This is my current flight schedule from DXB to DOH. There are about 15+ flights a day, all going 100% booked, being served mostly with 737s or 321s (and some with Q400s). Prices are fine as well, not too high.
Recently, I added A380s onto the route, but the plane is barely 50% booked. This is more strange because all the flights around that flight at fully booked!!
There is no issue with timing for connections either, seating and service is also the same as the 739s and 321s.
This leads me to believe that is better to have the airport served with smaller aircraft and more frequencies, than with less frequencies with larger aircraft. In this case, this defies logic as pax loads are 100% on all other flights!!
Exactly my question!! How is the A380 oversized for this route if I'm filling up 15 daily flights at 100% usually in the first, max second demand cycles?
I think the thought process the OP was going with is that since the rest of the flights fill up so quickly, that it would make sense the A380 should be filled as well since the rest of the flights are filling up so quickly. Now granted, if the A380 is filled purely with connections and no O & D, that would explain it.
I understand that the current booking system prefers smaller planes over bigger ones because they are being filled evenly if all other product factors (seating, service etc.) are equal. For example, if you use a 100-seater and a 150-seater of equal quality and without competition on a route where the demand is 200 Pax, the 100-seater will be full, the 150-seater only 2/3.
Exactly as Wrigt says. At each ORS elvel, there is a certain number of pax that is divided evenly among all the flights. So you get 1,000 pax at ORS 99, 800 pax at ORS 98, 700 pax at ORS 97, etc. If there are 10 flights @ ORS 99, all of the 10 would get the 1,000 divided equally among them. An A380 would be the 11th flight, so now the 1,000 would be divided equally among 11 flights.
This is for O/D demand, of course there are some strategies to affect the demand availability for interlining pax. For example, on heavy O/D routes, you may want to schedule a bigger plane around the hub wave times to accommodate connecting pax. Also, you might want to increase the price on those flight by couple of AS$... this would increase O/D ORS (so O/D pax would book other flights - maybe off peak times, first), for connecting pax a couple of AS$ increase in price would not affect much the ORS, as the total connection time and quality of product on individual legs is more important than the overall (and individual leg) price.
So in order to get the most loads, I should operate from a major airport with small aircraft?
Please allow me to be honest. We hear enough about airlines using small aircraft on big routes, therefore blocking slots.
Thank you for the revelation - in my 4 years on AS, I did not know this (though I did suspect it).
The wrong (if I may call it so) allocation of bookings is a known issue in AS.
With all the threads about slots and all the critics (me included) I wonder why this has never been approached. It’s the essential key in finally making widebodies more competitive.
It is the way as described above.
Just one example from my airline:
I operate 73J and 789 parallel on one route and within the same hub wave (timeframe).
Both 73J and 789 catch the same connecting flights with the 789 having the better rating (faster)
The result for cargo however is:
73J fully booked, 789 with 2/3 unsold capacity.
Normally you’d expect the 789 getting all the bookings and the 73J the surplus (zero in this case)
The wrong (if I may call it so) allocation of bookings is a known issue in AS.
With all the threads about slots and all the critics (me included) I wonder why this has never been approached. It’s the essential key in finally making widebodies more competitive.
It is the way as described above.
Just one example from my airline:
I operate 73J and 789 parallel on one route and within the same hub wave (timeframe).
Both 73J and 789 catch the same connecting flights with the 789 having the better rating (faster)
The result for cargo however is:
73J fully booked, 789 with 2/3 unsold capacity.
Normally you’d expect the 789 getting all the bookings and the 73J the surplus (zero in this case)
I'm not sure whether I got this right .. so basically the 789 ORS score in the given example is better than the 73J. Thus you would say pushing all cargo through the best ORS connection neglecting all other connections should be the right way to distribute demand. If that is your point, I would disagree, because not every customer would make use of the best scored connection because of its individual in-transparency and imperfectness. While not being satisfied with the current distribution, I think an only ORS based order would be better, i.e. closer to RL.
I'm not sure whether I got this right .. so basically the 789 ORS score in the given example is better than the 73J. Thus you would say pushing all cargo through the best ORS connection neglecting all other connections should be the right way to distribute demand. If that is your point, I would disagree, because not every customer would make use of the best scored connection because of its individual in-transparency and imperfectness. While not being satisfied with the current distribution, I think an only ORS based order would be better, i.e. closer to RL.
This essentially describes the problem: We can either "fill up" in descending order by rating or we can "distribute" by relative rating. Both have their pros and cons, but the former approach can cause extreme effects with minimal changes. Imagine a thin route with two providers, each offering more capacity than there is demand and both offering a very similar product (let's pick ratings of 90 for A and 89 for B). With a "fill up" approach, carrier A would get all the pax, carrier B would get none. Now imagine B bumps its rating from 89 to 91. From one ORS run to the other, A would go from 100% to 0% load and B from 0% to 100%. Doesn't sound quite right :)
Now, part of the problem is that we only have 3 passenger classes. In reality, people would have certain personal preferences that make them decide for one carrier/connection or another, even if the overall product offered is more or less the same. Since all our passengers have exactly the same preferences though, we see the effect described above. If we had - say - 10 or 20 more nuances types of passengers, the problem probably wouldn't be as bad unless one carrier is considerably better in all areas.
While not on our immediate list, we have something like this (demand diversification) on our roadmap.
I'm not sure whether I got this right .. so basically the 789 ORS score in the given example is better than the 73J. Thus you would say pushing all cargo through the best ORS connection neglecting all other connections should be the right way to distribute demand. If that is your point, I would disagree, because not every customer would make use of the best scored connection because of its individual in-transparency and imperfectness. While not being satisfied with the current distribution, I think an only ORS based order would be better, i.e. closer to RL.
Yes, you got this right. Though I was just making a fairly dramatic example.
I know that this way we'd have quite some troubles (as Martin explained) and what you say makes perfect sense.
I just wanted to express that the current distributon clearly favours smaller aircraft which is one major reason for the terribly congested airports in AS.
So I perfectly agree with both of you, but wish for a somehow improved distribution. Glad to see Martin has something on the to-do list already.
It's just a pain to see all the disadvantages I get when scheduling widebody aircraft.
I see. Referring to the basic problem with widebodies I totally agree. It's even worse when you are situated in China and have to deal with the low long-haul demand in parallel.
In terms of an only ORS-based distribution I want to add, that I dislike the current development, that more and more people developed a target pricing that would be impossible in RL. It's fair, clever, good play, whatever, but in the end, the target pricing is made by information that wouldn't be available in RL in that depth.
Probably it's worth to come back to another suggestion made earlier - I still think passengers (not cargo) prefer a widebody to a narrowbody, so that be one minor option to improve the situation of wide-bodies.
In terms of an only ORS-based distribution I want to add, that I dislike the current development, that more and more people developed a target pricing that would be impossible in RL. It's fair, clever, good play, whatever, but in the end, the target pricing is made by information that wouldn't be available in RL in that depth.
What do you mean by "target pricing"? Remembering the term from my business classes I "enjoyed" years ago, it means something like setting a sale price and then tuning production costs towards that price point. I assume you mean something else? If not, please elaborate.
What do you mean by "target pricing"? Remembering the term from my business classes I "enjoyed" years ago, it means something like setting a sale price and then tuning production costs towards that price point. I assume you mean something else? If not, please elaborate.
Ok, that was an improper use of that technical term and yes, I mean something different.
In general, I like the information given in the ORS, but from my experience, especially in competitive and saturated markets, people tend to adjust prices down to reach a 99 ORS score. Truly that's not always the best option, but it is a proper competitive action (that leads somewhere nowhere ...)
So the target in this pricing method is not costs but customer price/quality-perception. This method is from my point of view much to technical and doesn't really need pricing skills, but more a good portion of effort.
Let's just agree that the whole rating system needs an overhaul...in so many aspects ;)
The fact that ratings are capped at 100 points is somewhat nonsense. Of course the rating of a flight that stands at 100 points should get better when prices are lowered and/or service is improved. But then there are also all kinds of other problems beyond this one or the ones mentioned above...but replacing the current rating system will be massive undertaking and a painful thing to integrate into an existing game world.
Anyway, I don't want to give away/promise too much, but besides the "Demand Diversification" mentioned above, we also have other things within the same area on our roadmap, including an overhaul of the rating system and considerably extended functionality of the booking systems. That said, this stuff is on our "Roadmap", which means it's bullets on a long, kind-of ordered list of things planned for the future with no or incomplete concepts. If it weren't for the performance problems of the past months, we'd be considerably further down this list already. Nonetheless, we're trying to get through it one bullet after the other as fast as we can. I think we'll have some nice stuff to show late this or some time next year.
You won't be able to fill up your A380 more profitable than the 739.
If there is little competition you might choose to make your 739 less attractive (e.g. raise ticket price) and the A380 more attractive (lower ticket price), so that you finally get a lower ORS score for the 739. This would shuffle some PAX to the A380. But you see, that is only sensible if slots are rare and you are looking for capacity instead of margin. Besides, it's a lot of micro management, so I wouldn't recommend this option.