purpose but
my 2 cents behind the glass of wine :P

Quote from: Boot on July 06, 2012, 10:17:16 PM
If I understand correctly your current new formula favors A321 over A318, why is that?

that's not all... A318 will sell more than a 777/767/744 on short routes with heavy demand...

Quote from: Boot on July 06, 2012, 10:17:16 PM
5. Slot fees. Maybe in addition to cost of acquiring slots there should be also a fee for using it daily.

I sayed this on the other thread, but doing some forum search I've found it has been suggested several times, and all of them has been answered "it's not realistic"...

apparently making paxs select the aircraft acording to the seats it has is more realistic than leasing slots...

I just wanted to announce that I bought the oldest plane on the used market because why not? It's an 18-year old Tu-154M I rescued from the scrap heap. I'm going to fly it from Dulles to LAX in a ridiculous 50C/60Y configuration provided it doesn't crash on the delivery flight.

As for relevant discussion, looks like everyone crossing the pond into Dulles is flying narrow body except for Cream Airways with their 777s. I will look into getting some jumbos on those routes tomorrow to see how it runs. There's no competition right now.

Quote from: MA831 on July 06, 2012, 07:43:09 PM
Uh, no, he probably didn't. At least looking at my credit history, the refund was never applied and I did play to the end.

It is not visible in the history, just in the balance.

Quote from: ezzeqiel on July 06, 2012, 11:25:34 PM
that's not all... A318 will sell more than a 777/767/744 on short routes with heavy demand...

Incorrect.

On short routes they are equal.

Quote from: Boot on July 06, 2012, 10:17:16 PM
To sum it up - I think you need to create economic factors for players to prefer bigger planes on certain routes not some silly formulas like "if route has demand X and length Y then ideal plane size would be Z" and if it's smaller or bigger then it gets penalized...

Once again, do not focus on the techical aspects and what is happening at background but focus on the results.


And since people seem to misunderstand most of the concepts, I shall not (as already mentioned) discuss any more of the technical matters or formulas behind the system at the forum. The data on what is needed to know will be available in the manual once completed.

Quote from: l33ch86 on July 06, 2012, 08:13:16 PM
Seems that narrowbody vs widebody problem is still not solved

Cannot be judged alone from that image at all...  What is route image, what are the prices etc.

Quote from: Pukeko Airways on July 06, 2012, 08:16:12 PM
Still sitting at 100%, although flights with competition are suffering a bit. Time to increase by another 20%!!!

Forgot to mention yesterday. You are probably able to increase the prices a bit too much without hurting the sales. The effect of very high prices vs route overall demand may still be too small, but let me know when your sales start to decrease.

To continue my earlier experiment, I'm now running 2x tech-stopping 737 and 1x 772-ER on PVG > DUB.

YRD073/075 (737-700) averaging 76 pax per flight, average profit per return trip $31,292.
YRD037/037 (777-200ER) averaging 97 pax per flight, average loss per return trip $24,200.
YRD057/059 (737-700) averaging 76 pax per flight, average profit per return trip $31,292.

So despite the tech-stopping penalty and smaller aircraft penalty, an airline flying two 737s would be able to claim a 61% market share, whilst the airline flying the 772 would take 39% and be losing money hand over fist. And since the leasing cost for 2x 737s is less than for 1x 772ER I'm afraid the above figures would look even worse if you factored in all the other costs that aren't included on the route screen. So I don't think we've quite cracked this yet.

RI is 100. All this has been done with prices at default (or rather at what default was when I first set up the route, so probably now more like default minus 3 or 4%). I'm now going to experiment with cutting the price of tickets on the 772 to see how big a difference that makes.

UPDATE: 10% price cut led to 772 pax increasing by about 4, and each of the other two flights dropping by a couple. So a small change but not very sensitive to price. I'm going to try another 20% now to see if that has any impact.

Quote from: Pukeko Airways on July 06, 2012, 08:16:12 PM
Still sitting at 100%, although flights with competition are suffering a bit. Time to increase by another 20%!!!

Increasing to approx 180% of standard price was a bit too much... so i dropped price 20% and Lfs back to 100. Routes still have no competition and under served. Seems to work quite well.

Quote from: Boot on July 06, 2012, 10:17:16 PM
3. Does average Joe really care about what AC is he flying on? I think that most of the people do not check before buying the tickets what AC will they fly on. Even I as aviation enthusiast will check the AC AFTER I have already bought the tickets. And I'm pretty sure that average Joe will not even do that. At best they will know the AC type when they board. I think that after the WOW-factor has gone (which is after couple of flights) even aviation enthusiast does not really care if AC is new E195 or B777. After all - with smaller AC you get less waiting for boarding and de-boarding. Time and price are what matter, not AC size.

1. Specifically to your "aviation enthusiast" point: I check the AC before I buy the tickets; after, it's too late if I don't like the aircraft choice. This is especially important on either very long routes (because you will be on the plane for a long time) or on short routes (because airlines' increasing use of RJs means that you will not be able to roll a rollaboard aboard if it's an RJ, and on many routes, such as BOS-NYC, RJs are mixed in with standard narrowbodies so you have to be careful to check the fleet type).

2. Also to the aviation enthusiast point, FlyerTalk members frequently complain about airlines' use of narrowbodies across oceans.

3. To the Average Joe point, I think most probably won't know the name of an aircraft model, but will notice such things as whether it's an RJ (using high density seating which is standard on RJs), or whether the overhead bins are tiny. And more importantly, if they don't know the operational difference between an RJ and a standard narrowbody, they will get a bad impression of the whole airline, thinking "this airline has bad seats!"

4. In the specific context of narrowbodies across oceans, aircraft type obviously matters to Average Joes in this real-life situation, hurting the airline's CI and RI:
http://online.wsj.com/article/SB10001424052970203436904577152974098241982.html

That's it! The regulation of the market should work with price, passenger satisfaction, quality... all those factors... not artificial penalties.

Good idea - if you use RJ-s (or narrowbodies) on very long routes then your RI (and CI somewhat too) will suffer and that means you'll be able to sell less seats (if there is a competition with better planes and RI/CI). This way player will know that he is doing something wrong (sees that his RI is not climbing to 100).

Quote from: Boot on July 06, 2012, 10:17:16 PM
1. In AWS crew requirements are a bit unrealistic, salaries for Very Large airplane pilots are about 50% higher than for Large AC (that's realistic I guess), but for each Very Large AC there needs to be about 85% more sets of pilots than for each Large AC (based on my data from MT6, each AC has 2 pilot crew). That means that flying wide-bodies will get you a really dramatic pilot + all the overhead staff salary increase (this means that Very Large AC also need more sets of cabin crew). If this is sorted out in a more realistic manner then Very Large AC = wide-body will get a bit more competitive. One option is to increase pilot-set requirements for smaller AC, other option is to calculate crew requirements realistically based on scheduled flights (if I understand correctly currently it's done so that each AC class has it's static crew requirement which is absolutely not dependent on if it flies 1 or 18 hours per day).

Sami, I'd want to come back to this AC class vs crew requirements topic.

I made excel sheet based on my company in MT6:
http://www.orionis.net/~boot/pilot_costs.html

some comments:
1. planes column shows total number of planes of that type
2. pilots column is taken from "Personnel office" page
3. pilots per plane = pilots / planes
4. total pilot salary column is from "Personnel office" page
5. pilot cost per ac = total pilot salary / planes
6. cabin crew is from ac data page
7. total CC needed = cabin crew * planes * (pilots per plane / 2 = how many sets of pilots and CC are needed, every AC in my livery has 2-pilot crew)
8. CC cost per plane = total CC salary from Personnel page / total number of CC * total CC needed / planes
9. pilot + CC cost per plane = 5. + 8.
10. overhead to pilot + cabin crew cost ratio = for every dollar that is paid to pilot or CC $1.58 is paid to CEO/HR/whatever.

So this shows that direct salary costs for A330 are about 3.29 times bigger than for MD90 (which is roughly 2.3 times smaller than A330).
Additionally there are some really huge overhead costs but it's hard to guess how exactly are they distributed between different AC classes... Probably some portion of it is based on number of AC (which favors bigger AC) and some portion is based on number of pilots + CC associated with AC-s (which favors smaller AC).

Some easy questions:
1. Are my calculations correct?
2. If yes then do you think that numbers marked with red are reflecting real life situation?
3. If yes then do you plan to correct them?

Thanks!

Quote from: Glob-Al on July 07, 2012, 03:44:35 AM
To continue my earlier experiment, I'm now running 2x tech-stopping 737 and 1x 772-ER on PVG > DUB.

Settings tuning need some work still I see. Have to check closer later on.

Quote from: Boot on July 07, 2012, 12:42:46 PM
Sami, I'd want to come back to this AC class vs crew requirements topic.

Staff numbers are not under work here, they will be revisited perhaps for next version. So sorry, no time to check these any closer now. (The crew numbers are loosely based on real numbers though)

doh... is it so hard to understand that I have to spell it out for you?
You are trying to fix a problem - smaller aircraft beating bigger.
My calculated staff numbers show that this is one of the reasons why narrow-body AC are more profitable than WB - salary costs for 1 WB are higher than for 2 NB. IMO this is quite unrealistic.
If this is changed to more realistic numbers then it will make WB-s more competitive. Add tech-stop penalty (which you have already done as I understand) and maybe some additional little tweak and the problem is basically solved...

As I'm not participating in this beta and don't have credits to join it and not planning to buy any credits soon, can somebody test out the following scenario (or can you answer yourself Sami):
1. Demand 1000 pax/day.
2. 2 companies, one flying 5x A318 = 5*120 = 600 seats, other flying 5x A321 = 5*192 = 960 seats (so default configurations for both planes)
3. CI and RI for both companies = 100.

Will they sell same amount of tickets or will new system favor one of the planes/companies?
How does it depend on the route length (lets take for example 200, 500, 1000, 2000NM routes).

Quote from: Boot on July 07, 2012, 01:33:48 PM
My calculated staff numbers show that this is one of the reasons why narrow-body AC are more profitable than WB - salary costs for 1 WB are higher than for 2 NB. IMO this is quite unrealistic.

You seem to forgot completely that longhaul pilots are much less "efficient" than shorthaul pilots due to rest requirements etc. (and they actually use often 3 pilots in flights too, even with modern planes. ... not on most Trans-Atlantics though ..but let's say roughly, when flight time is over 8-9hrs, you need one more guy there, depending on company too)

So as already mentioned, staff is not an item on this update round. (But for example regional airline staff levels do need fixing and it is planned in future versions)


To your question. Both airlines should sell as much on the route, if all factors apart from plane type are the same. Regardless of route distance or any such thing. (seats sold is not a proxy of seats available)

Thanks for answers!

I know that LH pilots are less efficient but isn't 80% difference a bit too much?
And do LH flights IRL have also extra cabin crew members with them (in AWS it certainly seems to be modelled so).

I didn't have time to read all the other threads but now I found this:

QuoteI will also make a few changes to the calculations later today, by adding a factor if the route is dom/intl/longhaul which should improve this (and allowing "longhaul domestics" with narrowbodies) , and also adding a fleet group factor so 752 and 753 will be then equal (which they in pax'es eyes are). And it also makes the A321 worse since it's "comparable size" goes down since it will be similar to 320-319.

Does this mean that before this change A321 would have sold more tickets than A318 and after the change the numbers are equal?

Quote from: Boot on July 07, 2012, 02:07:40 PM
I know that LH pilots are less efficient but isn't 80% difference a bit too much?
And do LH flights IRL have also extra cabin crew members with them (in AWS it certainly seems to be modelled so).

Do not have any numbers here on my mobile browser, but a longhaul pilot flies only about 3-4 roundtrips a month, while a shorthaul guy does anything between 15-30 roundtrips monthly (depending on route lengths). (and add to that the possible 3 pilot need for longhauls too)

The problem is though that the system assumes a certain sized plane is used in a certain way, hence making it easier to use small plane in techstop-longhaul and bad to use big plane (like 777) in shorter routes. System is not that intelligent that it would calculate the exact schedules (and that may not be good either as players using manual staffing would need to make continuous changes, and other issues affecting playability).

so, an other disadvantage for widebodys... you know, the more I learn about the game engine, the more I think it's no "economy simulation" but an "airline enthusiast world simulation" ...
version 2 (if there will be ever a version 2) should be redesigned from scratch, otherwise you'll never be able to solve those problems...

Quote from: meiru on July 07, 2012, 04:06:16 PM
so, an other disadvantage for widebodys... you know, the more I learn about the game engine, the more I think it's no "economy simulation" but an "airline enthusiast world simulation" ...
version 2 (if there will be ever a version 2) should be redesigned from scratch, otherwise you'll never be able to solve those problems...

This IS an aviation enthusiast game, and that is what many others and I like about it. Should be kept that way ;)

Quote from: meiru on July 07, 2012, 04:06:16 PM
so, an other disadvantage for widebodys... you know, the more I learn about the game engine, the more I think it's no "economy simulation" but an "airline enthusiast world simulation" ...
version 2 (if there will be ever a version 2) should be redesigned from scratch, otherwise you'll never be able to solve those problems...

Its not just wide-bodies that appear to be dis-advantaged : staffing levels when running a fleet of small (-20) aircraft are a bit heavy too : but it is not a total simulation of an entire world-wide industry crossing every 't' and dotting every 'i', that would require something close to the entire computing power used for real in the entire industry and enough code-monkeys to make even AWS's hordes of route strategists seem a reasonable allocation of funds....

given the limitations : a browser-based game that doesn't require 24/7 presence or a T1 line to keep running : Sami has done an awesome job of even getting close to simulating this industry and, even when stuff can be irritating cos it messes up our own personal vision of the airline we would like to run, it's to Sami's credit he is working to improve it and address our concerns, which is way more than most online games ever even aspire to, never mind do.

Now, about these route strategisticalists....

Quote from: Boot on July 07, 2012, 02:07:40 PM
Thanks for answers!

I know that LH pilots are less efficient but isn't 80% difference a bit too much?
And do LH flights IRL have also extra cabin crew members with them (in AWS it certainly seems to be modelled so).

Yes, because the flight attendants sleep for part of the flight during LH flights (but not during SH flights), requiring an extra FA shift.

I have a suggestion to solve all many problems we have in just "not knowing" what we're doing wrong.  I think Route Image needs to be replaced altogether with a "Route Image Summation" metric.  Break down the route image into aircraft size, aircraft age/maintenance, flight frequency, comfort level, average flight duration, cancellation/delays, flight pricing, and all the other items that determine load factors.  Have each item worth 100 points (or weigh the most important metrics heavier) and then add them up for a total summation score.  On the route image page, show each metric and your score in each category.  This will help CEOs figure out their aircraft is too small on the route, which is something that could be found out in the real world by a CEO, but impossible in AWS without trial/error.  Once you have the summation metric, use this to determine load factors on routes instead of some complex polynomial with multiple "points of entry".  If there are issues in the algorithm, it will be easier to diagnose and will also help newbies figure out what they are doing wrong.

In the real world, CEOs have passengers complaining about seat comfort, cancellations, prices, etc. so why not make that data available to AWS CEOs to make better business decisions?

P.S. one of those metrics could be wide/narrow-body.  For those transatlantic flights, wide-bodies get extra points for the summation as a wide-body becomes the "appropriate" plane to use.

Quote from: meiru on July 07, 2012, 04:06:16 PM
so, an other disadvantage for widebodys... you know, the more I learn about the game engine, the more I think it's no "economy simulation" but an "airline enthusiast world simulation" ...
version 2 (if there will be ever a version 2) should be redesigned from scratch, otherwise you'll never be able to solve those problems...

Sorry, but if you really do not have anything more meaningful to post than this, then leave it.

You obviously have no information on how the background processes work, but you still keep posting these completely unrelated, false and disinformative messages. So stop derailing this topic now thanks - I shall not say this again anymore....

Don't let it get to you.  You will not be able to satisfy everyone.

Some people feel that you have a team of programmers shut in to a room in a nation where there is a Scandinavian/Nordic Cross Flag (of sorts) flying overhead.  They don't know that what we have in front of us now is the culmination of ~6 years (my best estimation) of really hard work by yourself and various dedicated gamers, fellow programmers, testers....  

With that said, are there holes?  Sure.   Are there things that need attention?  Yes.  Are there items missing from the game that should be in it?  Absolutely.  Hey, when City-Based Demand games with In-Air Entertainment and Club Lounges carrying tons of Cargo in their 772s start..... You will have 600 people trying to get into that game.  

I'm content knowing that you're trying to rebalance the sim.  I am a HUGE hater of frequency/narrow body wins.  Keep working at it.  It will take about 3-5 tweaks before we're close to a solution.  So many people fail to realize what permutations are.   Every time you add a variable to what is obviously a massive equation, you add an exponentially larger number of possible glitches/problems.  

Personally, I think you should have given an idea of mine more attention.  If you limit the number of slots that an airline can buy at a non-based airport, you will force them to use appropriate planes.  This is, as best as I can see, the best representation of what a real world airport would do.  They expect an airline to service demand while not taxing/burdening an airport with unnecessary air traffic.  

Anyway, get some sleep.  Its after midnight.

I'd lock the topic, and create the sim how you want it to be, and how you think it will be the most realistic and do what's best for the sim (like you've mentioned before).

(https://dl.dropbox.com/u/18530361/Screen%20Shot%202012-07-08%20at%203.06.22%20PM.png)
Bangkok, Test World Random Demand zero'd, then fixes itself days later????

With regards to FA numbers:

At BA, the 747s have 14 cabin crew, 777s (depending on cabin config and variant) have 10 to 13 crew and 767s have 10-11 crew.

These numbers don't change regardless of flight length. Most of the fleet (all 747s and most 777s) have crew bunks fitted. Those that don't have bunks fitted have high-comfort reclining jumpseats at the door areas. When it comes to crew rest mid-flight and between services, half the crew will rest while the other half works.

My point being, mid flight crew rest and ultra long haul flights don't affect cabin crew staffing levels as crew duties between services are lighter and don't require a full complement.

My $0.02

Cheers, Dan

Quote from: Nikliebz on July 08, 2012, 07:17:22 AM
Bangkok, Test World Random Demand zero'd, then fixes itself days later?

Being discussed at bugs thread already..  A bit of a mystery at the moment still.

Some more cash would be nice for us who are trying huge plane small route

Quote from: OldPilot on July 08, 2012, 05:02:31 PM
Some more cash would be nice for us who are trying huge plane small route

Making decent cash with B773/A333/A306 on all routes. Not a single loss maker. Though more cash is always good :) ;D


P.S.: exception: new routes with RI yet 0


BTW: Anyone who can beat this yet?
Flt #     Schedule          Load factors                        Avg LF %    Yield, ¢ / RPK    Revenue, ¢ / ASK    Aircraft    Profit     
LL389 EGLL - KMIA 0945-1435  Y: 100.0%C: 100.0%F: 100.0%   100.0%               12.96¢                12.96¢                         $ 546 743
LL390 KMIA - EGLL 1720-0730*Y: 100.0%   C: 100.0%F: 100.0%   100.0%    
               Days: ----5--

a/c: B777-300

Quote from: Jona L. on July 08, 2012, 05:08:07 PM
Making decent cash with B773/A333/A306 on all routes. Not a single loss maker. Though more cash is always good :) ;D


P.S.: exception: new routes with RI yet 0


BTW: Anyone who can beat this yet?
Flt #     Schedule          Load factors                        Avg LF %    Yield, ¢ / RPK    Revenue, ¢ / ASK    Aircraft    Profit     
LL389 EGLL - KMIA 0945-1435  Y: 100.0%C: 100.0%F: 100.0%   100.0%               12.96¢                12.96¢                         $ 546 743
LL390 KMIA - EGLL 1720-0730*Y: 100.0%   C: 100.0%F: 100.0%   100.0%    
               Days: ----5--

a/c: B777-300

Sure. Done. A332.

Has there been any recent discussion on departure punctuality? The most punctual airline is only at 78.2% while the median is only 56%. This is much lower than other scenarios.

About half of my delays are caused by weather, 25% by airport traffic and route restrictions, and 19% by scheduling.

Wild guess.. People are using old planes, lots of tech delays?

The calculation there is the same as before, it's just new code otherwise. So cannot rule out a bug either.

Good point Brian.  My airline is doing fine, except for punctuality, which drops to mid 50's at times.  This then effects my CI adversely.

My punctuality is currently 57%, and I am 21st out 75 players !!   ....so we are not the only ones with poor time keeping.

45% of my delays are WX related.  28% are due to busy airports, but all airports I fly to are more or less empty due to the lack of players.

Surprised nobody else has mentioned this with 50 players with 57% or below on time.

Quote from: sami on July 08, 2012, 08:19:46 PM
Wild guess.. People are using old planes, lots of tech delays?

The calculation there is the same as before, it's just new code otherwise. So cannot rule out a bug either.

A quick look at my punctuality shows only 55% on time with 0% due to technical problems, 67% due to weather. This does seem a little excessive to other games.

Punctuality does seem to be a problem and same as others it's weather affecting the most.... 55% of delays are WX related and 73.81% of cancellations are due to weather...

im guessing there's a bug in the code somewhere, hopefully a simple fix  :)

For me 44% was weather with scheduling at around 26% even though I have everything set to turn around at 1%, also after that I have 22% being Airport Traffic.

Similar issues here - I have a punctuality of 51.5% (!), with 72% weather conditions and 22% route and traffic restrictions. The remaining few % are divided between scheduling, tech problems and staff.

Same with flight cancellations - 7.6% with 70% (14) due to weather, 20% (4) due to scheduling, and 10% (2) due to tech.

Quote from: schro on July 08, 2012, 05:19:14 PM
Sure. Done. A332.

What is the config?!

okay, now I am late for work

Love the second one!

Quote from: Glob-Al on July 09, 2012, 12:59:18 PM
Love the second one!

I'm wondering how the E145s got to PHNL myself...

Quote from: schro on July 09, 2012, 01:37:27 PM
I'm wondering how the E145s got to PHNL myself...

Flew there?

I'm down to 53.7% punctuality and 30% percent of this is weather. My CI has also been affected by this and is at -1. 

Quote from: ARASKA on July 09, 2012, 02:10:39 PM
I'm down to 53.7% punctuality and 30% percent of this is weather. My CI has also been affected by this and is at -1.  

How about some advertising?

My CI is 92 and rising, despite 52.6% on time.... (71% of my delays [1931 mins] are weather).

Weekly result is +30M, spending 6.3M on Marketing.

Update:
Last week was +35M (if adjusted by newly leased a/c), spent 7.1M on marketing that week.

(another "for info".. Been stuck all day in MAD due to technical with the airplane, so updates to test world are delayed by few days... Cannot code with iPad and too many darn delays with this new game engine.. Lol..  :P )

Sadly I still don't think the efforts to change the pax allocation are working. On my PVG > DUB route, I have lowered the prices on my 777 flights significantly but still it's not enough to cancel out the frequency advantage. The 777 flight pulls in 105 pax per day whilst the two 737 flights between them get 150. This despite the fact that the 737s tech stop, take more than an hour longer each way, have lower pax comfort and cost about $200 more one way in economy class.

My observation from this is that a more radical overhaul of the allocation system is required. It seems to me that at the moment the system first divides pax demand by number of flights and then applies modifiers based on price, aircraft size, CI, RI etc. So if Airline A flies a route 3x daily and Airline B flies it 1x daily the system would default to giving Airline A 75% of the passengers, all other things being equal. It's very hard to set modifiers that coud ever cancel out that huge advantage for Airline A, without having a system that would make things unrealistically volatile when airlines were competing 1-1.

But how about if the starting point was flipped on its head and the initial allocation was made based on number of airlines flying the route (so with two airlines on a route all other things being equal both would take 50%) and then modified to take into account the other factors mentioned above PLUS frequency. Frequency would be more important the shorter the flight, pax comfort more important the longer the flight, and CI / price equally important no matter what length. In this way frequency could still be a consideration but it would be a modifiable variable that could be tweaked along with the others to find the best outcome - rather than something every other variable was trying to counteract.

I realise this might well require such a radical rewrite of the code that it couldn't be implemented now but I'm really starting to think it may be the way to go in the medium term. Perhaps in the meantime swiftus' proposal about the number of slots an "out" airport will allocate you could be used as a stop-gap solution for the next few GWs (presuming that would be easier to code, which I realise it might not be!) But if I'm right about how the system currently works (which I also accept I might not be!) then I fear tinkering with the current model can never be as effective as we'd all like.

The flight time, and seat comfort are not even modelled there yet. And further, I have not taken a look at the calculation on that route yet like mentioned earlier (ie. if there is something odd .. The results surely do not correspond to the other routes that well, s may be a bug too?), so nothing has changed in that sense during last three four days. And like I mentioned earlier, seems that the settings need tuning and I will attend to it during this week.

Also .. The calculation does not work in the way you thought. Instead it compares all flights equally based on their data and then distributes the passengers - instead of first allocating passengers based on number of available flights or so. But I will check these things later indeed.

Quote from: Glob-Al on July 09, 2012, 04:16:22 PM
Sadly I still don't think the efforts to change the pax allocation are working. On my PVG > DUB route, I have lowered the prices on my 777 flights significantly but still it's not enough to cancel out the frequency advantage. The 777 flight pulls in 105 pax per day whilst the two 737 flights between them get 150. This despite the fact that the 737s tech stop, take more than an hour longer each way, have lower pax comfort and cost about $200 more one way in economy class.

My observation from this is that a more radical overhaul of the allocation system is required.

I think the change is really simple:  If the aircraft is smaller than optimal, pax are allocated strictly on number of seats available.  Number of flights (frequency) should play no role.  

Despite a lot of tweaking, Sami did not turn off the frequency boost on aircraft that is too small for the route.  A lot of things were done to mask this frequency boost but the boost is still there.

The end result should be that if you fly "ideal" 300 pax aircraft and competition is flying 3x100, you should be allocated the same number of pax.  (same LF).  No matter how small the competitor aircraft, no matter how many flights he flies, his aircraft shoujld never exceed LF of the ideal aircraft...

As long as LF of smaller aircraft exceeds LF of larger aircraft, the frequency issues will be with us...

Sami... will you model people likes/deslikes regarding aircrafts on this new version? For instance using newer ac's vs 20yrs+ ac's competing on the same route? I want to believe that our virtual passengers prefer to fly brand new A330/340 or B777 to old DC-10 and B742/3.

Quote from: digifreak on July 10, 2012, 12:53:12 AM
Sami... will you model people likes/deslikes regarding aircrafts, for instance using newer ac's vs 20yrs+ ac's competing on the same route? I want to believe that our virtual passengers prefer to fly brand new A330/340 or B777 to old DC-10 and B742/3.

The amount of MD-80s I see flying out of Tampa alongside brand new A-320s says otherwise. People have no idea they are 20 years old with a fresh coat of paint and refurbished interiors with modern entertainment. The only thing that makes them stand out are the streaks of black exhaust they leave behind on takeoff and the resulting black stains behind the engines.

Quote from: digifreak on July 10, 2012, 12:53:12 AM
Sami... will you model people likes/deslikes regarding aircrafts on this new version? For instance using newer ac's vs 20yrs+ ac's competing on the same route? I want to believe that our virtual passengers prefer to fly brand new A330/340 or B777 to old DC-10 and B742/3.

Passengers in the real world tend to judge an aircraft's age by its cabin interior. You wouldn't beleive the number of people that I've had guess that a 35 year old DC-9-51 or a 20-24 year old MD-88 be less than 10 years old based on interior appearance.  I then proceed to tell them its real age, and they're white as a sheet for the rest of the flight.

I personally prefer the old Douglas jets. I'd take a Mad Dog over a Boeing or Airbii any day (in fact, I took a pair of MD-88's today - N919DL and N960DL).

Quote from: Troxartas86 on July 10, 2012, 01:04:16 AM
The amount of MD-80s I see flying out of Tampa alongside brand new A-320s says otherwise. People have no idea they are 20 years old with a fresh coat of paint and refurbished interiors with modern entertainment. The only thing that makes them stand out are the streaks of black exhaust they leave behind on takeoff and the resulting black stains behind the engines.

Hey now, the black stains are far more noticable on the DC-9-51's. The MD-80's are clean as a whistle in comparison!

INFO:

Route history data is cleared. Making sure the delay data bug is fixed...

Also, $100mil given again to anyone below that.

Found a bug that affected for example long and thin-demand routes (intl/longhaul). Also still tweaked some settings, and you shall see some changes in the longhaulers now where there is a small vs. big aircraft situation present. Further settings tweaking may still be needed (for example regarding C/F class), but could be taken care of with the addition of seat comfort (etc) variables later on.

Also the route overlapping rule is effective again.

And do not also wonder if the test world time jumps back and forth a bit. It's me running it manually.

Ouch. My A321's on KLAX-EGLL got nerfed with that change.  I was varying between ~80% loads and 100% loads before the change, now I'm seeing 50-60% loads.  Pricing is about 25% above standard and I haven't tweaked it yet.  Demand is right around being met...

Looks a lot better on LH to me Sami.  :) My PVG > DUB route is now seeing ~165 pax on the 777 versus 25 on each of the 737s. This is however with the triple seven prices much cheaper so I'll reset them to the same just to be sure. Also that means that only about 210 of the 290 daily demand are actually flying, despite the fact there's no competition, excess capacity, prices below default and RI of 100. That might just be a statistical glitch though - I'll take another look later on.

Cheers,
Glob-Al

Quote from: schro on July 10, 2012, 10:07:39 PM
Ouch. My A321's on KLAX-EGLL got nerfed with that change.  I was varying between ~80% loads and 100% loads before the change, now I'm seeing 50-60% loads.  Pricing is about 25% above standard and I haven't tweaked it yet.  Demand is right around being met...

I cant notice any change in my 340s & 380s flying that route - but then again my scheduling on that route is just crazy so it would be hard to find anyway

Sami,

Could you check my 2 routes?

RED EYE AIRLINE - https://www.airwaysim.com/game/Info/Airline/52/#AirlineInfo

Boston Logan - Guayaquil - RI 100, DEMAND ~160 - A321 - LF ~13%

Boston Logan - Quito Mariscal Sucre - RI 100, DEMAND ~ 150 - A320 - LF ~15%

Both routes 1 flight per day, normal time, no competition at all, default prices.

Quote from: markj23 on July 11, 2012, 01:53:49 AM
I cant notice any change in my 340s & 380s flying that route - but then again my scheduling on that route is just crazy so it would be hard to find anyway

You shouldn't. 340 or 380 ... Makes no difference.

Quote from: Mr Pepto on July 11, 2012, 08:17:36 AM
Could you check my 2 routes?

~3000 nm routes (= international longhaul) flown with a narrowbody plane is the main reason.

If the airlines are offering a "poor" product some passengers will elect not to fly at all, so you are not reaching the full demand despite of RI100.

That doesn't sound right, out of 160 people only ~25 decide to fly and 135 prefer to stay home? Just because this guy is flying A321?

The "true demand" set by the calculation module on that route is 95 (of 165), so about 60 decide not to go because of this. I'm checking where the rest have gone. :P

The passenger demand is not a line of passengers waiting at the airport. It's a number of maximum potential passengers.

Quote from: sami on July 11, 2012, 01:24:34 PM
The "true demand" set by the calculation module on that route is 95 (of 165), so about 60 decide not to go because of this. I'm checking where the rest have gone. :P

Sami, I wonder if the issue of your missing passengers is also affecting routes where there is competition. Since the last round of changes my PVG > DUB route has experienced a surprising trend:

- When there was 1x777 and 2x737s per day the total number of passengers flying was about 160 + 25 + 25 = 210.
- Now there is only 1x777 flying (I cancelled the 737s) the total number of passengers flying is about 245, despite the fact I also raised prices about 20% (to just below the default).

It sort of seems like before there were some passengers who were choosing not to fly the 737 because it would be inconvenient, but didn't "see" the 777 as an alternative option.

I still don't think this leads to a good end with the pax distribution...

Anyway... I wanted to ask something else: MD-80 has a cabin crew of 3, the MD-90 of 4 ... why? Don't they have the same capacity?

Quote from: meiru on July 11, 2012, 04:26:58 PM
MD-80 has a cabin crew of 3, the MD-90 of 4 ... why? Don't they have the same capacity?

There's no "MD-80" .. pls define the variant, since they are rather different sized machines (87 vs 83 for example).

May be a bug too though.


Edit: Confirmed to be an error. Fixed!  (since MD-83 for example can seat max 172, so that needs 4 Ca's ... )

Quote from: meiru on July 11, 2012, 04:26:58 PM
I still don't think this leads to a good end with the pax distribution...

Anyway... I wanted to ask something else: MD-80 has a cabin crew of 3, the MD-90 of 4 ... why? Don't they have the same capacity?

The M90 is a bigger plane than the M81/82/83/88 (all 4 are the same size with avionics differences) and is usually configured with 150-160 seats in the real world, even though it shares the same max certified 172 pax load as its smaller predecessors. In AWS, the max certified drives the number of seats available, so the M80's and M90's have the same capacity.  If the M90's had a proper capacity that reflects them being longer, then the 4th FA would be needed, however, due to how the seat config work, we're not going to get the M90 capacity buffed.. therefore... it seems to make sense to have 3 FA's on the 90 until then..

edit: and sami updated it to my suggestion before I could even post it.

Quote from: sami on July 11, 2012, 04:28:37 PM
There's no "MD-80" .. pls define the variant, since they are rather different sized machines (87 vs 83 for example).

When I'm comparing an MD-80 family member to the MD-90, it should be logical that I'm not talking about the MD-87... by the way, there's also an other bug, since it was possible to buy the MD-88 with the additional fuel tanks the MD-83 had. This variant has a higher takeoff weight (same as 83) but the other advantages of the 88... nobody ever ordered it, but it was offered. Just to mention that too...

... and about fuel consumption. I always hear/read/calculate, that A320's (not the neo) burn more fuel than in your game... so, my information show's this for a the same flight -> 738 100%, M90 102.75%, A320 106.02% ... in your game the A320 is the most fuel efficient aircraft.
I even found statement's of airlines, that they decided for the A320 alltough they have a little bit a disadvantage in fuel consumption against competitors but it's bether for other reasons... of course, it's always difficult to find out, if this information is correct. But to me it seems that it's most likely correct.
To give an example: the A320 with V2527 is heavier and faster than the MD-90. Experts say, the A320 fueselage has an aerodynamical disadvantage over the MD90 (because it's wider) and the MD-90 has with it's V2525 an engine that's 2 years younger and producing less thrust... so... there's not a single point that point's towards the fact, that the A320 could have a lower fuel consumption than the MD90.

By the way... I completely agree with the MD-11 consumption... but the MD-80's are way too high... take the 727 with 3 engines (old JT8D's) and compare it to the MD-88 ... there's at least 13 years between those engines and the MD-88's engines consume now 10% more than the old ones? ... it was almost a redesigned engine! That can't be correct... sorry...

The fuel usages of most modern planes in database are from reliable sources. For older models there is some 'artistic freedom' involved due to lack of full data. So if there is something wrong, some really hard facts are needed (= numbers, not just airline or plane manufacturer press release "lies").

(and not relative to this topic; so goes to bugs then)

ok, fine...

by the way, I didn't give you the exact numbers you should use, but I pointed out, that your number's can't be true, since it's impossible that a JT8D-200 series engine used MORE fuel than a JT8D-15 and that's what your game says... (the same for the MD-90s, I simply showed, why it can't be true)

that's no "lies" or press releases... I mean, why did MDD use the JT8D-200 series in the MD-80's and not the 17R, if the 17R was bether? and why did MDD and Boeing talk about a 15% more fuel efficient airplane (the MD-90 over the MD-80) and not about 20% like we see in your game? ... do you think their marketing guys were that stupid? ... but I see, you know it bether... like always ...

Read what I wrote previously and please do not argue yet again. If you have better facts available than the current data, then submit a bug report and I shall correct the data based on that. But, like on every other data correction, MUST have some facts on it to which I can base the corrections.

Quote from: meiru on July 11, 2012, 07:42:56 PM
ok, fine...

by the way, I didn't give you the exact numbers you should use, but I pointed out, that your number's can't be true, since it's impossible that a JT8D-200 series engine used MORE fuel than a JT8D-15 and that's what your game says... (the same for the MD-90s, I simply showed, why it can't be true)

that's no "lies" or press releases... I mean, why did MDD use the JT8D-200 series in the MD-80's and not the 17R, if the 17R was bether? and why did MDD and Boeing talk about a 15% more fuel efficient airplane (the MD-90 over the MD-80) and not about 20% like we see in your game? ... do you think their marketing guys were that stupid? ... but I see, you know it bether... like always ...

Bro, you're being borderline douchetastic.  Tone it down a notch.  There's been a long standing rule in the sim that you need concrete data to change anything (from seat max, fuel consumption and more). 

Quote from: swiftus27 on July 11, 2012, 08:49:04 PM
There's been a long standing rule in the sim that you need concrete data to change anything (from seat max, fuel consumption and more). 

of course... but he can use the 'artistic freedom' because he didn't find data...

Quote from: meiru on July 11, 2012, 09:05:31 PM
of course... but he can use the 'artistic freedom' because he didn't find data...

Well, since you still keep arguing and deliberately(?) misunderstanding everything I say:

Tell me for example the fuel consumption of a Baade 152 East German jet, or perhaps Vickers VC10 as that was bit more common. And to make it "easier" (= data what AWS requires, but makes it accurate in calculations once data is accurate), tell that plane's fuel consumption for climb between GND-FL100, FL100-FL240, and FL240->, and then cruise fuel usage at max cruise FL, and same step plan data for descent too. Cannot find it? Okay, that is what I thought too. And that is what I meant there by needing some freedoms to derive the data when no data is available.

It's easy to throw these comments since you do know what kind of data is even needed, and how impossible it is to find it to any historical aircraft. The fuel usage data for example most certainly is not just "500 kg / hour" type of thing like you could imagine but uses a more realistic approach into the calculations.

Any more messages of this issue in this topic will be removed, since it's now fully off-topic.

AWS isn't a democracy.   I probably have 20 idea suggestions with [-] next to them. 

I found some data... but it was too much to review it that quickly... I'll post the info in some days... I found very detailed information from different sources like ICAO...

OK, just  few observations. I'm not sure if any of these are implicitly due to the changes (I've come across similar before in other games worlds), but I'd thought I list them as I've gone from making $25 million a week to loosing $25 million a week. Of course this seems to correspond directly to the fuel surge, however the results still seem a bit strange to me.

I operate Big Dipper Air https://www.airwaysim.com/game/Info/Airline/65/#AirlineInfo out of New York. I've concentrated on a fleet of long haul - A340 and A380s with 737-900 serving the domestic market. I run all the long hauls on 7 day schedules and have consistently seen LF's in their 90s (many at 100) since I started (avg all up is 87%). All schedules are jam packed with optimal turn around (ie. 1% chance of delay) giving a fleet utilization of 18.7 hrs per 24.

I've mucked around with prices a bit, however have been running default prices for game year or so. With the recent surge in fuel, it appears the long haul planes have to be flying to 5 destinations a week to earn a profit, I have planes with 95% LF flying to 4 destinations (with a completely full schedule) and they are loosing money.

Now I know running Long Haul is difficult (in the real world and the game world), but it does look like plane ownership is the only way when fuel prices rise above $1000. However, I'm sticking to the leasing model to see if I can turn it around by increasing prices on routes with 100% LF - I have managed in this game world to increase prices to 160% of default with no detrimental effect on LF, passengers were not too keen on 180% though. Decreasing prices on competitive routes seem to have little effect on LF.

Here is some data for me, proving to me that it does really work the new system.

I'm based in Newcastle in the UK. There is 120 demand to Chicago and JFK and it's within 3400 nm. So I leased two 737-700 and fly there. Not only do I get the message that the airplane type is too small, load factors are only around 10% max.

So this feature of the new things really work. I really think this is a good thing.

Quote from: Dasha on July 12, 2012, 07:49:17 AM
Here is some data for me, proving to me that it does really work the new system.

I'm based in Newcastle in the UK. There is 120 demand to Chicago and JFK and it's within 3400 nm. So I leased two 737-700 and fly there. Not only do I get the message that the airplane type is too small, load factors are only around 10% max.

So this feature of the new things really work. I really think this is a good thing.

Agreed, however this also limits the viability of smaller airports like NCL as bases. With the fix now working like it should and a warning alerting the user to the fact that a plane may not pull a full load due to its size, can there be an exception where as long as an airline is the ONLY airline flying the route a full load is still possible? I am thinking of my MT6 airline out of JNB, for example. I had a fleet of over 100 737NG's. ~63 of them were -900ER's or -700's. The -900ER flights -could- have supported "proper" widebody equipment-the loads would have been lower, but probably still profitable. But, the 7 -700's and 63 -700ER flights I operated were using that equipment ONLY because a larger aircraft would not have been supported. I am talking routes like JNB-CAG and other small international airports in Europe, Western Asia, and a couple in Eastern Argentina and Brazil. The loss of ability to fly these routes would not adversely affect the playability of a city like JNB (may make it less interesting, but not ruin the city) but a city like NCL will benefit from the ability to fly these long, thin routes.

Don

Quote from: sami on July 11, 2012, 01:24:34 PM
The "true demand" set by the calculation module on that route is 95 (of 165), so about 60 decide not to go because of this. I'm checking where the rest have gone. :P

Found the missing passengers.  ;D

In other words, you can fly the long routes with smaller planes too provided there is no competition. However as mentioned some people will choose not to fly in that case, hence "true demand" is less what the graphs show. (and again, if you provide extra low prices you can attract more than the graphs show..)

INFO:

I'm slowly adding now also the effects of other variables into the calculations, so you will notice some changes.

Flight departure and arrival times, and penalty for too small frequency (ie. running a domestic destination like only once a week) are added now.

Quote from: sami on July 12, 2012, 05:10:40 PM
I'm slowly adding now also the effects of other variables into the calculations, so you will notice some changes.

is there such a thing as frequency too high?  Let's say I have 20xATR per day on a 1000 pax route ~ 300 nm and my competitor has 5xA320.

The old way would result in each aircraft getting 40 pax, with ATR winning by having ~66% LF, A320 in 25% range.

The above scenario is as much or more of a problem as frequency exploit on LH routes.  It leads to slot hogging and number of small aircraft that is 5x higher in MT games than in real world...

Quote from: JumboShrimp on July 12, 2012, 05:21:59 PM
is there such a thing as frequency too high?  Let's say I have 20xATR per day on a 1000 pax route ~ 300 nm and my competitor has 5xA320.

Yes, it's been modelled already.

For let's say 1500 / day demand, more than 7 daily flights is not optimal anymore.

So, it seems that the penalty for running 1x daily on a 120-140 passenger route is about 40-50%.. seems pretty significant.

Quote from: sami on July 12, 2012, 05:24:54 PM
Yes, it's been modelled already.

For let's say 1500 / day demand, more than 7 daily flights is not optimal anymore.

I'm not in beta, but I am wondering what will happen on routes like HND-ITM/CTS/FUK or HKG-TPE, having more than 10000 demand and 20+ daily in real life......

Quote from: schro on July 12, 2012, 05:42:34 PM
So, it seems that the penalty for running 1x daily on a 120-140 passenger route is about 40-50%.. seems pretty significant.

Pls be more specific so I can check.

Quote from: sami on July 12, 2012, 06:39:09 PM
Pls be more specific so I can check.

Here's one of routes - I was selling 116/day and it has dropped to 56/day after the change.  VGT and SCK have seen similar drops after the change. 

I'm worried about a few 2000-2500nm routes, specifically those from the northeast US/Canada to northern South America. In real life, many of these routes are operated by small AC (A320 and B737 families) but in AWS they receive the "Too small AC" warning.

I'm not in JFK/EWR, so I can't see if these routes have the same warning.

(https://www.airwaysim.com/forum/proxy.php?request=http%3A%2F%2Fi1015.photobucket.com%2Falbums%2Faf272%2FBryanIAH%2FAWS%2FScreenShot2012-07-12at15203PM.png&hash=426bf9d3207a84c8f023ab39160c12177743296f)

Take YYZ-BOG as an example. In real life, Air Canada uses a 767 for this route and AWS has a penalty to use an A320. That seems somewhat reasonable, given that a smaller AWS 767 would fill the 230 seats.

Then look at YYZ-CCS. In real life, Air Canada uses an A319 for this route but AWS has a penalty to use an A320. Since AWS demand is 360, it is reasonable to expect us to use a widebody.

I know it is impossible for Sami to compare every route to real life.

This new small AC penalty starts to fall apart with smaller routes. YYZ-GYE demand is only 160/day, but there is a penalty if I want to use an A320. This isn't a problem for me (the YYZ-based airline), but an airline based in GYE would need to operate this route with an A320 since their options are more limited in a smaller airport.

YYZ-POS is longer than YYZ-CCS but it doesn't have an A320 penalty even though YYZ-CCS does. In real life, B737s and B767s are used on YYZ-POS and AWS allows me to use an A320 without penalty. POS and CCS are both listed under AWS' South America category, so I don't understand exactly what is going on here.

AWS also penalizes A320s on YYZ-CLO, a route with only 100 demand/day. This would also harm smaller airlines in CLO if they wanted to operate longer routes with A320s or 737s.

Quote from: type45 on July 12, 2012, 05:52:55 PM
I'm not in beta, but I am wondering what will happen on routes like HND-ITM/CTS/FUK or HKG-TPE, having more than 10000 demand and 20+ daily in real life......

My guess is that you will get credit for any capacity you put on the route.  But you will not get extra LF boosting credit for more than 7 flights per day  The extra flights will be just capacity.

Well, at least that's my interpretation.  I am not sure how Sami implements that...

Quote from: schro on July 12, 2012, 06:59:00 PM
Here's one of routes - I was selling 116/day and it has dropped to 56/day after the change.  VGT and SCK have seen similar drops after the change. 

Try dep at 0700 instead?

Couple of observations today...

Red eye departures are now not viable....LF's of 10% and less....some even 0% with a 0115 departure.

ATL - DUB demand is around 170, a perfect 757 route IMO. With a RI of 100 LF is hovering around 45% and with no competition.  I do not have enough fingers and toes to count how many 757's cross the pond daily in real life, MAN alone gets 2 or 3.  Continental's (United)  european network is full of 757's.  There is no reason 757's in AWS should suffer as noted.  There is a big difference between a direct 757 and a tech-stopping 321.

Is there any disadvantage at scheduling a LH / ULH flight at 1-2am as it is in the current version? I'd love to enter the testing game... :P

Quote from: sami on July 12, 2012, 07:48:17 PM
Try dep at 0700 instead?

Drop of 133 to 72 on the VGT route which is at 9:30 on the same frame. SCK dropped from 96 to 54 with a 13:25 departure.

Just tried to open ATL-GLA, again with a 757 which is a perfectly reasonable a/c type choice, and get a "too small a/c" warning.  That's just daft !!

Just checked UA's real life schedule an there are at least 4 x 757's in LHR daily, 1 in BHX, 2 in MAN etc etc.   They go as far as Stuttgart and Hamburg from EWR, both around 4000nm.

757's are without doubt long haul a/c for thinner routes, if you call EWR-LHR a thinner route !   Or does frequency in real life really count....and 4 x 757's is better for the CUSTOMER than 1 daily 380, but not better for the players that insists 777's etc should rule the skies ?

Quote from: schro on July 12, 2012, 08:10:01 PM
Drop of 133 to 72 on the VGT route which is at 9:30 on the same frame. SCK dropped from 96 to 54 with a 13:25 departure.

Okay, what's the route demand and are you flying 1x daily or more?

Quote from: sami on July 12, 2012, 08:19:55 PM
Okay, what's the route demand and are you flying 1x daily or more?

VGT has 160 demand. Flying 1x daily. SCK is 110 demand, also 1x daily.

A route that I'm flying 3x daily (ELP) but overloaded in a similar manner exhibited no substantial load change when you added in the 1x daily nerf.... The main point is that the 1x daily penalty seems a bit high....

Many of my long haul flights that leave early in the morning and almost any flight for me that leaves at 22:00 - 5:00, were hit really hard, even though most long haul flights can be ran that way.

Quote from: libertyairlines on July 12, 2012, 08:25:34 PM
Many of my long haul flights that leave early in the morning and almost any flight for me that leaves at 22:00 - 5:00, were hit really hard, even though most long haul flights can be ran that way.

+1 Same here. It's really hard to operate 6000nm+ flights this way.

Quote from: schro on July 12, 2012, 08:23:31 PM
VGT has 160 demand. Flying 1x daily. SCK is 110 demand, also 1x daily.

Cannot test any closer now, but I disabled one part of the settings, any changes?  (after next day change)

Quote from: libertyairlines on July 12, 2012, 08:25:34 PM
Many of my long haul flights that leave early in the morning and almost any flight for me that leaves at 22:00 - 5:00, were hit really hard, even though most long haul flights can be ran that way.

The "normal" dep.time rules (night departures = bad, night arrivals = quite bad) were activated. = Normal settings.

Quote from: SAC on July 12, 2012, 07:50:34 PM

ATL - DUB demand is around 170, a perfect 757 route IMO. With a RI of 100 LF is hovering around 45% and with no competition.  I do not have enough fingers and toes to count how many 757's cross the pond daily in real life, MAN alone gets 2 or 3.  Continental's (United)  european network is full of 757's.  There is no reason 757's in AWS should suffer as noted.  There is a big difference between a direct 757 and a tech-stopping 321.

I am sorry.  I can not agree with this.  The 757 is not and was not made for this type of travel.  Sure, airlines are using these planes now.  Airlines, like Continited, had 57s in their fleet but since replaced them with 739s (which, is the plane that made Boeing quit making the 757).  They retasked these planes to fly some unique routes that fit inside its ETOPS rating.  So, yes, you have flights to Manchester.  You'll see them in Barcelona.  Now, Boeing makes a plane called the Dreamliner that is essentially designed to go to those markets.  

No major airport, up to its gills in planes and lack of slots, would allow an airline to swamp the taramac with these things.  These airports are designed to have people AND cargo delivered.   You are doing LHR/EWR/JFK/DUB a disservice by burdening it with a 57.

Quote from: sami on July 12, 2012, 08:29:52 PM
Cannot test any closer now, but I disabled one part of the settings, any changes?  (after next day change)

Yes - They bounced back to the pre-1x daily nerf levels.

VGT - 137 sold
SCK - 96 sold

Fair enough with JFK/EWR-LHR cannot argue with that and I was surprised to even see that UA has so many 757s in Heathrow, but what would you fly EWR-DUB/HAM/BHX etc with less than 200 demand....767/787's are too big are they not ?  739ER's won't make it without a tech stop...which is seen as an AWS no no.   If that is the case then these routes will become unflyable when 757's retire.

It is also important to note that 787's are only just coming online now in 2012.....MT games only have 7 years to play....but for the 10/15 years prior to that 757's did I as said very very regularly.   

Quote from: sami on July 12, 2012, 08:29:52 PM
The "normal" dep.time rules (night departures = bad, night arrivals = quite bad) were activated. = Normal settings.

But what about after midnight departures on long haul routes? IRL passengers don't seem to care to depart at 1am if a flight will last 10-12hrs, or am I wrong? Look at Emirates timetables for instance.

On a side note, does the 757 (or any "too small" airplane) on a given route have a poor LF despite being the only plane doing the route (I'm talking about a route with good demand - say 1000pax a day)?

Quote from: schro on July 12, 2012, 08:37:19 PM
Yes - They bounced back to the pre-1x daily nerf levels.

Okay, bug then. Will see to that later.

Also for the 757 discussion. You should be able to run 757's cross the pond with no problems if there is no competition. Given that the route demand is large enough (let's say 1.5x size the plane's seats). But narrowbodies have disadvantage over widebodies on longhaul routes.

Quote from: sami on July 12, 2012, 08:43:23 PM
Also for the 757 discussion. You should be able to run 757's cross the pond with no problems if there is no competition. Given that the route demand is large enough (let's say 1.5x size the plane's seats). But narrowbodies have disadvantage over widebodies on longhaul routes.

So if a MT game starts in 1995 what AWS a/c would we use to fly say EWR-BHX which is 3800nm and a 150 demand for example (not sure if figure accurate, but you know what I mean...these routes exsist  ;) )?    

A 767 would be the best option I think, but is far far too big for numerous routes across the pond.  East Coast USA to 2nd tier European airports would not be worth bothering with, yet there are dozens and dozens of these routes in RL....none really suitable for 767's.

My ATL-DUB route has no competition and a 45% LF.  I offer about the same seats as the shown demand.


Update :-

Something just changed again...LF up from 48% to 66% which is a little better....maybe still a little low though considering the lack of competition and maximum RI. 

Quote from: SAC on July 12, 2012, 08:51:59 PM
So if a MT game starts in 1995 what AWS a/c would we use to fly say EWR-BHX which is 3800nm and a 150 demand for example (not sure if figure accurate, but you know what I mean...these routes exsist  ;) )?    

A 767 would be the best option I think, but is far far too big for numerous routes across the pond.  East Coast USA to 2nd tier European airports would not be worth bothering with, yet there are dozens and dozens of these routes in RL....none really suitable for 767's.

My ATL-DUB route has no competition and a 45% LF.  I offer about the same seats as the shown demand.

I posted a question like this about a page back...

Basically, I propose that any route served non-stop by a "unsuitable" aircraft gets NO penalty if the route is flown competition free, or as long as there is no competition from a "suitable" aircraft.

In its current form, it would seem that a LOT of "thin" routes are going to be rendered unservable thus removing a significant amount of entertainment from AWS.

Don

Quote from: JetWestInc on July 12, 2012, 08:59:39 PM
Basically, I propose that any route served non-stop by a "unsuitable" aircraft gets NO penalty if the route is flown competition free, or as long as there is no competition from a "suitable" aircraft.

This is how it basically works already. Some of the pax still choose not to fly there, but 757 is not "so bad" compared to a 737 on similar route for example.

But have to see what it counts for that atlanta-dublin route for example to see if the settings are ok or need tuning.

Quote from: JumboShrimp on July 12, 2012, 05:21:59 PM
The above scenario is as much or more of a problem as frequency exploit on LH routes.  It leads to slot hogging and number of small aircraft that is 5x higher in MT games than in real world...

Interesting claim, I gathered some data for some popular AC from MT6 and wikipedia (of course you can say, that MT6 numbers are -2019 and we are still living in 2012 but there are couple of interesting facts nevertheless).
(https://www.airwaysim.com/forum/proxy.php?request=http%3A%2F%2Fwww.orionis.net%2F%7Eboot%2Fproduction.png&hash=ecb5c6e1b0c6b50a4a8947c02cca18eca794aedd)
You can clearly see AC types and how similar their AWS/RL ratio is.
Small AC (Q400, ATR, S2000) have it highest: 7.42-35.71.
Then there are big airplanes (A330, A340, B757, B767, B777) with medium ratio 3.50-5.55.
Then narrowbody jets (A320 & B737) with lowest ratio 2.41-2.55.
And then B747, which simply sucks in AWS :P
Did you notice that "insta-win" 757 has lower ratio (3.5) than "100% BK" 777 (4.25)?
Only problem that I see is really big numbers of big props, narrowbodies dont need any nerf against WB-s IMO...

Quote from: sami on July 12, 2012, 09:04:58 PM
This is how it basically works already. Some of the pax still choose not to fly there, but 757 is not "so bad" compared to a 737 on similar route for example.

But have to see what it counts for that atlanta-dublin route for example to see if the settings are ok or need tuning.

If the route is gathering a 45% load factor with a RI over 75, then it is bad.

From what I can see, as the final tweaks go in, there is far more good than bad with the new system, but in the end a few exceptions need to be programmed in to preserve playability. I understand people may not like to take a 737-700ER from JNB-GYD (a route I flew in MT6 at ~4200nm, IIRC) but in real life, if the alternate is double connecting through CAI and DXB they will take the non stop flight every time as long as the price is acceptable (read: average fare). All I am asking for is for the penalty to be completely turned off provided there is a) no competition or no competition from a "suitable" aircraft and b) the flight is flown non-stop. In other words, if another airline started that JNB-GYD route with another 73G, we would split the passengers 50/50. If they threw a 767-200 on the route, a "suitable" plane, I would lose my arse.

Don

Quote from: libertyairlines on July 12, 2012, 08:25:34 PM
Many of my long haul flights that leave early in the morning and almost any flight for me that leaves at 22:00 - 5:00, were hit really hard, even though most long haul flights can be ran that way.

I noticed this flight last time I was at JFK Terminal 1:

Departure: 0050 JFK
Arrival: 0400 ICN (Seoul Korea)
Duration: 14h:20m
Aircraft: 777-300ER

I think the LFs of a flight like this should be lower, but if the demand exceeds supply, and the daytime flight is fully booked, what can I do?  I would take the flight.

I think this nightime limitation should only kick in if there are alternatives.  If there are no alternatives, the reduction of LFs should be small.  No matter how bad unsuitable) the aircraft is, no matter how bad the time of day is...

Joe

Maybe just no penalty for an airlines first aircraft on a route? If they add a second, then it kicks in,regardless of competition?

Quote from: JumboShrimp on July 12, 2012, 09:57:40 PM
I noticed this flight last time I was at JFK Terminal 1:

Departure: 0050 JFK
Arrival: 0400 ICN (Seoul Korea)
Duration: 14h:20m
Aircraft: 777-300ER

I think the LFs of a flight like this should be lower, but if the demand exceeds supply, and the daytime flight is fully booked, what can I do?  I would take the flight.

I think this nightime limitation should only kick in if there are alternatives.  If there are no alternatives, the reduction of LFs should be small.  No matter how bad unsuitable) the aircraft is, no matter how bad the time of day is...

Joe

That truly is an awful flight timing! And I guess that's where Sami is coming from when he says that if we use the wrong/inappropriate aircraft or the schedule is poor, some people will opt not to travel at all resulting in lower than expected loads, regardless of competition.

I can safely say that if I didn't need to travel on the above flight (I.e. essential business) or there was no suitable alternative, then I wouldn't travel at all or I'd choose a different holiday destination  :o

Quote from: swiftus27 on July 12, 2012, 08:31:48 PM
I am sorry.  I can not agree with this.  The 757 is not and was not made for this type of travel.  Sure, airlines are using these planes now.  Airlines, like Continited, had 57s in their fleet but since replaced them with 739s (which, is the plane that made Boeing quit making the 757).  They retasked these planes to fly some unique routes that fit inside its ETOPS rating.  So, yes, you have flights to Manchester.  You'll see them in Barcelona.  Now, Boeing makes a plane called the Dreamliner that is essentially designed to go to those markets.  

No major airport, up to its gills in planes and lack of slots, would allow an airline to swamp the taramac with these things.  These airports are designed to have people AND cargo delivered.   You are doing LHR/EWR/JFK/DUB a disservice by burdening it with a 57.

Its the dis-connect between the ideal, as wished for by a/c designers/builders and airport operators on one side, and the airlines on the other : Not all airlines run big cargo sides, they may opt for fast turnaround of walk-on cargo instead with maybe some containers of high-value express freight, with others, the cargo can be as valuable as the passengers on the plane.

Why do some RL airlines fly one-type regardless, and take the knock when it is not the ideal for a specific route? Because bottom-line is that they save more overall than they lose in those cases : sure, ask them and they would love a fleet of Dreamliners, each specifically configged to max-out each specific route arriving at its pre-planned jet-way specifically designed for it : but that's fantasy-land.

The B757 was, and is, a good plane : Boeing replaced it, yes, but airlines still own them and I doubt they will scrap them/sell them off and buy a new fleet, just cos Boeing reckon their new bird is better. I read where an airline ordered new Boeings but the deal also included 10 re-furbished B757's : so they are still popular with airlines, are still keeping their resale value on the used markets : in short, the customer likes them, which may upset the planners but that happens in RL.

Not designed for how they are used, maybe : not optimal, sure : but Viagra was designed for treating heart problems, tho I doubt more than 1 in 10 patients ever use it for its 'designed' purpose...

Quote from: brique on July 13, 2012, 12:31:10 AM
with others, the cargo can be as valuable as the passengers on the plane.

Usually the cargo is far more valuable than the PAX... Recently -working as Ramp Agent- a Pilot asked me how much the to be carried cargo (2tn of Copper Wire) would bring as income. I couldn't answer it, as I don't know that special airlines' cargo payment rates, but he asked me to keep 5 PAX (and their baggage) at DUS airport, and instead carry the 2tn cargo from DUS to LPA on the 753 (enough hint to find the airline :P ). In the end it was his decision, thus "had" to be followed. My boss eventually asked me if that guy was completely sane, I just shrugged :P

Thus I had figured: Cargo > PAX as per financial outcome (only if PAX is Y, I assume...)

The Emirates B773, or the Etihad A333 both usually carry 15-30tn of Cargo into DUS, and outgoing. Probably enough to pay for the fuel and staff, while the PAX pay for the income of the airline, or for the lease fees.

Quote from: brique on July 13, 2012, 12:31:10 AM
Viagra was designed for treating heart problems, tho I doubt more than 1 in 10 patients ever use it for its 'designed' purpose...

Nice Comparison, mate, pretty much hit the spot ;D

First of all, I have missed the opportunity to join the test, but I kept my eyes close on it,

One thing I feel is that the intention of Sami to have this test to see if the machanism is implemented corrttly, and I see more and more posts are talking about the machanism rather than finding bugs.

Then my question is:
Should we go set an agreed machanism first? Or we should stick to bug findings?

Pai

ATL - DUB  update...

Still no competition and now creating a disappointing 61% LF.  I would expect 80-90% TBH, so the route is still suffering by maybe 20% less loads just because it is a narrow body, on a route that these a/c have flown for years, and years to come.

Also just launched ATL-GLA with another 757, but the route has slightly more demand than ATL-DUB.  Obviously RI is terrible, but as time passes it is another route to monitor.

Quote from: Pai on July 13, 2012, 07:28:35 AM
Should we go set an agreed machanism first? Or we should stick to bug findings?

This game world is about testing, the features and settings are not decided by forum "voting" or such.

But it is good that people point out potential issues with the routes, and I can then take a look of the results if they make any sense in regards to what is planned.

Red eye flights still getting hammered....

Quote from: SAC on July 13, 2012, 01:56:40 PM
Red eye flights still getting hammered....

Do these type of flights actually exist? Departing at 1am for a domestic flight of a few hours does not seem like a flight anyone would choose to fly on?

I totally understand a departure from US west coast at 10pm for an arrival on the East coast at 6am. But a 1am departure and 4am arrival... Who would fly that?

Quote from: NorgeFly on July 13, 2012, 03:56:35 PM
Do these type of flights actually exist? Departing at 1am for a domestic flight of a few hours does not seem like a flight anyone would choose to fly on?

I totally understand a departure from US west coast at 10pm for an arrival on the East coast at 6am. But a 1am departure and 4am arrival... Who would fly that?

UA runs multiple LAX/SFO-ORD/IAH flights leaving from 00:00-2:00 and arriving 4:00-6:00. I think Spirit operates even worse flight times on some of their shorter routes.

Quote from: BryanIAH on July 13, 2012, 04:01:20 PM
UA runs multiple LAX/SFO-ORD/IAH flights leaving from 00:00-2:00 and arriving 4:00-6:00. I think Spirit operates even worse flight times on some of their shorter routes.

Wow, I find that very surprising. It doesn't really happen in Europe with the exception of some charter airlines which run to typical holiday destinations at weird times. But scheduled carriers don't really operate shorthaul flights between 2330-0530. Maybe RyanAir and Easyjet have some after midnight arrivals but that's about it.

Hence why flights in AWS that depart after around 2300 or arrive before 0500 have always attracted a penalty.

I'm curious to the LH routes with no competition.

I fly a 737ER from TXL to ATL and suffer heavy loss.

I think that the way you solved for NB was not proper. Just because it is a smaller aircraft doesn't mean people won't fly it! It's the only direct flight and has no stop overs.

I (an avid flyer) would rather fly a 737 direct than have any stop over.

It's not a time difference either. The difference with a 767 is only 15 minutes. also The same penalty applies to the 757s

Quote from: NorgeFly on July 13, 2012, 04:08:25 PM
Hence why flights in AWS that depart after around 2300 or arrive before 0500 have always attracted a penalty.

True, they have NorgeFly, and I avoid 0000-0500 if at all possible, but as games progress inevitably flights start creeping past mid-night as the best slots dry up, flight spacing kicks in or just simply getting more flying hours from your fleet....LF's have always been a minimum of 40% and much more on many flights....but never seen a 0% before on AWS.

These flights in Europe are rare, for scheduled airlines....fly with the charter boys over at Thomas Cook or Thomsonfly etc and then 0230 departures are more common.   Red eye flights are much more common in the US, but as Old Pilot points out most are west > east coast cross country.

Quote from: SAC on July 13, 2012, 01:56:40 PM
Red eye flights still getting hammered....

I think this is great.  It will prevent flooding a route and force people to have realistic schedules...

Quote from: BryanIAH on July 13, 2012, 04:01:20 PM
UA runs multiple LAX/SFO-ORD/IAH flights leaving from 00:00-2:00 and arriving 4:00-6:00. I think Spirit operates even worse flight times on some of their shorter routes.

Also, I remember when Continited ran a LAS-IAH leaving at 0135 and arriving at 0635.  It usually was at least half full  on either a MD80 series or 733/735 :D

Quote from: lunchbox on July 13, 2012, 11:32:29 PM
Also, I remember when Continited ran a LAS-IAH leaving at 0135 and arriving at 0635.  It usually was at least half full  on either a MD80 series or 733/735 :D

Took that flight a few times as a non rev. Each time it was tight so near full.  As a passenger I love flights like this add it gets me a full day before I depart.

Don

Quote from: lunchbox on July 13, 2012, 11:32:29 PM
Also, I remember when Continited ran a LAS-IAH leaving at 0135 and arriving at 0635.  It usually was at least half full  on either a MD80 series or 733/735 :D

UA still runs LAS-IAH (737) from 1:16 to 6:09.

LAS-ORD is another route with a similar schedule. AA's flight is 12:50-6:10 (737) and NK's is 1:00-6:26 (A320).

I've been on most of UA's late night/early morning flights and most of them are 60%+ full.

Quote from: BryanIAH on July 14, 2012, 01:48:07 AM
UA still runs LAS-IAH (737) from 1:16 to 6:09.

LAS-ORD is another route with a similar schedule. AA's flight is 12:50-6:10 (737) and NK's is 1:00-6:26 (A320).

I've been on most of UA's late night/early morning flights and most of them are 60%+ full.

I think I'd rather stick pins in my eyes haha! Guess I'm one of those passengers Sami says would opt not to fly :)

I have just flooded the EDDF-LFPG route (Frankfurt-Charles de Gaulle) with my mauraduing horde of Dash 8s. I am flying every half-hour.

I believe the frequency issue has been addressed pretty well, but would still love to see it closer to reality. I don't think I should be doing as well as I am.

Pls keep the routes, I can check the specs on monday. (you may add more or change prices but if you leave them scheduled anyway makes my work faster)

No problem. Glad to be of help.

Quote from: Kazari on July 14, 2012, 06:42:10 PM
I believe the frequency issue has been addressed pretty well, but would still love to see it closer to reality. I don't think I should be doing as well as I am.

Well, you know, frequency is not bad... but it should be possible to offer smaller prices with larger and faster aircrafts and get the same chance on the market. But if someone flyes 8 x 736 he should of course get more pax than somebody flying 2 * 739. ... for the 10 * ATR vs. 2 * 739 it simply should be that people still select the 739 because of the shorter flight time and because of the smaller prices that could be offered... if the 10 ATR guy want's to sell it's ticket at the same price... well, that's pretty risky and maybe impossible, because fuel is not the only factor... (normally it only makes about a third of the costs as far as I saw now)
... so ... those penalties... also for longer flights... I don't think it's a good thing... it's better to allow competition over the price and quality (so, instead of adding penalties, bether add seat quality -> service could also be "intetrated" in seat quality so that you don't have to model it twice).

Quote from: sami on July 14, 2012, 07:17:15 PM
Pls keep the routes, I can check the specs on monday. (you may add more or change prices but if you leave them scheduled anyway makes my work faster)

This seems to be working just fine. But "test" is still incomplete. Since the competition does not supply the full demand, so the marketshare graph is not 'accurate' in that sense. You have 2500 seats, competition about 1200, and you have 54% marketshare now, when you provide 67% of the seats. Loadfactors on your flights are also 5-45% region, so it does not encourage to such frequency rape. (actually even a bit too high effect in my mind ..?  ..but all the variables are not factored in yet though)

If competition would up their flights to meet the demand the market share would be more to their favor even more. Maybe jumboshrimp could add some 737-sized equipment to that route to meet the demand on his side.

Quote from: Boot on July 12, 2012, 09:13:19 PM
Interesting claim, I gathered some data for some popular AC from MT6 and wikipedia (of course you can say, that MT6 numbers are -2019 and we are still living in 2012 but there are couple of interesting facts nevertheless).
(https://www.airwaysim.com/forum/proxy.php?request=http%3A%2F%2Fwww.orionis.net%2F%7Eboot%2Fproduction.png&hash=ecb5c6e1b0c6b50a4a8947c02cca18eca794aedd)
You can clearly see AC types and how similar their AWS/RL ratio is.
Small AC (Q400, ATR, S2000) have it highest: 7.42-35.71.
Then there are big airplanes (A330, A340, B757, B767, B777) with medium ratio 3.50-5.55.
Then narrowbody jets (A320 & B737) with lowest ratio 2.41-2.55.
And then B747, which simply sucks in AWS :P
Did you notice that "insta-win" 757 has lower ratio (3.5) than "100% BK" 777 (4.25)?
Only problem that I see is really big numbers of big props, narrowbodies dont need any nerf against WB-s IMO...

Some of the big LH jets were bought only for tax reasons, not for flying.  I suspect more of the big LH jets in MT7 will be used for flying.

But, as you pointed out (in a PM) there is no advantage profit-wise in flying full larger aircraft vs. flying full smaller aircraft.  Larger aircraft ges hit by higher pilot salary, and a higher multiple of staffing.

The changes put in place so far will have no influence on Small AC (Q400, ATR, S2000) that ended up with 7.42-35.71 x real life.

I see a guy flying 14x Q400 between LHR and CDG.  The route has ~110% supply.  If the Q400 has the same return on investment flying full planes as the guy flying 737, there will just be more Q400.  That's because Q400 is just completely competition proof, while 737/A200 is vulnerable to smaller aircraft.

In real life, I see only A320 class aircraft on the LHR-CDG route.  There is something about operating small aircraft that AWS is not capturing correctly.

My observation is that the smaller aircraft is generally on thinner routes with higher prices (per mile).  Larger aircraft (A320, 737) tend to be on higher density routes, with more competition, with lower ticket price per mile.  That to me means that a full A320 / 737 has much lower operating cost vs. smaller jets and props.  That's probably what we are missing in AWS.

If nothing is done that works against proliferation of Q400, ATR, S2000, they will just continue to proliferate...

High landing fees that are aircraft size independent is the only thing I can think of at this point...

Quote from: JumboShrimp on July 15, 2012, 02:35:19 PM


High landing fees that are aircraft size independent is the only thing I can think of at this point...

find a formula to limit the number of slots you can buy at the 'out' airport.  that will solve so much.

The ability to set fares significantly above default on routes without any/much competition is working well and a welcome change. I'm now charging default +15% on all routes (most if which have no competition) and there appears to be little or any impact in LF. I plan on nudging this up toward +20% to see how much I can squeeze out of my pax.

As I don't have any serious competition, I don't know how well the lowering if fares is helping attract pax on competitive routes.... Anyone have any comments on how it's working as a tactic?

Edit: if someone at LHR cares to experiment, we can coordinate a test by adding similar aircraft to the route between LHR and OSL and then when your reach CI of 100 coordinate prices to see what affect it has?

Quote from: swiftus27 on July 15, 2012, 02:50:38 PM

find a formula to limit the number of slots you can buy at the 'out' airport.  that will solve so much.

I don't think it will solve the problem.  Suppose a airline based at CDG gets 10 slot allocation at LHR.  And suppose the route CDG-LHR has 1500 pax demand and is already has 200% supply.

I can fly it with 10xA320, bringing supply to 300%, of which I have 33% and my LF is 33% and losing money.

Or suppose I fly 10x E-Jet or 10xQ400.  I supply only 50% of demand, bringing the supply from 200% to 250%.

But I will capture the same number of pax with 10x / day flights, resulting in 66% LF.  So while the 2 other guys with 737s and A320 are losing money with 33% LF, I am making money with E-Jet or Q400 with 66% LF.

Maybe there should be 2nd way to come up with "ideal" aircraft.
1. we already have the route length
2. ideal aircraft for route demand.

So suppose you determine that an ideal aircraft for 1500 pax per day route is 150 pax aircraft, the only way you can prevent a 75 pax aircraft from outcompeting everyone through frequency is if an aircraft smaller than 150 pax cannot be awarded LF larger than a 150 pax aircraft.

So you either deny a win to small aircraft by
- limiting its LF at routes where it is inappropriate
- increase the landing fees so that the aircraft too small for the route just does not make money
- or make other adjusments so that larger aircraft has some economies of scale.  Right now, the economies of scale are too insignificant to matter

Quote from: JumboShrimp on July 15, 2012, 02:35:19 PM
Some of the big LH jets were bought only for tax reasons, not for flying.  I suspect more of the big LH jets in MT7 will be used for flying.

But, as you pointed out (in a PM) there is no advantage profit-wise in flying full larger aircraft vs. flying full smaller aircraft.  Larger aircraft ges hit by higher pilot salary, and a higher multiple of staffing.

The changes put in place so far will have no influence on Small AC (Q400, ATR, S2000) that ended up with 7.42-35.71 x real life.

I see a guy flying 14x Q400 between LHR and CDG.  The route has ~110% supply.  If the Q400 has the same return on investment flying full planes as the guy flying 737, there will just be more Q400.  That's because Q400 is just completely competition proof, while 737/A200 is vulnerable to smaller aircraft.

In real life, I see only A320 class aircraft on the LHR-CDG route.  There is something about operating small aircraft that AWS is not capturing correctly.

My observation is that the smaller aircraft is generally on thinner routes with higher prices (per mile).  Larger aircraft (A320, 737) tend to be on higher density routes, with more competition, with lower ticket price per mile.  That to me means that a full A320 / 737 has much lower operating cost vs. smaller jets and props.  That's probably what we are missing in AWS.

If nothing is done that works against proliferation of Q400, ATR, S2000, they will just continue to proliferate...

High landing fees that are aircraft size independent is the only thing I can think of at this point...

Although what you say about the LHR-CDG route is true (being served by A320 sized aircraft) this is not the same for MAN-CDG.  This IS served by Q400s as well as A320 sized a/c (along with E195s FYI).  So you cannot use a blanket rule on small a/c for all routes as it is just NOT the case.  If there was an "ideal" aircraft, then there would only be a very limited amount of aircraft sold, when truth is there are MANY aircraft types sold for different kinds of routings.  You cannot limit the type of aircraft just because of it's size.  The only thing you can/should do is adjust the passengers preference to price, timing, and comfortability of the flight. 

the "ideal" aircraft is the one that allows you to operate the route with the smallest cost per seat... and if passengers select the flight according to values we understand (price, quality, flight length, departure-preference) and so on, everything will work automatically and then the game is a real game... so, someone could fly 380, an otherone 733 on the same route... if the passenger is happy... it works... and you'll get rid of all your pax-distribution problems... but as long as you try to add unrealistic things like penalties for "non ideal aircrafts", it would never work properly... so, why investing so much in the "ideal-aircraft"-machanism and not into quality-simulation? ... that would solve it in a highly realistic way...

Quote from: JumboShrimp on July 15, 2012, 03:45:47 PM
I don't think it will solve the problem.  Suppose a airline based at CDG gets 10 slot allocation at LHR.  And suppose the route CDG-LHR has 1500 pax demand and is already has 200% supply.

I can fly it with 10xA320, bringing supply to 300%, of which I have 33% and my LF is 33% and losing money.

Or suppose I fly 10x E-Jet or 10xQ400.  I supply only 50% of demand, bringing the supply from 200% to 250%.

But I will capture the same number of pax with 10x / day flights, resulting in 66% LF.  So while the 2 other guys with 737s and A320 are losing money with 33% LF, I am making money with E-Jet or Q400 with 66% LF.

Maybe there should be 2nd way to come up with "ideal" aircraft.
1. we already have the route length
2. ideal aircraft for route demand.

So suppose you determine that an ideal aircraft for 1500 pax per day route is 150 pax aircraft, the only way you can prevent a 75 pax aircraft from outcompeting everyone through frequency is if an aircraft smaller than 150 pax cannot be awarded LF larger than a 150 pax aircraft.

So you either deny a win to small aircraft by
- limiting its LF at routes where it is inappropriate
- increase the landing fees so that the aircraft too small for the route just does not make money
- or make other adjusments so that larger aircraft has some economies of scale.  Right now, the economies of scale are too insignificant to matter

Too many suppositions in here.  Not a fair way to disagree.   It all comes down to figuring out the correct algorythm

Quote from: JumboShrimp on July 15, 2012, 02:35:19 PM
Some of the big LH jets were bought only for tax reasons, not for flying.  I suspect more of the big LH jets in MT7 will be used for flying.

There were all types of AC bought only for tax reasons... American Southern in MT6 had more than 1000 planes listed for sale....153x 737, 120x 757, 80x 777, 151 Embraers, 78x MD90, 99x Mitsubishi RJ-s etc...

Quote from: JumboShrimp on July 15, 2012, 02:35:19 PM
But, as you pointed out (in a PM) there is no advantage profit-wise in flying full larger aircraft vs. flying full smaller aircraft.  Larger aircraft ges hit by higher pilot salary, and a higher multiple of staffing.
In real life, I see only A320 class aircraft on the LHR-CDG route.  There is something about operating small aircraft that AWS is not capturing correctly.

I think that "congestion charge" has been mentioned already couple of times but it hasn't really picked Sami's interest... This could be dependent on airport size and it's used slots per hour. No real reason to implement it for size 1-3 airports, but certainly for 5 and maybe also 4. If you have to pay 10k per landing at size 5 airport because it's slots are 95% full, then you'll think twice before scheduling a Q400 there...
I think that IRL you don't see Q400-s or E-jets flying LHR-CDG not because of their LF-s would be low, people would love 30 min interval flights... You don't see them because it's not economically not feasible...

Quote from: JumboShrimp on July 15, 2012, 03:45:47 PM
I don't think it will solve the problem.  Suppose a airline based at CDG gets 10 slot allocation at LHR.  And suppose the route CDG-LHR has 1500 pax demand and is already has 200% supply.

I can fly it with 10xA320, bringing supply to 300%, of which I have 33% and my LF is 33% and losing money.

Or suppose I fly 10x E-Jet or 10xQ400.  I supply only 50% of demand, bringing the supply from 200% to 250%.

But I will capture the same number of pax with 10x / day flights, resulting in 66% LF.  So while the 2 other guys with 737s and A320 are losing money with 33% LF, I am making money with E-Jet or Q400 with 66% LF.

Maybe there should be 2nd way to come up with "ideal" aircraft.
1. we already have the route length
2. ideal aircraft for route demand.

So suppose you determine that an ideal aircraft for 1500 pax per day route is 150 pax aircraft, the only way you can prevent a 75 pax aircraft from outcompeting everyone through frequency is if an aircraft smaller than 150 pax cannot be awarded LF larger than a 150 pax aircraft.

So you either deny a win to small aircraft by
- limiting its LF at routes where it is inappropriate
- increase the landing fees so that the aircraft too small for the route just does not make money
- or make other adjusments so that larger aircraft has some economies of scale.  Right now, the economies of scale are too insignificant to matter

But this may well happen in real life in which case you can't penalise the new competitor for using aircraft that are perfectly appropriate for the situation. If in the real world if an airline identifies that a route may be profitable with a regional jet but not with an A320, what will they use? The incumbent airlines on the route then have to respond accordingly, by either reducing fares (which should now have an impact whereas it didnt use to in AWS), reducing frequency or reducing aircraft size to maintain frequency. Or they maintain the status quo and haemorrige cash. This is how it works in the real world and how it should work in AWS.

The problem is that most players do not behave appropriately like a real business. Many insist on serving 100% or more of demand regardless of the level of competition and refuse to scale back when they face competition. Real airlines downsize or reduce capacity rather than point and cry about their competitor getting a better load factor on their Ejet.

I do however agree that landing fees and slot fees could be used more effectively to prevent regional aircraft being used at airports where slots are desperately short. Perhaps at hand picked airports which are problematic (LHR etc) should have much higher slot fees and landing fees right from the start. At the moment, the slot prices don't increase prohibitively until its really too late... I.e. most if the slots are already gone.

Artificially capping load factors is a really bad idea in my mind as it is so unrealistic.

Quote from: Pilot Oatmeal on July 15, 2012, 04:27:58 PM

Although what you say about the LHR-CDG route is true (being served by A320 sized aircraft) this is not the same for MAN-CDG.  This IS served by Q400s as well as A320 sized a/c (along with E195s FYI).  So you cannot use a blanket rule on small a/c for all routes as it is just NOT the case.  If there was an "ideal" aircraft, then there would only be a very limited amount of aircraft sold, when truth is there are MANY aircraft types sold for different kinds of routings.  You cannot limit the type of aircraft just because of it's size.  The only thing you can/should do is adjust the passengers preference to price, timing, and comfortability of the flight. 

Good points.  The one variable that changes between your and my examples is MAN vs. LHR.  And it could very will be that LHR airport authority just does not want to waste a slot on Q400 whereas MAN does not care.

Not sure how to model this in AWS other than through higher landing fees at high demand airports at high demand time slots...

Quote from: Boot on July 15, 2012, 05:01:09 PM
... You don't see them because it's not economically not feasible...

Yes! Let the economy decide! That's what I say all the time... just buying other aircraft because you like them or because the have a nice shape doesn't make sense! And it shouldn't be the idea of this game. I really love the MD-11... but hey... I can't build an all MD-11 fleet here, simply because it's not profitable! And that's what should be improved/tuned. The economical factors!

Quote from: JumboShrimp on July 15, 2012, 05:05:59 PM
Good points.  The one variable that changes between your and my examples is MAN vs. LHR.  And it could very will be that LHR airport authority just does not want to waste a slot on Q400 whereas MAN does not care.

Not sure how to model this in AWS other than through higher landing fees at high demand airports at high demand time slots...

Yes, higher slot costs and landing fees at high demand slots would be the way to do it.  But it should be by weight, so it would affect larger aircraft as much.  So were kind of going around in circles  :laugh:

Quote from: meiru on July 15, 2012, 04:56:38 PM
the "ideal" aircraft is the one that allows you to operate the route with the smallest cost per seat... and if passengers select the flight according to values we understand (price, quality, flight length, departure-preference) and so on, everything will work automatically and then the game is a real game... so, someone could fly 380, an otherone 733 on the same route... if the passenger is happy... it works... and you'll get rid of all your pax-distribution problems... but as long as you try to add unrealistic things like penalties for "non ideal aircrafts", it would never work properly... so, why investing so much in the "ideal-aircraft"-machanism and not into quality-simulation? ... that would solve it in a highly realistic way...

Well, it is all about improving the simulatin to better represent the real life, and fixing things that takes AWS far away from real life.
- One thing was excessive use of narrow bodies flying very long distance outcompeting widebodies.
- The second thing is small and medium aircraft outcompeting A320/737 sized aircraft, again through frequency.

I could not agree more with this statement:

Quote from: meiru on July 15, 2012, 04:56:38 PM
the "ideal" aircraft is the one that allows you to operate the route with the smallest cost per seat

It may just be that AWS does not capture this correctly.  A full 737 should be a far more profitable in AWS than a full Q400 at the same ticket prices.  That is the case in real life.  There is a reason we far more 737/A320s produced than E-Jets, ATR, Q400, Saabs.  Refer to post by Boot who summarized MT6 vs. real life.

And just because 737/A320 is more profitable than the smaller aircraft, it allows the operators to lower the prices on the busy routes served by 737/A320 and still maintain profitability.

Maybe the same issue was causing the LH problems.  Maybe the widebodies are just far more profitable in real life per passenger mile than narrowbodies (on same routes with same LF).  Otherwise, why would the airlines bother to buy anything other than 757?  (well, cargo change thngs somewhat here)

Quote from: Pilot Oatmeal on July 15, 2012, 05:13:45 PM
Yes, higher slot costs and landing fees at high demand slots would be the way to do it.  But it should be by weight, so it would affect larger aircraft as much.  So were kind of going around in circles  :laugh:

This doesn't necessarily need to be the case. London Gatwick airport recently increased fees disproptionately for small aircraft to discourage the use of such aircraft at peak times. Flybe then got upset (understandably) as they saw it as an assault in them and challenged the decision with the authorities (CAA I think). The challenge was rejected as the CAA believed that there was a a valid business case for LGW to encourage the use of larger aircraft at peak times and the fees wereincreased, resulting in Flybe scrapping two marginal routes which became unprofitable (LBA and DUS).

However, this should only be allowed at airports where there is a real problem with lack of slots, eg. LHR, LGW, FRA etc.

Quote from: NorgeFly on July 15, 2012, 05:01:25 PM
But this may well happen in real life in which case you can't penalise the new competitor for using aircraft that are perfectly appropriate for the situation. If in the real world if an airline identifies that a route may be profitable with a regional jet but not with an A320, what will they use?

Certainly true for many routes.  But it seems that when an airlines has a choice, and can fill up an A320, than A320 is used.  Why?  because it is more profitable than 2 flights smaller aircraft.  

Quote from: NorgeFly on July 15, 2012, 05:01:25 PM
The incumbent airlines on the route then have to respond accordingly, by either reducing fares (which should now have an impact whereas it didnt use to in AWS), reducing frequency or reducing aircraft size to maintain frequency. Or they maintain the status quo and haemorrige cash. This is how it works in the real world and how it should work in AWS.

The problem is that most players do not behave appropriately like a real business. Many insist on serving 100% or more of demand regardless of the level of competition and refuse to scale back when they face competition. Real airlines downsize or reduce capacity rather than point and cry about their competitor getting a better load factor on their Ejet.

I do however agree that landing fees and slot fees could be used more effectively to prevent regional aircraft being used at airports where slots are desperately short. Perhaps at hand picked airports which are problematic (LHR etc) should have much higher slot fees and landing fees right from the start. At the moment, the slot prices don't increase prohibitively until its really too late... I.e. most if the slots are already gone.

Artificially capping load factors is a really bad idea in my mind as it is so unrealistic.

I agree that far higher landing fees would be preferable way to limit the 14x day Q400 flight between CDG and LHR than artificiall limiting LFs.  I was just brainstorming...

But I would not limit it to LHR.  Any airport slot that is "green" should be normal rate, "orange" should be say 2x, "red" shold be something like 4x (or whatever it would take to make small aircraft unprofitable in that time slot).

Quote from: Pilot Oatmeal on July 15, 2012, 05:13:45 PM
Yes, higher slot costs and landing fees at high demand slots would be the way to do it.  But it should be by weight, so it would affect larger aircraft as much.  So were kind of going around in circles  :laugh:

For game playability, it should not be linearly by weight.  Especially before we have cargo modeled.

Something like c + weight * rate
where "c" is significant.

Quote from: NorgeFly on July 15, 2012, 05:27:54 PM
This doesn't necessarily need to be the case. London Gatwick airport recently increased fees disproptionately for small aircraft to discourage the use of such aircraft at peak times. Flybe then got upset (understandably) as they saw it as an assault in them and challenged the decision with the authorities (CAA I think). The challenge was rejected as the CAA believed that there was a a valid business case for LGW to encourage the use of larger aircraft at peak times and the fees wereincreased, resulting in Flybe scrapping two marginal routes which became unprofitable (LBA and DUS).

However, this should only be allowed at airports where there is a real problem with lack of slots, eg. LHR, LGW, FRA etc.

Really  ??? surely this is against the law as unfair competition rules.  Gatwick like many UK airports are owned independently (i.e. not owned by the government).  This case if it's true would be against the law (in the UK) by giving handicaps to operators operating smaller aircraft.  In fact I'm surprised that the government would allow that considering the CAA along with many other government organisations are calling for more fuel efficient aircraft.  - Off topic tho  :)


The ONLY real way to stop frequency is to have price, CI, RI, Comfortability and timing be higher ranked in the customers mind than frequency.  

 

Quote from: Pilot Oatmeal on July 15, 2012, 05:44:57 PM
The ONLY real way to stop frequency is to have price, CI, RI, Comfortability and timing be higher ranked in the customers mind than frequency.  

yup

Quote from: JumboShrimp on July 15, 2012, 05:40:40 PM
For game playability, it should not be linearly by weight.  Especially before we have cargo modeled.

Something like c + weight * rate
where "c" is significant.

Landing fees (in the UK) are charged by the weight of the aircraft, and then there are handling charges by the size of the aircraft.  I'm not sure on how to calculate this in AWS, we would need to find the most used calculation in most countries and use that.  It's the only way to really average it out.  

Quote from: NorgeFly on July 15, 2012, 05:27:54 PM
This doesn't necessarily need to be the case. London Gatwick airport recently increased fees disproptionately for small aircraft to discourage the use of such aircraft at peak times. Flybe then got upset (understandably) as they saw it as an assault in them and challenged the decision with the authorities (CAA I think). The challenge was rejected as the CAA believed that there was a a valid business case for LGW to encourage the use of larger aircraft at peak times and the fees wereincreased, resulting in Flybe scrapping two marginal routes which became unprofitable (LBA and DUS).

Good point, maybe it will convince Sami.  Sami's point about slot maintenance fees aka. increased landing fees is that it does not happen in real life.

Quote from: NorgeFly on July 15, 2012, 05:27:54 PM
However, this should only be allowed at airports where there is a real problem with lack of slots, eg. LHR, LGW, FRA etc.

There are far more airports that those 3.  There is JFK, and many asian airports that are slot limited.

In AWS, we manage to make many more airports slot constrained, by flying aircraft that is too small.  So I would make it a general formula based on existing color coding of take off slots.

This would mean that there is no problem what at all using small aircraft during "green" hours, the fees should be increased during "orange" hours, and should be cost prohibitive for small aircraft in "red" hours.

And still, the fee should really have really tiny impact on a full 777 taking off during the red time slot.

Quote from: Pilot Oatmeal on July 15, 2012, 05:44:57 PM
The ONLY real way to stop frequency is to have price, CI, RI, Comfortability and timing be higher ranked in the customers mind than frequency.  

But you can achieve the same CI, same RI, same comfort level on small aircraft as on larger aircraft.  So then everything is equal except frequency and we are back where we started.

Quote from: JumboShrimp on July 15, 2012, 05:52:42 PM
But you can achieve the same CI, same RI, same comfort level on small aircraft as on larger aircraft.  So then everything is equal except frequency and we are back where we started.

So we should have more variables added ;D

Quote from: NorgeFly on July 15, 2012, 05:01:25 PM
Artificially capping load factors is a really bad idea in my mind as it is so unrealistic.

(Will not happen, ever.)


Anyway. Focus on this test game world, and pls keep this topic for only those matters.

Quote from: JumboShrimp on July 15, 2012, 05:35:41 PM

But I would not limit it to LHR.  Any airport slot that is "green" should be normal rate, "orange" should be say 2x, "red" shold be something like 4x (or whatever it would take to make small aircraft unprofitable in that time slot).

I agree not only LHR, but it should not be applied to regional airports (definition a bit tricky I know) as they are not the problem.

And I see where you're coming from with the green/orange/red thing, but the problem with that is that by time it gets to red it's too late. Ideally the fees at problem airports should be high right from the start to prevent the initial influx of small aircraft which swamp popular airports early on. High fees once it's full is too late.

Quote from: sami on July 15, 2012, 06:06:51 PM
Anyway. Focus on this test game world, and pls keep this topic for only those matters.

We are keeping it on the test game.  There is an airline hapily flying 14x Q400 between LHR and CDG.  He could easily add another 14x.

Doing that is a great strategy in AWS.  But somehow, it is not what happens in RL.  So the discussion is how to make flying 30x Q400 between LHR and CDG not a viable strategy...

Quote from: JumboShrimp on July 15, 2012, 05:52:42 PM
But you can achieve the same CI, same RI, same comfort level on small aircraft as on larger aircraft.  So then everything is equal except frequency and we are back where we started.

yes, but... you can offer lower prices and larger seats (quality) and so on... and by the way, the RI ... what's that? Route-Image? it shouldn't always go up to 100 and reamain there... this should be the quality of the service and then it works

Quote from: JumboShrimp on July 15, 2012, 06:13:29 PM
We are keeping it on the test game.  There is an airline hapily flying 14x Q400 between LHR and CDG.  He could easily add another 14x.

Doing that is a great strategy in AWS.  But somehow, it is not what happens in RL.  So the discussion is how to make flying 30x Q400 between LHR and CDG not a viable strategy...

See my post earlier today on this, there was a rq for you too.

(he is NOT doing that great on those routes I believe, if you see my post about KazAir's data, so the problem you are disucssing is pretty much solved already according to that .. But needs further testing still)

I'm really thinking I'm gonna go build my own AirwaySim, with blackjack and hookers.

Quote from: NorgeFly on July 15, 2012, 06:10:36 PM
I agree not only LHR, but it should not be applied to regional airports (definition a bit tricky I know) as they are not the problem.

It is extremely rare to see orange at a regional airport.  I checked.  I flew to most airports in MT6.  It is rare to see anythign but green.  So this change would have almost zero effect on regional airports.

Quote from: NorgeFly on July 15, 2012, 06:10:36 PM
And I see where you're coming from with the green/orange/red thing, but the problem with that is that by time it gets to red it's too late. Ideally the fees at problem airports should be high right from the start to prevent the initial influx of small aircraft which swamp popular airports early on. High fees once it's full is too late.

With the current method of slot growth (starting at lower ~50% of normal for the year, growing to ~200% - I may be off on percentages), the slots run out very quickly at the most slot constrained airports.

The fees we are talking about are ongoing landing fees (or slot usage fees, or congestion surcharges), not slot acquisition fees, which are separate.
If the ongoing landing fees when slots are in red are high enough to make small aircraft unprofitable, then the opeators of these flights would disappear over time, freeing up the slots.  So it is never too late.

Quote from: sami on July 15, 2012, 06:17:08 PM
See my post earlier today on this, there was a rq for you too.

(he is NOT doing that great on those routes I believe, if you see my post about KazAir's data, so the problem you are disucssing is pretty much solved already according to that .. But needs further testing still)

Ok, will try to find the post.  And also test it myself.  I just got 4xQ400 delivered to do some more testing...

Pls use large planes against him (Q400) on that route, you had some 330s flying there but may wish to use something like 320.

FRA-CDG was the route .. me thinks.  :P

Quote from: meiru on July 15, 2012, 06:14:47 PM
yes, but... you can offer lower prices and larger seats (quality) and so on...

If all inclusive cost per passenger (including aircraft acquisition, staffing, other operational costs) of a full A320 is identical to a full smaller aircraft, would mean there are no economies of scale, and price cutting is completely a moot point.

Quote from: meiru on July 15, 2012, 06:14:47 PM
and by the way, the RI ... what's that? Route-Image? it shouldn't always go up to 100 and reamain there... this should be the quality of the service and then it works

Perhaps you are right, but it is not part of the current set of changes.

Quote from: sami on July 15, 2012, 06:30:58 PM
Pls use large planes against him (Q400) on that route, you had some 330s flying there but may wish to use something like 320.

FRA-CDG was the route .. me thinks.  :P

Ok, I will do that.  I just needed to have some route to make money against my Boston route that was losing money early on.

Quote from: JumboShrimp on July 15, 2012, 06:37:07 PM
Ok, I will do that.  I just needed to have some route to make money against my Boston route that was losing money early on.

I look forward to it. My planes are configured 68Y 0C, FYI.

And it's 250 flights a day among 7 planes.  ;)

Their current (un)profitability is shown below.

Quote from: Kazari on July 15, 2012, 06:53:01 PM
I look forward to it. My planes are configured 68Y 0C, FYI.

And it's 250 flights a day among 7 planes.  ;)

Their current (un)profitability is shown below.

Ok, I have 8 x A320 with 150 pax each, 138 Y, 10 C.  Just under 1200 total supply.  I need to wait 2 weeks to get the A320s delivered.  Good thing is that I am starting from 100 RI, so we should see the results right away.

I counted 38x day Q400, and Admin has 5x ATR.

Quote from: Kazari on July 15, 2012, 06:53:01 PM
I look forward to it. My planes are configured 68Y 0C, FYI.

And it's 250 flights a day among 7 planes.  ;)

Their current (un)profitability is shown below.

Just curious about flight spacing.  Do you have any flights that have nothing close to them within 1 hour?  I see most of your flights have spacing of 30 mins, which should be just fine in MT6 (even 20 min would have been fine).  I am not sure if that changed and 30 mins is no longer "safe".  If any of the low LFs are coming from that or from an algorithm adjustments.

Here is the route we are talking about:
https://www.airwaysim.com/game/Routes/Planning/LFPG/EDDF

That's fine... the DH8D guy should get most of passengers if he's offering the same price, since quality seems not to be modeled.

Quote from: meiru on July 15, 2012, 07:15:37 PM
That's fine... the DH8D guy should get most of passengers if he's offering the same price, since quality seems not to be modeled.

My aircraft is not flying yet.  But there is something else going on that is encouraging:

Sami is supplying only ~13% of capacity, and is getting 26% of demand.  Using roughly equivalent, actually slower aircraft (but I know speed is not modeled in yet).

That means that there is some penalty for flying too many times per day.  Of course the way to go about nuking your competitor is to fly the max per day, not any more, get maximum frequency benefit, maximum profitability, and not supply the route fully.  (Maybe something along the lines what NorgeFly was talking about...)

Quote from: JumboShrimp on July 15, 2012, 07:22:37 PM
My aircraft is not flying yet.  But there is something else going on that is encouraging:

Sami is supplying only ~13% of capacity, and is getting 26% of demand.  Using roughly equivalent, actually slower aircraft (but I know speed is not modeled in yet).

That means that there is some penalty for flying too many times per day.  Of course the way to go about nuking your competitor is to fly the max per day, not any more, get maximum frequency benefit, maximum profitability, and not supply the route fully.  (Maybe something along the lines what NorgeFly was talking about...)

So, over-cooking the frequency becomes self-destructive, profit-wise : that would be an elegant solution indeed :)

Wow, all I can say Impressive!

Recap:
Kazari: 37xQ400
Sami: 5xATR72
Jumbo: 8xA320

Supply = 150% Demand

Here are LFs of 8xA320.  Maybe Kazari can post his screen shot (maybe on 18th, to get full 7 days of me flying).

Any chance we can see kazari and sami's LFs?

Quote from: Kazari on July 15, 2012, 06:53:01 PM
I look forward to it. My planes are configured 68Y 0C, FYI.

And it's 250 flights a day among 7 planes.  ;)

Their current (un)profitability is shown below.

Hi Kazari,

Can you post a picture of a bunch of routes (not aircraft) on this route?  I would like to see if and how my addition of about 40% of demand changed things.  If you are being limited strictly by the too high frequency or if competition figures into it as well.

His sales have dropped somewhat, but it's a bit hard to judge as he has so many routes and planes in C checks etc. (and there's no overall indicator of this)

But on 0830 dep route at least had LF dropping from 45-50% to 25-35% over last 5 game days. And similar on another 1630 dep route.

Quote from: sami on July 15, 2012, 10:17:19 PM
His sales have dropped somewhat, but it's a bit hard to judge as he has so many routes and planes in C checks etc. (and there's no overall indicator of this)

But on 0830 dep route at least had LF dropping from 45-50% to 25-35% over last 5 game days. And similar on another 1630 dep route.

All I can say, interesting results.  Overall it looks great, but I see a slight problem.

I have ~40% capacity and I am allocated as of now, 50% of pax at LF of ~70%.  So I am supplying 35% of demand.  The slight problem is that it appears that all pax demanding to fly are not flying, but Kazari's flying is still affecting my LFs and your LFs.  (I am not sure if I am explaining it clearly, and if it is worth addressing).

Quote from: JumboShrimp on July 15, 2012, 06:13:29 PM
We are keeping it on the test game.  There is an airline hapily flying 14x Q400 between LHR and CDG.  He could easily add another 14x.

Quote from: sami on July 15, 2012, 06:17:08 PM
(he is NOT doing that great on those routes I believe, if you see my post about KazAir's data, so the problem you are disucssing is pretty much solved already according to that .. But needs further testing still)

Can agree to sami :) I fly 5 or 6x A306 and 1x 773, all having 100% C LFs and 60%+ Y LFs, all making a lot of cash...
In fact the A306 that flies 3 x daily (06, 12 and 18) makes over 500k/week ;)

P.S. Route is LHR-CDG against Cream's tons of DHC8

As requested, one screen of July 25.

I could try frequenting another large airport with my fleet if anyone wants to test another aspect of this. Maybe I'll fly only every 45 minutes?

Open to it. Got plenty of planes.

Quote from: Jona L. on July 15, 2012, 11:29:39 PM
Can agree to sami :) I fly 5 or 6x A306 and 1x 773, all having 100% C LFs and 60%+ Y LFs, all making a lot of cash...
In fact the A306 that flies 3 x daily (06, 12 and 18) makes over 500k/week ;)

P.S. Route is LHR-CDG against Cream's tons of DHC8

That route is not heavily over supplied.  I may put some A320s there to see what happens when there is some oversupply, who I would be taking pax from the most.  I am thinking you and Cream, less from London Jet.

Quote from: JumboShrimp on July 16, 2012, 01:43:41 AM
That route is not heavily over supplied.  I may put some A320s there to see what happens when there is some oversupply, who I would be taking pax from the most.  I am thinking you and Cream, less from London Jet.

i will add more 737s to oversupply if you wish...

Quote from: ARASKA on July 16, 2012, 01:51:10 AM
i will add more 737s to oversupply if you wish...

That would be cool.  But don't go overboard.  I don't know where the limit is, but maybe 15x per day might be ok.

Just trying to see where the limits are.  10 x ATR-72 = 640 pax on 720 pax route = no problem.  Slight fall off between 0500-0600 and 2300-0000 but everything else fully loaded.

10 x daily on 720 pax route seems to leave a plenty of room for frequency abuse, and only a small percentage routes (> 1000 pax) where high frequency with small aircraft would be penalized.
https://www.airwaysim.com/game/Routes/Planning/LFPG/EBBR

I wonder if the limit should not be lowered somewhat on a route like this.  Very shot distance (136 nm) is one thing making multiple ATR a little more tolerable on this route...  But I suspect the limit would not change a lot if the route was 300nm, making 10 x ATR more iffy.

In real life, there are 2xA320.  So real life demand is much loser than AWS, probably depressed due to a good rail connection.

As many (and I) sayed, I think it comes down to 2 issues... passenger aircraft preference and flight costs...

As a passenger I'll always prefer to take a 2hs jet trip than a 4hs turboprop trip... The longest turboprop scheduled commercial flight I ever took was a 45mins one (about 100nm) and only because there was a very very low demand route with only one airline operating it 3 times a week...

And, it's more profitable to take 200 pax on one A321 than 200 on 4 Q400... same with A320 compared to widebodies..

turboprops only flights on routes that won't fill any larger plane... A320s only flights on routes that won't fill any larger plane, and so on...

Quote from: JumboShrimp on July 15, 2012, 09:30:22 PM
Wow, all I can say Impressive!

Recap:
Kazari: 37xQ400
Sami: 5xATR72
Jumbo: 8xA320

Supply = 150% Demand

Here are LFs of 8xA320.  Maybe Kazari can post his screen shot (maybe on 18th, to get full 7 days of me flying).

Do I get it right, that E-170 for example would get the same kind of low LF-s as ATR & Q400? Because it's roughly the same size?
I can imagine many real-life people thinking "I dont want to fly a turbo-prop so I'll just wait 3 hours for that A320", but I'm having really-really hard time imagining the situation "I have to be at CDG 0800, but i'll still wait for Jumbos A320, because it's the right size for this route. And this capitalist pig E-170 owner is really oversupplying this route, that's a bad thing too, suffer biaatch!!!" :P

Quote from: Boot on July 16, 2012, 12:42:00 PM
Do I get it right, that E-170 for example would get the same kind of low LF-s as ATR & Q400? Because it's roughly the same size?
I can imagine many real-life people thinking "I dont want to fly a turbo-prop so I'll just wait 3 hours for that A320", but I'm having really-really hard time imagining the situation "I have to be at CDG 0800, but i'll still wait for Jumbos A320, because it's the right size for this route. And this capitalist pig E-170 owner is really oversupplying this route, that's a bad thing too, suffer biaatch!!!" :P

First, do you think it is a problem when airline flies a turbo prop 37x per day between 2 slot constrained airports?  If you agree that this is a problem (and I think it is seriously affecting AWS, as it became single dimensional, with players just stretching the frequency benefit further and further beyond reasonable limit).  Any way, if you agree that this is a problem some mechanism needs to be there to work against it, to limit how much frequency will help you gain higher LFs.

Second, I don't think in this case it is the size of the aircraft but number of times you fly.  37x daily is just too many.

I don't know where that limit is.  But I don't think it is really not that limiting.  10x day flights on a 720 pax route is not a problem, as I tested on CDG-BRU route ...

And since 10 x ATR not a problem on a 720 pax route,  this whole change will affect maybe 1% of all of the viable routes, so it will probably have extremely small overall effect.

Well, Sami is dead set against considering LF.  Yet, the algorithms increase the LFs of a seat way too much with frequency, then there are penalties that decrease the LFs (some probably too much).  I think a simple cap on LF would solve the frequency abuse both LH and SH.  A very simple proposition: If you fly small aircraft, smaller than some reasonable limit, you will not get any LF boost if you decrease the size of your aircraft below this limit.  So no harsh penalties just limiting the LF boost due to smallness of aircraft...

you're all still trying to solve the problem with penalties... why?

I remember when I flew (years ago) from Washington to New York, I used the shuttle of... was it Delta or United? ... anyway, we only got a ticket like a but-ticket, no time, nothing... and the flights went every 40 minutes... so, of course that is cool... and if this is economically possible... why not offering it? why adding a penalty? ... I will never understand that...

Quote from: JumboShrimp on July 16, 2012, 05:17:21 PM
First, do you think it is a problem when airline flies a turbo prop 37x per day between 2 slot constrained airports?  If you agree that this is a problem (and I think it is seriously affecting AWS, as it became single dimensional, with players just stretching the frequency benefit further and further beyond reasonable limit).  Any way, if you agree that this is a problem some mechanism needs to be there to work against it, to limit how much frequency will help you gain higher LFs.
Second, I don't think in this case it is the size of the aircraft but number of times you fly.  37x daily is just too many.
I don't know where that limit is.  But I don't think it is really not that limiting.  10x day flights on a 720 pax route is not a problem, as I tested on CDG-BRU route ...

Well, I've lived 12 years in a country where Communist Party said what the limit is and what to produce etc... And of course you needed to keep your mouth shut... Currently it's starting to feel a bit same.
Yes, I think that flying 37x small planes between 2 slot constrained airports is a problem. But it's airport's problem, not passenger's problem. Current solution seems to make it passenger's problem (I haven't heard any good explanation from Sami why should passengers prefer A320 over E170).

Quote from: JumboShrimp on July 16, 2012, 05:17:21 PM
Well, Sami is dead set against considering LF.  Yet, the algorithms increase the LFs of a seat way too much with frequency, then there are penalties that decrease the LFs (some probably too much).  I think a simple cap on LF would solve the frequency abuse both LH and SH.  A very simple proposition: If you fly small aircraft, smaller than some reasonable limit, you will not get any LF boost if you decrease the size of your aircraft below this limit.  So no harsh penalties just limiting the LF boost due to smallness of aircraft...

You are talking about LF but LF is simple a statistic, system does not use LF in any meaningful way (said by Sami), sold seats is what counts... Capping LF is 100% artificial.

Aviation is one of the riskiest businesses in the world. Margins are so tight that even a single 100lb bag added to every flight during a year may produce operating losses to the airline. For the contrary, AWS airlines showcase profit margins that are more suitable to oil sheiks.
If AWS economics where much more tight, then we wouldn't be dealing with this problem, as every airline will be running for every penny out there. There's a reason airlines try to avoid tech-stops, they're not that good economically. You'll have two takeoffs and climbs instead of one, thus adding to the fuel expense and engine wear, which is the main expense nowadays.
Someone previously said that is is better to run a single A321 than 4 Q400. And that's where I'm driving, currently airlines are swimming in dollar-filled pools and don't care if they run hundreds of turboprops instead of tens of jets because they can bear the economics of it, why? profit margins.
I don't actually know where's the difference in economics between a RL airline and an AWS's, but certainly must be there somewhere.

Quote from: Boot on July 16, 2012, 06:49:57 PM
Well, I've lived 12 years in a country where Communist Party said what the limit is and what to produce etc... And of course you needed to keep your mouth shut... Currently it's starting to feel a bit same.
Yes, I think that flying 37x small planes between 2 slot constrained airports is a problem. But it's airport's problem, not passenger's problem. Current solution seems to make it passenger's problem (I haven't heard any good explanation from Sami why should passengers prefer A320 over E170).You are talking about LF but LF is simple a statistic, system does not use LF in any meaningful way (said by Sami), sold seats is what counts... Capping LF is 100% artificial.

You are assuming that the passenger is currently perfectly modeled in AWS.  Which it is not, it is a work in progress.  The allocation model is not even passenger centric, rather flight centric.

Currently, in AWS as it was in MT6, the system would allocate > 75% of demand to those 37x daily Q400 flights.   Just because frequency is modeled to have an overwhelming effect.  Disregarding all the other variables, it is a tradeoff between:
a) capacity
b) capacity turbocharged by smallness of aircraft and frequency

We are all the way the side of b).
The only way to turn it down, we need to move to a)
That's how you offset the strong influence of b).  By letting capacity to be a stronger factor, frequency weaker factor.

Sami can't seem to be able smoothly turn the dial from b) to a).  Instead, a series of limits and penalties were introduced.  I may like the end result, but I don't necessarily like the way it is done.

The more the system moves from b) to a), the more the LFs between larger and smaller aircraft will converge.  Everyone seems to have extremely violent reaction against that, without really thinking it through...

But real life running airline is no fun as you rightly say...infact its nearly impossible to make a profit these days.    

....but is important we remember that AWS is a game, and it needs to be fun in some way.

I just hope these changes don't make the game sterile by not being able to meaningfully effect competing airlines...because the challenge of eliminating (or trying to (legally)) competitors is personally one of my main areas of fun...things that are not fun people don't do  :-\  

Making the game so you cannot meaningfully effect other airlines, which these changes seem to be doing, removes much of the fun / thrill AWS gives me.   Playing a game where all 500 airlines are able to happily survive until the end flying around full 777's with not a worry in the world sounds soooooo boring to me TBH.   People won't agree with that, but I am allowed my opinion.   Maybe AWS isn't the right place for me and I should start up AirforceSim so I can really have some fun with my planes :laugh: :laugh: ....I'll get me coat on the way out.

Quote from: SAC on July 16, 2012, 08:49:05 PM
But real life running airline is no fun as you rightly say...infact its nearly impossible to make a profit these days.    

....but is important we remember that AWS is a game, and it needs to be fun in some way.

I just hope these changes don't make the game sterile by not being able to meaningfully effect competing airlines...because the challenge of eliminating (or trying to (legally)) competitors is personally my main area of fun...things that are not fun people don't do  :-\  

Making the game so you cannot meaningfully effect other airlines, which these changes seem to be doing, removes much of the fun / thrill AWS gives me.   Playing a game where all 500 airlines are able to happily survive until the end flying around full 777's with not a worry in the world sounds soooooo boring to me TBH.   People won't agree with that, but I am allowed my opinion.   Maybe AWS isn't the right place for me and I should start up AirforceSim so I can really have some fun with my planes :laugh: :laugh: ....I'll get me coat on the way out.

It is the absence of price competition that moved all the competition to frequency.

I have not tested the price changes, but it is the price that is the driving factor to consumer preference and lack of profitability of RL airlines.  Competition will still be there, even more so, but there will just be another way to go about it in MT7

Quote from: JumboShrimp on July 16, 2012, 08:34:55 PM
You are assuming that the passenger is currently perfectly modeled in AWS.  Which it is not, it is a work in progress.  The allocation model is not even passenger centric, rather flight centric.

Then why are we wasting time on this "tweaks" instead of fixing the main system ??


I've tested price a little... on a 90% supplied route, a 50% LF plane went to 60% LF with 40% price reduction... so 10% LF = 40% price reduction with demand not 100% met. (It was a 747 on a 400nm route).

Quote from: ezzeqiel on July 16, 2012, 09:14:58 PM
Then why are we wasting time on this "tweaks" instead of fixing the main system

Sorry, but please read the first message of this topic again. Since you seem to have missed what this testing is all about. If I would be only making some tweaks, then a test world would not be running.

And yet again I have to also notify generally. This topic is only for discussion about the test world and matters related to it. Anything else is just unnecessary glogging of the topic...

Quote from: ezzeqiel on July 16, 2012, 09:14:58 PM
Then why are we wasting time on this "tweaks" instead of fixing the main system ??

I am sure there will be further changes when City Based Demand and passenger connectivity is introduced.  But since that may be a while, some tweaks are being done now...

Quote from: ezzeqiel on July 16, 2012, 09:14:58 PM
I've tested price a little... on a 90% supplied route, a 50% LF plane went to 60% LF with 40% price reduction... so 10% LF = 40% price reduction with demand not 100% met. (It was a 747 on a 400nm route).

The question is what the RI is.  If it is less than 100, that's just a temporay phenmenon, just live with it.

The bigger question is tha if you have a simple case.  Let's say 200 demand and 2 airlines supplying 150 pax with everything being identical.  So 100 pax each to each airline.  Now suppose one airline lowers the price to 90% of default ticket price what will be the pax breakdown?

In MT6, you might get 1 extra pax for 101 total and 9% revenue loss.  In real life, you might get extra 15 pax and higher revenue...  I have not tested how this works on test server...

Back to 757's. I am still monitoring them on short hops across the pond, and with no competition loads are still a little disappoint.

I have just ordered a 763 to try on ATL-GLA to see what LF that gets in comparison....will report back accordingly....

Quote from: SAC on July 16, 2012, 09:42:53 PM
Back to 757's. I am still monitoring them on short hops across the pond, and with no competition loads are still a little disappoint.

I have just ordered a 763 to try on ATL-GLA to see what LF that gets in comparison....will report back accordingly....

I am not too crazy about te concept of pax rather not flying than flying a narrowbody.  If there is no other choice, they should take narrowbody just fine.  But have a slighe preference to widebody if one is available....

For info:  Uploaded another revision of the code. And money and bankruptcy restarts are also reset for everyone.

But noticed also that lowering the ticket price may not have any effect at all in some cases; bug somewhere that I cannot find just now.

Quote from: SAC on July 16, 2012, 09:42:53 PM
Back to 757's. I am still monitoring them on short hops across the pond, and with no competition loads are still a little disappoint.

I have not looked into this at all yet.

I have a problem... this message appears:

"Error: You cannot schedule this route to this aircraft since the maximum allowed daily frequency between two airports has been exceeded (max 40 flights per day)."

But I have 30 daily flights scheduled...

Quote from: sami on July 16, 2012, 09:56:37 PM
But noticed also that lowering the ticket price may not have any effect at all in some cases; bug somewhere that I cannot find just now.

If it helps debug, I just opened LAX-IAD, discounted 50%, then went up to 100% and had no change in load factors between the two prices. Opened with 7x daily A321's with at least 1 hour spacing.

Another revision uploaded, that should fix the price issues. Please test; both low and high prices.

All other variables such as flight time, seat type, CI, flight departure/arrival times (etc) are now factored into the calculations. For the flight dep times, do note that it's not advisable to start early morning flights at 05, but make it 06 instead (slight change of settings compared to current worlds).

Quote from: sami on July 17, 2012, 10:18:02 AM
Another revision uploaded, that should fix the price issues. Please test; both low and high prices.

All other variables such as flight time, seat type, CI, flight departure/arrival times (etc) are now factored into the calculations. For the flight dep times, do note that it's not advisable to start early morning flights at 05, but make it 06 instead (slight change of settings compared to current worlds).

Pricing doesn't seem to be working. I opened KLAX -> BWI and CLE on the same day with A321's. I discounted CLE by 50% and left BWI at standard, yet, they're getting the same number of passengers carried. When the route first started (RI of less than 5), I was seeing maybe a one or two pax difference.

That must be the RI messing it then there, I did not test that (only RI 100). Gotta check tonight.

On a route with competition (LAX-FLL), where I'm supplying about 100% of demand and have about 100% of demand supplied by competitors, I was averaging 98-101Y and 9C seats sold per flight (on A321's) at standard pricing. I dropped pricing 25%, and that has increased sales to about 114Y and 10C. I just dropped them to 50% of standard pricing to see what happens.  I would assume this is working as designed, though, still from a player perspective, its more profitable to keep  standard prices still.

Quote from: sami on July 18, 2012, 04:26:45 AM
That must be the RI messing it then there, I did not test that (only RI 100). Gotta check tonight.

Well, the initial loads were about 2x higher than they were before the last tweak for both flights, so it seems like that changed as well.

Seems logical. But I am still thinking if the price reduction effect is too small?  (excluding the RI related bug of course)

However, I don't want to make it too large either, as it will be then very easy for established airlines to force newcomers out of business.

Quote from: sami on July 18, 2012, 01:43:59 PM
Seems logical. But I am still thinking if the price reduction effect is too small?  (excluding the RI related bug of course)

However, I don't want to make it too large either, as it will be then very easy for established airlines to force newcomers out of business.

The 50% price drop seems to have me carrying fewer passengers than the 25% price drop (only have a couple days of data, but its looking like 108Y/10C sold).

If one of the goals of this change is to make pricing a relevant decision point in competition, then I think it needs to have a greater effect than what I'm seeing at the 25% price drop level. However, from a new competitor perspective, its already going to be near impossible for them to fill planes with an RI=0, so I can see how price drops from the incumbant can essentially nuke them from orbit, though, if the new player has a lower cost basis than the incumbant, the lower prices/loads should be sustainable....

Perhaps there's a way to structure the preferences that would protect a new airline with discounted flights (i.e. RI < 100) from incumbant airlines (RI = 100) with discounted flights (by having less of an effect), but once RI for both airlines reaches 100, then the gloves come off and the price war really starts?  (though, this would need to be revisited if RI becomes more dynamic). The "real world" theory on this would be that new airlines in many towns often get free press in the news for starting flights which gives them a bit of a halo effect over the incumbants....

New starts would struggle if price had such a big effect.  If I had 400 a/c I could well afford to fly against a new start up airline with say just 5 a/c, with 5 of my a/c on the same routes but with 75% discount....they would go bust quicker than in MT6, as to have an effect in MT6 I would have to play the "fly more often card", which meant messing with schedules to outwit the competition, which is time consuming and a hassle.   Having to do nothing but reduce prices to kill off new airlines seems a little too easy for the big airlines....a one click way to BK little guys.

I can see Ryanair style $9.99 flights on the horizon....until competition is wiped out and then normal price plus +10% comes back !

Quote from: schro on July 18, 2012, 02:10:12 PM
Perhaps there's a way to structure the preferences that would protect a new airline with discounted flights (i.e. RI < 100) from incumbant airlines (RI = 100) with discounted flights (by having less of an effect), but once RI for both airlines reaches 100, then the gloves come off and the price war really starts?  (though, this would need to be revisited if RI becomes more dynamic). The "real world" theory on this would be that new airlines in many towns often get free press in the news for starting flights which gives them a bit of a halo effect over the incumbants....

Or have price changes for airlines with low RI have more effect than price changes for airlines with high RI (if that is possible?)

IRL new airlines often do need to offer low prices to attract customers when they first start up, in order to steal pax away from older airlines and not be BK'ed. Low prices are one way that people find out about your route if they are not familiar with your company.

Quote from: schro on July 18, 2012, 02:10:12 PM
The 50% price drop seems to have me carrying fewer passengers than the 25% price drop (only have a couple days of data, but its looking like 108Y/10C sold).

I saw it too after looking your airline. Possibly some changes over at competition's routes. I will 'debug' that one anyway to see what it counts on background.

Quote from: sami on July 18, 2012, 02:40:45 PM
I saw it too after looking your airline. Possibly some changes over at competition's routes. I will 'debug' that one anyway to see what it counts on background.

FYI, I've moved the route to standard + 25% now. I'm seeing 91Y/8C sold, which makes it FAR more profitable than selling at standard...

Yes, if you are the only airline supplying a route (or many airlines, but unmet demand), or otherwise you have full planes already, you can now safely increase prices quite much even. There's no steep "cliff" in the price sensitivity anymore. In other words default price is only a rough suggestion, instead of something you'd need to reset to every now and then.

Quote from: sami on July 18, 2012, 02:54:54 PM
Yes, if you are the only airline supplying a route (or many airlines, but unmet demand), or otherwise you have full planes already, you can now safely increase prices quite much even. There's no steep "cliff" in the price sensitivity anymore. In other words default price is only a rough suggestion, instead of something you'd need to reset to every now and then.

Right - this is a route where it is supplied to about 200% of demand between myself (100%) and others (the other 100%). the lack of price sensitivity is going the other way as well - just wanted to alert you the route change incase you start debugging thinking the price is still at -50%  ;D

Question, could it be that the other airlines are discounting as well?

Quote from: ArcherII on July 18, 2012, 03:38:09 PM
Question, could it be that the other airlines are discounting as well?

That's entirely possible - though, I was making and observing changes so quickly that at least my price changes were likely done against a constant price supplied by the other airlines...

I wish to tune this properly, so here are some numbers. Test results on LAX-FLL route:
- 3 airlines operating, demand approx 900-1000, seats avail approx 2200. 11 flights for this given day.
- each airline uses proper equipment for the route, RI 100.
- variables of CI, dep/arr times and seat quality eliminated, as well as cancellations and delays (= none).
- every route & airline set to default prices at beginning.

Standard prices:
- each route (11) sells 86 seats (Y class)

Airline 1 (6 routes) lowers prices 50% from default ($362 > $181):
- others sell 67 (Y) while this airline sells 103 per flight.

Airline 1 (6 routes) lowers prices 20% from default ($362 > $290):
- others sell 77 (Y) while this airline sells 95 per flight.

Airline 1 (6 routes) lowers prices significantly from default ($362 > $30):
- others sell 46 (Y) while this airline sells 120 per flight.

Airline 1 (6 routes) ups prices 10% from default ($362 > $398):
- others sell 88 (Y) while this airline sells 85 per flight.

Airline 1 (6 routes) ups prices 30% from default ($362 > $471):
- others sell 108 (Y) while this airline sells 68 per flight.

Airline 1 (6 routes) ups prices 50% from default ($362 > $543):
- others sell 126 (Y) while this airline sells 54 per flight.

Airline 1 (6 routes) doubles prices from default ($362 > $724):
- others sell 161 (Y) while this airline sells 25 per flight.

Every airline (11 routes) changes prices +50%:
- each route (11) sells 62 seats (Y class).


Note that extreme price decreases will not capture the entire market, which is intended. And the price increase seems to work logically too, you are losing revenue that big increases are not any good if you have competition. All of the above is one example case only still, it varies according to demand and amount of competition (= you cannot make any direct calculation formula of the data above ...)...

Overall I would like to note that passengers are all of similar type still. There are no penny hunting leisure people with no hurry and no "no matter what it costs" type businessman. I would like to introduce the different traveller types and "necessity of travel" in the future to make the demand more dynamic; but for now each passenger represents a sort of average joe..

So anyway, thoughts.. Too big or too small effect on this? This effect is something that is rather easily tuned.

I like the differences but need to make sure that an airline at minus fifty can not profit on that line (including fixed costs)

High prices should lower ri growth.  Inverse true for lower prices.

On that route using A320 his income would be $18643 (103 * $181; -50% price) and direct route expenses (fuel price $700) about $19360. + non-direct costs (staff, marketing, maintenance ..etc). So I assume a "break-even pricing point" is, in this case, somewhere around "default price -20%"?  (rough estimate?)


edit; actually direct route expenses are too small, as the passenger fees go up when more passengers are transported, that $19360 was from standard priced day.

Here's a quick graph on the revenue effect between the variables.

In terms of net revenue between the two different airlines on a per flight basis, it's rather moot between the -20% to +10% price points, so overall, pricing has little to no effect on profitability in that range (with all other factors being equal).

Quote from: swiftus27 on July 18, 2012, 05:10:38 PM
I like the differences but need to make sure that an airline at minus fifty can not profit on that line (including fixed costs)

High prices should lower ri growth.  Inverse true for lower prices.

In this scenario (2x demand level of competition), with a 2000nm stage length flight with MT era fuel, no airline will be profiting even at standard pricing in this scenario once you factor in staff and plane ownership.

So anyway, any actual opinions on that, before I "lock in" the settings. :P

Happy, now let's add the next variable

Will we be able to see what prices competitors are charging otherwise how will we be able to see that it's pricing and not RI we're being beaten on? It's something that you'd easily be able to see in the real world. You could argue RI being public too because it's how 'well established' someone is - which is again easy to see in the real world. Both of the above would mean we could really focus on the metrics of individual routes.

I missed the beta as I didn't see it in time but I am following this with huge interest - just thoughts, (I know Sami doesn't want thoughts just specifics so apologies Sami!)

Quote from: nedson on July 18, 2012, 06:55:47 PM
It's something that you'd easily be able to see in the real world.

Not at all.

Airline ticket pricing is something that changes every day, or every hour even, depending on bookings and other "optimization". And depends even from the location (country) you are buying them and dozens of other factors. It for sure is not just "hey, let's look from the web how much XYZ charges and cut it by five bucks". (to which it would lead here!)

(making such complicated revenue management systems here is quite difficult, so pricing model is simplified)

Quote from: sami on July 18, 2012, 04:43:03 PM
Standard prices:
- each route (11) sells 86 seats (Y class)

Airline 1 (6 routes) lowers prices 50% from default ($362 > $181):
- others sell 67 (Y) while this airline sells 103 per flight.

Total pax, standard prices: 946
Total pax while one airline offers %50% off: 953
Total pax increase when ticket at 50% off: 0.7% (!!!) increase

I think the total pax demand should increase 50% or more with tickets 50% off.

If the pax went up 50% to 1419, the Airline 2 could stay where you have it, but Airine 1 should go to 180 pax.

So some gains for Airline 1 would come at expense of Airline 2, but most would come from extra pax.

Quote from: swiftus27 on July 18, 2012, 05:10:38 PM
High prices should lower ri growth.  Inverse true for lower prices.

Good idea.

Quote from: sami on July 18, 2012, 04:43:03 PM
Airline 1 (6 routes) doubles prices from default ($362 > $724):
- others sell 161 (Y) while this airline sells 25 per flight.

Somehow, this one does not add up, or the total pax demand dependendent on price level is not really modeled in.  Here is what I get:

Total pax, standard prices: 946
Total pax while one airline doubles prices: 955
Total pax increase when 2x: 0.9% (increase!!!)

Quote from: JumboShrimp on July 18, 2012, 08:14:20 PM
Somehow, this one does not add up, or the total pax demand dependendent on price level is not really modeled in.  Here is what I get:

Total pax, standard prices: 946
Total pax while one airline doubles prices: 955
Total pax increase when 2x: 0.9% (increase!!!)

You advertised Punch and Pie....  gotta include all facts.

What I would do is I would get a "blended" ticket price offered:

Flight 1 capacity * flight 1 price + flight n capacity * flight n price / total capacity.

I would multily the total demand by default price / blended price.

If blended price offered is 20% below standard ticket price, demand would increase 20% over standard demand.

Quote from: sami on July 18, 2012, 04:43:03 PM
Every airline (11 routes) changes prices +50%:
- each route (11) sells 62 seats (Y class).

So there is a reduction of total pax from 946 to 682 (28% reduction).  Overall revenue is up, which is probably ok.

So the reduction of pax seems to be modeled in when prices are higher, but but not the other way.

I think the increase of overall demand when one airline lowers price should be one of the factors softening the blow on the competing airline (at standard prices).

Quote from: sami on July 18, 2012, 04:43:03 PM
Overall I would like to note that passengers are all of similar type still. There are no penny hunting leisure people with no hurry and no "no matter what it costs" type businessman. I would like to introduce the different traveller types and "necessity of travel" in the future to make the demand more dynamic; but for now each passenger represents a sort of average joe..

I think the price elasticity should be reasonably easy to implement (compared to all the other changes made), and would add to the realism of price changes.  It would not just be a zero sum game...

I think the penalty for increasing prices above the standard should be harsher...if the airline increases the price 10%, then their passengers go from 86 to 85 and a 30% increase drops the PAX to 68 (roughly a 30%) drop.  I think there should be harsher penalty for increasing price above the default in a competitive route.

For the decreasing price scenario, it seems closer to balancing the need to increase the PAX vs. not allowing bigger airlines to crush little airlines.

Also, as an aside, I assume that there is demand destruction as the price increases, the PAX willing to fly decreases (i.e. if demand is 1000 PAX, and the price increases by 50%, the true demand may only be 600 now).

Thoughts?

Quote from: sami on July 18, 2012, 07:08:23 PM
Not at all.

Airline ticket pricing is something that changes every day, or every hour even, depending on bookings and other "optimization". And depends even from the location (country) you are buying them and dozens of other factors. It for sure is not just "hey, let's look from the web how much XYZ charges and cut it by five bucks". (to which it would lead here!)

(making such complicated revenue management systems here is quite difficult, so pricing model is simplified)

Airline ticketing pricing does change every day, but that doesn't mean that it's impossible for competitors to monitor; it just means that it requires effort to monitor.

If airlines did not know each other's prices, then why do we see often fare battles resulting in matches of exact prices to the penny? Obviously, airlines are monitoring or hearing about each other's fare sales, and are matching (or beating) them accordingly.

For example, when American Airlines, British Airways, and Iberia introduced the recent Diamond Jubilee fare sale ($1,952 for business class anywhere in the US to anywhere in the UK), United Airlines matched the fare within hours, for the exact same price and the exact same validity dates--or at least a close enough approximation of the dates that Expedia/Kayak users searching a given date would see the exact same fares listed for UA and AA.

And that's just one high-profile example; lower-profile, route-specific fare matching also happens all the time on highly competitive routes such as LGA-ORD and BOS-LGA, for example.

Quote from: dmoose42 on July 18, 2012, 09:27:14 PM
I think the penalty for increasing prices above the standard should be harsher...if the airline increases the price 10%, then their passengers go from 86 to 85 and a 30% increase drops the PAX to 68 (roughly a 30%) drop.  I think there should be harsher penalty for increasing price above the default in a competitive route.

For the decreasing price scenario, it seems closer to balancing the need to increase the PAX vs. not allowing bigger airlines to crush little airlines.

Also, as an aside, I assume that there is demand destruction as the price increases, the PAX willing to fly decreases (i.e. if demand is 1000 PAX, and the price increases by 50%, the true demand may only be 600 now).

Thoughts?

No, I think pax demand should be reasonably elastic (and roughly linear, at least in the middle range). I don't think that there should be a specific "penalty" for exceeding the default price, because an industry-wide default price is arbitrary and not realistic.

I think that increasing prices should lower the number of tickets sold, obviously; but increasing from 100% to 105% of default should not cause a steeper drop in pax numbers than increasing from 95% to 100% does.

If you are charging a huge amount (e.g. 150%) then it makes sense to have a steeper drop for that. But I think the current "death cliff" at the default price level (in the current, non-beta universes) is unrealistic, and annoying when the default price itself has sudden drops (e.g. on Jan. 1), making everyone's fares "overpriced" even if they did not actively raise prices.

Why should PAX demand for a given price be a linear function of price on highly competitive routes?  The PAX would just leave the higher price airline for the others.  Obviously this has to be balanced to prevent larger airlines from crushing their smaller neighbors who may be just starting.  That's why I think the non-symmetrical approach is a reasonable compromise between real world behavior and gameplay considerations.

Agree that the death cliff is not appropriate either and a smoother function is required.  However, I do think the customer demand elasticity should be such that for highly competitive routes (seats >= 200% demand) that the % loss in customers when using a price greater than default should be more than the change in the ticket price.

Quote from: dmoose42 on July 18, 2012, 11:35:00 PM
Agree that the death cliff is not appropriate either and a smoother function is required.  However, I do think the customer demand elasticity should be such that for highly competitive routes (seats >= 200% demand) that the % loss in customers when using a price greater than default should be more than the change in the ticket price.

If it's not too complicated to program, then I agree that price elasticity should be higher (more benefit/loss from changing prices) on routes where seats supplied exceed total demand.

I would hope its more than JUST price at work in these cases, that seat comfort, frequency, price, speed, flight times, etc are also acting in 'balance' : otherwise, we will have swapped 'frequency rape' for 'price rape' :  a better seat offering should attract a better price 'ceiling' before numbers fall away, whereas a lousy arrival time should lower it : I guess until the whole thing is in the wild and subject to a full game environment it will be a bit hit and miss until we figure out where the tipping point is now.

Quote from: brique on July 19, 2012, 12:05:42 AM
I would hope its more than JUST price at work in these cases, that seat comfort,

We are talking and comparing just ONE thing here now (price). The other things mentioned are factored in as well with the same style, just excluded from this particular test now (like I posted).

perhaps badly phrased, I was responding to what seemed like a slightly blinkered focus on price and price alone and wanted to add that other factors would come into play when the re-worked game goes 'live'

but, nuff said, apologies, I'll shut up and go play quietly in the corner...

Quote from: JumboShrimp on July 18, 2012, 08:01:10 PM
Total pax, standard prices: 946;  Total pax while one airline offers 50% off: 953

Adjusted, new total demand when 5 flights std price, 6 flights -50% ==> 1143. (+20%)

If all 11 flights are -50%, the new total demand is 1427. (+50%)

Quote from: JumboShrimp on July 18, 2012, 08:14:20 PM
Total pax, standard prices: 946
Total pax while one airline doubles prices: 955

The roundings are messing that up; not feasible to compare single passengers. Anyway; the demand will decrease when the overall supply of products is bad; not if only one airline of many is supplying a bad product (in that case the pax just moves to the other guy). As it would make no sense to count this based on the average - jack up the prices and competition would lose sales too in that case.



To all; With these few adjustments, the pricing effects are finalized for now (for this update). Feel free to post in feature rq if changes are needed. Gotta focus on finalizing the systems before MT#7.

Did you squash the RI != 100 bug with this update?  If so, I'd like to go test....

Quote from: schro on July 19, 2012, 07:27:13 PM
Did you squash the RI != 100 bug with this update?  If so, I'd like to go test....

Working on it. (and the updates above are not uploaded yet either)


Edit: Also the credit refund (2 Cr) has been issued to all players who are currently members of the test world (as per "terms" stated in the first instructions posting). Visible in credit balance only, not on credit history page.


Edit 2: Seems to be working now:

New route; true demand (Y class) is 510 seats. We have 0 route image and two daily flights (no "penalties" of adverse flight dep/arr times, or such), only airline serving the route. With default prices sales are 51 seats per flight (total 102). With -50% prices sales are 99 per flight (198 tot.). With -20% prices sales are 71 per flight (142). When prices are +20% (RI 0 still), sales are 37 per flight (74 total).

Making a new post of this to be sure it's seen.

- The latest revision is now active in test world.

- This includes changes / adjustments to effects in pricing, and also price vs. RI (and low RI generally).

- Demand vs. price effect is also adjusted.

- All other flight variables (seats, flight times etc) are also active.

- Has to be noted that the "cabin comfort" factor is modelled poorly, like in existing game worlds too; since paxes do not really understand yet the difference of sitting in ATR for 4 hours or having a seat in A320 for 2 hours on the same flight, comfortwise (my ears hurt after 30 mins in ATR .. :P) - since we do not have a global pax comfort factor here yet. This is being factored in with different variables such as route flight time effect, and certain aircraft type pax comfort factor... etc. So in effect 4 hr ATR is worse than 2 hr Airbus flight, but not as much as it would be in real life. But something to be added for future. Just for info. (= not really a change)


Please test "everything" as much as you can (also some "silly" things please). My goal is to launch MT#7 on Sunday evening, and my time is rather limited for the next few days, as I shall be away during whole saturday and sunday morning/afternoon (have to attend to a wedding .. will be quite drunk, hopefully. :P). So if you have any feedback, please post them asap.

Minor adjustments are possible later on too, but at this time all of the main aspects are done. And I am focusing on fixing any problems and inconsistencies that could still arise.

Quote from: sami on July 19, 2012, 07:54:58 PM
- All other flight variables (seats, flight times etc) are also active.

- Has to be noted that the "cabin comfort" factor modelled poorly, like in existing game worlds too; since paxes do not really understand yet the difference of sitting in ATR for 4 hours or having a seat in A320 for 2 hours on the same flight, comfortwise (my ears hurt after 30 mins in ATR .. :P) - since we do not have a global pax comfort factor here yet. This is being factored in with different variables such as route flight time effect... etc. So in effect 4 hr ATR is worse than 2 hr Airbus flight, but not as much as it would be in real life. But something to be added for future. Just for info. (= not really a change)

Just to clarify re "cabin comfort": you mean that seat quality (e.g. benefit of using premium vs standard seats) is now fully modeled, and it's just "plane type" (e.g. propeller noise) that is weakly modeled / not really modeled, right?

Seat quality itself is modelled. But it makes no difference if you have a HD seat in ATR or HD seat in A319.

The ATR vs. A319 (example) difference comes then from a) flight duration and b) fleet group related "aircraft type vs. pax comfort index" (where new jets have a higher rating over props or old jet models). But "b)" is a rather small factor overall, since generally passengers make no difference between a 737 or 757 for example, so it cannot be very dominant factor.

(In the future this is an area to be improved.)

For same type of seat pax won't notice then. But what about a full-flat C cabin against a standard C layout? Would it be worthwhile in order to gain market?

Quote from: ArcherII on July 19, 2012, 08:16:24 PM
For same type of seat pax won't notice then. But what about a full-flat C cabin against a standard C layout? Would it be worthwhile in order to gain market?

Have not tested that in effect, the formula for seat comfort is the same as before.

I have no idea what you have done, or what has happened, but all of my load factors have all of a sudden (from last week to this week) dropped by at least 15%-points (going from 90+% eco to 83%, from 90% to 70% in C-class, and from 85 to 69% in F-class).

For trying to get it "fixed", I reset prices to default, and then lowered them by 5% (was at standard before, refreshed last about 3 ingame weeks ago).

cheers,
Jona L.

Quote from: sami on July 19, 2012, 08:23:46 PM
Have not tested that in effect, the formula for seat comfort is the same as before.

Premium or luxury seats should provide an actual, measurable benefit over standard seats, shouldn't they? The manual/game text state that it does, so that should be modeled in the formula... otherwise, it's just a sucker option that kills noobs by lowering aircraft capacity when better seats are installed.

Quote from: EsquireFlyer on July 20, 2012, 01:47:05 AM
Premium or luxury seats should provide an actual, measurable benefit over standard seats, shouldn't they? The manual/game text state that it does, so that should be modeled in the formula... otherwise, it's just a sucker option that kills noobs by lowering aircraft capacity when better seats are installed.

The problem is that the load factor will increase because there's less seats, and you might sell more seats than you would with standard seating, but your total revenue will be lower because you burned half the plane on a single lie flat seat instead of 300 sardine tins....

Quote from: schro on July 20, 2012, 02:29:48 AM
The problem is that the load factor will increase because there's less seats, and you might sell more seats than you would with standard seating, but your total revenue will be lower because you burned half the plane on a single lie flat seat instead of 300 sardine tins....

The problem is that what you said above is a problem.

For comparison, changing Y seats into F or C seats also lowers capacity (and potentially raises load factor for that reason) but does not lower total revenue (provided that there is F/C demand to fill your seats) even if you "burn half the plane on a single lie flat seat instead of 300 sardine tins."

If pax are willing to pay more for F and C seats, they should also be willing to pay more (to a lesser extent) for better seats within Y (or within F or C).

Similarly, when a route is oversupplied, improving seat quality should cause more passengers to want to fly with that airline (an entirely realistic outcome), rather than causing less passengers fly with that airline because the system only looks at total number of seats supplied and the airline is now doing less market-flooding than before (the current outcome, and totally unrealistic). In reality, passengers do not care how many empty seats an airline is flying around, but they do care about the quality of the seat that they are sitting in. So, for an airline upgrading its seats to cause it to lose passengers (in absolute numbers, not LF%) in an oversupply/competition situation makes no sense at all.

What the game does not need is another sucker option (such as "aircraft insurance") which all experienced players ignore and which serves only to put newbies at a disadvantage to experienced players (because newbies rely on misleading text in-game and in the manual, and may purchase the sucker option as a result).

If premium and luxury seats are not going to make a difference, then they should be eliminated altogether (leaving only HD and standard). If premium and luxury are put in the game as options, they should actually provide a competitive advantage under some circumstances (when appropriately used), rather than being simply a newbie deathtrap.

Quote from: EsquireFlyer on July 20, 2012, 03:32:43 AM
If premium and luxury seats are not going to make a difference, then they should be eliminated altogether (leaving only HD and standard). If premium and luxury are put in the game as options, they should actually provide a competitive advantage under some circumstances (when appropriately used), rather than being simply a newbie deathtrap.

They _do_ provide a competitive advantage. The problem is that they don't provide a profitable advantage.

If you want to pull the real world card, take a look at American's venture into offering  (http://seatexpert.com/blogs/ask_the_seat_expert/2009/05/20/american-airlines-more-room-throughout-coach-product/) economy+ in the whole cabin....

Sami, could you check one thing in my Airline: https://www.airwaysim.com/game/Info/Airline/151/#AirlineInfo ?

I have ended 2 marketing campaigns: one worth 900k/week and one 800k/week. I opened instead 2, one campaign worth 3,5mln/week. And suddenly my CI droped, from CI 70 to CI 62. Also my LF droped around 5-6% (not sure if it is related with CI drop).

Last game scenario I played, there were not much difference what kind of campaign I used, all matter was total money spend on it. Did you adjust that lately?

Quote from: Jona L. on July 19, 2012, 10:21:30 PM
I have no idea what you have done, or what has happened, but all of my load factors have all of a sudden (from last week to this week) dropped by at least 15%-points (going from 90+% eco to 83%, from 90% to 70% in C-class, and from 85 to 69% in F-class).

Ive seen similar sort of results - the consistency of the LFs (which i thought was realistic) seems to have gone

Quote from: EsquireFlyer on July 20, 2012, 01:47:05 AM
Premium or luxury seats should provide an actual, measurable benefit over standard seats, shouldn't they? The manual/game text state that it does, so that should be modeled in the formula...

Pls read again what I wrote. It is modelled, and I have not tested it particulaly for this update since it has not changed from the prev version.


The variations in loadfactors is probably also caused by the fact that random daily effects are on there at the moment (by mistake though...).

Quote from: Mr Pepto on July 20, 2012, 06:29:37 AM
Sami, could you check one thing in my Airline

You CI is in steady decline. Which hints that delays have grown or marketing is not efficient as before.

And if latter has changed only recently, then I'd look there. (it has not changed for this update)

Small comparison on seat comfort:

1000 nm route, demand around 520, equipment A300-600 (scheduled time 3 hours), other flight (of the same airline) uses HD seats and other standard seat (seats are Y351 vs Y279), standard prices, no competition. Ticket price $170.

Route image 0: HD sells 45 seats ($7650), while STD sells 54 ($9180).
Route image 100: HD 233 ($39610), STD 274 ($46580).

So in this case it pays off to fit less seats to the plane since flight is longer than the "suggested" 2 hours with HD seats.

Since the last change, my LF on some routes with no competition and suitable aircraft have dropped to the 50% level. The flights have at least one hour apart so I don't understand why? Any ideas?

Take a look at OSL-BGO https://www.airwaysim.com/game/Routes/Planning/X/ENGM/ENBR/?TB_iframe=true&width=800 it's a high frequency route but the flights are spaced appropriately, but load factors have fallen off a cliff.

Will be adjusted, found an issue there (due to you having 14x flights (or so..) a day).


The random effects on sales are also turned back OFF in this world.

Quote from: sami on July 20, 2012, 10:07:08 AM
Will be adjusted, found an issue there (due to you having 14x flights (or so..) a day).


The random effects on sales are also turned back OFF in this world.

Back to normal now with LFs of 80%+ again.

Quote from: schro on July 20, 2012, 03:41:52 AM
They _do_ provide a competitive advantage. The problem is that they don't provide a profitable advantage.

A loss-making advantage is, by definition, not a competitive advantage. If the only effect of an option is to kill newbies who don't know better, the option should not be offered.

If the option is offered, it should be something that provides a competitive (profitable) advantage at least under some circumstances, when appropriately used, not something that only newbies fall for and all experienced players avoid like the plague.

Quote from: schro on July 20, 2012, 03:41:52 AM
If you want to pull the real world card, take a look at American's venture into offering  (http://seatexpert.com/blogs/ask_the_seat_expert/2009/05/20/american-airlines-more-room-throughout-coach-product/) economy+ in the whole cabin....

Interesting example, but you realize that there are at least 10 counterexamples, right? Your own article link cites two of them (United and JetBlue).

Many European and Asian airlines (e.g. British Airways, Virgin, Singapore, Thai, Emirates etc.), routinely charge more than US airlines and lower-cost Asian/European competitors do on the same routes, by offering much better seats (Emirates only in F/C; their Y seats are craptastic). And most of these are the airlines that remained profitable when the entire US airline industry was in bankruptcy status a few years ago. Providing better seats must provide a profitable advantage, when done correctly.

Now, as for American. The key problem was that American did not charge more than its competitors did on the same routes. This could be for a variety of reasons, such as: (1) poor company image, compared to BA, VS, SQ, EK, etc. means less ability to raise prices, and (2) inadequate marketing of the "More Room Throughout Coach" campaign. Anyone can tell you that charging the same price for fewer seats is going to lower your profits. What I am saying is that you should be able, in some cases, to improve profits by charging a higher price for fewer seats.

For example, when United introduced "premium" seats in economy (just more legroom, no more width or any better service) as "Economy Plus," they allocated the seats to pax who would pay more for them (either implicitly, in the form of repeat business on UA to earn elite status, or explicitly, in the form of a small % upcharge to sit in E+, but much less than the price of a cabin upgrade). This model was so successful and profitable that it has been copied by Continental, Delta, and now American as well (doing it the right way this time, even though MRTC was a disaster before).

In effect, the AWS method of modeling seat quality forces players to use the American model if they want to improve seat quality (a losing strategy); I am saying that it should allow for the possibility of using the British Airways or United model, by making pax willing to pay (a little bit) more for better seats.

Or at least more willing to choose the airline with better seats in an oversupply situation. Which is the exact opposite of what happens now, because in an oversupply situation, between premium and standard seating (holding extraneous variables such as price and scheduling constant, for comparison purposes) passengers will flock to the airline offering worse seating because oversell demand is allocated primarily by capacity, and the airline offering crap seating has higher capacity. This is a totally unrealistic result, because there is no reason why passengers would actively prefer the airline that jams more people into smaller seats, for the same ticket price. Why would any passenger do that?

And the above is just about economy. The effect of using premium or luxury seats in C and F in the real world is even more obvious. Singapore and Emirates, for example, make tons of money while doing in C and F exactly what you said not to do in AWS, i.e., "burn half the plane on a lie flat seat rather than 300 sardine tins."

So, from both a realism perspective (pax prefer better seats, especially in F and C, but to a lesser extent in Y) and a user experience perspective (not helpful to have always-losing options that experienced players won't touch and only newbies fall for), passenger preference re: premium and luxury seats should be modeled, at least on long flights. Like the way their preferences re: HD vs. standard seats are currently (and IMO appropriately) modeled.

Quote from: sami on July 20, 2012, 07:49:01 AM
Pls read again what I wrote. It is modelled, and I have not tested it particulaly for this update since it has not changed from the prev version.

The variations in loadfactors is probably also caused by the fact that random daily effects are on there at the moment (by mistake though...).

Thanks, sami! But I did read what you wrote, and what you wrote before (and even what you wrote after) only directly addressed HD vs. standard seats, which I think are modeled well.

My post was specifically about premium and luxury seats, and I was saying that they should be modeled more like the HD/standard seats. I understand that they were not changed in this version, but I was suggesting that they should be changed (if not in this update, then in a future update). For example, maybe on flights longer than 12 hours (or maybe 6-8 hours in F/C), pax preference for premium over standard seats should be as noticeable as pax preference for standard over HD seats at the 2 hour mark.

Or, in an oversupply situation, pax should prefer the airline with better seats (opposite of current model, where more pax actually prefer the airline with worse seats, since that airline has more capacity and pax are allocated based on capacity).

So that premium seats aren't just a guaranteed money loser.

Yea, gotta test the lux seat too when I can.

Quote from: sami on July 20, 2012, 12:55:06 PM
Yea, gotta test the lux seat too when I can.

Thank you very much! Sorry for complaining.  ;D

I seem to be getting an un warrented bankruptcy warning, no idea how or why

(https://www.airwaysim.com/forum/proxy.php?request=http%3A%2F%2Fwww.freewebs.com%2Fjetairliners%2FUntibanktled.jpg&hash=2273fca3e6936f5f2db4ad20965a7a75574cff8b)

"final" revision is active, with daily randomization on too (= you may see a small drop in sales).

I'm seeing some different results with my TATL flights right now that seem a little odd - just want to make sure that they are consistent with the intended design/behavior.

A day or two ago, I reset pricing to standard, then raised all fares to +20% of standard. RI = 100 and CI = 100.

I'm running a number of A330 planes to Euro destinations that are fairly well matched with daily demand. At the +20% price point, I'm seeing loads in Y of around 50-55% against no competition, and C/F seem to be doing better, some in the 80-90% range, some back down in the 50's. At the same price point, KLAX-RCTP is flying with full planes....

To see if this is a pricing issue, I lowered LAX-ORY down to standard pricing to see if that has any significant impact on loads (but based on the pricing variable discussion, I don't think that I will end up with a significant boost).  I'd like to think that this is related to the 1x day frequency penalty, but I'm also seeing similar things happening to LAX-LEMD that is 2x daily (though, I do have a 5am landing in the mix there). It'll take a bit to see if ORY changes... but something smells funny to me.

A question that just crossed my head...How's going to be for the very large demand routes like HND-Sapporo or or any other Japanese route for that matter, with the size of aircrafts? In some instances if you operate 753s over that route you'd eventually hit the maximum frequency and start to lose pax accordingly?

If so, how is it gonna be for a route like Congonhas-Santos Dumont where up to 12,000 pax are willing to cross? Mind that the Dumont's runway is only 1320mts (barely possible for a 700NG, or A319). Would an airline have a "cap" on the amount of flights and therefore not meeting demand?   

Is the test game going to come back or is it done for good?

There´s a regular new world based on the new formulas (MT7), so why should it?

Quote from: BryanIAH on July 13, 2012, 04:01:20 PM
UA runs multiple LAX/SFO-ORD/IAH flights leaving from 00:00-2:00 and arriving 4:00-6:00. I think Spirit operates even worse flight times on some of their shorter routes.

??? Hmm, that didn't sound right, so I just checked UA's LAX and SFO to ORD schedules and found nothing leaving after 00:00 or arriving before 06:00. The "worst" flight time I found was UA828 leaving SFO at 23:59 and arriving in ORD at 06:01.

The CO side, on the other hand, has traditionally had some bad flight times to IAH. From SFO, it's still not bad; the worst is UA480 leaving SFO at 23:59 and arriving IAH at 05:42. From LAX, there is UA1170 leaving LAX at 01:12 and arriving IAH at 06:24. But CO kind of runs like a regional carrier in a global airline's clothing.

However, there is still nothing leaving as late as 2am or arriving as early as 4am between those cities. UA's lounges typically aren't even open that late (or early), so the elite/premium pax would be p***ed.

Quote from: sami on July 20, 2012, 09:25:03 PM
"final" revision is active, with daily randomization on too (= you may see a small drop in sales).

I did not get to test the final revision, I just glanced at some of the flights I was monitoring.

One general concern that I have had in < MT7 and the last glance at the test server is 100% LF - or more precisely, rarity of it.  Once all the variables were enabled, I think all of my flights that had 100% LF prior to all the variables being enabled they all dropped below 100%.

In real life, full flights are not that uncommon.  Everytime I hear an announcement asking for volunteers to not fly, I know it is a full flight.  And I hear it quite often.  I don't know about the rest of the world, but in the US, overbooking flights is quite common.  Maybe 10-20% of the flights I take.  In AWS, it is a rarity, even when flying a perfect aircraft, perfect time, and the route is undersupplied.

I am not sure where at the allocation process these variable get applied.  But my thinking is that on an undersupplied route of say 200 pax demand, 150 supply, the system should allocate to the only flight serving it, than apply all the things depressing that, resulting in say 165 pax allocated.  Given 150 supply, the LF should be 100% most of the days of the week, if not every day.  So let's say all the variables are great, I should be seeing 100% LF 100% of time.

My experience with routes like this from MT6 is that on a route like this, with perfect supply, I end up with 100% LF maybe 1% of the time...  I think it should be adjusted.