Airline Schedule De-peaking
Airports go through peak periods when demand for departures and arrivals exceeds capacity. These peaks cause congestion on both the surface and in the terminal area, and can create delays and inefficiencies in operations. These peaks form mostly due to the way airlines create their schedules in response to their own business goals. For example, many airlines schedule short commuter runs between New York and Washington, or between Los Angeles and San Francisco, at the beginning of the business day. While individual airline schedules rarely exceed airport capacities, collectively, their schedules form peaks.
The optimization-based de-peaking algorithm groups flights into time intervals of a given length. The de-peaking algorithm considers all the departures and arrivals at the OEP 34 airports and smoothes the operations simultaneously. This prevents smoothing of the demand at one airport from causing peaks at other airports. The block times (block time starts when an aircraft moves under its own power for the purpose of flight and ends when the aircraft comes to rest after landing) are assumed fixed, so when a flight is shifted, both the departure and arrival times are updated. The main decision in the de-peaking algorithm is to determine the best possible departure/arrival interval for each flight given the feasible set of available intervals for that flight within a ±15-minute shift window of the original departure/arrival time. Flights that are not scheduled in the conventional sense (e.g., general aviation and military) are exempt from de-peaking.
Metron Aviation has conducted research on this phenomenon, promoting a mechanism by which airlines create their schedules such that the aggregate outcome of their schedules better matches airport capacities. The initial benefits of this mechanism fall into four categories: passenger delay, fuel burn, emission of pollutants, and direct operating cost. Our research shows that de-peaking can result in an average daily saving of $548,000 in passenger value of time. On a busy day, this value can reach up to $1,321,000. De-peaking can also save up to 450,000 kg of fuel on a busy day. Reduction in fuel burn can in turn lead to a 7000 kg reduction in CO emissions and a 1.5 million kg reduction in CO2 emissions on a busy day. Finally, our mechanism reduces aggregate direct operating costs—$706,000 reduction in average direct operating costs per carrier and up to $1,500,000 per carrier on a busy day.