Originally Posted by
chrisl137
If you do the simplest possible estimate and assume that the need for rides (13,500/day) is distributed uniformly over 16 hours of the day and that the rate limiting step is the ability to load the passengers into an Uber, get that Uber out of the space, and get a new Uber into the space, then there's about 5.5 minutes for those activities per ride. At PDX it takes much less than that to move a vehicle in, load it, and get it out, though it's just a straight line queue rather than slots. There will obviously be spikes and lulls, but the rough estimate says it shouldn't be disastrous. It would be interesting to see a discrete model using real data all the way back to aircraft arrivals to estimate the likelihood of choke periods when the system gets completely overloaded due to multiple banks of flights arriving at once and things like that.
Continuing the fun with numbers exercise, let's assume at a peak hour there is a bus arriving every 5 minutes with 40 people aboard. Add another 120 people an hour walking to the lot (from T-1 and elsewhere) and we're at a 600 person queue in an hour. With strong dispatch oversight by LAWA, I guess it could work. A couple of necessities would be near instant availability of vehicles and an equal number of riders splitting the business 50/50 between the two providers.