I stayed in a hotel recently. This isn’t news, as I travel a lot. I’m usually pretty chill about most things that happen when living an airport-plane-uber-hotel-uber-office-airport-plane lifestyle, but something happened on this trip that really annoyed me.
Picture this: A hotel of 23 floors with five elevators with only two of them only go to Floor 18 or above. First, I’m baffled by why that happened in the first place, but I don’t have time to think about that. I was staying on Floor 19, but there’s only one Call button, meaning that if the 1-18 elevator arrives first I’m not being served but I can’t press the Up button again, because that will stop the 1-18 car from closing its door and ascending. By the way, I’m sure 50% of guests on floors 19-23 found out the hard way about the limitation of those two cars, and that’s a customer experience problem for the hotel.
Here’s the trouble. The peak hours for downward elevator travel are 7am to 10am but the elevator bank runs on inefficient software. The way this manifests itself is in delays and in non-value-added activity. The car loads quickly from the top floor, and by the time it gets to 15, it’s full. Since people on every floor have pressed the down button, it’s obliged to stop. At each floor there’s that charmingly human activity of looking apologetically at the folks who can’t board, and then sometimes a repeat of pressing the Door Open button, preventing the doors from closing. It’s annoying for everyone. Given that these people are checking out, they mostly have suitcases and cannot take the stairs.
What’s happening here is a case of old-fashioned rules-based software. The software essentially tells the cars to start at the bottom and go all the way to highest requested floor, stopping wherever someone has pressed up. And then do the opposite — start at the highest floor with a down request and stop at every down request on the way to the lowest requested floor.
Imagine a simple scenario: two cars, three floors, and five passengers. Four passengers are at the bottom: two wish to go to Floor 2, and two wish to go to Floor 3. The remaining passenger on Floor 3 wants to come down to Floor 1. Both cars are at the bottom because they’re programmed to do that at this time of day.
The call button is pushed simultaneously. Two people enter each car, each pressing for Floors 2 and 3. This means that both cars are in use, and both will call at every floor. The person on 3 is going to have a wait. It won’t be long in this simple example, but imagine being in this situation on Floor 23, or 50, or — at the Burj Khalifa — 160.
There’s a better system available, but it’s not widespread yet. The fundamental shift is from elevators with the floor buttons on the inside, to placing the floor request buttons on the outside. That way, before anyone boards a car, the system already knows all of the requests for all of the cars. Using my example above, now one car would zoom to the top, to collect that descending passenger, while the other car would fulfil the request for 2 and 3 ascending.
It’s the same in pricing software. Rules-based pricing engines aren’t good enough anymore. Expressing pricing in “IF” statements is too simplistic. True optimization is something else entirely, and it’s game changer — especially if you’re on the 23rd Floor!
Get in touch (firstname.lastname@example.org) and I’ll tell you more about it. I’ll have plenty of time while I wait for an elevator.
About the AuthorMore Content by Ben Blaney