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

Author Topic: Time for another experiment  (Read 5877 times)

Offline Sanabas

  • Members
  • Posts: 2161
Re: Time for another experiment
« Reply #20 on: December 23, 2012, 09:15:21 AM »
This is meant to be an experiment, so that's my excuse for some of my more questionable route choices.  :laugh:

Stat geekery ahead, including a look at ASK/RPK, why I think they're pointless, and what I'd use as a replacement stat, if we were able to put it in. And as always, I'm making it up as I go along, pointing out things I notice, and may not be hugely coherent. Feel free to skip to the conclusions at the end...

I need to be making about 250k/plane/month to even think about breaking even, and I need to be making about 400k/plane/month to think about breaking even once I factor in leasing & maintenance. I'll call it 360k/month, 12k/day, $500/hour. If a route can conceivably make me $500/hour, then breaking even is somewhat feasible, if I fly 24 hours/day.

So, my current routes...

Plovdiv: 74 NM. 3 flights. 2:50 round trip. At 100% full, $1600 profit. $565/hour. Fuel is about $250, other costs about $1400. Actual profit across all 3 is $416/hour. RPK is 132, ASK is ~115.

Varna: 197NM, 5 flights, 4:10 round trip. At 100% full, $1265 profit. $304/hour. Fuel is $520, other costs $1600. Actual profit ignoring the night flights is $275/hour. RPK is 52, ASK 50.

Burgas: 181NM, 4 flights, 4:00 round trip. At 100% full, $1300 profit. $325/hour. Fuel is $500, other costs $1560. Actual profit $220/hour. RPK is 56, ASK 49.

Thessaloniki: 132NM, 3 flights, 3:35 round trip. At 100% full, $1490 profit, $416/hour. Fuel is $400, other costs $1530. Actual profit $155/hour. RPK is 78, ASK 56.

Athens: 287 NM, 3 flights, 5:05 round trip. At 100% full, $971 profit, $191/hour. Fuel is $700, other costs $1880. Actual profit, on 89% full plane, is $114/hour. That one extra seat each way is an extra $398, $78/hour. RPK 37, ASK 32

Kavala: 120 NM, 1 flight, 3:05. 100% full, $1750 profit, $568/hour. Fuel is $337, other costs $1284. Actual profit $142. RPK 85, ASK 49

The rest of the flights aren't full enough to be making much profit at all. They're all below $100/hour, or in a lot of cases, below 0. But I've still got ASK numbers of 58, 42, and a couple of others above what Athens has.

Bucharest HC: 162NM, 3:50. 100% full, $1481, $386/hour. $450 fuel, $1542 others.

Pogdorica: 185 NM, 3:45. 100% full, $1700, $454/hour, $460 fuel, $1330 others.

Istanbul-S: 285 NM, 4:55. 100% full, $257/hour, $650 fuel, $1670 others

I'll just list the longer, stupider routes, leave out the other 150-300 NM ones, as they're similar.


Odessa, 383 NM, 5:45, potential profit of $205/hour. Even though the round trip length is double Plovdiv's, the potential revenue is just 15% higher.

Ganja, 1030 NM, 11:15. 4 times the length of Varna, less than 25% increase in ticket prices. I don't think that disparity will be as glaring in bigger planes, with longer turn times. Think I'll run at least two more test airlines in the rest of MT, see what I can find out. Fuel is $2300, and what I find really strange is that despite there being just 2 people on the plane, vs 7-9 for the other flights I've listed, and despite Ganja being a much smaller airport than the others, the other costs of ground handling/pax/landing/nav fees are $2000, 40% higher than the route to Plovdiv. I'd assume handling/pax/landing are significantly less, and am surprised nav fees make that much difference. The game clearly breaks down those 4 individually somewhere, as they're separate entries on your income statement. Time for a feature request to make them viewable on a route by route basis. Unless I can do it already, and haven't noticed. Anyway, my best case for Ganja is actually to lose about $30/hour. My best case just for revenue is $350/hour.

Lahore, 2491 NM, 26:20. I haven't found a single, virtual AWS pax silly enough to buy a ticket for this yet. I just dropped the price to $10, curious if anyone will pay for that. There is ~50 demand, and I was nice enough to put in 8 standard seats, not 9 HD ones. fuel is 5000, other costs are 4200, presumably so high thanks to the tech stop & the length. So my best case is losing $60/hour. But if I had a profitable airline, I'd leave it flying.  ;D *update* I have just discovered the system sets minimum prices. 100/200/500 in this case, when default is 483/1308/2190. But 2 people paid $100 each to fly from Bulgaria to Pakistan on an 8 seat plane. ;D :laugh: I'm officially a LH airline.  8)


So, looking at all that, there are a few conclusions:

Shorter flights are significantly better. The only flights with a hope of approcahing my $500/hour benchmark are those under 125NM, plus a few under 200NM to smaller airports, where the fees are less. Anything beyond that, and I'd need full planes and significantly upgraded prices to approach my break-even point. Also, if I don't actually fly 24 hours/day, that break-even point gets higher.

Even on the best route, the expenses are just over half the potential revenue. That means if I only sell 8 instead of 9 seats, I'm looking at 22% of my route profit gone, taking me from ~10% above break even to ~10% below. That's a significant difference.

The fuel expenses aren't really an issue. The ground handling, etc are significant. Flying into my size 4 HQ, even on the shortest of routes, costs about $800 in fees, $100 per pax on a $180-190 ticket.

Therefore, what I really need is a smaller HQ airport, with quite a few very short, very empty routes. US might have it, but staff costs are going to be expensive. Though my current level of 200k/month/plane works out to be under $30/hour. Even if US wages are double, that could easily be offset by more short routes. So, time to find myself an airport, and restart, try and keep staff a bit lower, stick to short routes only, and see what happens. If I can actually show long term ability to break even, I'll be pretty happy.

And on revenue/ASK: As you can see, the order of RPK/ASK does not correspond to which is the most profitable flight. Which is one reason I don't think it is useful. to me, profit/hour is a much more useful stat, because it actually allows you to directly compare routes. Can compare them across planes if you want to, as well, if you work out the plane's expenses/hour. Lets you see the optimal leg length, lets you see how much you're punished for a non-optimal leg. Which RPK/ASK can't do. If you sort all routes on a given plane type by RPK, you're basically just sorting by leg length. Sort by ASK, and it's just a combination of leg length & LF. If you could show profit/hour, that would actually be helpful to me. I think it would be an easy stat to generate. Profit is already in the route stats, and for the time, I'd use flight time + 2 x what you have set the turn time for that plane to. So that would offer me useful data, and if for some reason I habitually used short turns, that would be reflected in the numbers too.

Offline Sanabas

  • Members
  • Posts: 2161
Re: Time for another experiment
« Reply #21 on: December 23, 2012, 09:20:09 AM »
Fleet common hasn't changed much since public AWS launch. If you look back through the posts about fleet common, you'll find a central theme to all of them.

"Rapid expand" out of the initial fleet common penalty. Once you go past 10 aircraft of the same fleet type, you'll see the costs start to even out.


Perhaps Sami will change it. However I suspect it's not high on the list right now.

Yeah. But what I found surprising is that while the plane side of commonality shows a slight, steady decrease as you grow (31k/plane for 2, 26k/plane for 8), the engine commonality is much more dramatic. From double the plane side at 2 planes, to barely half at 8 planes, to ~15% for 58 planes in DOTM.

Offline Sanabas

  • Members
  • Posts: 2161
Re: Time for another experiment
« Reply #22 on: December 23, 2012, 10:10:30 AM »
Best two choices I can find are Bauerfield, Vanuatu and Tagbilaran, Phillipines. Both have roughly 1000 daily pax under 200 NM, spread over 23 & 21 airports, and all routes are at least 20 pax. Both are size 3 with no competition. Vanuatu wins because it has slightly more under 100 NM (370 vs 260), all of the destinations are size 1 or 2, and it doesn't have curfews. Suspect the staff price might be a little better, too.

Offline Sanabas

  • Members
  • Posts: 2161
Re: Time for another experiment
« Reply #23 on: December 23, 2012, 11:03:07 AM »
Smaller airports also mean less taxi time. No leg under 100 NM takes more than 2:40 for the round trip. That means 9 legs each for my first two planes, 18 legs to 10 different airports between the two of them.

So, 2 planes, and the costs are:

Staff: 116 staff, 268k, 134k per plane. 30% more staff than last time, thanks to the extra destinations & number of legs. But actual cost is cheaper. The 3 high level managers needed to run my 2 planes cost me 28,278 per month. The 6 high level managers I had for 5 planes in Bulgaria cost 78,082. Across the board, wages are 71-72% of what they were. I'd love to have a searchable database of staffing salary weights, or have it included next to the tax rate. i.e. Singapore = 1, US = 0.92, Vanuatu = 0.57, whatever it happens to be, and whatever country is used as the base.

Plane commonality: 60,501
Engine commonality: 116,659 - near enough to identical for both. Might actually be identical, don't know.
Insurance: 11,645

So overall, roughly 230k/plane/month. A little better than before, add in maintenance and slightly cheaper leases, and I'm looking at ~$450/hour as a break-even target, and hopefully that drops further as more planes arrive.

Offline Sanabas

  • Members
  • Posts: 2161
Re: Time for another experiment
« Reply #24 on: December 23, 2012, 11:20:42 AM »
The 244 slots I bought cost me just 4,080.  ;D Not that it makes any real difference long term, the 12 mill start up cash is more than enough to get a dozen planes and all the slots for them. But it still beats paying 1 mill for a set of 7.

Promising start on day 1. Best 2 routes sold 15 of 18 seats, made $483 and $514/hour, and would be $686 & $728 if all 18 sold.

Offline Sanabas

  • Members
  • Posts: 2161
Re: Time for another experiment
« Reply #25 on: December 23, 2012, 03:18:16 PM »
First full week, I only lost 160k, and 130 of that was leases & loan payments. Looks promising...

Offline Sanabas

  • Members
  • Posts: 2161
Re: Time for another experiment
« Reply #26 on: December 23, 2012, 09:19:27 PM »
4 planes scheduled now, and I may have got a little carried away, I've actually got 11 planes arrived/ordered in total. Possible I should have stopped at 8 or 9, to undersupply demand and crank up prices.

However, since this will hopefully be the entirety of my fleet, and I'm now scheduling for the rest of the game and leaving it, and I have exactly 7 planes left to schedule, I may as well get carried away. 7 day schedule here I come.  :laugh:

Offline Sanabas

  • Members
  • Posts: 2161
Re: Time for another experiment
« Reply #27 on: December 23, 2012, 11:18:06 PM »
7 day schedule done. 49 routes per plane. I've got 15.1 hours utilisation out of a 9 seater.

I'm actually going to grab a 12th & final plane. That should leave all my routes about 70-90% supplied, allowing me to raise the prices a bit once RI hits 100.

Other benefit of 7 day scheduling when you're cutting things as fine as this is I can control the prices of individual days if I want to. Monday & Friday tickets can be more expensive. Don't think I'll go that far, because of the tedium. Hopefully I can make a little money without it...

5 planes:

Personnel: 428 staff, 988k/month, 198k/plane

Plane comm: 185k/month, 37k/plane

Engine comm: 127k/month, 25k/plane

Insurance: 28k/month, 7k/plane

Total: 267k/plane

Offline Sanabas

  • Members
  • Posts: 2161
Re: Time for another experiment
« Reply #28 on: December 24, 2012, 12:12:09 AM »
7 planes:

580 staff, 1350k/month, 193k/plane

plane comm: 199k/month, 28k/plane

engine: 135k/month, 19k/plane

insurance: 39k/month, 6k/plane

246k/plane.

brique

  • Former member
Re: Time for another experiment
« Reply #29 on: December 24, 2012, 12:15:09 AM »
Given that Mon & Fri demand can be significantly higher, noticably so on low demand routes as a proportion, you should have room for some major price-hiking on those days : its a fair bit of micro-management, yes, but they are also 28%-ish of your total flights, and any extra revenue is going straight to the bottom line : so the extra % squeezed out there is going to be a more than useful contribution to overall profitability.

I do reckon, to make the small birds work, you have to actively go after those extra percentages on the fares, even to the extent of planning your route structure to allow day-to-day price differentials. Maybe not 7-day scheduling, but splitting the week into 2 routes, 'busy days', (Mo,Th,Fr) and 'slow days' (Tu,We,Sa,Su) is a do-able strategy.


Offline Sanabas

  • Members
  • Posts: 2161
Re: Time for another experiment
« Reply #30 on: December 24, 2012, 12:28:54 AM »
If I make money without it, I doubt I'll bother. Especially as if I reset prices to default, I have to redo it.

But if doing it's what gets me over the line of profitability, then I will.

8 planes now:

Staff: 658, 1526k, 191 (32 managers, 72 seats.)

Plane: 205, 26

Engine: 138, 17

Insurance: 45, 6

240k/plane.

Will be at 11 or 12 planes after some sleep, should hopefully drop a bit more.

Offline Sanabas

  • Members
  • Posts: 2161
Re: Time for another experiment
« Reply #31 on: December 24, 2012, 09:13:01 AM »
I have routes that are making enough to make me profitable. But I'm dragged down by the newer, and longer (200-300 NM) routes. See what happens as CI goes up. I'm optimisitc it can work. But if it doesn't, there'll be a 3rd attempt, only a couple of planes, and the shortest routes possible. Which I think will definitely make money.

11 planes now:

staff: 894, 2072k, 188k

plane: 226, 21

engine: 147, 13

insurance: 70, 6

Down to 228k/plane.

The older planes are 25k/month lease, 45k/month maintenance.

Newer planes are 40k lease, 37k/month maintenance.

So ~300k/plane, $420/hour as the benchmark.

12th & final plane arrives in a week, that one I actually bought.

Offline Sami

  • Administrator
  • Members
  • Posts: 14535
    • AirwaySim - Are you the next Richard Branson?
Re: Time for another experiment
« Reply #32 on: December 24, 2012, 09:57:45 AM »
btw. After done, I would like to get a summary (short list of items) of things that need to be worked to improve this.

Offline Sanabas

  • Members
  • Posts: 2161
Re: Time for another experiment
« Reply #33 on: December 24, 2012, 10:06:26 AM »
No worries, will do.


Offline Sanabas

  • Members
  • Posts: 2161
Re: Time for another experiment
« Reply #34 on: December 24, 2012, 03:53:13 PM »
Haven't got as far as scheduling plane #12 yet.

Just finished a week, with some B-checks. 1.24 mill revenue, at 64% LF or so. So there's about another 600k in potential revenue if all planes fly full.

Lost 380k, but did actually make 25k or more profit at the end of the day I saw change, which was the Sunday. So not enough to offset the 480k staffing bill, or the daily midday expenses, but hopefully in week to week profit once loads improve. Whether that will be enough to pay for B & C checks remains to be seen. As does whether I run out of money before I get there. 5.3 mill in the bank, 4.3 mill of secured loans available, so should be good for at least 6 months, by which time RI is 100 and I can see exactly where I'm at.

Offline Sanabas

  • Members
  • Posts: 2161
Re: Time for another experiment
« Reply #35 on: December 24, 2012, 04:14:57 PM »
Those daily expenses (rent, marketing, commonality) are 23k.

Random graph of one route. It's sold 8 tickets on one day. A lot of 7 & 6. As you can see, the expenses are basically constant at $750. 6 tickets is $1098 in revenue, $350 profit. 7 tickets is $530 profit, 8 tickets is $710, 9 tickets will be $900. Takes 2:55 for the round trip, so 1.5 hours for that leg. 8 or 9 tickets is above my profit threshhold. 7 isn't. But 7 tickets at +17% would give me $500/hour profit.

So, I think I've made a mistake by getting 12 planes. It's only decreased the cost per plane by roughly 10% (330 down to 300k/month). But it's harder to fill the planes offering 36 seats on a 50 seat route, especially with RI at 18, compared to only offering 9 or 18. Would have been better off undersupplying and overcharging while RI grows. I should get near 100% LF once RIs are all 100 though.
« Last Edit: December 24, 2012, 04:36:07 PM by Sanabas »

Offline Sanabas

  • Members
  • Posts: 2161
Re: Time for another experiment
« Reply #36 on: December 24, 2012, 04:34:36 PM »
I made 61k when the day changed.  ;D

That's 38k profit for the day.  ;D

That's nearly 300k/week.  ;D

That's only 200k short of paying the staff.  :'(

Want RI to increase faster....

Offline Sanabas

  • Members
  • Posts: 2161
Re: Time for another experiment
« Reply #37 on: December 24, 2012, 08:34:31 PM »
300k loss for the week. 1.3 mill in revenue. 500k each for staff & ground handling.

Average ticket price is $186. $70 of that is needed to pay for ground handling. Roughly $400 per flight for a 9 seat plane.

I don't know about RL, but $400 to deal with a 9 seater every time it lands at a tiny airport seems really high. Most of the times I've been in a plane, it's been that size, landing at a very small airport, and I've been jumping out of the thing. There definitely wasn't that sort of money getting spent each landing.

Offline Troxartas86

  • Members
  • Posts: 877
Re: Time for another experiment
« Reply #38 on: December 25, 2012, 07:00:12 AM »
I thought I'd chime in here as a long-time user of the PC-12 in this round (you may actually be flying some of my former planes). I originally ordered a few just to get my numbers up so I could expand to a 2nd base but I got tired of watching them rust so I started giving them routes. That seemed to sort of work so eventually I wound up with ten that actually fly all over French Polynesia. At one point before fuel went over 1000 I could actually boast that I was serving every airport in the country thanks to these tiny birds.

In recent years I have downsized the fleet a little and closed a few hopeless routes but I also upgraded the entire fleet to NGs which has noticeably improved profitability. Considering that they don't all have ideal schedules thanks to much of the country being closed from 18-6, they range from $6k to $43k/week. I also have a single aircraft that flies a full schedule of hopelessly unprofitable flights for $-12k/wk.

All the usual expenses are significantly lower in Tahiti as I assume they are in neighboring Vanuatu which makes these planes probably somewhat viable. I had planes with 100%LFs making much better profits at one point but I found that they were hurting the LFs on my E120s so I stopped mixing them together. It's likely that without the 120s, my PCs would be more numerous and probably profitable. As it stands, all I really know is that they haven't caused me to BK and at this point I know they aren't even putting me at risk. However, I was in the top 3 for profit margin before I started flying PCs and now that's a distant memory.

Comparing profit reports to commonality, I'd say this operation probably loses six figures every month but I've had it going for years. Realistically, it would probably be kept up with subsidies to ensure I kept serving every airport like I used to. I'm sure staffing isn't doing me any favors either.

Talentz

  • Former member
Re: Time for another experiment
« Reply #39 on: December 25, 2012, 07:17:22 AM »
Quote
All the usual expenses are significantly lower in Tahiti as I assume they are in neighboring Vanuatu which makes these planes probably somewhat viable.

This is true outside of most major aviation countries.

Would you believe that in the DR (Dominican Republic), I pay more per week for Navigation fees then I do for Staff salaries. I only spend 1.5m more to maintain CI100 vs weekly navigation fees.

Smaller countries or Island hoping is more viable then doing the same in the US or Euroland.


Talentz

 

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

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