Scheduling Theory
Regarding scheduling, there, too, are two extremes with their own pros and cons, the trade-off being connectivity vs. a/c utilization.
- departure waves
- randomly distributing flights throughout the day
transfer time
every airport has a minimum transfer time. Pax can only transfer to flights that leave no earlier than after the minimum transfer time (plus one minute for rounding, AS works with seconds, but only displays minutes) and no later than eight hours (16 hours on Devau (I think it was Devau) after arrival (arrival, not when the a/c is ready for departure again). Pax that will continue on the same a/c have no minimum transfer time as they just remain seated. Those pax can also remain insight the aircraft and continue travel at non-transfer airports.
departure wave
for an airport with a minimum transfer time of 1 hour, the departure wave structure might look like that:
all a/c arrive at your hub as close to but no later than 10:59am and depart again as close to but no earlier than 12am. This way, all arriving flights offer connections to all departing flights allowing for maximum connectivity. However, you will have to keep a/cs on the ground longer than necessary to actually meet the "hub times" (the arrival and departure times for that wave), leading to a decresed utilization. A/cs only make money when in the air, thou. You will obviously use more than one departure wave, depending on the distances of your destinations. You usually end up with three of four daily primary waves.
Since slots for landing and take-off are a major limitation, you will soon see, that slots best meeting your waves will become sparse. Also, you need available slots at your destinations, as well. This isn't a problem for medium and small destinations, but you can't assume to freely plan your schedule when flying to Atlanta, JFK, Heathrow, etc, if you can fly there at all.
random scheduling
alternatively, you can schedule routes randomly, meaning once an a/c is ready for departure again, you have it depart, maximizing utlization. This way, connectivity suffers as you might just miss a transfer opportunity or only have the next one available 5 hour after arrival at your hub. Since you are not bound to time blocks for your waves, you will not encounter slot use congestions. (you might, if another airline is using departure waves at your airport).
things to keep in mind
- when using departure waves, you can introduce secondary waves, so when there are no slots left for your primary waves, just introduce waves in-between. Since they will be within the eight our max transfer time limit, you will still get some transfer pax between the different waves
- when using random scheduling, don't make it completely random. If you offer at least four daily flights to a destination, space them out evenly over the day, if you have one flight roughly every 6 hours, there will always be a connection to every destination within the eight hour max transfer time, although it might be a longer wait.
- make yourself aware of the geographic situation. Who would even consider transfering at your airport? AS pax only consider connections that are no longer than twice the direct distance between origin and destination. so you might want to consider sorting flights geographically as to offer a better connectivity for traffic from east to west, north to south, etc.
- while transfer times should be as low as possible, you have to consider the over-all travel time. If your direct travel time is 6 hours and three of them are for transfer, that is a very different story than a pax arriving from a 10 hours longhaul flight waiting for a regional connection.
- keep the market situation in mind. if there is no competition offering faster connections, pax will have to use your flights (unless your ORS rating is negative). if there is strong competition, you need every single transfer pax you can get
there are many more ways to optimize scheduling.