Community
Wiki Posts
Search

is CO yield management individualized?

 
Thread Tools
 
Search this Thread
 
Old Dec 13, 2007 | 11:58 pm
  #1  
Original Poster
All eyes on you!
15 Years on Site
 
Join Date: Jun 2007
Programs: CO-plat, SPG-plat
Posts: 1,658
is CO yield management individualized?

In the late hours of the evening, I obviously have too much spare time if I'm musing about CO's yield management software. xyzzy's comment in another thread touched on CO's excellent yield management, and I've long wondered whether the software is individualized.

For example, if 5 passenger on a given EWR-SEA flight are connecting from BWI, does CO look only at historical data and conclude that 80% of BWI-EWR-SEA passengers show up? Alternatively, does CO look at profiles of passengers A - E and conclude that passenger A shows up 20% of the time whereas passengers B-E show up 100% of the time? Depending on the numbers, the results could be rather different.
Totoro is offline  
Old Dec 14, 2007 | 12:16 am
  #2  
 
Join Date: Feb 2007
Location: SFO/SJC/OAK
Programs: CO Plat (thanks to FT), UA 1P, AA
Posts: 338
Originally Posted by Totoro
In the late hours of the evening, I obviously have too much spare time if I'm musing about CO's yield management software. xyzzy's comment in another thread touched on CO's excellent yield management, and I've long wondered whether the software is individualized.

For example, if 5 passenger on a given EWR-SEA flight are connecting from BWI, does CO look only at historical data and conclude that 80% of BWI-EWR-SEA passengers show up? Alternatively, does CO look at profiles of passengers A - E and conclude that passenger A shows up 20% of the time whereas passengers B-E show up 100% of the time? Depending on the numbers, the results could be rather different.
As interesting as the idea is, I do not think that this (individualized yield management) is the case. And there are several reasons for this:

1. The confidence levels would be too low. Most individual do not fly enough on a single airline to have significant statistics. And individual flying patterns can change very suddenly: change of jobs, locations, life circumstances...
2. It would be extremely expensive, and maybe provide only a marginal improvement over global statistics.
3. I just cannot imagine CO's IT being able to handle this!

Just my $0.02
enmascarado is offline  
Old Dec 14, 2007 | 12:27 am
  #3  
20 Countries Visited500k30 Nights15 Years on Site
 
Join Date: Dec 2007
Location: Austin, TX
Programs: UA S; Marriott LG; IHG P; Hertz PC; AA, WN, Pan Am!
Posts: 820
Originally Posted by enmascarado
As interesting as the idea is, I do not think that this (individualized yield management) is the case. And there are several reasons for this:

1. The confidence levels would be too low. Most individual do not fly enough on a single airline to have significant statistics. And individual flying patterns can change very suddenly: change of jobs, locations, life circumstances...
2. It would be extremely expensive, and maybe provide only a marginal improvement over global statistics.
3. I just cannot imagine CO's IT being able to handle this!

Just my $0.02
You could certainly weight individual experience to mix it with global experience. The more an individual flies with CO, the more weight shifts towards the individual. General change of circumstances could be accounted for via changes in address by more than 50 miles or simply some change in flyership algorithm.

But as an improvement over global stats, you're probably right (and definitely right RE cost, unless they could convince the other carriers to license it and try to recover some of the costs).
texd is offline  
Old Dec 14, 2007 | 1:25 am
  #4  
 
Join Date: Jan 2007
Location: The high ground in Swamp 205 miles east of the Dungeon (IAH B-84)
Programs: CO Plat / Hilton Gold / RCL Diamond Plus
Posts: 144
Originally Posted by enmascarado
As interesting as the idea is, I do not think that this (individualized yield management) is the case. And there are several reasons for this:

1. The confidence levels would be too low. Most individual do not fly enough on a single airline to have significant statistics. And individual flying patterns can change very suddenly: change of jobs, locations, life circumstances...
2. It would be extremely expensive, and maybe provide only a marginal improvement over global statistics.
3. I just cannot imagine CO's IT being able to handle this!

Just my $0.02
4. I just cannot imagine CO's main frame IBM 360 being able to handle it.
Lafayetteflyer is offline  
Old Dec 14, 2007 | 6:10 am
  #5  
All eyes on you!
20 Years on Site
 
Join Date: May 2002
Location: Moreland Hills (CLE)
Programs: Over-entitled UA 1.3MM Gold, AA Gold, Hilton Diamond, Marriott L-T Plat, Hertz PC
Posts: 5,526
Talking

Originally Posted by Lafayetteflyer
4. I just cannot imagine CO's main frame IBM 360 being able to handle it.

Yes... the one they bought from Ross Perot when he worked for IBM in the 1960s.

LOL
Billiken is offline  
Old Dec 14, 2007 | 7:13 am
  #6  
FlyerTalk Evangelist
10 Countries Visited
20 Countries Visited
30 Countries Visited
20 Years on Site
 
Join Date: Aug 2002
Location: Bay Area, CA
Programs: UA Plat 2MM; AS MVP Gold 75K
Posts: 35,092
I don't know about this, but they most certainly take connections (and where they're coming from ) into account.

For example, the late afternoon/evening flights out of IAH have a lot of Caribbean or Latin American connections; and the evening flights out of EWR have a lot of Europe connections.

They certainly take the int'l segment no-show factor into account for the domestic flights when overbooking.
channa is offline  


Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service -

This site is owned, operated, and maintained by MH Sub I, LLC dba Internet Brands. Copyright © 2026 MH Sub I, LLC dba Internet Brands. All rights reserved. Designated trademarks are the property of their respective owners.