Method, system and processor-readable media for estimating airport usage demand

• US 9,715,695 B2
• Filed: 06/01/2015
• Issued: 07/25/2017
• Est. Priority Date: 06/01/2015
• Status: Active Grant

• Current Assignee: Conduent Business Services, LLC (Conduent, Inc.)
• Original Assignee: Conduent Business Services, LLC (Conduent, Inc.)
• Inventors: de Souza, Cesar Roberto, Ulloa Paredes, Luis Rafael, Chidlovskii, Boris, Ciriza, Victor
• Primary Examiner(s): CASS, JEAN PAUL
• Application Number:
  US14/726,887
• Publication Number:
  US 20160350770A1
• Time in Patent Office: 785 Days
• Field of Search: 701120
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities

Abstract

Methods and systems for estimating airport usage demand. Airport parking traffic usage data and flight-time table data can be compiled with respect to an airport (or more than one airport). The airport parking traffic usage data and flight-time table data can be analyzed using an efficient time matching approach (e.g., a time segment matching algorithm). An efficient method to match passengers and flights is introduced. Passenger behavior can be estimated with respect to the airport based on the airport parking traffic usage data and flight-time table data.

Claims

1. A method for estimating airport usage demand utilizing an electronic database to improve the functioning of an airport system, comprising:

compiling in an electronic database, airport parking traffic usage data and flight-time table data with respect to at least one airport, wherein said airport parking traffic usage data includes a distribution of vehicle parking events associated with said at least one airport and wherein said flight-time table data includes flight data comprising arrival flight information including a total number of arriving flights, flight arrival times, departure flight information including a total number of departing flights, flight departure times, and departure flight maximum capacities including a total number of passengers for each departing flight, said vehicle parking events associated with vehicles including cars and trucks;

analyzing said airport parking traffic usage data and said flight-time table data contained in said electronic database using a time matching approach comprising processing an efficient time matching algorithm comprising a segment matching algorithm with said flight-time table data and said traffic usage data utilizing a processor of a data-processing system that is operably connected to said electronic database, said time matching approach including performing an initial matching between parking users and potential flight passengers in linearithmic average time complexity; and

estimating with a processor of a data-processing system, passenger behavior including passenger flow with respect to said at least one airport based on said airport parking traffic usage data and said flight-time table data, in response to analyzing said airport parking traffic usage data and said flight-time data contained in said electronic database, wherein said estimating includes inferring an amount of passengers that took a flight in response to analyzing said distribution of parking events, and wherein said passenger flow as said at least one airport is modeled as a mixture of components during a given time period, wherein each component among said mixture of components is associated with a departure flight and wherein said each component in said mixture is modeled as a finite interval distribution and in some cases of said mixture as triangular distributions.

2. The method of claim 1 further comprising determining an alignment of said flight data and parking data comprising said distribution of parking events in response to analyzing said airport parking traffic usage data and said flight-time table data using said time matching approach and wherein said distribution of parking events associated with said at least one airport comprises a mixture distribution with respect to said mixture of said components that is approximated using said triangular distributions or Beta distributions and wherein weights of said mixture distribution are estimated.

3. The method of claim 2 further comprising estimating said passenger behavior with respect to said at least one airport based on said alignment of said flight data and said parking data, wherein both component distributions and said weights are learned using a maximum likelihood technique from matched flights and said parking events.

4. The method of claim 2 further comprising:

deriving said parking data from at least one off-street parking lot composed of distinct parking zones with different pricing strategies and physical characteristics with respect to said at least one airport;

determining an alignment of said flight data and said parking data in response to analyzing said airport parking traffic usage data and flight-time table data using said time matching approach; and

estimating said passenger behavior with respect to said at least one airport based on said alignment of said flight data and said parking data.

5. The method of claim 3 wherein said efficient time matching algorithm comprises said segment matching algorithm with said linearithmic average time complexity in an average case.

6. A method of modeling airport usage demand to improve the functioning of an airport system, comprising:

providing an electronic database of past and future departing and arriving flights from at least one airport and an electronic database of historical parking data with respect to said at least one airport, wherein said historical parking data includes a distribution of parking events associated with said at least one airport and said past and future departing and arriving flights includes information comprising a total number of arriving flight, flight arrival times, departure flight information including a total number of departing flights, flight departure times, and departure flight maximum capacities including a total number of passengers for each departing flight;

storing said electronic database in a memory of a computer; and

generating a model of airport usage demand including passenger with respect to said at least one airport by processing information contained in said database of past and future departing and arriving flights and in said database of historical parking data and based on a processed match between said historical parking data and said past and future departing and arriving flights using a time matching approach comprising a segment matching algorithm, and wherein said passenger flow as said at least one airport is modeled via said model as a mixture of components during a given time period, wherein each component among said mixture of components is associated with a departure flight and wherein said each component in said mixture is modeled as a finite interval distribution and in some cases of said mixture as triangular distributions and wherein said time matching approach includes performing an initial matching between parking users and potential flight passengers in linearithmic average time complexity.

7. The method of claim 6 wherein said model after being processed by a processor infers how many passengers took a particular flight by analyzing a distribution of parking events with respect to said at least one airport and based on said information contained in said database of past and future departing and arriving flights and said database of historical parking data involving vehicles including cars and trucks parked at said at least one airport and at parking lots surrounding said at least one airport.

8. The method of claim 7 further comprising estimating for a particular flight with respect to said at least one airport, and based on said model, a percentage of users who likely parked said vehicles in airport parking at said at least one airport and a percentage of users who did not.

9. The method of claim 7 further comprising developing said match based on viewing a flux of passengers within said at least one airport as a mixture distribution wherein each component of said mixture distribution represents at least one flight with respect to said at least one airport and wherein said at least one flight is either an arrival, a departure, or both and wherein said mixture distribution is approximated utilizing triangular distributions or Beta distributions.

10. The method of claim 9 wherein said historical parking data is derived from an off-street parking lot composed of distinct parking zones with different pricing strategies and physical characteristics, surrounding said at least one airport.

11. The method of claim 9 wherein data associated with said match is employed to create a generative model of parking events that is queried to generate new parking event samples, wherein said model comprises a generative model.