I am new to AWS, and still getting the hang of the intricacies of the game, but I do have extensive RL experience with airline scheduling and slots which I think may be of relevance to this discussion and Sami’s initial suggestion.
My background is in Europe and whilst there may be some differences in North America, I don’t think there is much variation given that the rules and regs for allocating slots are managed by IATA. These rules are set out in the Worldwide Slot guidelines (WSG) http://www.iata.org/policy/slots/Documents/wsg-5.pdf
although airports and the locally based airlines can agree certain minor variations (Local rules) which apply to that airport only, these do not contradict the fundamental principles of slot allocation.
Ignoring uncongested airports, where airlines can come and go as they please, for major airports which suffer congestion at peak times (classed as Level 3 or SCR airports in IATA terminology), slots are allocated according to WSG rules. Airlines can apply to change their schedules or apply for new slots at any time during the year, however, they have more success in achieving their aims if they organise their major schedule changes around the biannual slot conferences. In effect all the airlines operating to an airport revise their planned schedules for the coming season at the same time and the airport coordinator matches demand to capacity as best they can. The order of priority that the Coordinators apply in determining which airline gets what is as follows:
1. Historic slots – that is if you have operated a flight at a given time last year, you are entitled to the same timed slots again this year.
2. Historic Retimes – if an airline with historic slots wishes to reschedule a flight, the coordinator attempts to move the old slot onto the newly requested time (or close to it) before addressing other demands
3. The slots that are left (any unclaimed Historic slots and any new capacity that the airport or ATC have agreed can be released) are allocated out, with 50% being reserved for “New Entrant” airlines. The definition of “New Entrant” has varied over the years, but is currently defined as an airline which holds less than 5 slots on the day for which they are applying for the new slot – note that an arrival counts as a slot as does a departure so this means a carrier with less than two daily rotations from an airport can request New Entrant status. In reality this means that New Entrant status only applies to non-based airlines. The remaining 50% of slots are available to all other carrier requests.
Over and above the formal rules for allocating capacity, airlines can and do trade slots between themselves. Note that IRL, the value of slots defers to the airline, not the airport, much to the airports’ displeasure. I can apply and obtain a slot at an unconstrained airport and I pay nothing. Or I can sell my slots at a constrained airport to another airline and pocket the cash. If I want to fly out of JFK at peak time, it is not the airport I need to pay, but an existing operator with a slot/gate and stand at the times I want. Or I can apply to the JFK coordinator year after year for a slot and get told that nothing is available in the hours that I have requested.
IRL there are airline business models which involve operating cheap turboprops into congested airports in order to build up a slot portfolio which has an inherent value and can then be sold on in part or in pieces. Note that the more historic slots an airline has, the better its ability to access peak time slots as over time, through repeated (bi-annual) historic retime requests, the airline will eventually get its 04:00 departure moved to 06:00 or even 07:00 or 08:00 as other carriers go bust, lose their rights to historic slots or request moves out of the congested hours.
In terms of adding a “market based” approach to slots, slot trading between airlines is it. The price a carrier wishing to operate a long haul wide-body is prepared to pay for a slot sets the opportunity cost for an existing carrier operating small aircraft on regional services. This is why you rarely see an ATR 42 at LHR…..
Another point which may be too complex for AWS now, but for which the basic infrastructure already exists is the performance and use it or lose it rules – airlines which routinely operate late can lose their rights to historic slots, a major blow at a congested airport. Likewise if an airline operates less than 80% of the slots it holds, it will also lose its historic rights to the slots. This is a basic method of avoiding slot hoarding. A simple rule on repeated poor punctuality leading to loss of slots would add realism to the game and force players to give due weight to their turn-round times and scheduling as well as occasionally freeing up slots at congested airports in peak hours.
Based on the real life rules, I would suggest that a system which reserves 50% of any new capacity being released for the non- based carriers at a given airport would go some way to solving the problem of early players being able to dominate a given airport, as long as capacity is slowly and steadily increased over time. IRL capacity is slowly increased, even at airports which aren’t having new terminals or runways built. Often capacity improvements come in the form or one or two movements per hour as ATC review the delay statistics, the airport adds some rapid exit taxiways or reorganise the layout of the terminal to allow more international (or domestic) passenger throughput. Note that this new capacity is first available to retime an existing carriers slots. So if an airport adds one movement in every hour, the request of an existing airline to move a slot from 09:00 to 10:00 would be met first and then the new entrants could pick up a slot in 09:00 or 11:00 or hour other than 10:00.
I’m not sure how you would devise a system which allows an airline to move its slots between capacity constrained hours over time without requiring immense processing behind the scenes in order to avoid busting the set hourly capacity limits. In essence, what you need is a wait list, with carriers bidding to move their existing slots into a new time as and when capacity becomes available, either through expansion or other carriers going bust or moving their slots out of the hours in question. The system Sami proposed of allowing airlines to bust the limits on certain hours whilst maintaining the overall cap on movements seems likely lead to certain hours being heavily (and unrealistically) oversubscribed over time.