US-12626602-B2 - Forecasting using real-time flight leg data

Abstract

A method, apparatus, system, and computer program product for managing flight leg data for flight legs flown by a plurality of aircraft. A computer system receives the flight leg data for the flight legs flown by the plurality of aircraft in real-time. The computer system associates aircraft operation data and airport data with the flight leg data such that the aircraft operation data and airport data associated with the flight leg data forms flight activity data for the flight legs. The computer system displays a selected view of the flight activity data in real-time in a graphical user interface on a display system in response to a query for the flight activity data using the selected view.

Inventors

• Shervin Beygi
• Jiawen LIANG
• DUANE M. DAVIS
• Veronica MACINNIS
• Andrew J. Parker

Assignees

• THE BOEING COMPANY

Dates

Publication Date

20260512

Application Date

20220923

Claims (20)

1. 1 . A method for managing flight leg data for flight legs flown by a plurality of aircraft, the method comprising: receiving, by a computer system, the flight leg data from a plurality of flight activity data sources for the flight legs flown by the plurality of aircraft in real-time, wherein the flight leg data includes missing data, erroneous data, and other incorrect information received from the plurality of flight activity data sources; transforming, by the computer system in order to improve accuracy of the flight leg data, the flight leg data by performing at least one of removing erroneous records, completing partial records, backfilling missing data, and correcting any inaccurately formed flight legs that may be received; receiving, by the computer system, airport data from a plurality of airport data sources, wherein the plurality of airport data sources are different than the plurality of flight activity data sources and data type of the airport data is different than the data type of the flight leg data; associating, by the computer system, aircraft operation data and airport data with the flight leg data such that the aircraft operation data and the airport data associated with the flight leg data forms flight activity data for the flight legs; generating, by a machine learning model in the computer system, a prediction of the flight activity data for future flight legs using the flight activity data determined in real-time as an input to the machine learning model, wherein the machine learning model has been trained using historical flight activity data; and displaying, by the computer system, a future view of the prediction of the flight activity data for the future flight legs in real-time in a graphical user interface on a display system in response to a query for the flight activity data using the future view.
2. 2 . The method of claim 1 further comprising: mapping, by the computer system, the flight activity data to geographic regions, wherein the selected view of the flight activity data displayed on the graphical user interface is based on a number of geographic regions selected for the selected view.
3. 3 . The method of claim 1 further comprising: comparing, by the computer system, a forecast for a flight volume for a time period with the flight volume that actually occurs using the flight activity data for a set of the flight legs for the time period wherein a comparison is formed from the forecast for the flight volume being compared with the flight activity data that actually occurs for the set of the flight legs for the time period; and generating, by the computer system, a set of actions in response to at least one of the flight volume being below a threshold, a change in a market trend, or a change in the flight volume in a regional market crossing a threshold.
4. 4 . The method of claim 1 , wherein the flight leg data is received from a plurality of flight activity data sources over a communication interface during flights of the plurality of aircraft.
5. 5 . The method of claim 1 , wherein associating, by the computer system, the aircraft operation data with the flight leg data such that the aircraft operation data associated with the flight leg data forms the flight activity data comprises: joining the flight leg data with the aircraft operation data; joining the flight leg data with the airport data; identifying international flights in the aircraft operation data; identifying commercial flights in the aircraft operation data; and joining the flight leg data with current market outlook data.
6. 6 . The method of claim 5 , wherein the flight leg data is located in flight leg records and wherein joining the flight leg data with the aircraft operation data comprises: removing the flight leg records in the flight leg data having missing registry numbers; searching the aircraft operation data associated with registry numbers in the flight leg records, wherein identified aircraft operation data for the registry numbers are identified from searching the aircraft operation data using the registry numbers in the flight leg records; combining the identified aircraft operation data with corresponding flight leg records using the registry numbers; identifying the registry numbers and flight times during which the registry numbers of the flight legs are present in the flight leg records; and removing duplicate flight leg records having same registry numbers at same flight times using an aircraft type.
7. 7 . The method of claim 5 , wherein the flight leg data is located in flight leg records and wherein joining the flight leg data with the airport data in the aircraft operation data comprises: identifying the flight leg records with missing airport identifiers; adding the missing airport identifiers using aircraft location information; searching the airport data associated with airport identifiers in the flight leg records, wherein identified airport data for the airport identifiers are identified from searching the airport data using the airport identifiers in the flight leg records; and combining the identified airport data identified with corresponding flight leg records using the airport identifiers.
8. 8 . The method of claim 5 , wherein the flight leg data is located in flight leg records and wherein identifying the international flights in the aircraft operation data comprises: determining origination country codes and destination country codes in the flight leg records; identifying domestic flights as flights having a match between the origination country codes and the destination country codes; and identifying the international flights as the flights having in which the match is absent between the origination country codes and the destination country codes; wherein identifying the commercial flights in the aircraft operation data comprises: determining aircraft usage type from flight leg records; identifying the commercial flights as flights having aircraft usage types that are commercial aircraft usage types; and identifying non-commercial flights as the flights having aircraft usage types that are non-commercial aircraft usage types.
9. 9 . The method of claim 5 , wherein the flight leg data is located in flight leg records and wherein joining the flight leg data with the current market outlook data comprises: determining a number of countries for flight leg data using airport identifiers in the flight leg data; determining a number of regions in current market outlook region data that corresponds to the flight leg data from the current market outlook region mapping data using the number of countries identified; and adding the number of regions identified to the flight leg data.
10. 10 . The method of claim 1 , wherein associating, by the computer system, the aircraft operation data with the flight leg data such that the aircraft operation data associated with the flight leg data forms the flight activity data comprises: removing flight leg records in the flight leg data having missing registry numbers; searching the aircraft operation data associated with registry numbers in the flight leg records, wherein identified aircraft operation data for the registry numbers are identified from searching the aircraft operation data using the registry numbers in the flight leg records; combining the aircraft operation data with corresponding flight leg records using the registry numbers; identifying the registry numbers and flight times during which the registry numbers of the flight legs are present in the flight leg records; removing duplicate flight leg records having same registry numbers at same flight times using an aircraft type; identifying the flight leg records with missing airport identifiers; adding the missing airport identifiers using aircraft location information; searching the airport data associated with airport identifiers in the flight leg records, wherein identified airport data for the airport identifiers are identified from searching the airport data using the airport identifiers in the flight leg records; combining the identified airport data with corresponding flight leg records using the airport identifiers; determining origination country codes and destination country codes in the flight leg records; identifying domestic flights as flights having a match between the origination country codes and the destination country codes; identifying international flights as the flights having in which the match is absent between the origination country codes and the destination country codes; determining aircraft usage type from the flight leg records; identifying commercial flights as flights having aircraft usage types that are commercial aircraft usage types; identifying non-commercial flights as the flights having aircraft usage types that are non-commercial aircraft usage types; determining a number of countries for flight leg data using airport identifiers in the flight leg data; determining a number of regions in current market outlook region data that corresponds to the flight leg data from the current market outlook region mapping data using the number of countries identified; adding the number of regions identified to the flight leg data to form a training data set for training a machine learning model; and training the machine learning model using the training data set, wherein the machine learning model predicts aircraft operation data for flight legs in response to being training using the training data set.
11. 11 . A method for training a machine learning model to manage aircraft operations, the method comprising: determining, by a computer system, aircraft operation data, airport data, and flight leg records having flight leg data for a plurality of aircraft; associating, by the computer system, the aircraft operation data and the airport data with the flight leg data to create flight activity data for flight legs; searching, by the computer system, for missing aircraft operation data and missing airport data within the flight activity data associated with the plurality of aircraft; adding, by the computer system, at least one of the missing aircraft operation data or the missing airport data into the flight activity data for the plurality of aircraft to create historical flight activity data; and training the machine learning model to predict the flight activity data for future flight legs for the plurality of aircraft using the historical flight activity data.
12. 12 . The method of claim 11 , wherein training the machine learning model to predict the flight activity data for the future flight legs for the plurality of aircraft using the historical flight activity data further comprises: inputting, by the computer system, real-time flight activity data received by the computer system into the machine learning model; and predicting, by the machine learning model, additional flight activity data for the future flight legs for the plurality of aircraft from the real-time flight activity data.
13. 13 . The method of claim 12 further comprising: comparing, by the computer system, the real-time flight activity data with the historical flight activity data for the plurality of aircraft; and training the machine learning model using the real-time flight activity data when a comparison value of a comparison between the real-time flight activity data and the historical flight activity data exceeds a threshold.
14. 14 . The method of claim 12 , wherein, associating, by the computer system, the aircraft operation data with the airport data to create the flight activity data for the flight legs further comprises: searching, by the computer system, for missing registry numbers within the flight leg records; removing, by the computer system, the flight leg records having the missing registry numbers; searching, by the computer system, the aircraft operation data using registry numbers in the flight leg records and the airport data using airport identifiers in the flight leg records, wherein identified aircraft operation data and identified airport data are identified from searching the aircraft operation data and the airport data using the registry numbers in the flight leg records and the airport identifiers in the flight leg records; and combining, by the computer system, at least one of the identified aircraft operation data or the identified airport data with corresponding flight leg records using the registry numbers and the airport identifiers.
15. 15 . An aircraft information system for managing flight leg data for flight legs flown by a plurality of aircraft, the aircraft information system comprising: a computer system; and an aircraft data manager in the computer system, wherein the aircraft data manager operates to: receive the flight leg data from a plurality of flight activity data sources for the flight legs flown by the plurality of aircraft in real-time, wherein the flight leg data includes missing data, erroneous data, and other incorrect information received from the plurality of flight activity data sources; transform, in order to improve accuracy of the flight leg data, the flight leg data by performing at least one of removing erroneous records, completing partial records, backfilling missing data, and correcting any inaccurately formed flight legs that may be received; receive airport data from a plurality of airport data sources, wherein the plurality of airport data sources are different than the plurality of flight activity data sources; associate aircraft operation data and airport data with the flight leg data such that the aircraft operation data associated with the flight leg data forms flight activity data for the flight legs; generate, by a machine learning model in the computer system, a prediction of the flight activity data for future flight legs using the flight activity data received in real-time as an input to the machine learning model, wherein the machine learning model has been trained using historical flight activity data; and display a future view of the prediction of the flight activity data for the future flight legs in real-time in a graphical user interface on a display system in response to a query for the flight activity data using the future view.
16. 16 . The aircraft information system of claim 15 , wherein the aircraft data manager operates to: compare a forecast for a flight volume for a time period with the flight volume that actually occurs using the flight activity data for a set of the flight legs for the time period wherein a comparison is formed from the forecast for the flight activity data for the set of the flight legs being compared with the flight activity data that actually occurs for the set of the flight legs for the time period; and generate a set of actions in response to at least one of the flight volume being below a threshold, a change in a market trend, or a change in the flight volume in a regional market crossing a threshold.
17. 17 . The aircraft information system of claim 15 , wherein the flight leg data is received from a plurality of flight activity data sources over a communication interface during flights of the plurality of aircraft.
18. 18 . An aircraft information system for generating training data to manage aircraft operations, the aircraft information system comprising: a computer system; and an aircraft data manager in the computer system, wherein the aircraft data manager operates to: determine aircraft operation data, airport data, and flight leg records having flight leg data for a plurality of aircraft; associate the aircraft operation data and the airport data with the flight leg data to create flight activity data for the flight legs; search for missing aircraft operation data and missing airport data within the flight activity data associated with the plurality of aircraft; add at least one of the missing aircraft operation data or the missing airport data into the flight activity data for the plurality of aircraft to create historical flight activity data; and train a machine learning model to predict the flight activity data for future flight legs for the plurality of aircraft using the historical flight activity data.
19. 19 . The aircraft information system of claim 18 , wherein in training the machine learning model to predict the flight activity data for future legs for the plurality of aircraft using the historical flight activity data, the aircraft data manager operates to: input real-time flight activity data received by the computer system into the machine learning model; predict, by the machine learning model, additional flight activity data for the future flight legs for the plurality of aircraft from the real-time flight activity data; compare the real-time flight activity data with the historical flight activity data for the plurality of aircraft; and train the machine learning model using the real-time flight activity data when a comparison value of a comparison between the real-time flight activity data and the historical flight activity data exceeds a threshold.
20. 20 . The aircraft information system of claim 18 , wherein in associating the plurality of aircraft with the aircraft operation data and the airport data to create the flight activity data for the flight legs, the aircraft data manager operates to: search for missing registry numbers within the flight leg records; remove the flight leg records having the missing registry numbers; search the aircraft operation data using registry numbers in the flight leg records and the airport data using airport identifiers in the flight leg records, wherein identified aircraft operation data and identified airport data are identified from searching the aircraft operation data and the airport data using the registry numbers in the flight leg records and the airport identifiers in the flight leg records; and combining at least one of the identified aircraft operation data or the identified airport data with corresponding flight leg records using the registry numbers and the airport identifiers.

Description

BACKGROUND INFORMATION 1. Field The present disclosure relates generally to improved computer system and in particular, to a method, apparatus, system, and computer program product for forecasting flight activity using flight leg data. 2. Background Flight volume is a type of flight activity that is a useful indicator of the needs of commercial airlines. Flight volume activity can be used to identify services for airlines. For example, forecasts of flight volume activity can be useful in determining future maintenance needs for airlines. For example, the flight volume for a commercial airline can be used to plan maintenance and manage part inventories for its aircraft fleet. Forecasts of flight volume can be used by the commercial airline to anticipate the need for parts and maintenance personnel based on these forecasts. Further, the commercial airline can also plan on the location of positioning of aircraft in its fleet with respect to maintenance facilities using these forecasts. Current forecasting techniques provide forecasts using historical flight volume activity data to make projections of future flight volume activity. For example, current analysis techniques use historical airplane utilization data that may be obtained on a quarterly or monthly basis. Current forecasts, however, are not in real-time and may not provide a desired level of accuracy. As result, these forecasts may not be as useful to a commercial airline for planning purposes. Therefore, it would be desirable to have a method and apparatus that take into account at least some of the issues discussed above, as well as other possible issues. For example, it would be desirable to have a method and apparatus that overcome a technical problem with forecasting flight activity for commercial airlines. SUMMARY An embodiment of the present disclosure provides a method for managing flight leg data for flight legs flown by a plurality of aircraft. A computer system receives the flight leg data for the flight legs flown by the plurality of aircraft in real-time. The computer system associates aircraft operation data and airport data with the flight leg data such that the aircraft operation data and airport data associated with the flight leg data forms flight activity data for the flight legs. The computer system displays a selected view of the flight activity data in real-time in a graphical user interface on a display system in response to a query for the flight activity data using the selected view. According to other illustrative embodiments, a computer system, and a computer program product for managing flight leg data are provided. In another illustrative embodiment of the present disclosure, a method for training a machine learning model to manage aircraft operations is provided. A computer system determines aircraft operation data, airport data, and flight leg records having flight leg data for a plurality of aircraft. The computer system associates the aircraft operation data and the airport data with the flight leg data to create flight activity data for the flight legs. The computer system searches for missing aircraft operation data and missing airport data within the flight activity data associated with the plurality of aircraft. The computer system adds at least one of the missing aircraft operation data or the missing airport data into the flight activity data for the plurality of aircraft to create historical flight activity data. The computer system trains the machine learning model to predict the flight activity data for future flight legs for the plurality of aircraft using the historical flight activity data. According to other illustrative embodiments, a computer system, and a computer program product for training a machine learning model to manage aircraft operations are provided. The features and functions can be achieved independently in various embodiments of the present disclosure or may be combined in yet other embodiments in which further details can be seen with reference to the following description and drawings. BRIEF DESCRIPTION OF THE DRAWINGS The novel features believed characteristic of the illustrative embodiments are set forth in the appended claims. The illustrative embodiments, however, as well as a preferred mode of use, further objectives and features thereof, will best be understood by reference to the following detailed description of an illustrative embodiment of the present disclosure when read in conjunction with the accompanying drawings, wherein: FIG. 1 is a pictorial representation of a network of data processing systems in which illustrative embodiments may be implemented; FIG. 2 is an illustration of a block diagram of an aircraft information environment in accordance with an illustrative embodiment; FIG. 3 is an illustration of a block diagram of an aircraft data manager displaying flight activity data in accordance with an illustrative embodiment; FIG. 4 is an illustration of a block diagram o