We should probably start a new thread for this, but I'll bite.

I have quite a bit of experience with the pairing construction and PBS process at a large regional airline which I assume is close to what VX has.

First and foremost... Garbage in Garbage out! Alaska suffers from having a larger percentage of their block being shorter flights (like a regional). Virgin on the other hand has a higher percentage of the total block they fly as longer flights(Transcons, etc.). This will inherently make the pairings (and credit per duty/day) better for Virgin, and also allow pilots more days off. Listening to a VX crew member say (while true) the most junior line holder has 16 days off, isn't indicative of much. So It's a little hard to compare PBS vs Line bidding in their current forms.

That being said, there is complex software made by different companies that will build the pairings. The system works by creating a cost solution for each run you do. I am familiar with the Jeppesen System and It works by loading all flights into the system and then assigning penalties to whatever you want to avoid. You can assign a penalty for almost anything that has a cost to it. So a company will assign varying penalties to the system for obvious things like deadheads, hotel nights, soft time, or anything that costs the company they want to avoid. This gives the user the ability to assign penalties for thing they want to avoid. For example, we would assign penalties for four day pairings for some domiciles that didn't want them, or penalties for long sits to try to massage the system to not have 4 hours sits. From what I can tell looking at the Alaska pairings, it looks like they put little emphasis on anything and just let the system build the most efficient solution, possibly small penalties on pairings of certain lengths to get a good (subjective) distribution of pairing lengths. This is great for the company but not for a commuter. From my experience the most efficient solution if you assigned almost no penalties for the whole system would be five and six day parings that show early and end late. This, as one would know, would allow the system to work the pilot to absolute max and get the most work out of them. We, of course, limited the system to four days.

Looking at the Alaska lines it looks to me that they aren't built by hand. This would be an absolute waste of time when a program could build the lines. I'm also skeptical of your process. There are a few things that are quite suspect. The first thing is that theres a magic seniority in each domicile where you go from 14 to 15 days off a month which seems to be 50%. Also there seems to be the top few lines in every domicile where the best two days or most efficient pairings are hand picked to get 20-22 days off a month. The bottom end is absolutely dismal at 12 days off a month and the most inefficient trips all strung together.

There are other completely asinine rules in the contract that handicap ones productivity like having to have four 48 hours blocks off in a month. Which I understand, is protect the poor guy with only 12 days off a month, but wouldn't be needed if the lines were built by PBS.

Now on to PBS. The pairings are already built by the pairing optimizer and then put into the system for crew members to bid on. What PBS does is essentially allow you to build a line with preferences that you assign and not what someone who is cherry picking prime trips for the top 10 lines while leaving the crap for everyone else wants. Alaska system is pretty much PBS anyway except you have to spend have your month trolling the trade board looking for what you would have just bid for in the first place. It allows the person who can hold commutable trip have them, someone who wants two days have them, someone can bid two day back to back trips instead of 4 days, etc. From what I could gather the a line at Alaska is a hodgepodge of everything 4 days, 3 days, 2 days, three day back to back, there was no logic in it other than the program was screwed because it was allowing for the four 48 hour periods.

I am an advocate of PBS, as you could probably tell, but I must stress that there needs to be bullet proof language behind it. It can very easily be manipulated to the companies advantage and it is inherently better for the company, we should get something for it! I'm sure PBS is coming but there needs to be a group of people highly familiar with PBS, pairing construction, etc. that are making the rules behind it. The current scheduling group will be hard pressed to understand the nuances of PBS if their experience is with line bidding. Things like globalization, ability to manipulate the required average line credit monthly, days off between work blocks, etc. are very important.

The people who are calling for pilots to build the pairings are delusional, suspecting this would be better. What's good for you might not be good for the next guy. So careful what you wish for. Oversight yes, but a few senior people building pairings will be good for them, not you.