Market Mechanisms for
Near-Real-Time Landing and Takeoff Slot Allocation
John Musacchio (PI) , Pat Mantey (Co-PI)
Supported by UC Santa Cruz / NASA University Affiliated Research Center
Abstract/Summary: We propose to develop market based mechanisms for allocating landing and takeoff slots
in near real time. Weather can greatly affect the landing capacity of a number
of major airports, which can lead to severe delays. By using a market mechanism
to reallocate slots, the flights that have the greatest need for arriving on or
near their scheduled time can be prioritized, as operators of such flights
would be willing to bid more. (We use the word "slot" in the sense of a
landing/takeoff opportunity given current conditions, rather than in the
traditional sense of a long-term right to have a scheduled arrival/departure at
a particular time.) For instance
flights with larger aircraft or flights with a large number of connecting
passengers suffer more economic losses per unit of time delay. This concept
contributes to the vision of Green
Aviation for a number of reasons. Since larger aircraft ought to see their
delays reduced significantly in days with adverse weather, the proposed scheme
has the potential to reduce fuel consumption. Furthermore, by ensuring that
runways are used much more efficiently, fewer new runways will need to be
built. For instance adding a new runway to SFO would require filling a large
area of San Francisco bay, which would likely be environmentally damaging.
Finally, airlines that have to pay for each aircraft landed will have an
incentive to switch to larger aircraft, which have a lower
per-passenger fuel consumption.
Background: Slot controls have existed in the United States since the late 1960s
in order to relieve congestion in certain congested airports like New YorkÕs
LaGuardia (LGA) airport. However, the slots were not originally allocated based
on market based mechanisms. Currently the FAA is
planning to phase in a new scheme at LaGuardia to force some number of slots to
go on the market every year and also to encourage larger aircraft by demanding
a minimum average aircraft size from each airline [1, 2]. Slot controls are
also used in European airports, and the responsible government agencies are planning
to encourage more market turnover of these slots [3, 4]. While these examples
of slot controls have achieved some degree of success in managing congestion,
they have not been entirely successful. For instance LGA has some of the worst
delays in the United States, despite its slot controls. Part of the problem is
that basic slot control schemes do not adapt to changes in airport operational
capacity due to weather. Real-time or near real-time market mechanisms have the
potential to offer this added adaptability, but such schemes are not currently
used today. The research community has also studied the problem of managing
airport congestion with pricing and market mechanisms [5, 6, 7, 8, 9, 10]. Our
work will complement the existing literature. Our initial investigation will
focus on a novel idea using a bidding language based on temporal utility
functions in the context of a Vickery-Clarke-Grove (VCG) mechanism. The scheme allows aircraft operators to express their utility
for each of a series of slots with a single bid. VCG type mechanisms work by
solving an optimization problem and our choice for the form of the temporal
utility function makes the solution easy to compute, even with a large number
of bids. Our mechanism works in an "on-line" way so that as aircraft enter and
depart the auction, the slots can be optimally reallocated with minimal
computation.
Objectives:
Further
study the properties of our proposed mechanism from a game-theory/mechanism
design perspective. For instance: What are the exact computational
requirements? The mechanism may elicit truthful bids from individual bidders;
what happens when an airline is bidding for a large number of aircraft? The
scheme would generate revenues; could these revenues be a replacement for
existing landing fees?
Evaluate
time-scale of market, and possible implementation. The scheme should allocate slots
with a short-enough time scale to adapt to changes in capacity due to weather.
For landings, a reallocation of slots would require re-sequencing aircraft as
they arrive. This would be difficult or impossible with a very short lead-time,
but could it perhaps be done with a lead-time of a few hours by using ground holds
and changes en-route speeds. For takeoffs the re-sequencing problem is probably
easier. These are issues we expect to understand better by the end of the
project.
Evaluate
the environmental effect of the proposed scheme. What is the potential savings in
fuel and resulting reduction CO2 emissions? After considering how
the proposed scheme would incentivize airlines to use larger aircraft and avoid
peak times, how much more passenger traffic could airports like SFO handle
without having to build new, environmentally damaging runways?
Preliminary
Study As a way of quantifying the potential passenger delay savings of such a
scheme, we considered the delay of aircraft arriving to SFO according to the
published schedule in Summer 2006, with only runway operating. Under the
assumption that only 1 aircraft could land every 2 minutes, we considered the
number of passengers "waiting in the air" if either: i)
the aircraft landed in a First Come First Served (FCFS) manner (according to
the published schedule, or ii) the largest aircraft in queue lands first
(BPFS). Policy (ii) approximates what might happen with a market mechanism. The
results show that the passenger hours of delay can be cut in more than half.
[1] D. Bond, "A New Day at LGA; The
FAA's demand-management plans for LaGuardia call for bigger aircraft,
market-based slot turnover," Aviation
Week and Space Technology, Sept. 4, 2006.
[2]
M. Wald, "Airlines at LaGuardia Fight Bush Administration Proposal to Auction
Off Landing Rights," The New York Times,
Feb, 18, 2007.
[3]
Turner, "EC near to solution on slot trading," Flight International, Mar., 7, 2006.
[4]
D. Knibb, "Slot Shake-up," Airline Business, June 27, 2006.
[5]
M. E. Levine, "Landing Fees and the Airport Congestion
Problem," Journal of Law and Economics,
vol. 12, no. 1. Apr.1969, pp. 79-108.
[6]
S. Morrison, "The equity and efficiency of runway pricing," Journal of Public
Economics, vol 34, no. 1, Oct. 1987, pp. 45-60.
[7]
T. Johnson and I. Savage, "Departure Delays, the pricing of
congestion, and expansion proposals at Chicago O'Hare Airport," Journal of Air
Transport Management, vol. 12, no. 4, July 2006, pp. 182-190.
[8]
J. Daniel, K. T. Harback, "(When) Do hub airlines
internalize their self-imposed congestion delays?," Journal of Urban Economics, In Press.
[9]
S. R. Mehndiratta and M. Kiefer, "Impact of slot controls with a market-based allocation mechanism at San
Francisco International Airport," Transportation
Research Part A: Policy and Practice, vol. 37, no. 7, Aug 2003, pp 555-578.
[10]
M. Ball, G. Donohue, and K. Hoffman, "Auctions for the Safe, Efficient and
Equitable Allocation of Airspace System Resources," Chapter in: Combinatorial Auctions, MIT Press, 2005,
pp. 507-538.