Hi. ElevatorBank testing continues. In the attached model, there is one ElevatorBank with three cars in it. Travelers arrive at 0, 1, 2 and 3 sec. My question is, why the car #2 and car#3 are sent down (for Traveler#4) at time ~40 sec? Can I modify the logics somehow?
I guess that the reason is that car #2&3 can't be sure if the traveler will get into car#1. So it's better to send also car#2&3 as the request remains in the queue. This is a good default logic. But if I'd like to modify it so that when a car takes the request/one car is sent, the other car is not sent. But if the traveler can't get in, s/he orders a new car. I'd say that most of the elevators work that way: 1) press the button to order a car 2) one car is sent 3) if the traveler for example doesn't fit into the car, s/he orders second car by pressing the button again
The default logic works fine if the utilization rate is low (well, at least for the capacity calculation, not utilization rate calculation). But I think it will strengthten the high utilization periods: queue is built up-> all cars are busy -> all cars are sent toward one request (only one is actually needed) -> even more queue is built up-> elevators are even more busy
So is it possible to modify the logic so that only one car per request would be sent and new request is created if the request cannon be served (while the car arrives)?