Flight operations scheduling method and apparatus, and device

Abstract

A flight operations scheduling method and apparatus, and a device. The method comprises: obtaining an estimated time of arrival and an estimated taxi‑in time of each inbound flight, and obtaining an earliest cleared pushback time and an estimated taxi‑out time of each outbound flight (step 1003); and on the basis of the estimated time of arrival of each inbound flight, the estimated taxi‑in time of each inbound flight, the earliest cleared pushback time of each outbound flight and the estimated taxi‑out time of each outbound flight, determining a suggested take-off time and / or a suggested pushback time of each outbound flight (step 1004). The method can perform relatively more rational scheduling on operation moments of airport flights.

Description

Flight operation scheduling method, device and equipment

[0001] This invention claims priority to the Chinese patent application filed with the State Intellectual Property Office of China on January 29, 2024, with application number 202410121036.9 and application name “Flight Operation Scheduling Method, Device and Equipment”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the technical field of flight operation scheduling, and in particular to a flight operation scheduling method, device and equipment. Background Art

[0003] Airports prioritize the regular arrival and departure of flights and the regular movement of passengers. Flight regularity directly impacts the passenger experience and is a key performance indicator for airport performance. Therefore, rationally scheduling flight schedules is a crucial function for airport application systems.

[0004] Summary of the Invention

[0005] The embodiments of the present application provide a flight operation scheduling method, apparatus, and device that can schedule the operation times of airport flights in a relatively more reasonable manner.

[0006] In a first aspect, the present application provides a flight operation scheduling method, including: obtaining the estimated arrival time and estimated arrival taxi time of each incoming flight, and obtaining the earliest pushback time and estimated departure taxi time of each outgoing flight; determining the recommended take-off time and / or recommended pushback time of each outgoing flight based on the estimated arrival time of each incoming flight, the estimated arrival taxi time of each incoming flight, the earliest pushback time of each outgoing flight, and the estimated departure taxi time of each outgoing flight. The method determines the recommended take-off time and / or recommended pushback time of each outgoing flight based on the estimated arrival time of each incoming flight, the estimated arrival taxi time of each incoming flight, the earliest pushback time of each outgoing flight, and the estimated departure taxi time of each outgoing flight, thereby enabling a relatively more reasonable scheduling of the recommended take-off time and / or recommended pushback time of the outgoing flights, that is, enabling a relatively more reasonable scheduling of the operating times of airport flights.

[0007] In one possible implementation, determining a recommended takeoff time and / or recommended pushback time for each outbound flight based on the estimated arrival time of each inbound flight, the taxi time of each inbound flight, the earliest pushback time for each outbound flight, and the taxi time of each outbound flight includes: inputting the estimated arrival time of each inbound flight, the estimated arrival taxi time of each inbound flight, the earliest pushback time for each outbound flight, and the estimated departure taxi time of each outbound flight into a preset MIP model to obtain the recommended takeoff time and / or recommended pushback time for each outbound flight.

[0008] In one possible implementation, obtaining the estimated arrival time and taxi time of each incoming flight and obtaining the earliest pushback time and taxi time of each outgoing flight includes: obtaining the estimated arrival time of each incoming flight and the earliest pushback time of each outgoing flight; and determining the estimated arrival taxi time of each incoming flight and the estimated departure taxi time of each outgoing flight based on the estimated arrival time of each incoming flight and the earliest pushback time of each outgoing flight.

[0009] In one possible implementation, obtaining the estimated arrival time of each inbound flight includes: obtaining the real-time location of each inbound flight; inputting the real-time location of the inbound flight into a preset remaining flight time prediction model, and using the output of the remaining flight time prediction model as the remaining flight time of the inbound flight; and determining the estimated arrival time of the inbound flight based on the remaining flight time of the inbound flight.

[0010] In a possible implementation, the method further includes: obtaining weather information of each waypoint between the real-time position of the incoming flight and the target position, equipment information of the incoming flight, and / or weather information of the target position of the incoming flight;

[0011] The real-time positions of incoming flights are fed into a pre-set remaining flight time prediction model, including:

[0012] The real-time position of the incoming flight, weather information of each waypoint between the real-time position and the target position, equipment information of the incoming flight, and / or weather information of the target position of the incoming flight are input into a preset remaining flight time prediction model.

[0013] In one possible implementation, obtaining the estimated arrival time of each inbound flight includes: for each inbound flight, obtaining real-time location information of the inbound flight and weather information associated with the inbound flight; inputting the real-time location information and weather information of the inbound flight into a preset remaining flight time prediction model, and using the time output by the remaining flight time prediction model as the remaining flight time of the inbound flight; and determining the estimated arrival time of the inbound flight based on the remaining flight time of the inbound flight.

[0014] In a possible implementation, the method further includes: obtaining equipment information of incoming flights; inputting the real-time location information and weather information of incoming flights into a preset remaining flight time prediction model, including: inputting the real-time location information, weather information and equipment information of incoming flights into a preset remaining flight time prediction model.

[0015] In one possible implementation, obtaining the earliest possible push-out time for each outbound flight includes: for each outbound flight, obtaining real-time status information of the outbound flight and the time interval between the target take-off time of the outbound flight and the real-time time; the real-time status information of the outbound flight is used to record the real-time status of the outbound flight at each flight support node; a flight support node is a node in the support process from arrival to departure of a flight; the real-time status information and the time interval of the outbound flight are input into a preset remaining support time prediction model, and the time output by the remaining support time prediction model is obtained as the remaining support time of the outbound flight; and the earliest possible push-out time of the outbound flight is determined based on the remaining support time.

[0016] In one possible implementation, the estimated arrival taxi time of each arriving flight and the estimated departure taxi time of each departing flight are determined based on the estimated arrival time of each arriving flight and the earliest pushback time of each departing flight, including: inputting the estimated arrival time of each arriving flight and the earliest pushback time of each departing flight into a preset taxi time prediction model, and obtaining the estimated arrival taxi time of each arriving flight and the estimated departure taxi time of each departing flight output by the preset taxi time prediction model.

[0017] In one possible implementation, a method for training a taxi time prediction model includes: generating a taxi path network diagram based on historical taxi paths and stand runway threshold information of inbound and outbound flights; clustering the stands and runway thresholds based on the taxi path network diagram to obtain runway stand groupings where taxi conflicts may occur; and estimating the obstacle-free taxi time and congestion factor for each inbound and outbound flight.

[0018] In a second aspect, an embodiment of the present application provides a flight operation scheduling device, comprising:

[0019] An acquisition unit, configured to obtain the estimated arrival time and estimated arrival taxi time of each incoming flight, and obtain the earliest pushback time and estimated departure taxi time of each outgoing flight;

[0020] a determining unit, configured to determine a recommended take-off time and / or a recommended pushback time for each outbound flight based on the estimated arrival time of each inbound flight, the estimated arrival taxi time of each inbound flight, the earliest pushback time of each outbound flight, and the estimated departure taxi time of each outbound flight.

[0021] In a third aspect, an embodiment of the present application provides an electronic device comprising: a processor and a memory; wherein one or more computer programs are stored in the memory, and the one or more computer programs include instructions, which, when executed by the processor, enable the electronic device to execute any one of the methods described in the first aspect.

[0022] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, which, when executed on a computer, enables the computer to execute any of the methods described in the first aspect.

[0023] In a fifth aspect, an embodiment of the present application provides a computer program, which, when executed by a computer, is used to execute the method of the first aspect.

[0024] In one possible design, the program in the fifth aspect may be stored in whole or in part on a storage medium packaged with the processor, or may be stored in whole or in part on a memory not packaged with the processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] FIG1 is a schematic diagram of the flight arrival and departure procedures and departmental collaboration relationships provided in an embodiment of the present application;

[0026] FIG2 is a schematic diagram of key factors based on a flight support process according to an embodiment of the present application;

[0027] FIG3 is a schematic diagram showing the relationship between the ACDM system and other systems provided in an embodiment of the present application;

[0028] FIG4 is a functional diagram of a national traffic management system provided in an embodiment of the present application;

[0029] FIG5 is a schematic diagram of an airport operation application architecture provided by an embodiment of the present application;

[0030] FIG6 is a schematic diagram of the operation of the flight operation suggestion system provided in an embodiment of the present application;

[0031] FIG7 is a schematic diagram of the operation architecture of the flight operation scheduling method provided in an embodiment of the present application;

[0032] FIG8 is a flow chart of a flight operation scheduling method provided in an embodiment of the present application;

[0033] FIG9 is a schematic diagram of the MIP model architecture provided in an embodiment of the present application;

[0034] FIG10 is another schematic flow chart of a flight operation scheduling method provided in an embodiment of the present application;

[0035] FIG11 is a schematic diagram showing the principle of an estimated time of arrival prediction model provided in an embodiment of the present application;

[0036] FIG12 is a schematic diagram of a relationship network between flight support nodes provided in an embodiment of the present application;

[0037] FIG13 is a schematic diagram showing the principle of the earliest available time prediction model provided in an embodiment of the present application;

[0038] FIG14 is a schematic diagram showing the principle of a glide time prediction model provided in an embodiment of the present application;

[0039] FIG15 is a schematic diagram of the structure of the flight operation scheduling device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0040] The terms used in the implementation section of this application are only used to explain the specific embodiments of this application and are not intended to limit this application.

[0041] First, the terms that may be involved in the embodiments of the present application are exemplified.

[0042] Punctuality: also known as on-time rate or flight regularity rate, refers to the ratio of the number of flights whose actual departure times are relatively consistent with the planned departure times (i.e. normal flights) to the total number of flights when the air passenger transport department implements the transport plan, which represents the carrier's transport efficiency and quality.

[0043] Normal release rate: normal release flights / total release flights. The judgment rule for normal release flights is that the flight has not been canceled, returned, or diverted and meets the following conditions:

[0044] 1) Landing no later than 10 minutes after the scheduled door opening time, i.e. actual landing time minus the planned arrival time ≤ 10;

[0045] 2) Take off within the specified airport ground taxi time after the scheduled door closing time, that is, actual take-off time - planned departure time - ground taxi time ≤ 0.

[0046] Scheduled Time of Departure (STD): The flight departure time approved by the flight schedule management department.

[0047] Estimated Time of Arrival (ETA): The predicted time a flight will arrive at the airport.

[0048] Variable Taxi Time (VTT): The estimated time a flight takes to taxi from the departure gate to the takeoff runway, or from the landing runway to the arrival gate.

[0049] Estimated Taxi-In Time (VTT-IN / Estimated Taxi-In Time, EXIT): The time from the time an aircraft lands to the time it leaves the wheel block.

[0050] Estimated Taxi-Out Time (EXOT): The time between an aircraft leaving the block and taking off.

[0051] Estimated Off-Block Time (EOBT): The time when a flight is released from block.

[0052] Calculated Take-Off Time (CTOT): The estimated take-off time of a flight calculated and assigned by the flight flow management system.

[0053] Airports prioritize the regular arrival and departure of flights and the regular movement of passengers. Flight regularity directly impacts the passenger experience and is a key metric in airport performance. According to statistics, in 2021, national passenger airlines operated a total of 3.7859 million flights, of which 3.3314 million were regular, with an average regularity rate of 88.00%. In 2022, national passenger airlines operated a total of 2.3938 million flights, of which 2.2735 million were regular, with an average regularity rate of 94.98%.

[0054] Key factors affecting flight regularity include weather, airlines, and air traffic control. Taking 2022 as an example, the percentage of flight irregularities due to various factors is shown in Table 1 below, with severe weather accounting for the majority. In late March 2023, heavy rain in a certain area caused hundreds of flight cancellations at the local airport, with nearly half of the flights delayed. See Table 2 below for details. At the end of May 2023, continuous thunderstorms at a certain airport also significantly reduced the on-time flight clearance rate. See Table 3 below for details.

[0055] Table 1 Classification of reasons for flight irregularities in 2022

[0056] Table 2: Airport flight regularity statistics from March 23 to 25, 2023

[0057] Table 3. Airport flight regularity statistics from May 25 to June 3, 2023

[0058] Airport operations primarily revolve around flight arrival and departure procedures. As shown in Figure 1, normal flight takeoffs and landings involve resource coordination and trade-offs across multiple departments. For example, air traffic control controls runway use for takeoff and landing, airlines manage gate entry and exit, and open and close doors, the apron tower controls gate usage and taxiing, and ground handling companies provide support from gate entry to exit. The operations command center coordinates information from all aspects and the work of various departments to ensure that flights follow the proper arrival and departure procedures.

[0059] Based on the key nodes of the flight support process, the key influencing factors of each department can be quantified into key data as shown in Figure 2. These mainly include: the estimated time of arrival (ETA) and estimated arrival taxi time (EXIT / VTT-IN) of inbound flights, the estimated off-block time (EOBT), estimated departure taxi time (EXOT / VTT-OUT), and scheduled departure time (STD) of outbound flights, as well as the runway's arrival and departure hourly capacity and air traffic control.

[0060] In one example, an airport collaborative decision making (A-CDM) system such as that shown in FIG3 is used to realize information sharing between the airport and the collaborative departments, thereby assisting the collaborative decision making among the departments.

[0061] According to the "A-CDM Implementation Specifications (Trial)", A-CDM is an operational mechanism led by airports and jointly participated by airlines, air traffic control, and ground service companies, based on information sharing and centered on collaborative decision-making. A-CDM achieves comprehensive improvements in airport ground operation efficiency by effectively controlling aircraft ground operation support nodes, optimizing ground resource allocation, and improving flight ground transit and departure sequencing. Expected goals include: sharing operational data, achieving transparency of preceding flight information, achieving process control of flight ground operation support, and implementing ground-to-air monitoring to ensure ground operation safety and improve ground support efficiency.

[0062] The current A-CDM system serves more as an IT system and information integration platform, enabling information sharing between airports and collaborative departments and assisting in collaborative decision-making among departments. It does not have built-in operational optimization algorithms and cannot help optimize decisions such as flight release sequencing.

[0063] In another example, flight scheduling is performed using the National Air Traffic Management System (NATM). Launched in 2021, the system automatically generates flight sequences based on flow control information. It utilizes communication, navigation, and surveillance technologies to monitor and control aircraft flight activities, providing various air traffic management services for civil aircraft, effectively maintaining air traffic order, promoting air traffic safety, and ensuring smooth air traffic.

[0064] The National Traffic Management System, as shown in Figure 4, primarily uses traffic control information from national airspace to schedule all flights passing through controlled airspace, waypoints, and corridor entrances, and provides a calculated takeoff time (CTOT). However, the National Traffic Management System only provides recommended takeoff times for controlled flights. Uncontrolled flights must still schedule their departure times based on actual airport conditions. In particular, in the event of flight delays, airports must rationally adjust the release sequence for uncontrolled flights based on overall capacity, limiting the number of delayed flights within capacity limits and improving the on-time release rate of flights.

[0065] Based on the above description, it can be seen that reasonable scheduling of airport flight operating times is an important function that the airport application system needs to achieve.

[0066] To this end, the embodiments of the present application also provide a flight operation scheduling method and system, which more accurately optimizes airport flight operation recommendations by integrating flight key node prediction and time scheduling optimization.

[0067] Furthermore, in the pre-adjustment scenario, the embodiment of the present application can adjust the flight plan N hours in advance. By comprehensively considering future flow control requirements, ground support capabilities, runway release capabilities, etc., a version of the flight plan that is fair, stable and balanced can be output, thereby improving the flight release regularity rate in the next N hours from the plan, while minimizing flight schedule changes and improving passenger experience; in the backlog scenario, by coordinating the goals of all parties (operation command center, air traffic control, ground services, airlines, etc.) and ground dynamics, flights that have been delayed or are about to be delayed can be quickly and reasonably reduced, thereby improving the release speed, avoiding the occurrence of mass malicious incidents, and ensuring the stable operation of the airport.

[0068] An example of an airport operation architecture applicable to the embodiment of the present application is shown in Figure 5. The airport operation architecture may adopt a layered structure, for example, including: a business application layer, a data base layer, a cloud layer, a network layer, and a data source layer.

[0069] The business application layer includes a series of applications, such as flight operation status, flight operation prediction, flight support reminder, backlog release strategy, etc. In some embodiments, the flight operation scheduling method of the present application embodiment can be implemented as a business application of the business application layer, for example, it can be called flight operation scheduling suggestion, involving content including flight key support node prediction and flight release strategy suggestion.

[0070] In some embodiments, the prediction results of the flight operation scheduling method of the embodiment of the present application for key nodes can be used as reference data for applications such as flight guarantee reminders, flight operation status prediction, and flight operation prediction. The prediction of relevant times of outbound flights, such as the recommended take-off time and recommended deport time of outbound flights, can be used as reference data for pre-adjustment and dynamic adjustment of flight plans. Specifically, it can be used as reference data for release strategies and plan adjustments in pre-adjustment scenarios and backlog scenarios.

[0071] The data base layer includes services such as next-day flight planning, flight dynamic planning, and runway mode allocation.

[0072] The cloud layer includes: HCS cloud platform.

[0073] The network layer includes: wired network, wireless network, etc.

[0074] The data source layer includes: production data, management data, external data, etc., providing data sources for the data base layer and business application layer.

[0075] Airport operations primarily focus on flight flow. This application is primarily used for flight operation scheduling recommendations, including predictions of critical flight support nodes and flight release strategy recommendations. The application system architecture is shown in Figure 5. Key node prediction results can be used in scenarios such as flight support progress reminders and flight operation status prediction. Combined with the time scheduling model, they can be used for both pre-adjustment and dynamic adjustment of flight plans, including release strategy and plan adjustments in pre-adjustment and backlog scenarios.

[0076] The core of this application is the flight operation suggestion calculation engine, the core of which is the key node prediction algorithm (including ETA / EOBT / VTT prediction algorithm) and the time scheduling optimization algorithm. The specific schematic diagram is shown in Figure 6. Among them, the key node prediction algorithm may include: a prediction algorithm for the estimated arrival time (ETA) of incoming flights, a prediction algorithm for the earliest push-out time of outgoing flights, and a prediction algorithm for the taxiing time of incoming and outgoing flights. After integrating into the flight plan and dynamic data, as well as flow control, runway, taxiing and other data, the user can run the key node prediction algorithm to obtain the ETA of incoming flights, the earliest push-out time of outgoing flights, and the taxiing time of all flights. These output results can be used to guide ground service support work. At the same time, as shown in Figure 7, the above time information can also be provided to the time scheduling algorithm to refresh the flight queue. The algorithm comprehensively considers the flight regularity target and the ground service support balance target, as well as the airline fairness target, optimizes the release sequence, and outputs the recommended take-off time and / or recommended push-out time for each planned outbound flight.

[0077] As shown in FIG7 , in some embodiments, the above-mentioned key node prediction algorithm can be implemented using a corresponding prediction model.

[0078] As shown in Figure 7, in some embodiments, the time scheduling algorithm can implement outbound flight operation scheduling in different scenarios, including but not limited to: pre-scheduled scheduling scenarios and dynamic scheduling scenarios. Pre-scheduled scheduling scenarios can optionally include flow control scheduling and ice and snow removal, while dynamic scheduling scenarios can include backlog scheduling scenarios and regular scheduling scenarios.

[0079] FIG8 is a flow chart of a flight operation scheduling method according to an embodiment of the present application. As shown in FIG8 , the method may include:

[0080] Step 801: Obtain the estimated time of arrival (ETA) and estimated arrival taxi time (EXIT) of each incoming flight, and obtain the earliest possible push-out time and estimated departure taxi time (EXOT) of each outgoing flight.

[0081] Optionally, the estimated arrival taxi time may be the time between the arrival of the inbound flight and the arrival of the wheel block.

[0082] Alternatively, the estimated departure taxi time may be the time from the earliest possible pushback time to the departure of the departure flight.

[0083] In some embodiments, the estimated gate closing time or estimated off-block time (EOBT) of the outbound flight may be used as the earliest pushback time of the outbound flight. The earliest pushback time of the outbound flight may also be referred to as the scheduled time of departure (STD) of the outbound flight.

[0084] In some embodiments, in this step, the estimated arrival time and estimated arrival taxi time of each inbound flight within the first time period, as well as the earliest possible push-out time and estimated departure taxi time of each outbound flight can be obtained. The specific value of the above-mentioned first time period is not limited in this embodiment of the present application. The above-mentioned information on which inbound and outbound flights within the first time period is obtained can be determined based on the planned arrival time of inbound flights and the planned departure time of outbound flights in the future second time period, etc., which is not limited in this embodiment of the present application. Optionally, the starting time of the above-mentioned first time period can be determined based on the execution time of step 801, for example, using the execution time of step 801 as the starting time, etc. Optionally, the starting time of the above-mentioned second time period can be determined based on the execution time of step 801, for example, using the execution time of step 801 as the starting time, etc.

[0085] Step 802: Determine a recommended take-off time and / or a recommended pushback time for each outbound flight based on the estimated arrival time of each inbound flight, the estimated arrival taxi time of each inbound flight, the earliest pushback time of each outbound flight, and the estimated departure taxi time of each outbound flight.

[0086] In some embodiments, a mixed-integer programming (MIP) model may be pre-constructed in this step. The estimated arrival time and estimated arrival taxi time of the inbound flight, the earliest pushback time and estimated departure taxi time of the outbound flight obtained in the above step, etc. are input into the MIP model to obtain the recommended take-off time and / or recommended pushback time of each outbound flight.

[0087] In some embodiments, other information of each inbound flight and outbound flight may also be input into the MIP model in this step, which is not limited in the embodiment of the present application.

[0088] With reference to FIG9 , the method for constructing the MIP model is exemplarily described.

[0089] Step S11: Define the decision variables of the model according to the timing decision.

[0090] Optionally, the decision variables may include: an estimated arrival time of each incoming flight, an estimated arrival taxi time of each incoming flight, an earliest pushback time of each outgoing flight, and an estimated departure taxi time of each outgoing flight.

[0091] In some embodiments, in order to improve the accuracy of the MIP model, referring to FIG9 and Table 4, the decision variables may also include: the current time CurrentTime, the recommended departure time AiTOT of the outbound flight i (Continuous variable), scheduled delay time, whether the outbound flight is recommended to take off at a certain hour DIH it (0-1 variable), whether the incoming flight is recommended to land at AIH at a certain hour jt IsNormal (0-1 variable), whether the outbound flight is released normally i (0-1 variable), whether the outbound flight is flow controlled, whether the outbound flight is canceled IsCancelled i (0-1 variable), the variable RunwaySeqInd for the order of queues on the runway ij (0-1 variables) etc.

[0092] The above-mentioned debugging strategies may include: backlog reduction within a certain period of time (for example, 7 hours), flight pre-adjustment within a certain period of time (for example, 24 hours), etc.

[0093] Step S12: Convert the time adjustment business rules into constraints of the model.

[0094] In some embodiments, as shown in FIG9 and Table 4, the aforementioned constraints may include hourly capacity constraints, time interval constraints (also referred to as takeoff interval constraints), normal dispatch determination constraints (also referred to as normal dispatch constraints), flow control parking constraints, and the like. For specific implementations, see the constraint-related formulas shown in Table 4, which are generated based on the aforementioned decision variables. It is understood that in actual applications, the MIP model may have more or fewer constraints than those shown in FIG9 and Table 4, and this embodiment of the present application does not impose any limitations thereon.

[0095] Step S13: Write the timing target and priority rules into the objective function of the model, and express multiple targets into a single target in a weighted manner.

[0096] As shown in Figure 9, the objective function of the MIP model in the embodiment of the present application may include: time adjustment penalty cost, ground agent adjustment ratio, difference cost, flight delay cost, flight cancellation cost, etc. In some embodiments, as shown in Table 4, the above objective function may specifically include: maximizing the normal release rate (i.e., minimizing the delay ratio), minimizing the number of uncontrolled flight delays, minimizing passenger delay time, minimizing flight delay time, minimizing flight cancellations, and minimizing the weighted target value.

[0097] Through the above processing, a multi-objective optimization MIP model under different rule constraints can be constructed.

[0098] Based on the pre-built MIP model, in this step, real-time data related to the decision variables of each flight within the target time can be input into the MIP model, and the MIP model can output the recommended take-off time and / or recommended push-back time of each outbound flight.

[0099] Table 4 below shows a specific implementation of the decision variables, key constraints, and objective function in the MIP model.

[0100] Table 4

[0101] Among them, i represents outbound flights, j represents inbound flights, t represents hours, p represents different flight priorities (such as the priority of closed doors is higher than that of open doors), STD i Indicates the planned departure time of outbound flight i, EXOT i Indicates the estimated departure taxi time of outbound flight i, EOBT i Indicates the estimated off-block time of outbound flight i (corresponding to the earliest available time), CurrentTime indicates the current time, AiTOT i Indicates the recommended departure time of outbound flight i, DIH itIndicates whether outbound flight i is recommended to take off at hour t, AIH jt Indicates whether the incoming flight is recommended to land at hour t, GapMinutes r Indicates the time interval between flights, AiTOT j Indicates the recommended landing time of inbound flight j, RunwaySeqInd ij Indicates the queue order variable of flights on the runway (including incoming flight j and outgoing flight i), IsCancelled i Indicates whether the outbound flight i is canceled, IsCancelled j Indicates whether the incoming flight j is canceled, IsNormal i Indicates whether the departure flight i is released normally, NormalTOT i Indicates the normal release time of outbound flight i, Flights p Represents a set of flights with priority p, DepartureFlights is a set of departure flights, ArrivalFlights is a set of arrival flights, ImpactedFlights is a set of controlled flights, LockedFlights is a set of locked flights, UnlockedFlights is a set of unlocked flights, and Capacity t is the total capacity of inbound and outbound ports at hour t, CapacityOut t is the outbound capacity at hour t, NormalTOT i is the normal release time of flight i, PassengersNum i is the number of passengers on outbound flight i (approximate number of available seats), DelayPenalty p is the penalty coefficient for different flight priorities.

[0102] This step can comprehensively consider the interests of multiple parties such as airports, ground handling, airlines, and passengers, integrate air and ground resource constraints, establish an MIP model, and solve flight scheduling problems in different scenarios such as pre-scheduled and dynamic scenarios. This model can be used to provide operational suggestions for different scenarios. In some embodiments, statistics show that in the flight delay scenario, the normal release rate of all flights throughout the day can be increased by 3%-5%, and the average delay time per flight can be shortened by more than 5%.

[0103] In the method shown in FIG8 , the MIP model is used to determine the recommended take-off time and / or recommended push-back time of each outbound flight based on the estimated arrival time of each inbound flight, the estimated arrival taxi time of each inbound flight, the earliest push-back time of each outbound flight, and the estimated departure taxi time of each outbound flight. This enables a relatively more reasonable scheduling of the recommended take-off time and / or recommended push-back time of the outbound flights, and thus enables a relatively more reasonable scheduling of the operating times of airport flights.

[0104] FIG10 is a flow chart of a flight operation scheduling method according to an embodiment of the present application. As shown in FIG10 , the method may include:

[0105] Step 1001: Predict the estimated arrival time of an incoming flight.

[0106] In some embodiments, the estimated arrival time of the incoming flight may be predicted based on the real-time location information of the incoming flight and weather data and other information.

[0107] In some embodiments, the estimated arrival time of the incoming flight may be predicted based on information such as the real-time location information of the incoming flight, weather data, and equipment information of the incoming flight.

[0108] Optionally, the real-time location information of the incoming flight may include: the latitude and longitude, altitude, flight direction, etc. of the incoming flight.

[0109] Optionally, the inbound flight weather data may include weather information for each waypoint between the inbound flight's location and the target location, as well as weather information for the target location. Optionally, the waypoint weather information may specifically be the weather level at the waypoint. Optionally, the target location weather information may specifically be the weather level at the target location. The above method for classifying weather levels is not limited in this embodiment of the present application.

[0110] Optionally, the equipment information of the incoming flight may include: the model of the incoming flight, etc.

[0111] Taking the example of predicting the estimated arrival time of an inbound flight based on the real-time location information and weather data of the inbound flight, in some embodiments, a prediction model for the remaining flight time can be preset. As shown in FIG11 , the input of the prediction model may include the real-time location information and weather data of the inbound flight, etc., and the output may include: the remaining flight time of the inbound flight.

[0112] During model training, the historical real-time location information of incoming flights, historical weather data, and the time difference between the recording time of the historical real-time location information and the actual landing time of the flight (i.e., the remaining flight time) can be used to generate training samples, and the remaining flight time prediction model of incoming flights can be trained to obtain a trained remaining flight time prediction model. The above-mentioned historical weather data may include: historical weather information (such as historical weather levels) for each waypoint between the historical real-time location and the target location. The specific training method of the remaining flight time prediction model can be implemented using the model training method in the relevant technology, which will not be repeated here.

[0113] Then, this step may include:

[0114] Input the real-time location information and weather data of incoming flights into the trained remaining flight time prediction model to obtain the remaining flight time of incoming flights;

[0115] Determine the estimated time of arrival (ETA) of the incoming flight based on the remaining flight time of the incoming flight.

[0116] Specifically, the current time may be added to the remaining flight time of the incoming flight to obtain the estimated arrival time of the incoming flight. The above-mentioned current time may be the real time when this step is executed.

[0117] It is understood that if the estimated arrival time of an incoming flight is predicted based on the incoming flight's real-time location information, weather data, and equipment information, then when training the remaining flight time prediction model, the remaining flight time prediction model can be trained using the incoming flight's historical real-time location information, historical weather data, equipment information, and the time difference between the recorded time of the historical real-time location information and the flight's actual landing time (i.e., the remaining flight time) to generate training samples. The remaining flight time prediction model for the incoming flight can then be trained to obtain a trained remaining flight time prediction model. Accordingly, in this step, the incoming flight's real-time location information, weather data, and equipment information can be used as input data into the remaining flight time prediction model to obtain the remaining flight time of the incoming flight as output by the model, thereby determining the expected arrival time of the incoming flight.

[0118] Predicting the estimated arrival time of inbound flights based on the above method can improve the accuracy of the predicted estimated arrival time of inbound flights. In some embodiments, according to statistics, the proportion of flights with a prediction result deviation of less than 3 minutes when predicted 30 minutes before the actual landing of the inbound flight exceeds 90%.

[0119] Step 1002: Predict the earliest available launch time of the outbound flight.

[0120] In some embodiments, dynamic node data of outbound flights can be obtained, and the earliest available launch time of the outbound flights can be predicted based on the dynamic node data. The dynamic node data may include real-time status information of the outbound flights at various flight support nodes.

[0121] Specifically, flight support nodes can be determined based on each work node in the flight support process. Then, based on the relationships between the various flight support nodes, a flight support node network can be pre-constructed, as shown in Figure 12. Referring to Figure 12, flight support nodes may include: flight arrival, wheel block, refueling start, refueling completion, wheel block removal, takeoff, cabin door opening, boarding start, boarding completion, cabin door closing, cargo door opening, cargo door closing, etc. Among these, the three branches after wheel block and before wheel block removal, refueling start and refueling completion, cabin door opening, boarding start, boarding completion, cabin door closing, and cargo door opening and closing can be executed in parallel. It is understood that the above flight support node configuration is merely an example, and the specific configuration of flight support nodes may differ from the above example, and this embodiment of the application is not limiting.

[0122] Based on the above flight support node network, a remaining support time prediction model can be preset. The input of the remaining support time prediction model may include: the real-time status information of the outbound flight at each flight support node, and the time interval between the target departure time of the flight and the current time. The output of the remaining support time prediction model may be the time interval between the earliest possible launch time of the outbound flight and the current time, which is also referred to as the remaining support time of the outbound flight in the embodiment of the present application. The above-mentioned current time may be the execution time of this step or the recording time of the real-time status information of each flight support node. Accordingly, based on the output remaining support time of the remaining support time prediction model, combined with the above-mentioned current time, the earliest possible launch time of the outbound flight can be determined.

[0123] In some embodiments, the outbound flight's off-block time or door closing time can be used as the earliest pushback time for the outbound flight. The earliest pushback time for the outbound flight refers to the earliest time when the outbound flight completes flight support and can be pushed back to the runway for takeoff.

[0124] Specifically, when inputting the real-time status information of the outbound flight at each flight support node into the remaining support time prediction model, the real-time status information of the flight support node whose real-time status has occurred or is occurring can be recorded as 1, and the real-time status information of the flight support node whose real-time status has not occurred can be recorded as 0. In this way, a character string consisting of the real-time status information of the outbound flight at each flight support node can be obtained as the real-time status information of the outbound flight at each flight support node, and input into the remaining support time prediction model. For example, the real-time status information of the outbound flight at each flight support node is recorded in the order of flight arrival, wheel block, start of refueling, end of refueling, cabin door opening, start of boarding, end of boarding, cabin door closing, cargo door opening, cargo door closing, wheel block removal, and takeoff. Assuming that the five flight support nodes of the outbound flight, namely, flight arrival, wheel block, start of refueling, end of refueling, and cabin door opening, have occurred, and the other flight support nodes have not yet occurred, a character string consisting of the real-time status information of the outbound flight at each flight support node, 111110000000, can be obtained. It is understandable that the order of each flight support node in the character string can be adjusted, and this embodiment of the present application does not limit this.

[0125] When training the remaining support time prediction model, the historical real-time status information of the outbound flights, the time interval between the target take-off time and the historical real-time time (for example, the recording time of the historical real-time status information), and the time interval between the actual push-back time of the outbound flights (for example, the actual cabin door closing time or the actual block withdrawal time, etc.) and the historical real-time time can be used to generate training samples to train the remaining support time prediction model and obtain a trained remaining support time prediction model. For example, for an outbound flight, if it is in the refueling stage, the wheel blocking, passenger door opening, and cargo door opening have occurred, and the cargo door closing, boarding start, and boarding end flight support nodes have not occurred, the historical real-time status information of the outbound flight at each flight support node can be recorded as: 111010001000. Assuming that the time interval between the target take-off time of the outbound flight and the time of refueling start (corresponding to the above historical real-time time) is t1, and the time interval between the actual passenger door closing time and the time of refueling start (corresponding to the above historical real-time time) is t2, then the above time intervals t1, t2 and the real-time status information string 111010001000 can be used as training samples to train the remaining support time prediction model. The specific training method of the remaining support time prediction model can be implemented using the model training method in the relevant technology, which will not be repeated here.

[0126] Predicting the earliest pushback time of an outbound flight based on the above method can improve the accuracy of the prediction of the earliest pushback time of an outbound flight. In some embodiments, according to statistics, the proportion of flights with a deviation of less than 5 minutes when the prediction result is predicted 20 minutes before the actual door closing of the outbound flight can reach 80%.

[0127] Step 1003: Predict the estimated taxi time (VTT-IN / EXIT) of the incoming flight and the estimated taxi time (VTT-OUT / EXOT) of the outgoing flight based on the estimated arrival time of the incoming flight and the earliest available taxi time of the outgoing flight.

[0128] In some embodiments, a taxi time prediction model may be preset. The input of the preset taxi time prediction model may be the estimated arrival time of the incoming flight and the earliest push-out time of the outgoing flight. The taxi time prediction model may predict and output the estimated arrival taxi time of the incoming flight and the estimated departure taxi time of the outgoing flight based on the estimated arrival time of the incoming flight and the earliest push-out time of the outgoing flight. In other words, the output of the taxi time prediction model may be the estimated arrival taxi time of the incoming flight and the estimated departure taxi time of the outgoing flight.

[0129] Optionally, referring to FIG14 , the specific working principle of the glide time prediction model is exemplarily described as follows:

[0130] The historical taxi paths and runway head information of each flight (including inbound and outbound flights) are used as input to the initial taxi time prediction model. The initial taxi time prediction model can be trained by the following steps:

[0131] Step S21: Generate a taxiing path network diagram.

[0132] The initial taxi time prediction model generates a directed graph based on each flight's historical taxi paths and stand runway threshold information. This graph, known as a taxi path network, includes points and edges from the runway threshold to the stand (for inbound flights) and from the stand to the runway threshold (for outbound flights). Points can include key locations in the taxi path, such as runway thresholds and stand gates, while edges represent the taxi paths between these points.

[0133] Step S22: Clustering the aircraft stands and runway ends according to the taxiing path network diagram to obtain runway and aircraft stand groups where taxiing conflicts may occur.

[0134] A runway stand group may include stands and runway ends where taxiing conflicts may occur. The runway stand group may also be referred to as a set of points where taxiing conflicts may occur.

[0135] Specifically, in this step, it is possible to determine whether a node is shared based on the paths passed by different aircraft stands and runway heads, identify whether there is route overlap between taxi paths of different flights, and obtain runway stand groupings that may cause taxiing conflicts.

[0136] Step S23: Estimate the obstacle-free taxiing time of each flight (including inbound and outbound flights) and the congestion coefficient of each flight (including inbound and outbound flights).

[0137] The above steps can complete the training of the taxiing time prediction model.

[0138] Afterwards, in this step, the estimated arrival time of each incoming flight and the earliest available departure time of the outgoing flight can be input into the taxiing time prediction model. The taxiing time prediction model can then perform the following steps to output the estimated arrival taxiing time of the incoming flight and the estimated departure taxiing time of the outgoing flight:

[0139] Step S24: Calculate the actual taxiing time of each flight based on the congestion coefficient and obstacle-free taxiing time of each flight.

[0140] Step S25: Determine the arrival time of each flight at the runway end based on the estimated arrival time of each inbound flight and the earliest available push-out time of each outbound flight. Build a queue based on the runway operation mode and the arrival time of the flight at the runway end, and estimate the queue time of each flight at the runway end.

[0141] Step S26: Determine the taxiing time of each flight according to the actual taxiing time of each flight and the queuing time at the runway head, that is, the estimated arrival taxiing time of the incoming flight and the estimated departure taxiing time of the outgoing flight.

[0142] Predicting the estimated arrival taxi time of inbound flights and the estimated departure taxi time of outbound flights based on the above graph theory method can improve the prediction accuracy of the estimated arrival taxi time of inbound flights and the estimated departure taxi time of outbound flights. In some embodiments, according to statistics, the proportion of flights with a deviation of less than 3 minutes in the prediction results 20 minutes before the actual taxiing of the flight is increased by more than 20% compared with the system.

[0143] Step 1004: Determine a suggested take-off time and / or suggested pushback time for each outbound flight based on the estimated arrival time of each inbound flight, the estimated arrival taxi time of each inbound flight, the earliest pushback time of each outbound flight, and the estimated departure taxi time of each outbound flight.

[0144] Please refer to step 802 for the execution of this step, which will not be described here in detail.

[0145] The method shown in Figure 10 further provides a prediction method for the estimated arrival time of each inbound flight, the estimated arrival taxi time of each inbound flight, the earliest pushback time of each outbound flight, and the estimated departure taxi time of each outbound flight, thereby improving the prediction accuracy of the above times, and furthermore, determining the recommended departure time and / or recommended pushback time of each outbound flight based on the above times more accurately, thereby enabling the embodiment of the present application to perform relatively more reasonable scheduling of the recommended departure time and / or recommended pushback time of outbound flights, that is, to perform relatively more reasonable scheduling of the operating times of airport flights.

[0146] FIG15 is a schematic diagram of the structure of a flight operation scheduling device provided in an embodiment of the present application. As shown in FIG15 , the device 1500 may include:

[0147] An acquisition unit 1501 is configured to obtain an estimated arrival time and an estimated arrival taxi time of each inbound flight, and obtain an earliest available pushback time and an estimated departure taxi time of each outbound flight;

[0148] The determining unit 1502 is configured to determine a recommended take-off time and / or a recommended pushback time for each outbound flight based on the estimated arrival time of each inbound flight, the estimated arrival taxi time of each inbound flight, the earliest pushback time of each outbound flight, and the estimated departure taxi time of each outbound flight.

[0149] It should be noted that all relevant contents of each step involved in the above method embodiment can be referred to the functional description of the corresponding functional module and will not be repeated here.

[0150] The device provided in this embodiment is used to execute the above-mentioned thread leakage detection method, and thus can achieve the same effect as the above-mentioned method.

[0151] An embodiment of the present application also provides an electronic device, including a processor and a memory, wherein the processor is used to execute the method provided in the embodiment of the present application.

[0152] An embodiment of the present application further provides a computer-readable storage medium, which stores a computer program. When the computer-readable storage medium is run on a computer, the computer executes the method provided by the embodiment of the present application.

[0153] An embodiment of the present application also provides a computer program product, which includes a computer program. When the computer program product is run on a computer, it enables the computer to execute the method provided by the embodiment of the present application.

[0154] In the embodiments of the present application, "at least one" refers to one or more, and "more" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent the existence of A alone, the existence of A and B at the same time, and the existence of B alone. Among them, A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b and c can represent: a, b, c, a and b, a and c, b and c or a and b and c, where a, b, c can be single or multiple.

[0155] Those skilled in the art will appreciate that the various units and algorithm steps described in the embodiments disclosed herein can be implemented using a combination of electronic hardware, computer software, and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0156] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0157] In the several embodiments provided in this application, if any function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of this application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory; hereinafter referred to as: ROM), random access memory (Random Access Memory; hereinafter referred to as: RAM), magnetic disk or optical disk, and other media that can store program code.

[0158] The above description is merely a specific embodiment of the present application. Any person skilled in the art may easily conceive of variations or substitutions within the technical scope disclosed in this application, and such variations or substitutions shall be within the scope of protection of this application. The scope of protection of this application shall be subject to the scope of protection of the claims.

Claims

1. A flight operation scheduling method, characterized in that: include: Get the estimated arrival time and estimated taxi time of each incoming flight, and get the earliest pushback time and estimated taxi time of each outgoing flight; A suggested take-off time and / or a suggested pushback time for each outbound flight is determined based on the estimated arrival time of each inbound flight, the estimated arrival taxi time of each inbound flight, the earliest pushback time of each outbound flight, and the estimated departure taxi time of each outbound flight.

2. The method according to claim 1, characterized in that Determining the recommended take-off time and / or recommended pushback time of each outbound flight based on the estimated arrival time of each inbound flight, the taxi time of each inbound flight, the earliest pushback time of each outbound flight, and the taxi time of each outbound flight includes: The estimated arrival time of each inbound flight, the estimated arrival taxi time of each inbound flight, the earliest pushback time of each outbound flight and the estimated departure taxi time of each outbound flight are input into a preset MIP model to obtain a recommended take-off time and / or recommended pushback time of each outbound flight.

3. The method according to claim 1 or 2, characterized in that The method of obtaining the estimated arrival time and taxi time of each incoming flight, and obtaining the earliest available push-out time and taxi time of each outgoing flight, includes: Get the estimated arrival time of each incoming flight and the earliest available departure time of each outgoing flight; According to the estimated arrival time of each incoming flight and the earliest pushback time of each outgoing flight, an estimated arrival taxiing time of each incoming flight and an estimated departure taxiing time of each outgoing flight are determined.

4. The method according to claim 3, characterized in that Get the estimated arrival time for each incoming flight, including: For each incoming flight, obtaining real-time location information of the incoming flight and weather information associated with the incoming flight; Inputting the real-time location information of the inbound flight and the weather information into a preset remaining flight time prediction model, and using the time output by the remaining flight time prediction model as the remaining flight time of the inbound flight; An estimated arrival time of the inbound flight is determined based on the remaining flight time of the inbound flight.

5. The method according to claim 4, characterized in that Also includes: Obtaining equipment information of the incoming flight; The step of inputting the real-time location information of the incoming flight and the weather information into a preset remaining flight time prediction model includes: The real-time location information of the incoming flight, the weather information, and the equipment information are input into a preset remaining flight time prediction model.

6. The method according to claim 1 or 2, characterized in that The earliest available launch time for each outbound flight is obtained, including: For each outbound flight, obtain the real-time status information of the outbound flight and the time interval between the target departure time of the outbound flight and the real-time time; the real-time status information of the outbound flight is used to record the real-time status of the outbound flight at each flight support node; the flight support node is a node in the support process from arrival to departure of the flight; Inputting the real-time status information of the outbound flight and the time interval into a preset remaining guarantee time prediction model, and obtaining the time output by the remaining guarantee time prediction model as the remaining guarantee time of the outbound flight; The earliest possible launch time of the outbound flight is determined based on the remaining guaranteed time.

7. The method according to claim 3, characterized in that Determining the estimated arrival taxi time of each inbound flight and the estimated departure taxi time of each outbound flight based on the estimated arrival time of each inbound flight and the earliest pushback time of each outbound flight includes: The estimated arrival time of each incoming flight and the earliest available departure time of each outgoing flight are input into a preset taxiing time prediction model to obtain the estimated arrival taxiing time of each incoming flight and the estimated departure taxiing time of each outgoing flight output by the preset taxiing time prediction model.

8. The method according to claim 7, characterized in that The training method of the glide time prediction model includes: Generate a taxi path network diagram based on the historical taxi paths and runway head information of the inbound flights and the outbound flights; Clustering the aircraft stands and runway ends according to the taxiway path network diagram to obtain runway and aircraft stand groups where taxiing conflicts may occur; Estimate the clear taxi time and congestion factor for each incoming flight and each outgoing flight.

9. A flight operation scheduling device, characterized in that: include: An acquisition unit, configured to obtain the estimated arrival time and estimated arrival taxi time of each incoming flight, and obtain the earliest pushback time and estimated departure taxi time of each outgoing flight; a determining unit, configured to determine a recommended take-off time and / or a recommended pushback time for each outbound flight based on the estimated arrival time of each inbound flight, the estimated arrival taxi time of each inbound flight, the earliest pushback time of each outbound flight, and the estimated departure taxi time of each outbound flight.

10. An electronic device, characterized in that: include: processor, memory; One or more computer programs are stored in the memory, and the one or more computer programs include instructions, which, when executed by the processor, enable the electronic device to perform the method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which, when executed on a computer, enables the computer to execute the method according to any one of claims 1 to 8.

Images