Interesting comparison perhaps to software projects with ever changing design requirements. At least in software we have the option to add on later, but it’s more difficult to expand an airport terminal once people start using it. I imagine that motivates some of the obsession for getting it “perfect” at launch.
… and the importance of feature freezing. One of the reasons that software projects often seem to go awry is the constant temptation to add one more feature during development, and I always tell clients that this is like changing the floorplan half way through building a house.
So this is nice as an example of what happens if you do try and change requirements part way through development … building projects turn out like death march software projects!
Aren’t most airports (in growth areas, at least) ever-expanding? I can think of countless airports that are regularly adding terminals, and if they have the land, runways.
Adding terminals, but probably seldom expanding them. Of course, a terminal can operate in an unfinished form (say, with half of the gates and without a mall upstairs), but that would require the “core” of the airport to be functional, and in case of BER they haven’t built a minimal viable system yet.
What the article doesn’t mention is that BER overlaps with the Schönefeld airport, and one of the two runways planned for BER is one of two SXF’s runways. That runway has been integrated into BER, or at least disconnected from SXF, some years ago, and apparently it’s not trivial to plug it back in. Apparently, the other SXF runway has been cynically closed.
So now SXF is operating at half capacity, and the tiny and efficient TXL with its two runways handles more flights than it can. Both are scheduled to close when BER opens, though now they say it won’t be able to handle the projected amount of flights. But at least there’ll be a mall.
Yes, but without some careful planning, you get crazy airports where it takes a minimum of 30 minutes to make a connection.
Personally, I don’t do much shopping at airports, nor consider them a culinary destination, but if I’m going to be stuck in an airport for two hours waiting for a delayed flight, I’d much rather have the “mall” there than nothing but a Dunkin Donuts and a Nathan’s. But that requires a certain upfront commitment to build a large hall.
The other Berlin airports managed to avoid this by building new terimnals adjacent to the main building. (Don’t know how much of it was pre-planned.) I’m particularly impressed with Tegel: it stayed small and efficient, but its main hexagonal building still contains all the necessities (coffee, food, pseudo-food, a bit of shopping) without a proper mall.
The last paragraph of the linked article is relevant. “Lovely and plush and modern as Brandenburg airport will be when it finally opens, allegedly in 2012 […]”
PHL is the counter example. Lots of terminals, but they’re just connected by hallways. There’s no “main building” to speak of. Instead of a hub and spoke, it’s more like a line of TTTTT. There’s a few food options more interesting than Starbucks, but they’re scattered about such that they are hard to find. If each individual T terminal were an independent small airport, it’d actually make a lot of sense. But they’re strung together quite poorly.
Of course, this also reminds me of the requirements gathering phase escalation. Three meetings later, your second paragraph has been translated:
This is the same airport where, approximately two years ago, no one could figure out how to turn any of the lights off.