Probably both. And the CNN report on this also mentioned reduced time between flights. I’ve got a SW flight on Friday from Indy through Las Vegas to Oakland. No status yet. Coming out we had an only 45 minute layover in Vegas, but our flight came in early and the departing gate was right across from our arrival gate.
We’ll see, but we’re at our daughter’s and since I don’t have to worry about work, no problem if we get delayed. But I’d rather not.
That’s exactly what my wife posited, that Southwest’s non-hub system, along with being the largest domestic carrier by domestic flights (so not counting United, Delta, American’s international flights), were among the major components that led to this boondoggle.
Latest CNN article about the delays on Southwest:
What’s weird is that from the ground here in Chicago, Friday’s storm did not seem bad at all. We got like 4 inches maybe in my part of the city, right by Midway Airport. It was cold as fuck, though. It was a disappointment as far as snow storms go – we were hoping for a lot more for the kids (and myself, I like a good foot plus of snow). Perhaps the wind and visibility were particularly bad. I just don’t remember it being particular hard to get around town or even shovel. It seemed like just another normal snowfall. But this is from the ground, of course.
My bad. Midway snow totals were all of 2".
Disclaimer: I haven’t talked to anyone at SWA. I don’t have any special insight besides general industry knowledge & what’s on the news.
Here’s some disjointed ideas in no particular order that don’t form a coherent essay.
-
As suggested upthread, SWA’s non-hub scheduling approach can do better when a localized storm screws up one city while some other airline using that same city as a hub suffers a (partial) meltdown from the same storm. But the SWA advantage turns into a disadvantage for more regional weather problems.
-
SWA schedules their aircraft and their people very hard, harder than the other majors do. They plan to get more work per day out of their stuff, whether that’s gates or jets or workers. Which is fine on a sunny day but fails and fails hard when there is so little slack that there is negligible chance to catch up for any glitches during the day.
If they also plan the overnights tight (which they do), then any delay problems today don’t self-recover overnight. The late arrival tonight forces a late departure tomorrow and the problem perpetuates or snowballs. Other operators don’t push quite as hard, so an e.g. 30 or 45 minute delay at the end of the workday is just absorbed by a little less sleeping time at the hotel instead of flowing minute for minute into a late start tomorrow. -
SWA is shorthanded (as are we all) and they have been running their people ragged. Meanwhile as the Kelleher era fades into ancient legend and the modern bean-counter era becomes the lived reality for years now, the employees are just about done going the extra mile every week while management continues for years to fiddle about maybe considering pay raises and such. All while crowing to Wall Street about high and growing profits. The front-line folks be burnt.
-
SWA in general is (or at least has been) way behind the other carriers in IT capability as their airline has grown and expanded and become a LOT more complicated. What works for 200 airplanes in the good weather low-traffic US southwest breaks down with 600 airplanes plying the high-density shit-weather US northeast. Corporate culture can get in the way when what used to be wonderful “secret sauce” has now become in changed circumstances an obsolete “sacred cow”. SWA is coming up hard against a lot of now harmful “secret sauce” verities.
-
Putting Humpty Dumpty back together is a very hard problem, requiring many millions of dollars of high-tech resources to manage. It is still not computationally tractable. It gets exponentially worse by the minute. The only way to win that game is to get ahead and stay ahead early or not to play at all. Once behind enough, you’re going to face-plant the whole show into grid-lock before you can regain control of the situation.
-
Each carrier gets burned in their turn. A couple of the other majors grossly overscheduled their June. And duly had meltdowns when they simply lacked the assets (mostly people) to fly their fantasy plan even on sunny days. Having been burned and burned hard, they played this winter in general, and the Christmas schedule in particular, fairly safe.
OTOH, SWA got through June & July pretty well. There was some crowing from their spokes-execs about their superior methods, superior people, etc. So not having been recently burned, they decided to go “full-bore or bust” over the Christmas holiday. Oops. Hubris meet nemesis. Again.
Thanks for the detailed and intelligent commentary. A lot of what you wrote I have heard in bits and pieces here and there, so I trust your analysis is pretty spot-on.
The discussion about IT reminds me of the several incidents that have been in the news over the last decade or so - Canadian airlines where the computers failed one day. It must be so frustrating - the planes are there, the people are there, the weather is clear, but they have no way to issue boarding passes or luggage tags, so everyone sits and waits or flights get cancelled. I presume the computers too take care of things like crew schedules or fueling?
I assume then the issue with SW is flexibility. “We planned plane A to go to X,Y,Z and plane B to go to Q, Y , R. So the plane B is at Y now and R is socked in but Z is open. Can we reschedule the plane B and crew to go to Z?” “If we can get Bob and Alice to Y they can pilot the next leg… are they trained for that aircraft type?” If the computers have a problem with that, the plane ain’t gonna go. And then fun times undoing the shuffles.
As to computers …
At a major airline everything is computerized. There are back-up plans for doing most of the essential tasks using alternate systems or semi-manual processes if necessary. But everything is databases and text messages and websites and tablets and modern 21st Century Techno-Magic.
All of which preparation is secondary to the basics of reliable modern IT like multiple geo-dispersed data centers, redundant everything everywhere from electricity to network connections to water supplies, continuous real-time backup & sync to the back-up systems etc.
But all the back-up stuff is much slower, more error prone, and far more people intensive. The FAA or whoever of course is quite interested in making sure no corners are cut along the way. And that all required records are meticulously kept with no lost chunks.
In general we can take battle damage to one or another system and keep on running mostly normally. If the regular fuel management system is out there’s a ready alternative. Ditto flight planning or baggage control or boarding pass issuance. What we can’t fight through (much) is all that crap going down at once. Or the major connectivity backbone dying and we’re back to voice phones and post-it notes while the backup computer systems sit quietly in their dark rooms unable to communicate to/from anyone or anything.
Meanwhile of course if/when somebody does suffer a multi-hour major IT outage, all the schedule disruption grows exponentially over time while the ability to manage it has fallen straight into the shitter.
Without our computers it’s about like a memorable quote from the original Star Trek. Paraphasing as best I can recall from 55 years later:
Shame I can’t remember the episode or find the clip on YouTube. Maybe some real Trekkie can. But “wallow like a garbage scow against a warp-driven starship” is a very apt description.
IIRC one of SWAs ‘special sauces’ was that they basically flew only one type of airplane - so they didn’t have to fight that particular issue. Don’t know if that is still true, though.
More or less still true. Vast majority are currently 737-700s and 800s with a few MAX 8’s as well.
Southwest has traditionally flown the 737-300, -500, -700, and -800. The (now-retired by Southwest) -300 and -500 both have the same pilot-in-charge (PIC) certificate, but the active -700 and -800 have different certificates, as doe the MAX 8. The real advantage of flying mostly one or two types of aircraft is in the maintenance, as the airline needs a smaller inventory of spares and can swap out whole planes with the same seating configuration so as to not screw up reserved seats.
I imagine pilots are encouraged to be cross certified in all types that the airline flies but bear in mind that all airlines have undergone a contraction of staff in the last three years, often retiring off senior and more experienced pilots and crew, and have been restaffing with less experienced pilots who may not have had the opportunity to be certified in all aircraft. The cabin crew also have to be certified in a type so they are familiar with all of the safety and egress systems.
It is easy to complain that the airline should be prepared for disruption and maintain sufficient excess capability to rectify problems but that comes at large cost against the overhead, which is something that Southwest pioneered minimizing in the airline industry by using a different model. They’ve also had issues with their reservation system which seems to be obsolescent and not robust against large scale rescheduling. This is what you get when you run a minimum cost airline on a continental scale. About a decade or so ago I used to catch $59 each way same day flights from Burbank to San Jose and back by making a reservation on the drive to the airport or walking up to the counter, which was super-reliable until SFO and OAK got fogged in and they redirected everyone to SJC and all flights would be blocked out for the next day and a half to catch up.
Stranger
The 737 NG = -700, -800, -900 versions and the corresponding MAX -7, -8, -9, & -10 are all common ratings. There is a 5-hour vidoe training program to add MAX capability to any existing 737-NG driver’s authorizations. AFAIK, all SWA pilots are qualified on both NGs & MAXes.
So SWA only has 2 kinds of pilots: Captains & FOs. Each pilot can fly their seat in any airplane in SWA’s entire fleet.
Which is an operational and overhead cost advantage versus other carriers with 4, 5, or even 7 different kinds of jets needing both Captains & Co-pilots specializing in each.
Over my 30+ year career I’ve been on 7 types of airplanes. That’s hardly unusual. But it does represent 6 schools and probably 10 unproductive months I’d have avoided had my employers only flown 1 type the whole time.
Interesting comment about software issues at Southwest:
My daughter used to be a project manager for an airline, and her job involved new software and upgrades to software to do things like pilot scheduling. (Not one of the majors.)
The CEO said he considered the company an IT company that happened to own airplanes.
I have been pondering the current Southwest meltdown and the acres of luggage separated from their owners that keeps appearing on the news. How, exactly, does luggage get separated from people like that, on such a large scale? I thought the luggage accompanies the passenger, no? It seems like a systemic problem across the country, and not limited to one airport. How does this occur?
Perhaps because the software that directs the luggage doesn’t know how to direct it from a plane which hasn’t taken off (or hasn’t even arrived) back to carousels?
Not just the US - my daughter was caught in the Heathrow meltdown a few years back and didn’t see her luggage for days. She bought stuff and got reimbursed eventually.
BTW, as for using the phone to rebook, BritishAir was impossible to contact from London or from the US. Her boyfriend finally had the idea of connecting to the number in South Africa which finally worked. Not too useful for SWA however.
Airlines don’t really control the luggage logistics once it goes through TSA inspection. That is generally provided by the airport itself, and each airport has different systems. When flights are cancelled or rescheduled with little notice, the luggage basically gets dumped in sideline that isn’t readily accessible, and the airline can’t pull those bags unless they cancel the flight, and really have little control over where it goes even if they redirect the passenger. I once got diverted last moment to John Wayne International as the last flight out and ended up having to drive to Ontario (my original destination) the next day to pick up my bags because the airline refused to deliver bags further than a 30 mile radius from the airport. Don’t even get me started on the time that Delta lost a $15k theodolite because I got bumped from one flight to another 40 minutes later and I had to expense it as “lost luggage”, only to have it show up two weeks later in Chattanooga of all places. I’ve since learned to ship all tools via FedEx and make the loss their problem even if they do bill an enormous surcharge for anything marked “Delicate Instrumentation—Handle with Care”.
Stranger
Yeah apparently from most reports there is an issue with that SW’s systems for such a scenario were lagging behind where they should be to support an outfit this big at this point in time. Like @LSLGuy said, the watchword is “impress with the quarterly report” and that dissuades from spending big on upgrades as long as under normal circumstances it works.
But who can blame them, as mentioned, the market is relentless and price is king – otherwise how do you explain how someone keeps buying tickets on (break your)Spirit Airlines.
(And BTW, come to think of it… an airline that routes a flight San Jose-Phoenix-Dallas-Chicago-Birmingham-Baltimore and makes it to Dallas, will it appear as having had three cancellations, while one that nonstops San Francisco-Dulles and scratches will appear as having had only one?)
Sounds likely, I mean, some of the luggage may be already in airplanes that hoped to leave but did not, some may be already in the carts to be taken to airplanes not yet there but most would be somewhere in the bowels of the handling line waiting for the jam to clear out… and I suspect the system is not designed to simply reverse 'cause you still have more stuff being checked in. ISTM you’d have to get all the luggage out the “far end” of the outgoing handling line and then lateral it to the receiving line… where now all the barcodes are all wrong for this place. And it’s a different thing when one flight gets cancelled, where it is just a matter of “ok, find one idle belt in baggage claim and tell everyone to go get it there”, vs. when you get dozens of cancellations.
From a co-worker -
“Be glad if you are not trying to fly. Just heard from a friend who was suppose to Fly back from NY today. Apparently 75% of southwest flights were canceled. they are now driving cross country to avoid the $3000 plane tickets or having to wait until next week”.
That would be NY → Denver.
Seems kinda strange though. That’s a long haul in a car. At least 2 days, probably 3. So they are not saving much money or time if they where to just wait it out instead. But then they would have to rent hotel space in NY for a week. 