This idea just came to mind while I was reponding to that "roll call" thread.

Starting today, I'm going to start tracking C class on flights that I usually take i) 72 hours prior ii) day of departure. That will give me a pretty good indication on which flight will give me no problems when the 72 hours window comes around.

I used to do this when I was a Gold/ExP with CP and acheived 100% on upgrades but stopped last year when we were able to upgrade at the time of booking.

Do you guys think this is an effective way of "predicting" C class or just a waste of time? I think it's worth the effort especially if we have to pay that much more for a shot at an upgrade. As we all know, airlines use historical data as part of their yield management system and I was thinking along the same line when I came up with this idea.

Yup that will give you a very good idea about how it should look come the 72hr window.

Trouble is you need to watch constantly as the inventory levels adjust with each booking made and again at each judgment point made by the computor program [30 days, 14 days, 7 days, 72 hours, 24 hours etc], and again when the controllers look at their reports and make random manual adjustments.

Take into account; Local holidays at origin/ destination, time off year, day of week, block bookings for conventions and tour groups, and any other variance that may affect loads [full moon?].

When you have your system perfected then write a computor program that relects your hypothesis and sell it back to the airline for a few millions.

Seriously though, you are on the right track and barring a few unknowns you can get pretty good at predicting loads if you are only tracking a few flights or routes.

We would do this periodically to check how the PROS system would forecast the loads.

The only problem is that I don't have access to what the true loads are. [Well... unless I pull up seat maps on PTO and start counting seats everyday then i might get close to what the true loads are]

The only data that I can use are the ones from ITN and those info already factor in what AC yield management thinks. So if I am developing my own hypothesis based on the ITN data, it's nothing more than what AC already thinks. Those data are effective for helping me look at which flight I should pick for upgrade purposes but I am not sure whether it'll be useful beyond that.

Do you think there is a better way of doing it or acheive what you are suggesting based on those ITN/PTO data?

It occurs to me that if AC was really clever, they'd start publishing lists of flights on which you are most likely to get upgraded, or get a premium Y seat. This would help load balancing, result in fewer bumps and their attendant costs, and be a big customer service benefit. All at essentially no cost.

This could direct away from heavily or overbooked flights to lighter loaded flights, such as planes that have to positioned overnight regardless of loads.

I just created a database for YVR-HKG for the next 30 days using the "judgement point" that exAC stated. [30 days, 14 days, 7 days, 72 hours, 24 hours] Anything outside of 30 days won't do me any good as i don't ever book that far out.

I can already see a trend in C class. Almost nothing 30 & 14 days out, always available on on flights within the 7 days window, zeros again once the 72 hours comes around. I'll need more data to comfirm this simple hypothesis but give me about a month and I think I'll get real good on this.

Moving on to other routes now... [I am doing YVR/HKG, YYZ/YVR as starters, any other routes you guys are interested? I won't bother with commuter routes for now as there are hourly flights and not worth wasting time on it since most of us won't get upgradeable fare on those routes come March.]

This looks like it might provide some useful data, YVR-NRT would be of interest to me.

Notice how this board is already shifting from "How could they do this to us?" mode to "OK. How do we work the system now?" mode.

Empress.....

How about YYZ-GRU ?

AC 48 leaves once a day.

Empress, instead of you managing all these yourself, I wonder if it would be possible to set up a database where each of us picks a route, does a quick daily check, and posts the info?

andrew

Agree with Andrew about helping out Empress. Noticed the YVR-HKG trend to Zero out C up to the 7 day window which makes it a bit of purchase lottery on the H fares ie. waiting to purchase at the 7 day minimum window and hoping that C opens up. I guess this is the boat we find ourselves in so let's start paddling. YYZ-YVR seems to be more open with C9s across the board well in advance but diminishing on the early morning late afternoon "rush hour" flights as flight day approaches. Again another lottery which suggests we should aim for shoulder and off-peak times -- although usually the dreaded domestic-J config 67s are on those flights.

Have noted that the V to H shift for SE international upgrades really "hurts" $ wise on flights to mainland China, Taiwan and Korea where the price differential is pushing $1,000 or more. On flights to HKG and Japan the lowest fares (save for seat sales) was usually H anyways. Have booked two separate long weekends in Shanghai and Beijing out of SEA ($1,028 and $1,139 CDN in refundable(!) V) for February as a last "hoorah".

Hmmm... I thought we'd shifted from "How could they do this to us?" mode to "OK. How do we do this to them?"

This is a good idea... I've done YYZ-YVR last night for the next 30 days and started doing YVR-YYZ. [Early trends shows that don't pick the 8:45a flight from YYZ and don't pick the 12:30p flight from YVR.]

Should we set up a website for this or just set up a seperate free bulletin board where we can post all this info and then I could update the database and post them to a website on a daily basis?

They want to nickel and dime us, we'll work their system to the max!!

SO far we'll have YVR/HKG, YYZ/YVR, YVR/NRT, YYZ/GRU... I'll continue to work on that when I have time later on tonight. Once the database is set up [I'm just using Microsoft Access], it's a matter of checking once a day and inputting the data.

If you guys have any better suggestions on doing this, let me know.

I don't have any specific suggestions, but perhaps a website where we each enter our own info? (Like the Lounge Project)

andrew

I'm happy to contribute data for the YVR/NRT route -- the Lounge Project approach occured to me too.

This could be a fun project

Good... I'll work on that later. Let's see what AC will do this time.

hey Empress, thanks for implementing this resource concept! Hopefully someone out there will be able to assist with programming an online version...(?)

I'd offer, but my hacking is limited to my ability to type while taking directions!

I'd like to submit some of my main routes:

YVR-SYD (via HNL)
YVR-YYC-FRA (AC metal to Europe...)
YVR-LHR