Flight adjustment method, flight adjustment system and electronic device

Abstract

This disclosure relates to a flight adjustment method, a flight adjustment system, and electronic equipment. The method includes: receiving information; extracting flight adjustment intentions and constraints based on the information via a first intelligent agent, and generating task instructions based on the flight adjustment intentions and constraints; determining a business scenario via the first intelligent agent based on the task instructions; performing operations: in response to a first business scenario, obtaining flight plan data from a flight plan database via a second intelligent agent based on the task instructions and determining an adjustment scheme based on the task instructions, the flight plan data, and rules for flight adjustment; or in response to a second business scenario, obtaining flight plan data via the first intelligent agent based on the task instructions and processing the flight plan data to obtain neighborhood data based on the task instructions, and determining an adjustment scheme via a third and fourth intelligent agent based on the task instructions, the neighborhood data, and rules for flight adjustment; and outputting the adjustment scheme.

Description

Technical Field

[0001] This disclosure relates to the field of civil aviation technology, and more specifically, to a flight adjustment method, a flight adjustment system, and related electronic equipment, computer-readable storage media, and computer program products. Background Technology

[0002] With the rapid development of the air transport industry, flight operation management has become a core part of airline production organization, and the adjustment of flight schedules (i.e., flight adjustment) is a high-frequency, dynamic, and complex systemic project. Affected by multiple factors such as market demand fluctuations, capacity resource optimization, seasonal time slot changes, and emergencies (such as severe weather, air traffic control, aircraft malfunctions, or crew overtime), airlines often need to adjust their flight schedules. Summary of the Invention

[0003] A brief overview of this disclosure is given below to provide a basic understanding of some aspects of it. However, it should be understood that this overview is not an exhaustive summary of this disclosure. It is not intended to identify key or essential parts of this disclosure, nor is it intended to limit the scope of this disclosure. Its purpose is merely to present certain concepts of this disclosure in a simplified form as a prelude to the more detailed description that follows.

[0004] According to a first aspect of this disclosure, a flight adjustment method is provided, comprising: receiving information related to flight adjustment input by a user; extracting flight adjustment intentions and flight adjustment constraints based on the information via a first intelligent agent, and generating a task instruction based on the flight adjustment intentions and flight adjustment constraints, wherein the flight adjustment intentions indicate a business scenario for flight adjustment, and the flight adjustment constraints indicate a target time range and a target airport; determining the business scenario for flight adjustment via the first intelligent agent based on the task instruction; and performing the following operations: in response to the determined business scenario being a first scenario, obtaining flight plan data from a flight plan database via a second intelligent agent based on the task instruction, and adjusting the flight plan data based on the task instruction, the flight plan data, and the information related to flight adjustment. The adjustment rules determine the adjustment scheme, or in response to the determined business scenario (the second scenario), the first intelligent agent obtains flight plan data from the flight plan database based on task instructions and processes the flight plan data to obtain neighborhood data based on task instructions. The third and fourth intelligent agents determine the adjustment scheme based on task instructions, neighborhood data, and rules for flight adjustment. The fourth intelligent agent is configured to generate candidate adjustment schemes based on task instructions, neighborhood data, and rules for flight adjustment and provide the candidate adjustment schemes to the third intelligent agent. The third intelligent agent is configured to filter the candidate adjustment schemes based on task instructions to determine the adjustment scheme; and output the determined adjustment scheme.

[0005] In some embodiments, flight planning data includes at least one of the following: flight entity data for multiple flights, aircraft type layout data, operating cost and revenue data, reservation data, predicted reservation data, execution rate threshold, VIP reservation data, flight execution rate data, operational approval data, flight special restriction data, paired flight data, mandatory connection data, minimum ground time (MGT), overbooking allowance, and MGT exemption rules.

[0006] In some embodiments, the flight adjustment constraint in the task instruction further indicates the target aircraft type. Determining the adjustment scheme by the second intelligent agent based on the task instruction, flight plan data, and rules for flight adjustment includes: obtaining neighborhood data by the second intelligent agent based on the flight plan data and identifying multiple flight pairs in the neighborhood data; determining, by the second intelligent agent based on the task instruction, that the flight adjustment in the first scenario includes aircraft type change of a single flight pair and determining the target aircraft type; adjusting the aircraft type of each of the multiple flight pairs to the target aircraft type by the second intelligent agent to obtain multiple candidate adjustment schemes; verifying each candidate adjustment scheme in the multiple candidate adjustment schemes based on the first type of rules and the basic verification rules in the rules for flight adjustment, and calculating the marginal contribution change of the verified candidate adjustment scheme, wherein the marginal contribution change indicates the difference between the marginal contribution of the candidate adjustment scheme after adjustment and the marginal contribution before adjustment; and requiring the output of the adjustment scheme by the second intelligent agent based on the task instruction, the marginal contribution change of the verified candidate adjustment scheme, and the marginal contribution change.

[0007] In some embodiments, the first type of rules includes at least one of the following: operation review verification rules, overbooking verification rules for sold reservations, overbooking verification rules for predicted reservations, MGT verification rules, flight special restriction verification rules, scheduled maintenance flight constraint verification rules, and mandatory connection constraint verification rules.

[0008] In some embodiments, determining an adjustment scheme via a second intelligent agent based on task instructions, flight schedule data, and rules for flight adjustment includes: obtaining neighborhood data via the second intelligent agent based on the flight schedule data, and identifying multiple flight pairs in the neighborhood data; determining via the second intelligent agent based on the task instructions that flight adjustment in the first scenario includes the cancellation of a single flight pair; canceling each of the multiple flight pairs via the second intelligent agent to obtain corresponding multiple candidate adjustment schemes; verifying each candidate adjustment scheme in the corresponding multiple candidate adjustment schemes based on the second type of rules and the basic verification rules in the rules for flight adjustment, and calculating the marginal contribution change of the verified candidate adjustment scheme, wherein the marginal contribution change indicates the difference between the marginal contribution of the candidate adjustment scheme after adjustment and the marginal contribution before adjustment; and requiring the output of the adjustment scheme via the second intelligent agent based on the task instructions, the marginal contribution change of the verified candidate adjustment scheme, and the marginal contribution change.

[0009] In some embodiments, the second type of rules includes at least one of the following: time slot execution rate verification rules, VIP reservation verification rules, scheduled maintenance flight cancellation restriction rules, and forced connection cancellation verification rules.

[0010] In some embodiments, obtaining neighborhood data based on flight plan data by the second agent includes at least one of the following: filtering flight plan data based on the target airport by the second agent to obtain flight plan data related to the target airport as neighborhood data; filtering flight plan data based on the target time range by the second agent to obtain flight plan data within the target time range as neighborhood data.

[0011] In some embodiments, obtaining neighborhood data based on flight schedule data via a second agent includes: directly using flight schedule data as neighborhood data via a second agent.

[0012] In some embodiments, in response to the determined business scenario being a second scenario, processing flight plan data by a first intelligent agent based on task instructions to obtain neighborhood data includes: performing the following operations by the first intelligent agent: filtering flight plan data based on a target time range to obtain first flight plan data within the target time range and determining a first flight with the earliest departure time and a second flight with the latest arrival time in the first flight plan data; determining a third flight forming a flight pair with the first flight and a fourth flight forming a flight pair with the second flight based on the flight plan data; in response to at least one of the third and fourth flights not being in the first flight plan data, using at least one of them and the first flight plan data as second flight plan data; advancing and delaying the earliest departure time and the latest arrival time of the flights in the second flight plan data by a specified amount of time as an extended time range, and obtaining neighborhood data based on the extended time range and the flight plan data.

[0013] In some embodiments, the fourth intelligent agent includes a flight intelligent agent and an aircraft intelligent agent. The fourth intelligent agent is configured to generate candidate adjustment schemes based on task instructions, neighborhood data, and rules for flight adjustment, including: performing a first operation via the flight intelligent agent, the first operation including: identifying at least one flight pair operated by each of a plurality of aircraft in the neighborhood data, the plurality of aircraft including a first aircraft and a second aircraft; transferring each flight pair operated by the first aircraft to the second aircraft to generate a corresponding first candidate adjustment scheme; verifying the corresponding first candidate adjustment scheme based on a third type of rule and a basic verification rule in the rules for flight adjustment; and providing the verified first candidate adjustment scheme as a second candidate adjustment scheme to the aircraft intelligent agent; performing a second operation via the aircraft intelligent agent, the second operation including: in response to receiving the second candidate adjustment scheme, verifying the second candidate adjustment scheme based on a fourth type of rule in the rules for flight adjustment; providing the verified second candidate adjustment scheme as a third candidate adjustment scheme; determining whether the third candidate adjustment scheme meets the flight adjustment constraints based on the task instructions; and in response to determining that the third candidate adjustment scheme meets the flight adjustment constraints, providing the third candidate adjustment scheme as a candidate adjustment scheme to the third intelligent agent.

[0014] In some embodiments, the third type of rules includes at least one of the following: operation review verification rules, overbooking verification rules for sold reservations, overbooking verification rules for predicted reservations, flight special restriction verification rules, and mandatory connection constraint verification rules; the fourth type of rules includes at least one of the following: MGT verification rules and scheduled maintenance flight constraint verification rules.

[0015] In some embodiments, the basic verification rules include: the candidate adjustment scheme does not include multiple flight pairs whose flight time periods at least partially overlap; and the departure airport of the outbound flight in the adjusted flight pair in the candidate adjustment scheme is the same as the arrival airport of the return flight in the adjacent preceding flight pair of the adjusted flight pair in the candidate adjustment scheme, and the arrival airport of the return flight in the adjusted flight pair is the same as the departure airport of the outbound flight in the adjacent following flight pair of the adjusted flight pair in the candidate adjustment scheme.

[0016] In some embodiments, the second operation further includes: in response to determining that the third candidate adjustment scheme does not meet the flight adjustment constraints, sending a prompt instruction to the flight agent, wherein the prompt instruction instructs the flight agent to repeat the first operation based on the third candidate adjustment scheme until the third candidate adjustment scheme meets the flight adjustment constraints or no new second candidate adjustment scheme is generated.

[0017] In some embodiments, the third agent is configured to screen candidate adjustment schemes based on task instructions to determine an adjustment scheme, including: receiving multiple candidate adjustment schemes from a fourth agent, and performing the following operations for each candidate adjustment scheme: determining the number of adjustment steps for the candidate adjustment scheme based on the number of adjusted flight pairs in the candidate adjustment scheme, and determining the step score of the candidate adjustment scheme based on the number of adjustment steps; determining whether the current operating aircraft of the adjusted flight pairs in the candidate adjustment scheme is consistent with the original operating aircraft, and determining the aircraft type score of the candidate adjustment scheme based on the determination result; determining the change in the marginal contribution of the candidate adjustment scheme based on the marginal contribution after adjustment and the marginal contribution before adjustment, and determining the marginal contribution score of the candidate adjustment scheme based on the change in the marginal contribution; determining user preferences based on the task instructions and determining the soft constraint score of the candidate adjustment scheme based on the user preferences; determining the comprehensive score of the candidate adjustment scheme based on the step score, aircraft type score, marginal contribution score, and soft constraint score; and screening based on the comprehensive score of each candidate adjustment scheme to determine an adjustment scheme.

[0018] According to a second aspect of this disclosure, a flight adjustment system is provided. The system includes a first intelligent agent, a second intelligent agent communicatively coupled to the first intelligent agent, a third intelligent agent communicatively coupled to the first intelligent agent, and a fourth intelligent agent communicatively coupled to the third intelligent agent. The first intelligent agent is configured to: receive flight adjustment-related information input by a user; extract flight adjustment intentions and constraints based on the information; generate task instructions based on the flight adjustment intentions and constraints; and transmit the task instructions to the second intelligent agent. The flight adjustment intentions indicate the business scenario of the flight adjustment, and the flight adjustment constraints indicate the target time range and the target airport. The second intelligent agent is configured to: receive the task instructions from the first intelligent agent and determine the flight adjustment business scenario based on the task instructions; and, in response to the determined business scenario being a first scenario, retrieve flight plan data from a flight plan database based on the task instructions. The system determines and outputs an adjustment scheme based on task instructions, flight plan data, and rules for flight adjustments. The first agent is further configured to: in response to the determined business scenario being a second scenario, retrieve flight plan data from the flight plan database based on task instructions, process the flight plan data to obtain neighborhood data based on the task instructions, and transmit the neighborhood data to the third and fourth agents. The third and fourth agents are configured to determine the adjustment scheme based on task instructions, neighborhood data, and rules for flight adjustments. Specifically, the fourth agent is configured to generate candidate adjustment schemes based on task instructions, neighborhood data, and rules for flight adjustments and provide these candidate schemes to the third agent. The third agent is configured to, in response to receiving candidate adjustment schemes, filter the candidate schemes based on task instructions to determine the adjustment scheme and output the determined adjustment scheme.

[0019] According to a third aspect of this disclosure, an electronic device is provided, comprising: a processor; and a memory storing computer-executable instructions, which, when executed by the processor, cause the processor to perform a flight adjustment method according to any embodiment of the first aspect of this disclosure.

[0020] According to a fourth aspect of this disclosure, a computer storage medium having computer-executable instructions stored thereon is provided, which, when executed by a processor, cause the processor to perform a flight adjustment method according to any embodiment of the first aspect of this disclosure.

[0021] According to a fifth aspect of this disclosure, a computer program product is provided, the computer program product including instructions that, when executed by a processor, implement the flight adjustment method according to any embodiment of the first aspect of this disclosure. Attached Figure Description

[0022] The foregoing and other features and advantages of this disclosure will become clear from the following description of embodiments illustrated in conjunction with the accompanying drawings. The drawings, incorporated herein and forming a part of the specification, are further used to explain the principles of this disclosure and to enable those skilled in the art to make and use it. Wherein: Figure 1 This is a flowchart illustrating a flight adjustment method according to some embodiments of the present disclosure; Figure 2 This is a schematic diagram illustrating a flight schedule under a single-flight aircraft type change scenario according to some embodiments of the present disclosure, wherein, Figure 2 (A) is a schematic diagram of the flight schedule before the change. Figure 2 (B) is a schematic diagram of the revised flight schedule; Figure 3 This is a schematic diagram illustrating a flight schedule under a single flight cancellation scenario according to some embodiments of the present disclosure, wherein, Figure 3 (A) is a schematic diagram of the flight schedule before cancellation. Figure 3 (B) is a schematic diagram of the flight schedule after cancellation; Figure 4 This is a schematic diagram illustrating flight schedules under capacity clearance scenarios according to some embodiments of the present disclosure, wherein, Figure 4 (A) is a schematic diagram of the flight schedule before the adjustment. Figure 4 (B) is a schematic diagram of the flight schedule after the first step of adjustment. Figure 4 (C) is a schematic diagram of the flight schedule after the second step of adjustment; Figure 5 This is a schematic block diagram illustrating a flight adjustment system according to some embodiments of the present disclosure; Figure 6 This is a schematic block diagram illustrating an electronic device according to some embodiments of the present disclosure; Figure 7 This is a schematic block diagram illustrating a computer system on which embodiments of the present disclosure may be implemented.

[0023] Note that in the embodiments described below, the same reference numerals are sometimes used across different figures to denote the same parts or parts with the same function, and repeated descriptions are omitted. In some cases, similar reference numerals and letters are used to denote similar items, so once an item is defined in one figure, it does not need to be discussed further in subsequent figures.

[0024] For ease of understanding, the positions, dimensions, and extents of the structures shown in the accompanying drawings and other materials may not represent actual positions, dimensions, and extents. Therefore, this disclosure is not limited to the positions, dimensions, and extents disclosed in the accompanying drawings and other materials. Detailed Implementation

[0025] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the present disclosure.

[0026] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the scope of this disclosure or its application or use. That is, the structures and methods herein are shown in an exemplary manner to illustrate different embodiments of the structures and methods in this disclosure. However, those skilled in the art will understand that they merely illustrate exemplary ways that can be used to implement this disclosure, and not exhaustive ways. Furthermore, the drawings are not necessarily drawn to scale, and some features may be enlarged to show details of specific components.

[0027] In addition, techniques, methods and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods and equipment should be considered part of the specification.

[0028] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0029] In related technologies, airlines primarily employ a manual decision-making model for flight adjustments in scenarios such as capacity shortages, flight cancellations, flight delays, and last-minute aircraft changes. Specifically, when flight adjustments are necessary, business personnel manually retrieve and integrate heterogeneous data from multiple sources, including flight plans, aircraft information, route revenue, and crew resources, to formulate adjustment plans based on their own professional experience.

[0030] As the demand for capacity in the civil aviation industry continues to increase, airline fleets are also expanding rapidly. This has led to a significant increase in the amount of data related to flight schedules and adjustments, and the number of feasible adjustment solutions grows exponentially with the scale of flights. While airlines' flight operation control systems, used for digital management, can provide data such as marginal contribution predictions for specific routes and aircraft types, operational staff still need to expend considerable effort processing this massive amount of information, making it difficult to determine adjustment solutions within a limited timeframe. Furthermore, operational staff can only manually search for adjustable flights based on experience, resulting in inefficient searches and a high risk of overlooking feasible solutions, leading to a low success rate for flight adjustments.

[0031] While airlines' flight operation systems possess basic data processing capabilities, flight adjustments involve complex constraints and spatiotemporal coupling. Adjustment plans generated by these systems cannot automatically integrate multiple constraints. Human decision-making is limited by cognitive computing power; even when an adjustment plan is determined, it's difficult to rely solely on experience for localized trial and error, making it impossible to select the optimal solution from a massive pool of feasible options within a reasonable timeframe. Furthermore, human decision-making makes scheduling highly dependent on the personal experience of operational personnel, which not only easily leads to decision-making errors but also hinders the improvement of overall operational efficiency.

[0032] To this end, this disclosure provides a flight adjustment method. After receiving flight adjustment-related information input by a user, the method processes the information through a corresponding agent in a multi-agent system to determine the type of business scenario desired by the user. Under the determined business scenario, the method calls the corresponding agent in the multi-agent system to handle flight adjustments for that type of business scenario to determine the adjustment plan. This enables targeted automated flight adjustments for different types of business scenarios, improves the efficiency of flight adjustments, the feasibility and practicality of adjustment plans, and enhances the airline's flight operation and maintenance capabilities.

[0033] The following will combine Figure 1 A flight adjustment method 100 according to various embodiments of the present disclosure is described in detail. It will be understood that the actual flight adjustment method 100 may include other steps, but in order to avoid obscuring the essential points of the present disclosure, these other steps will not be discussed herein and are not shown in the accompanying drawings.

[0034] like Figure 1 As shown, the flight adjustment method 100 may include steps S102 to S110.

[0035] At step S102, information related to flight adjustments input by the user is received. As a non-limiting example, the user can input in natural language "I want to cancel a flight on September 25th from 08:00 to 13:00 SHA (SHA is the International Air Transport Association (IATA) code for Shanghai Hongqiao International Airport)".

[0036] In step S104, the first intelligent agent (also known as the "monitoring intelligent agent") extracts the flight adjustment intention and flight adjustment constraints based on information, and generates a task instruction based on the flight adjustment intention and flight adjustment constraints. Here, the flight adjustment intention indicates the business scenario of flight adjustment, and the flight adjustment constraints indicate the target time range and the target airport.

[0037] Referring again to the aforementioned non-restrictive example, the first intelligent agent can extract the flight adjustment intention "cancel a flight," the target time range "September 25th 08:00-13:00," and the target airport "SHA" from the above information. The generated task instruction can be a JSON-formatted instruction including the extracted flight adjustment intention "cancel a flight," the flight adjustment constraint "September 25th 08:00-13:00," and the SHA.

[0038] In some embodiments, the first agent may deploy a large language model configured to perform semantic parsing on user input to identify flight adjustment intentions and constraints within the information. In some examples, the first agent may deploy multiple large language models, each with different prompts to process the user input in different ways to extract relevant content.

[0039] As a non-limiting implementation, the large language model can be Qwen3-8B, Qwen3-32B, Qwen-72B, or DeepSeek 671B, etc. The prompt words of the large language model can embed terminology definitions and output format constraints from the flight adjustment domain to output task instructions with a preset format. Furthermore, the first intelligent agent can select different large language models based on the amount of information to meet different efficiency and capability requirements.

[0040] Therefore, the flight adjustment method 100 disclosed herein can receive flight adjustment requests in natural language format and automatically complete intent parsing and structured instruction generation using a large language model. Business personnel do not need to master professional query languages ​​or be familiar with underlying data structures to initiate complex adjustment tasks using everyday expressions, reducing the operational threshold and communication costs for business personnel.

[0041] In step S106, the first intelligent agent determines the business scenario for flight adjustment based on the task instruction. The first intelligent agent can determine the business scenario for flight adjustment based on the flight adjustment intention in the task instruction or by further combining flight adjustment constraints.

[0042] Continuing with the aforementioned non-restrictive example, the first agent can determine the business scenario of flight adjustment as a "single flight cancellation scenario" based on the flight adjustment intent "cancel a flight" in the task instruction. Regarding the first and second scenarios described below, the first scenario can indicate a single flight cancellation scenario or a single flight aircraft type change scenario, while the second scenario can indicate a multi-flight coordinated adjustment scenario, or it can also be called a capacity freeing-up scenario. Capacity freeing-up can be understood as follows: when there are no available aircraft within the user's desired time frame, the user expects the original flights operated by those aircraft within that time frame to be transferred to other aircraft to free up aircraft capacity within that time frame, thus creating available capacity for other tasks that need to be performed during that period.

[0043] In step S108 ( Figure 1 At a location (not shown), an operation is performed, which includes steps S1081 and S1082.

[0044] In step S1081, in response to the determined business scenario being the first scenario, the second intelligent agent (also known as the "capacity coordination intelligent agent") obtains flight plan data from the flight plan database based on the task instructions and determines the adjustment plan based on the task instructions, flight plan data, and rules for flight adjustment.

[0045] In step S1082, in response to the determined business scenario being the second scenario, the first intelligent agent retrieves flight plan data from the flight plan database based on task instructions and processes the flight plan data to obtain neighborhood data. Then, the third intelligent agent (also referred to as the "scheduling intelligent agent") and the fourth intelligent agent determine an adjustment scheme based on the task instructions, the neighborhood data, and rules for flight adjustment. Here, the fourth intelligent agent is configured to generate candidate adjustment schemes based on the task instructions, the neighborhood data, and the rules for flight adjustment, and provide these candidate adjustment schemes to the third intelligent agent. The third intelligent agent is configured to filter the candidate adjustment schemes based on the task instructions to determine the appropriate adjustment scheme.

[0046] In step S110, the determined adjustment scheme is output.

[0047] Therefore, this disclosure classifies flight adjustment business scenarios into different business scenarios (i.e., the first business scenario and the second business scenario) based on a multi-agent structure including the first to fourth agents, according to the actual needs of users. Different agents are invoked for personalized processing of different categories of business scenarios. This allows single flight adjustments to be quickly processed by a lightweight rule engine (i.e., the second agent), while complex multi-flight coordination enables the first, third, and fourth agents to work together. This disclosure avoids the waste of resources by "making a mountain out of a molehill" and the timeout for solving large-scale problems while ensuring the accuracy of flight adjustments, thus adapting to the real-time decision-making needs of aviation operations.

[0048] In some examples, the flight schedule database can store flight schedule data for multiple days. The first and second agents can retrieve the full (i.e., 24-hour) flight schedule data for the current day from the flight schedule database based on the date within the target time range in the task instruction.

[0049] In some embodiments, flight planning data may include at least one of the following: flight entity data for multiple flights, aircraft layout data, operating cost and revenue data, reservation data, projected reservation data, execution rate threshold, VIP reservation data, flight execution rate data, operational approval data, flight special restriction data, paired flight data, mandatory connection data, MGT, overbooking allowance limit, and MGT exemption rules.

[0050] As a non-restrictive example, the flight main data (also referred to as "D1") indicates the main data of the flight schedule currently used by staff operating multiple flights; the aircraft layout data (also referred to as "D2") indicates the number of available seats (including the number of business / first / economy class layouts) for each aircraft type used to operate multiple flights; the operating cost and revenue data (also referred to as "D3") indicates the hourly operating cost of each aircraft type used to operate multiple flights and the current projected revenue for that flight; the reservation data (also referred to as "D4") indicates the number of currently sold reservations for each class of service on each flight; the projected reservation data (also referred to as "D5") indicates the projected number of reservations for each class of service on each flight; the execution rate threshold (also referred to as "D6") indicates the minimum acceptable execution rate threshold for each flight at the corresponding airport; the VIP reservation data (also referred to as "D7") indicates whether VIP passenger reservations exist for each class of service on each flight; and the flight execution rate data (also referred to as "D8") indicates the historical execution rate of each flight (e.g., the ratio of actual flights to total planned flights); The approved flight data (also known as "D9") indicates the set of available aircraft types for routes within a specified date range; the special flight restriction data (also known as "D10") indicates the constraint relationship between specific aircraft types and routes (e.g., the A327 cannot operate the route from Airport A to Airport B); the paired flight data (also known as "D11") indicates the flight pair relationship consisting of outbound and return flights. Therefore, two flights with a flight pair relationship can be called a flight pair, and any adjustment to any flight in a flight pair requires simultaneous adjustment to the other flights in the flight pair; the mandatory connection data (also known as "D12") indicates the preceding and following flight pairs marked as fixed connections (e.g., fixed configuration, connecting flights, round-trip relationships); the MGT (also known as "D13") indicates the minimum turnaround time requirements for each airport and each aircraft type combination; the overbooking limit (also known as "D14") indicates that the overbooking limit for business class / first class is 0, and the overbooking limit for economy class is a first preset number (e.g., 5); the MGT exemption rules (also known as "D15") indicate the list of exemption scenarios such as scheduled maintenance flights and flights with vacant slots.

[0051] In some embodiments, the rules for flight adjustments may include at least one of a first type of rule, a second type of rule, a third type of rule, a fourth type of rule, and a basic verification rule.

[0052] In some examples, basic verification rules can be used for single-flight aircraft type change scenarios, single-flight aircraft type cancellation scenarios, and capacity relocation scenarios. Basic verification rules may include: candidate adjustment schemes do not include multiple flight pairs whose operating time periods at least partially overlap; and the departure airport of the outbound flight in the adjusted flight pair in the candidate adjustment scheme is the same as the arrival airport of the return flight in the adjacent preceding flight pair of the adjusted flight pair in the candidate adjustment scheme, and the arrival airport of the return flight in the adjusted flight pair is the same as the departure airport of the outbound flight in the adjacent following flight pair of the adjusted flight pair in the candidate adjustment scheme.

[0053] In some examples, the first type of rule can be used for single-flight aircraft type change scenarios. The first type of rule may include at least one of the following: operation review verification rules, sold reservation overbooking verification rules, predicted reservation overbooking verification rules, MGT verification rules, flight special restriction verification rules, scheduled maintenance flight constraint verification rules, and mandatory connection constraint verification rules.

[0054] In some examples, the second type of rule can be used for single-flight aircraft type cancellation scenarios. The second type of rule can include at least one of the following: time slot fulfillment rate verification rule, VIP reservation verification rule, scheduled maintenance flight cancellation restriction rule, and forced connection cancellation verification rule.

[0055] In some examples, the third and fourth types of rules can be used in capacity relocation scenarios. The third type of rules may include at least one of the following: operation review verification rules, overbooking verification rules for sold reservations, overbooking verification rules for predicted reservations, flight special restriction verification rules, and mandatory connection constraint verification rules; the fourth type of rules may include at least one of the following: MGT verification rules and scheduled maintenance flight constraint verification rules.

[0056] As an example of non-restrictive explanation, the Operations Review Verification Rule (also known as "R1") indicates that the adjusted aircraft type must be among the available aircraft types for that route in the Operations Review data; the Overbooking Verification Rule (also known as "R2") indicates that the cabin layout of the adjusted aircraft type must cover the sold bookings; the Forecast Overbooking Verification Rule (also known as "R3") indicates that the cabin layout of the adjusted aircraft type must cover the forecast bookings; the MGT Verification Rule (also known as "R4") indicates that the connections of each flight after adjustment must meet the MGT requirements; the Flight Special Restriction Verification Rule (also known as "R5") indicates that the adjusted aircraft type must not violate the aircraft type-route constraint relationship defined in D10; the Scheduled Maintenance Flight Constraint Verification Rule (also known as "R6") indicates that if a flight is a scheduled maintenance flight (which can be obtained based on D1 parsing), then only its binding is allowed. The rules are as follows: Fixed aircraft; Mandatory connection constraint verification rule (also known as "R7") indicates that if flights are in a mandatory connection relationship such as fixed configuration, connecting flights, or round-trip (D12), then the connecting flights must simultaneously perform the same aircraft type modification operation, and all affected flights must pass R1 to R5; Time availability verification rule (also known as "R8") indicates that after a flight is cancelled, the flight's availability rate must still meet the regulatory threshold requirements; VIP reservation verification rule (also known as "R9") indicates that the flight containing the VIP cannot be cancelled; Scheduled maintenance flight cancellation restriction rule (also known as "R10") indicates that if a flight is a scheduled maintenance flight, then the flight cannot be cancelled; Mandatory connection cancellation verification rule (also known as "R11") indicates that if flights are in a mandatory connection relationship, then the connecting flights must be cancelled simultaneously, and all must pass R8 and R9. Therefore, the first type of rules can include R1 to R7, the second type of rules can include R8 to R11, the third type of rules can include R1, R2, R3, R5, and R7, and the fourth type of rules can include R4 and R6.

[0057] In some embodiments, the flight adjustment constraints in the task instruction may also indicate the target aircraft type. In this embodiment, determining the adjustment scheme by the second intelligent agent based on the task instruction, flight plan data, and rules for flight adjustment includes: obtaining neighborhood data by the second intelligent agent based on the flight plan data and identifying multiple flight pairs in the neighborhood data; determining by the second intelligent agent based on the task instruction that the flight adjustment in the first scenario includes aircraft type change of a single flight pair and determining the target aircraft type; adjusting the aircraft type of each of the multiple flight pairs to the target aircraft type by the second intelligent agent to obtain multiple candidate adjustment schemes; verifying each candidate adjustment scheme in the multiple candidate adjustment schemes based on the first type of rules and the basic verification rules in the rules for flight adjustment, and calculating the marginal contribution change of the verified candidate adjustment schemes; and outputting the adjustment scheme by the second intelligent agent based on the task instruction, the marginal contribution change of the verified candidate adjustment schemes, and the marginal contribution change.

[0058] In this disclosure, marginal contribution indicates the difference between a flight's revenue and cost, and the marginal contribution of a candidate adjustment scheme is determined based on the marginal contributions of each flight in that scheme. The change in marginal contribution indicates the difference between the adjusted marginal contribution and the original marginal contribution of the candidate adjustment scheme. Here, the adjusted marginal contribution of the candidate adjustment scheme is the sum of the marginal contributions of the adjusted flight (i.e., the two flights in the adjusted flight pair) and the marginal contributions of the other flights; the original marginal contribution of the candidate adjustment scheme is the sum of the original marginal contribution of the adjusted flight and the marginal contributions of the other flights. If the change in marginal contribution is negative, it indicates that the flight adjustment is unprofitable; if the change in marginal contribution is positive, it indicates that the flight adjustment is profitable.

[0059] In some embodiments, obtaining neighborhood data via a second intelligent agent based on flight plan data may include directly using the flight plan data as neighborhood data. In this embodiment, although the flight plan data is not filtered, the second intelligent agent can subsequently verify candidate adjustment schemes based on task instructions to select those that meet the flight adjustment constraints. Therefore, the process of filtering flight plan data can be omitted, improving adjustment efficiency and ensuring the completeness of candidate adjustment schemes, which is beneficial for obtaining the optimal solution.

[0060] In some embodiments, obtaining neighborhood data based on flight plan data via the second agent may include: filtering flight plan data based on the target airport to obtain flight plan data related to the target airport as neighborhood data; and / or filtering flight plan data based on a target time range to obtain first flight plan data within the target time range as neighborhood data. This reduces the subsequent solution space search range for the second agent, lowers computational load, and improves adjustment efficiency.

[0061] As a non-restrictive example, when a user inputs "There's an empty flight from Northwest 320 to Hongqiao at 08:15 tomorrow, how can I optimize it?", the first intelligent agent can transmit a JSON-formatted task instruction to the second intelligent agent. This instruction includes the extracted flight adjustment intent "single flight aircraft type change," the target time range in the flight adjustment constraints "May 27th 08:00-15:00," the target airport "Shanghai Hongqiao," and the target aircraft type "320." The first intelligent agent can then determine the current flight adjustment business scenario as the first scenario based on the flight adjustment intent "single flight aircraft type change" in the task instruction. It can be understood that the agent can confirm via the network that the date indicated "today" is May 26th, thereby determining that the date indicated "tomorrow" is May 27th.

[0062] Responding to the identified business scenario as the first scenario, the second intelligent agent loads structured data D1 to D15 from the flight plan database for May 27th, and filters related flight plan data as neighborhood data based on "Shanghai Hongqiao" in the task instruction, identifying 7 flight pairs in the neighborhood data. (Refer to reference...) Figure 2 This is a schematic diagram illustrating a flight schedule under a single-flight aircraft type change scenario according to some embodiments of the present disclosure, wherein... Figure 2 (A) is a schematic diagram of the flight schedule before the change. Figure 2 (B) is a schematic diagram of the revised flight schedule. For example... Figure 2 As shown in (A), the seven flight pairs include {P1-1, P1-2}, {P2-1, P2-2}, and {P3-1, P3-2} operated by aircraft of type 318 (which can be simply referred to as aircraft 318), {P4-1, P4-2}, and {P5-1, P52} operated by aircraft 319, and {P6-1, P6-2} and {P7-1, P7-2} operated by aircraft 320. The second agent also determines that the flight adjustment in the first scenario is a change in aircraft type for a single flight pair and identifies the target aircraft type "320".

[0063] Note that in this disclosure, regardless of the first or second scenario, the basic unit for flight adjustments is the flight pair. When scheduling flights, typically one aircraft is assigned to operate both the outbound and return flights. Therefore, adjusting only one of the outbound and return flights without adjusting the other would render the other unusable. Thus, using the flight pair as the basic adjustment unit improves the practicality and feasibility of the adjustment plan. For a single flight, since it cannot constitute a flight pair, flight adjustments in either the first or second scenario will not affect the individual flight.

[0064] After determining the target aircraft type, the second agent reassigns each of the seven flight pairs to an aircraft of type 320, resulting in seven candidate adjustment schemes. Each candidate adjustment scheme includes a distinct modified flight pair. Note that in this disclosure, regardless of whether it's the first or second scenario, when adjusting flight pairs, the flight time interval of the flight pair (i.e., from the departure time of the outbound flight to the arrival time of the return flight) must remain unchanged to avoid impacting the overall flight schedule and improve the practicality and feasibility of the adjustment scheme.

[0065] The second agent verifies the seven candidate adjustment schemes based on the first type of rules (i.e., R1 to R7) and the basic verification rules. For example, it verifies whether the flight pair assigned to aircraft 320 has overlapping flight time periods with the original flight pair or whether there is a time conflict, resulting in two candidate adjustment schemes A and B.

[0066] like Figure 2 As shown in (B), in candidate adjustment scheme A, the aircraft type for {P2-1, P2-2} is changed from 318 to 320. The flight time slots for {P2-1, P2-2} are "9:40, 15:00" (where P2-1 departs at 9:40 and P2-2 arrives at 15:00). Therefore, the flight time slots for {P2-1, P2-2} do not conflict with the original flight time slots of the 320 aircraft. Furthermore, P2-1 is an INC (Yinchuan Hedong International Airport) - SHA (Shanghai Hongqiao International Airport) flight (i.e., P2-1's departure airport is INC, and its destination airport is SHA). The return flight P6-2 of the preceding pair of flights {P6-1, P6-2} arrives at INC. P2-2 is an SHA-INC flight, and the departure flight P7-1 of the following pair of flights {P7-1, P7-2} departs from INC. Therefore, {P2-1, P2-2 can connect with adjacent preceding and following flight pairs (i.e., {P6-1, P6-2} and {P7-1, P7-2}).

[0067] In candidate adjustment plan B, the aircraft type for {P4-1, P4-2} is changed from A319 to A320, and the flight time slots for {P4-1, P4-2} are set to "8:40, 14:15", where P4-1 is INC-SHA and P4-2 is SHA-INC. Therefore, the flight time slots for {P4-1, P4-2} do not conflict with the original flight time slots of the A320 aircraft and can be connected with adjacent preceding and following flight pairs.

[0068] The second agent calculates the marginal contribution before and after adjustment, as well as the change in marginal contribution, for candidate adjustment schemes A and B, respectively. Here, the change in marginal contribution is required to be maximized. Alternatively, the change in marginal contribution can be a threshold; if the change in marginal contribution of a candidate adjustment scheme exceeds or does not exceed this threshold, the candidate adjustment scheme can be output as the adjustment scheme.

[0069] The second agent determines the candidate adjustment scheme A, which has the largest change in marginal contribution, based on the changes in marginal contribution of candidate adjustment schemes A and B, sorted in ascending order. It then verifies the adjusted flight pairs {P2-1, P2-2} within candidate adjustment scheme A to ensure they meet the flight adjustment constraints in the task instruction. Therefore, the output is as follows: Figure 2 The adjustment scheme shown in (B)

[0070] In some embodiments, determining an adjustment scheme via a second intelligent agent based on task instructions, flight plan data, and rules for flight adjustment includes: obtaining neighborhood data via the second intelligent agent based on the flight plan data, and identifying multiple flight pairs in the neighborhood data; determining via the second intelligent agent based on the task instructions that flight adjustment in the first scenario includes the cancellation of a single flight pair; canceling each of the multiple flight pairs via the second intelligent agent to obtain corresponding multiple candidate adjustment schemes; verifying each candidate adjustment scheme in the corresponding multiple candidate adjustment schemes based on the second type of rules and the basic verification rules in the rules for flight adjustment, and calculating the marginal contribution change of the verified candidate adjustment schemes; and outputting an adjustment scheme via the second intelligent agent based on the task instructions, the marginal contribution change of the verified candidate adjustment schemes, and the marginal contribution change.

[0071] As a non-restrictive example, when a user enters "I want to cancel a flight on September 25th from 08:00 to 13:00 SHA", the first intelligent agent can transmit a JSON-formatted message to the second intelligent agent, including the extracted flight adjustment intention "single flight cancellation", the target time range in the flight adjustment constraints "September 25th 08:00-13:00", and the target airport "SHA". The first intelligent agent can determine the current flight adjustment business scenario as the first scenario based on the flight adjustment intention "single flight cancellation" in the task instruction.

[0072] Responding to the identified business scenario as the first scenario, the second intelligent agent loads structured data D1 to D15 from the flight schedule database for September 25th as neighborhood data and identifies eight flight pairs within the neighborhood data. (Refer to reference...) Figure 3 This is a schematic diagram illustrating a flight schedule under a single flight cancellation scenario according to some embodiments of the present disclosure, wherein, Figure 3 (A) is a schematic diagram of the flight schedule before cancellation. Figure 3 (B) is a schematic diagram of the flight schedule after cancellation. For example... Figure 3 As shown in (A), the eight flight pairs include {P1-1, P1-2}, {P2-1, P2-2}, and {P3-1, P3-2} operated by the first aircraft of type 319 (which can be referred to as aircraft 319-1), {P4-1, P4-2}, {P5-1, P52}, and {P6-1, P6-2} operated by aircraft 319-2, and {P7-1, P7-2} and {P8-1, P8-2} operated by aircraft 319-3.

[0073] The second agent cancels each of the eight flight pairs, resulting in eight candidate adjustment schemes. Each candidate adjustment scheme has one distinct canceled flight pair. Next, the second agent verifies each of the eight candidate adjustment schemes based on the second type of rules (R8 to R11) and the basic verification rules, resulting in one candidate adjustment scheme. In this candidate adjustment scheme, {P8-1, P8-2} is canceled. The times for {P8-1, P8-2} are "8:10, 13:00". P8-1 is HIA (Huai'an Lianshui International Airport) - SHA, and P8-2 is SHA - HIA. The return flight P7-2 of the adjacent preceding flight pair {P7-1, P7-2} arrives at HIA. {P8-1, P8-2} has no adjacent following flight pair.

[0074] It is understandable that for an adjusted flight pair without an adjacent subsequent flight pair, it defaults to satisfying the basic verification rule that "the arrival airport of the return flight in the adjusted flight pair is the same as the departure airport of the outbound flight in the adjacent subsequent flight pair of the adjusted flight pair in the subsequent adjustment scheme"; for an adjusted flight pair without an adjacent preceding flight pair, it defaults to satisfying the basic verification rule that "the departure airport of the outbound flight in the adjusted flight pair in the candidate adjustment scheme is the same as the arrival airport of the return flight in the adjacent preceding flight pair of the adjusted flight pair in the candidate adjustment scheme".

[0075] The second agent calculates the adjusted marginal contribution and the original marginal contribution of the candidate adjustment scheme, as well as the change in marginal contribution, and determines that the change in marginal contribution of the candidate adjustment scheme meets the requirement for change in marginal contribution. Furthermore, the second agent verifies that the times of the cancelled flight pairs {P8-1, P8-2} in the adjustment scheme fall within the target time range, and that {P8-1, P8-2} are related to the target airport, thus satisfying the flight adjustment constraints. Therefore, the output is as follows: Figure 3 The adjustment scheme shown in (B)

[0076] In some embodiments, in response to the determined business scenario being a second scenario, processing flight plan data based on task instructions by a first intelligent agent to obtain neighborhood data includes: performing the following operations by the first intelligent agent: filtering flight plan data based on a target time range to obtain first flight plan data within the target time range and determining a first flight with the earliest departure time and a second flight with the latest arrival time in the first flight plan data; determining a third flight forming a flight pair with the first flight and a fourth flight forming a flight pair with the second flight based on the flight plan data; in response to at least one of the third and fourth flights not being in the first flight plan data, using the at least one flight and the first flight plan data as the second flight plan data; advancing and delaying the earliest departure time and the latest arrival time of the flights in the second flight plan data by a specified amount of time as an extended time range, and obtaining neighborhood data based on the extended time range and the flight plan data.

[0077] In some embodiments, under the first business scenario, the second intelligent agent obtaining neighborhood data based on flight plan data may further include processing the flight plan data based on task instructions to obtain neighborhood data. The "processing" of the second intelligent agent can refer to the aforementioned operations included in the first intelligent agent processing flight plan data based on task instructions to obtain neighborhood data, and will not be repeated here.

[0078] As a non-restrictive example, when a user inputs "September 26th, Shanghai Hongqiao, 08:00-13:30, need capacity, preferably wide-body aircraft.", the first intelligent agent can transmit a JSON-formatted task instruction to the second intelligent agent, including the extracted flight adjustment intention "capacity relocation", the target time range in the flight adjustment constraints "September 26th, 08:00-13:30", and the target airport "Shanghai Hongqiao". The first intelligent agent can determine the current flight adjustment business scenario as the second scenario based on the flight adjustment intention "capacity relocation" in the task instruction.

[0079] As an example of non-restrictive explanation, see reference. Figure 4 It is a schematic diagram illustrating flight schedules under capacity clearance scenarios according to some embodiments of the present disclosure, wherein, Figure 4 (A) is a schematic diagram of the flight schedule before the adjustment. For example... Figure 4As shown in (A), in response to the determined business scenario being the second scenario, the first agent filters flight plan data based on the target time range "08:00-13:30 (i.e., t1)" to obtain the first flight plan data including P2-1, P4-1, P4-2, and P6-2 within t1. It then determines the first flight P4-1 with the earliest departure time and the second flight P2-1 with the latest arrival time from the first flight plan data. Next, it determines the third flight P4-2 and the fourth flight P2-2, which form flight pairs with the first and second flights P4-1, respectively. Since the fourth flight P2-2 does not belong to the first flight plan data or does not fall within t1, P2-2 and the first flight plan data are combined as the second flight plan data. Therefore, the second flight plan data includes P2-1, P2-2, P4-1, P4-2, and P6-2. Next, based on the departure time of P4-1 at 8:20 (the earliest departure time) in the second flight plan, which is 3 hours earlier and the arrival time of P2-2 at 16:00 (the latest arrival time), which is 3 hours later, the extended time range t2 "5:20-19:00" is obtained. Finally, based on t2 and the flight plan data, the neighborhood data is obtained. At this time, the neighborhood data includes P2-1, P2-2, P4-1, P4-2, P6-1, and P6-2.

[0080] Therefore, in the second scenario, although the full flight plan data is reduced to the flight plan data within the target time range in the first step, the neighborhood data obtained by expanding the target time range in the subsequent step includes more data than the flight plan data within the target time range. This helps to obtain feasible solutions (i.e. candidate adjustment schemes), improves the success rate of solution on the basis of high solution efficiency, and thus improves the feasibility of flight adjustment.

[0081] In some embodiments, the fourth intelligent agent includes a flight intelligent agent and an aircraft intelligent agent. The fourth intelligent agent is configured to generate candidate adjustment schemes based on task instructions, neighborhood data, and rules for flight adjustment, including: performing a first operation via the flight intelligent agent, the first operation including: identifying at least one flight pair operated by each of a plurality of aircraft in the neighborhood data, the plurality of aircraft including a first aircraft and a second aircraft; transferring each flight pair operated by the first aircraft to the second aircraft to generate a corresponding first candidate adjustment scheme; verifying the corresponding first candidate adjustment scheme based on a third type of rule and a basic verification rule in the rules for flight adjustment; and providing the verified first candidate adjustment scheme as a second candidate adjustment scheme to the aircraft intelligent agent; performing a second operation via the aircraft intelligent agent, the second operation including: in response to receiving the second candidate adjustment scheme, verifying the second candidate adjustment scheme based on a fourth type of rule in the rules for flight adjustment; providing the verified second candidate adjustment scheme as a third candidate adjustment scheme; determining whether the third candidate adjustment scheme meets the flight adjustment constraints based on the task instructions; and in response to determining that the third candidate adjustment scheme meets the flight adjustment constraints, providing the third candidate adjustment scheme as a candidate adjustment scheme to the third intelligent agent.

[0082] In some embodiments, the second operation may further include: in response to determining that the third candidate adjustment scheme does not meet the flight adjustment constraints, sending a prompt instruction to the flight agent, wherein the prompt instruction instructs the flight agent to repeat the first operation based on the third candidate adjustment scheme until the third candidate adjustment scheme meets the flight adjustment constraints or no new second candidate adjustment scheme is generated.

[0083] Here, "based on the third candidate adjustment scheme" means that all subsequent newly generated first candidate adjustment schemes include the adjusted flight pairs from the third candidate adjustment scheme. Whenever the flight agent provides a new second candidate adjustment scheme to the aircraft agent, the aircraft agent automatically executes the steps in the second operation upon receiving the second candidate adjustment scheme. Furthermore, if the flight agent does not generate or provide a new second candidate adjustment scheme to the aircraft agent, it indicates that there are no other feasible solutions, and the user's desired flight adjustment cannot be achieved.

[0084] As an example of a non-restrictive statement, please refer to [reference needed]. Figure 4 ,in, Figure 4 (B) is a schematic diagram of the flight schedule after the first step of adjustment. Figure 4 (C) is a schematic diagram of the adjusted flight schedule after the second step. For example... Figure 4As shown in (A), the flight agent can identify the flight pairs {P1-1, P1-2} and {P2-1, P2-2} operated by aircraft 319-8, the flight pairs {P3-1, P3-2}, {P4-1, P4-2} and {P5-1, P5-2} operated by aircraft 319-9, and the flight pairs {P6-1, P6-2} and {P7-1, P7-2} operated by aircraft 320-1 in the neighborhood data.

[0085] In this adjustment, flights {P1-1, P1-2} and {P2-1, P2-2} operated by aircraft 319-8 (i.e., the first aircraft) were transferred to aircraft 319-9 (i.e., the second aircraft) to obtain two corresponding first candidate adjustment schemes. In one candidate adjustment scheme, {P1-1, P1-2} will be operated by the second aircraft, and in the other candidate adjustment scheme, {P2-1, P2-2} will be operated by the second aircraft. The rest are the same as those in the previous scheme. Figure 4 The flight schedule shown in (A) is the same. The flights {P1-1, P1-2} and {P2-1, P2-2} operated by aircraft 319-8 (i.e., the first aircraft) will be transferred to aircraft 320-1 (i.e., the second aircraft) to obtain two corresponding first candidate adjustment schemes.

[0086] Flights {P3-1, P3-2}, {P4-1, P4-2}, and {P5-1, P5-2} operated by aircraft 319-9 (i.e., the first aircraft) were transferred to aircraft 320-1 (i.e., the second aircraft) to obtain three corresponding first-candidate adjustment schemes. Flights {P3-1, P3-2}, {P4-1, P4-2}, and {P5-1, P5-2} operated by aircraft 319-9 (i.e., the first aircraft) were also transferred to aircraft 319-8 (i.e., the second aircraft) to obtain three corresponding first-candidate adjustment schemes.

[0087] The flights {P6-1, P6-2} and {P7-1, P7-2} operated by aircraft 320-1 (i.e., the first aircraft) were transferred to aircraft 319-8 (i.e., the second aircraft) to obtain two corresponding first candidate adjustment schemes. The flights {P6-1, P6-2} and {P7-1, P7-2} operated by aircraft 320-1 (i.e., the first aircraft) were also transferred to aircraft 319-9 (i.e., the second aircraft) to obtain two corresponding first candidate adjustment schemes.

[0088] It is understandable that the first and second aircraft are relative rather than fixed, used to distinguish between the aircraft that originally operated the adjusted flight pair and the aircraft currently operating the adjusted flight pair.

[0089] Next, the flight agent verifies the 14 candidate adjustment schemes based on the third type of rules (i.e., R1, R2, R3, R5, and R7) and the basic verification rules (the specific verification process can be found in the example under the first business scenario, and will not be repeated here), and determines the appropriate adjustment scheme. Figure 4 (B) The first candidate adjustment scheme C passed the verification. In the first candidate adjustment scheme C, {P2-1, P2-2}, originally operated by aircraft 319-8, was adjusted to be operated by aircraft 320-1. The flight agent sent the second candidate adjustment scheme (i.e., the first candidate adjustment scheme C) to the aircraft agent.

[0090] Upon receiving the second candidate adjustment plan, the aircraft agent verifies it based on the fourth set of rules (R4 and R6) and uses the verified second candidate adjustment plan as the third candidate adjustment plan. The aircraft agent then determines whether the third candidate adjustment plan meets the flight adjustment constraints based on the mission instructions. In the third candidate adjustment plan, although {P2-1, P2-2}, originally operated by aircraft 319-8, is changed to be operated by aircraft 320-1, the arrival time of P1-2 operated by aircraft 319-8 is 08:05, while the target time range is 08:00-13:30, therefore it does not meet the flight adjustment constraints. In response to determining that the third candidate adjustment plan does not meet the flight adjustment constraints, the flight agent sends a prompt instruction.

[0091] Upon receiving the prompt instruction, the flight agent repeats the first operation to generate a new flight schedule, such as {P2-1, P2-2}, which was originally operated by aircraft 319-8, but is now operated by aircraft 320-1. Figure 4 (C) shows the first candidate adjustment scheme D, and the first candidate adjustment scheme D is verified to obtain the second candidate adjustment scheme that passes the verification. Figure 4 As shown in (C), in the first candidate adjustment scheme D, the {P2-1, P2-2} originally operated by the 319-8 aircraft are adjusted to be operated by the 320-1 aircraft, and the {P4-1, P4-2} originally operated by the 319-9 aircraft are adjusted to be operated by the 319-8 aircraft.

[0092] Upon receiving the second candidate adjustment scheme, the aircraft agent verifies it based on the fourth type of rule and uses the verified second candidate adjustment scheme as the third candidate adjustment scheme. Based on the mission instruction, it determines that the third candidate adjustment scheme meets the flight adjustment constraints, thus obtaining the candidate adjustment scheme provided to the third agent (i.e., the first candidate adjustment scheme D). In the third candidate adjustment scheme, aircraft 319-9 has no flights scheduled between 08:00 and 13:30, therefore it has available capacity.

[0093] Therefore, this disclosure achieves collaborative verification of multi-dimensional hard constraints such as operation review rules, overbooking limits, MGT, scheduled inspection binding, and forced connection by distributing verification rules to flight agents and aircraft agents. Candidate adjustment schemes that do not meet any constraint are marked as rejected, reducing the probability of infeasible schemes entering the candidate set of the third agent and ensuring the completeness of hard constraints of the final output adjustment scheme.

[0094] In some embodiments, the third agent is configured to screen candidate adjustment schemes based on task instructions to determine an adjustment scheme, including: receiving multiple candidate adjustment schemes from a fourth agent, and performing the following operations for each candidate adjustment scheme: determining the number of adjustment steps for the candidate adjustment scheme based on the number of adjusted flight pairs in the candidate adjustment scheme, and determining the step score of the candidate adjustment scheme based on the number of adjustment steps; determining whether the current operating aircraft of the adjusted flight pairs in the candidate adjustment scheme is consistent with the original operating aircraft, and determining the aircraft type score of the candidate adjustment scheme based on the determination result; determining the change in the marginal contribution of the candidate adjustment scheme based on the marginal contribution after adjustment and the marginal contribution before adjustment, and determining the marginal contribution score of the candidate adjustment scheme based on the change in the marginal contribution; determining user preferences based on the task instructions and determining the soft constraint score of the candidate adjustment scheme based on the user preferences; determining the comprehensive score of the candidate adjustment scheme based on the step score, aircraft type score, marginal contribution score, and soft constraint score; and screening based on the comprehensive score of each candidate adjustment scheme to determine an adjustment scheme.

[0095] In some examples, the soft constraint score can also be determined based on the degree of overbooking of economy class in candidate adjustment schemes, wherein the degree of overbooking of economy class in candidate adjustment schemes is determined based on at least one of economy class booking data and predicted booking data in the candidate adjustment schemes, and a preset economy class booking threshold.

[0096] As a non-restrictive example, the overall score of candidate adjustment scheme S. ,in, Represents the number of steps as a fraction. Indicates the model score. Represents the marginal contribution score. This represents the soft constraint score.

[0097] Steps score for The product of the maximum adjustment step threshold and the difference between the actual adjustment steps. The step weights are positive and sufficiently large to ensure that the solution with the fewest steps is selected.

[0098] Model score for The product of this and the adjusted number of flights that did not undergo aircraft type changes. The model weight is significantly less than The positive weights are used to ensure that the solution with fewer model changes is selected when the number of steps is the same.

[0099] Marginal contribution score Changes in marginal contribution and The product of The marginal contribution weight is significantly less than The positive weights ensure that, given the same number of steps and the same number of adjusted flights without aircraft type changes, the solution with the largest change in marginal contribution is selected.

[0100] Soft constraint score This can be understood as increasing the soft constraint score when the adjusted flight meets the user's preferences. For example, when the user prefers "to use wide-body aircraft as much as possible," each time a narrow-body aircraft flight is adjusted to be operated by a wide-body aircraft, a specified score is added, thus obtaining... In addition, soft constraint scores It can also be the sum of user preference score and overbooking score. The overbooking score is determined based on the number of economy class overbookings in the candidate options that exceed a preset threshold, and the overbooking score is a negative value.

[0101] Therefore, the scoring of adjustment schemes in this disclosure aligns with the human decision-making logic of flight schedulers: the primary optimization objective is the number of adjustment steps, with fewer steps resulting in higher scores; when the number of steps is the same, the secondary optimization objective is the retention rate of the same aircraft type to minimize the disruption of existing support resources caused by aircraft type changes; when the first two are consistent, the tertiary optimization objective is the change in marginal contribution, while also incorporating soft constraints such as overbooking control and wide-body aircraft preference for fine-tuning. Thus, the scheme ranking output by Flight Adjustment Method 100 is highly consistent with the judgment logic of experienced schedulers, improving the practicality and feasibility of the adjustment schemes.

[0102] In some embodiments, the third agent can be configured to screen candidate adjustment schemes based on a simulated annealing algorithm. Of course, it is understood that this application does not limit the specific algorithm used by the third agent to screen candidate adjustment schemes; with future technological advancements, AI coding (i.e., artificial intelligence-assisted code generation) can be used to tailor screening algorithms for each flight adjustment.

[0103] As a non-limiting implementation, the scheduling agent's screening of candidate adjustment schemes based on the simulated annealing algorithm may include the following steps: (1) In the initialization phase, the scheduling agent calculates the list of gap time slots between adjacent flights for each aircraft based on neighborhood data, and constructs an initial solution S0; sets the initial temperature T0 and the termination temperature T minAnd the cooling coefficient α; (2) Hierarchical neighborhood operation construction: In the fixed temperature iteration layer, the scheduling agent initiates neighborhood exploration to generate candidate solutions. The operation mode is reconnection operation: randomly select one or more flight pairs and move them from the original aircraft to another aircraft that can operate the route. The flight time remains fixed: reconnection operation: if the new aircraft is the same as the original aircraft, then the flight only changes the aircraft and the aircraft type remains unchanged; aircraft type replacement operation: if the new aircraft is different from the original aircraft, then the flight simultaneously changes the aircraft and the aircraft type. At this time, the constraints of the route operation approval range and overbooking must be met; (3) Agent negotiation and evaluation: the scheduling agent receives candidate adjustment schemes from the aircraft agent through the Hypertext Transfer Protocol (HTTP) and determines the comprehensive score of the candidate adjustment schemes; (4) Probability acceptance and state update: the scheduling agent executes the Metropolis acceptance criterion: if the candidate solution The objective function value is better than the current solution. (Right now If the solution is positive, it is directly accepted as the new current solution; otherwise, it is determined by probability. Accept suboptimal solutions, where T is the current temperature; update the current solution after each acceptance, and dynamically update the historical best solutions. (i.e., the solution with the highest F value during the search process); (5) Convergence output: when the temperature drops to T min If no improvement is found after several consecutive iterations, the scheduling agent outputs the current solution as an adjustment plan and returns it. If no feasible solution is found during the entire search process ( If the initial infeasibility state is still present or the F value is extremely low, the normal output will be "No solution", indicating that under the current data and hard constraints, the required capacity cannot be squeezed out through flight adjustments.

[0104] This disclosure also provides a flight adjustment system in another aspect. (See reference) Figure 5 This is a schematic block diagram illustrating a flight adjustment system 200 according to some embodiments of the present disclosure. Figure 5 As shown, the flight adjustment system 200 includes a first agent 202, a second agent 204 communicatively coupled to the first agent 202, a third agent 206 communicatively coupled to the first agent 202, and a fourth agent 208 communicatively coupled to the third agent 206.

[0105] The first intelligent agent 202 is configured to receive information related to flight adjustment input by the user; extract flight adjustment intentions and flight adjustment constraints based on the information; and generate task instructions based on the flight adjustment intentions and flight adjustment constraints and transmit the task instructions to the second intelligent agent 204, wherein the flight adjustment intentions indicate the business scenario of flight adjustment, and the flight adjustment constraints indicate the target time range and the target airport.

[0106] The second intelligent agent 204 is configured to receive task instructions from the first intelligent agent 202 and determine the business scenario of flight adjustment based on the task instructions. In response to the determined business scenario being the first scenario, it retrieves flight plan data from the flight plan database based on the task instructions, determines the adjustment scheme based on the task instructions, the flight plan data, and the rules for flight adjustment, and outputs the determined adjustment scheme.

[0107] The first intelligent agent 202 is also configured to, in response to the determined business scenario being the second scenario, obtain flight plan data from the flight plan database based on task instructions, process the flight plan data based on task instructions to obtain neighborhood data, and transmit the neighborhood data to the third intelligent agent 206 and the fourth intelligent agent 208.

[0108] The third agent 206 and the fourth agent 206 are configured to determine an adjustment scheme based on task instructions, neighborhood data, and rules for flight adjustment. The fourth agent 208 is configured to generate candidate adjustment schemes based on task instructions, neighborhood data, and rules for flight adjustment and provide the candidate adjustment schemes to the third agent 206. The third agent 206 is configured to, in response to receiving the candidate adjustment schemes, filter the candidate adjustment schemes based on the task instructions to determine an adjustment scheme and output the determined adjustment scheme.

[0109] Continue to refer to Figure 5 In some embodiments, the fourth intelligent agent 208 may include a flight intelligent agent 210 and an aircraft intelligent agent 220. Here, the flight intelligent agent 210 is configured to perform a first operation, which includes: identifying at least one flight pair operated by each of a plurality of aircraft in neighborhood data, the plurality of aircraft including a first aircraft and a second aircraft; transferring each flight pair operated by the first aircraft to the second aircraft to generate a corresponding first candidate adjustment scheme; verifying the corresponding first candidate adjustment scheme based on a third type of rule in the rules for flight adjustment and a basic verification rule; and providing the verified first candidate adjustment scheme as a second candidate adjustment scheme to the aircraft intelligent agent 220. The aircraft intelligent agent 220 is configured to perform a second operation, which includes: in response to receiving a second candidate adjustment scheme, verifying the second candidate adjustment scheme based on a fourth type of rule in the rules for flight adjustment and providing the verified second candidate adjustment scheme as a third candidate adjustment scheme; determining whether the third candidate adjustment scheme meets the flight adjustment constraints based on a task instruction; and in response to determining that the third candidate adjustment scheme meets the flight adjustment constraints, providing the third candidate adjustment scheme as a candidate adjustment scheme to the third intelligent agent 206.

[0110] As a non-limiting example, the flight adjustment system 200 can be deployed on an airline's internal agent platform. Each agent within the flight adjustment system 200 can operate as an independent service unit, with the agents being relatively decoupled. For example, if an agent needs to modify its internal code due to a technology upgrade, as long as its output data is normal, it will not affect the normal operation of other agents. Therefore, this decoupled modular design of the flight adjustment system 200 of this application offers high flexibility. Furthermore, the agents communicate and exchange data via HTTP, and the message format can be JSON. The functional modules within each agent (such as data loading, rule validation, and scheme filtering) can transfer data and handle logical flow through in-process local function calls, without incurring additional network overhead.

[0111] The flight adjustment system 200 can be used to execute various embodiments of the aforementioned flight adjustment method 100. Therefore, embodiments of the flight adjustment system 200 can be referenced to the various embodiments of the aforementioned flight adjustment method 100, and will not be repeated here.

[0112] Therefore, the flight adjustment system 200 disclosed herein covers the entire flight adjustment process, from natural language intent parsing, flight plan data processing, business scenario classification, generation of candidate adjustment schemes to final output of adjustment schemes, assisting business personnel to quickly implement flight adjustments and achieving end-to-end automated flight adjustment.

[0113] This disclosure also provides an electronic device in another aspect. (See reference...) Figure 6 This is a schematic block diagram illustrating an electronic device 300 according to some embodiments of the present disclosure. Figure 6As shown, electronic device 300 includes processor 302 and memory 304 storing computer-executable instructions that, when executed by processor 302, cause processor 302 to perform the flight adjustment method 100 according to any of the foregoing embodiments. Processor 302 may be, for example, a central processing unit (CPU) of electronic device 300. Processor 302 may be any type of general-purpose processor or may be a processor specifically designed for flight adjustment, such as an application-specific integrated circuit (“ASIC”). Memory 304 may be coupled to processor 302 and may include various computer-readable media accessible by processor 302. In various embodiments, memory 304 described herein may include volatile and non-volatile media, removable and non-removable media. For example, memory 304 may include any combination of: random access memory (“RAM”), dynamic RAM (“DRAM”), static RAM (“SRAM”), read-only memory (“ROM”), flash memory, cache memory, and / or any other type of non-transient computer-readable media. The memory 304 may store instructions that, when executed by the processor 302, cause the processor 302 to execute the flight adjustment method 100 according to any of the foregoing embodiments of this disclosure.

[0114] The electronic device 300 is configured to perform the flight adjustment method 100 described in any of the foregoing embodiments, and therefore can be referred to the description of the various embodiments of the flight adjustment method 100 above, which will not be repeated here.

[0115] This disclosure also provides a computer-readable storage medium having computer-executable instructions stored thereon, which, when executed by a processor, cause the processor to perform the flight adjustment method 100 according to any of the foregoing embodiments of this disclosure.

[0116] This disclosure also provides a computer program product that may include instructions that, when executed by a processor, implement the flight adjustment method 100 according to any of the foregoing embodiments of this disclosure. The instructions may be any set of instructions that will be executed directly by one or more processors, such as machine code, or any set of instructions that will be executed indirectly, such as a script. The instructions may be stored in an object code format for direct processing by one or more processors, or stored in any other computer language, including scripts or sets of independent source code modules that are interpreted on demand or compiled in advance.

[0117] Figure 7This is a schematic block diagram illustrating a computer system 400 on which embodiments of the present disclosure may be implemented. The computer system 400 includes a bus 402 or other communication mechanism for transmitting information, and a processing means 404 coupled to the bus 402 for processing information. The computer system 400 also includes a memory coupled to the bus 402 for storing instructions to be executed by the processing means 404; the memory may be random access memory 406 or other dynamic storage device. The memory may also be used to store temporary variables or other intermediate information during the execution of instructions to be executed by the processing means 404. The computer system 400 also includes a read-only memory 408 or other static storage device coupled to the bus 402 for storing static information and instructions for the processing means 404. A storage device 410, such as a magnetic disk or optical disk, is provided and coupled to the bus 402 for storing information and instructions. The computer system 400 may be coupled via the bus 402 to an output device 412 for providing output to a user, such as, but not limited to, a display (such as a cathode ray tube (CRT) or liquid crystal display (LCD)), a speaker, etc. Input devices 414, such as a keyboard, mouse, and microphone, are coupled to bus 402 for transmitting information and command selections to processing device 404. Computer system 400 can execute embodiments of this disclosure. Consistent with certain implementations of this disclosure, results are provided by computer system 400 in response to processing device 404 executing one or more sequences of one or more instructions contained in memory. Such instructions may be read into memory from another computer-readable medium, such as storage device 410. Execution of the sequence of instructions contained in memory causes processing device 404 to perform the methods described herein. Alternatively, the teachings may be implemented using hardwired circuitry in place of or in combination with software instructions. Therefore, implementations of this disclosure are not limited to any particular combination of hardware circuitry and software. In various embodiments, computer system 400 may be connected across a network via network interface 416 to one or more other computer systems, such as computer system 400, to form a networked system. This network may include a private network or a public network such as the Internet. In a networked system, one or more computer systems may store data and supply data to other computer systems. As used herein, the term "computer-readable medium" refers to any medium that participates in providing instructions to processing device 404 for execution. Such media can take many forms, including but not limited to non-volatile media, volatile media, and transmission media. Non-volatile media include, for example, optical discs or magnetic disks such as storage device 410. Volatile media include dynamic memory such as memory. Transmission media include coaxial cables, copper wires, and optical fibers, including wiring that includes bus 402.Common forms of computer-readable media or computer program products include, for example, floppy disks, flexible disks, hard disks, magnetic tapes, or any other magnetic media, CD-ROMs, digital video discs (DVDs), Blu-ray discs, any other optical media, thumb drives, memory cards, RAM, PROMs and EPROMs, fast EPROMs, any other memory chips or cartridges, or any other tangible media from which a computer can read. Various forms of computer-readable media may be involved when carrying one or more sequences of one or more instructions to processing device 404 for execution. For example, instructions may initially be carried on a disk of a remote computer. The remote computer may load the instructions into its dynamic memory and transmit the instructions over a telephone line using a modem. A modem local to computer system 400 may receive data over a telephone line and convert the data into an infrared signal using an infrared transmitter. An infrared detector coupled to bus 402 may receive the data carried in the infrared signal and place the data on bus 402. Bus 402 carries the data to memory, and processing device 404 retrieves the instructions from memory and executes the instructions. Optionally, instructions received from the memory may be stored on the storage device 410 before or after execution by the processing device 404.

[0118] According to various embodiments, instructions configured to be executed by a processing device to perform a method are stored on a computer-readable medium. The computer-readable medium may be a device for storing digital information. For example, a computer-readable medium includes a compact disc read-only memory (CD-ROM) as known in the art for storing software. The computer-readable medium is accessed by a processor adapted to execute the instructions configured to be executed.

[0119] The foregoing has described one or more exemplary embodiments of this disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0120] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or physical entities, or by products with certain functions. A typical implementation device is a server system. Of course, this disclosure does not exclude the possibility that, with the future development of computer technology, the computer implementing the functions of the above embodiments can be, for example, a personal computer, a laptop computer, an in-vehicle human-machine interaction device, a cellular phone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or any combination of these devices.

[0121] While one or more embodiments of this disclosure provide the method operation steps as described in the embodiments or flowcharts, more or fewer operation steps may be included based on conventional or non-inventive means. The order of steps listed in the embodiments is merely one possible execution order among many and does not represent the only execution order. In actual device or terminal product execution, the methods shown in the embodiments or drawings can be executed sequentially or in parallel (e.g., in a parallel processor or multi-threaded processing environment, or even a distributed data processing environment).

[0122] The terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, product, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, product, or apparatus. Without further limitation, the presence of other identical or equivalent elements in the process, method, product, or apparatus that includes said elements is not excluded. For example, the use of terms such as "first" or "second" to denote names does not indicate any particular order.

[0123] For ease of description, the above devices are described in terms of function, divided into various modules. Of course, when implementing one or more embodiments of this disclosure, the functions of each module can be implemented in one or more software and / or hardware, or a module that performs the same function can be implemented by a combination of multiple sub-modules or sub-units. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0124] This disclosure is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in one or more blocks of the flowchart illustrations and / or one or more blocks of the block diagrams.

[0125] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means that implement the functions specified in one or more flowcharts and / or one or more blocks of a block diagram.

[0126] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, such that the instructions, which execute on the computer or other programmable apparatus, provide steps for implementing the functions specified in one or more flowcharts and / or one or more blocks of a block diagram.

[0127] Those skilled in the art will understand that one or more embodiments of this disclosure may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, one or more embodiments of this disclosure may take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0128] One or more embodiments of this disclosure can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a particular task or implement a particular abstract data type. One or more embodiments of this disclosure can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In a distributed computing environment, program modules can reside in local and remote computer storage media, including storage devices.

[0129] The same or similar parts between the various embodiments of this disclosure can be referred to mutually, and each embodiment focuses on describing the differences from other embodiments. In particular, for the apparatus embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and relevant parts can be referred to the description of the method embodiments. In the description of this disclosure, the descriptions of terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., mean that the specific feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of this disclosure. In this disclosure, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. Furthermore, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this disclosure and the features of the different embodiments or examples.

[0130] Additionally, when used in this disclosure, the terms “here,” “above,” “below,” “below,” “in the following,” “overall,” and similar terms should refer to the entirety of this disclosure and not any particular part thereof. Furthermore, unless expressly stated otherwise or otherwise understood in the context in which they are used, conditional language used herein, such as “may,” “possibly,” “for example,” “like,” etc., is generally intended to express that certain embodiments include, while other embodiments do not, certain features, elements, and / or states. Therefore, such conditional language is not generally intended to imply that one or more embodiments require features, elements, and / or states in any way, or whether such features, elements, and / or states are included or performed in any particular embodiment.

[0131] The above description is merely an embodiment of one or more embodiments of this disclosure and is not intended to limit the scope of the one or more embodiments of this disclosure. Various modifications and variations can be made to the one or more embodiments of this disclosure by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this disclosure should be included within the scope of the claims.

Claims

1. A method for adjusting flight schedules, comprising: Receive user input related to flight adjustments; The first intelligent agent extracts flight adjustment intentions and flight adjustment constraints based on the information, and generates task instructions based on the flight adjustment intentions and flight adjustment constraints, wherein the flight adjustment intentions indicate the business scenario of flight adjustment, and the flight adjustment constraints indicate the target time range and target airport; The first intelligent agent determines the business scenario of flight adjustments based on the task instructions; Perform the following operations: In response to the identified business scenario being the first scenario, a second intelligent agent retrieves flight plan data from the flight plan database based on the task instruction and determines an adjustment plan based on the task instruction, the flight plan data, and rules for flight adjustments. In response to the determined business scenario being the second scenario, the first intelligent agent retrieves flight plan data from the flight plan database based on the task instruction and processes the flight plan data to obtain neighborhood data. Then, the third and fourth intelligent agents determine an adjustment scheme based on the task instruction, the neighborhood data, and the rules for flight adjustment. The fourth intelligent agent is configured to generate candidate adjustment schemes based on the task instruction, the neighborhood data, and the rules for flight adjustment, and provide these candidate adjustment schemes to the third intelligent agent. The third intelligent agent is configured to filter the candidate adjustment schemes based on the task instruction to determine the final adjustment scheme. Output the determined adjustment plan.

2. The method according to claim 1, wherein, The flight planning data includes at least one of the following: flight entity data, aircraft layout data, operating cost and revenue data, reservation data, predicted reservation data, execution rate threshold, VIP reservation data, flight execution rate data, operational approval data, flight special restriction data, paired flight data, mandatory connection data, minimum turnaround time (MGT), overbooking allowance, and MGT exemption rules.

3. The method according to claim 1, wherein, The flight adjustment constraints in the task instruction also specify the target aircraft type. The adjustment plan, determined by a second intelligent agent based on the task instruction, the flight schedule data, and the rules used for flight adjustment, includes: The second intelligent agent obtains neighborhood data based on the flight plan data and identifies multiple flight pairs in the neighborhood data; The second intelligent agent determines, based on the task instructions, that flight adjustments in the first scenario include aircraft type changes for a single flight pair and identifies the target aircraft type. The second intelligent agent adjusts the aircraft type of each of the multiple flight pairs to the target aircraft type to obtain multiple candidate adjustment schemes. The second intelligent agent verifies each of the corresponding multiple candidate adjustment schemes based on the first type of rules and the basic verification rules in the rules for flight adjustment, and calculates the marginal contribution change of the candidate adjustment scheme that passes the verification. The marginal contribution change indicates the difference between the marginal contribution of the candidate adjustment scheme after adjustment and the marginal contribution before adjustment. The second intelligent agent outputs an adjustment scheme based on the task instructions, the marginal contribution change of the verified candidate adjustment schemes, and the marginal contribution change requirement.

4. The method according to claim 3, wherein, The first category of rules includes at least one of the following: operation review verification rules, overbooking verification rules for sold reservations, overbooking verification rules for predicted reservations, MGT verification rules, flight special restriction verification rules, scheduled maintenance flight constraint verification rules, and mandatory connection constraint verification rules.

5. The method according to claim 1, wherein, The determination of the adjustment plan by the second intelligent agent based on the task instructions, the flight plan data, and the rules for flight adjustment includes: The second intelligent agent obtains neighborhood data based on the flight plan data and identifies multiple flight pairs in the neighborhood data; The second intelligent agent determines, based on the task instructions, that flight adjustments in the first scenario include the cancellation of individual flight pairs; The second intelligent agent cancels each of the multiple flight pairs to obtain multiple candidate adjustment schemes. The second intelligent agent verifies each of the corresponding multiple candidate adjustment schemes based on the second type of rules and the basic verification rules in the rules for flight adjustment, and calculates the marginal contribution change of the candidate adjustment scheme that passes the verification. The marginal contribution change indicates the difference between the marginal contribution of the candidate adjustment scheme after adjustment and the marginal contribution before adjustment. The second intelligent agent outputs an adjustment scheme based on the task instructions, the marginal contribution change of the verified candidate adjustment schemes, and the marginal contribution change requirement.

6. The method according to claim 5, wherein, The second category of rules includes at least one of the following: time slot execution rate verification rules, VIP reservation verification rules, scheduled maintenance flight cancellation restriction rules, and forced connection cancellation verification rules.

7. The method according to claim 3 or 5, wherein, The neighborhood data obtained by the second intelligent agent based on the flight plan data includes at least one of the following: The second intelligent agent filters the flight plan data based on the target airport to obtain flight plan data related to the target airport as the neighborhood data; The second intelligent agent filters the flight plan data based on the target time range to obtain flight plan data within the target time range as the neighborhood data.

8. The method according to claim 3 or 5, wherein, Obtaining neighborhood data based on the flight plan data via the second intelligent agent includes: directly using the flight plan data as neighborhood data via the second intelligent agent.

9. The method according to claim 1, wherein, In response to the determined business scenario being the second scenario, the first intelligent agent processes the flight plan data based on the task instructions to obtain neighborhood data, including: Perform the following operations via the first intelligent agent: Based on the target time range, the flight plan data is filtered to obtain the first flight plan data within the target time range, and the first flight with the earliest departure time and the second flight with the latest arrival time are determined from the first flight plan data. Based on the flight schedule data, a third flight that forms a flight pair with the first flight and a fourth flight that forms a flight pair with the second flight are determined; In response to the fact that at least one of the third and fourth flights is not in the first flight schedule data, the at least one flight and the first flight schedule data are used as the second flight schedule data; The earliest departure time and latest arrival time of the flight in the second flight plan data are advanced and delayed by a specified amount of time as extended time ranges, and neighborhood data is obtained based on the extended time ranges and the flight plan data.

10. The method according to claim 1 or 9, wherein, The fourth intelligent agent includes a flight intelligent agent and an aircraft intelligent agent. The fourth intelligent agent is configured to generate candidate adjustment schemes based on the task instructions, the neighborhood data, and the rules for flight adjustment, including: The first operation is performed via a flight intelligent agent, the first operation including: Identify at least one flight pair operated by each of a plurality of aircraft in the neighborhood data, the plurality of aircraft including a first aircraft and a second aircraft. Each flight pair operated by the first aircraft is transferred to the second aircraft to generate a corresponding first candidate adjustment scheme. Based on the third type of rules and basic verification rules in the rules for flight adjustment, the corresponding first candidate adjustment scheme is verified, and the first candidate adjustment scheme that passes the verification is provided to the aircraft agent as the second candidate adjustment scheme. The second operation is performed via the aircraft intelligent agent, the second operation including: In response to receiving the second candidate adjustment scheme, the second candidate adjustment scheme is verified based on the fourth type of rule in the rules for flight adjustment, and the verified second candidate adjustment scheme is used as the third candidate adjustment scheme. Based on the task instruction, determine whether the third candidate adjustment scheme meets the flight adjustment constraints. In response to determining that the third candidate adjustment scheme satisfies the flight adjustment constraint, the third candidate adjustment scheme is provided to the third intelligent agent as the candidate adjustment scheme.

11. The method according to claim 10, wherein, The third category of rules includes at least one of the following: operation review verification rules, overbooking verification rules for sold reservations, overbooking verification rules for predicted reservations, flight special restriction verification rules, and mandatory connection constraint verification rules. The fourth category of rules includes at least one of the MGT verification rules and the scheduled maintenance flight constraint verification rules.

12. The method according to any one of claims 3, 5, or 10, wherein, The basic verification rules include: The candidate adjustment plan does not include multiple flight pairs whose operating times overlap at least partially; and The departure airport of the outbound flight in the adjusted flight pair in the candidate adjustment scheme is the same as the arrival airport of the return flight in the adjacent preceding flight pair in the candidate adjustment scheme, and the arrival airport of the return flight in the adjusted flight pair is the same as the departure airport of the outbound flight in the adjacent following flight pair in the candidate adjustment scheme.

13. The method according to claim 10, wherein, The second operation also includes: In response to determining that the third candidate adjustment scheme does not meet the flight adjustment constraint, a prompt instruction is sent to the flight agent, wherein the prompt instruction instructs the flight agent to repeat the first operation based on the third candidate adjustment scheme until the third candidate adjustment scheme meets the flight adjustment constraint or no new second candidate adjustment scheme is generated.

14. The method of claim 10, wherein, The third agent is configured to filter the candidate adjustment schemes based on the task instructions to determine the adjustment scheme, including: The fourth agent receives a plurality of candidate adjustment schemes and performs the following operations on each candidate adjustment scheme: The number of adjustment steps for a candidate adjustment plan is determined based on the number of adjusted flight pairs in the candidate plan, and a step score is determined based on the number of adjustment steps. Determine whether the current aircraft type of the adjusted flight pairs in the candidate adjustment plan is the same as the original aircraft type, and determine the aircraft type score of the candidate adjustment plan based on the determination result. The change in the marginal contribution of the candidate adjustment scheme is determined based on the adjusted marginal contribution and the original marginal contribution, and the marginal contribution score of the candidate adjustment scheme is determined based on the change in the marginal contribution. Based on the task instruction, user preferences are determined, and based on these user preferences, the soft constraint score of the candidate adjustment scheme is determined. The overall score of the candidate adjustment plan is determined based on the step score, model score, marginal contribution score, and soft constraint score. The adjustment scheme is determined by screening based on the comprehensive score of each candidate adjustment scheme.

15. A flight adjustment system, the flight adjustment system comprising a first intelligent agent, a second intelligent agent communicatively coupled to the first intelligent agent, a third intelligent agent communicatively coupled to the first intelligent agent, and a fourth intelligent agent communicatively coupled to the third intelligent agent, wherein, The first intelligent agent is configured as follows: Receive user input related to flight adjustments. Based on the aforementioned information, the flight adjustment intentions and constraints are extracted. Based on the flight adjustment intention and the flight adjustment constraints, a task instruction is generated and transmitted to the second intelligent agent, wherein the flight adjustment intention indicates the business scenario of flight adjustment, and the flight adjustment constraints indicate the target time range and the target airport; The second intelligent agent is configured as follows: Receive the task instruction from the first intelligent agent and determine the business scenario for flight adjustment based on the task instruction, and In response to the determined business scenario being the first scenario, the system retrieves flight plan data from the flight plan database based on the task instruction, determines an adjustment scheme based on the task instruction, the flight plan data, and the rules for flight adjustment, and outputs the determined adjustment scheme. The first intelligent agent is also configured as follows: In response to the determined business scenario being the second scenario, flight plan data is retrieved from the flight plan database based on the task instruction. The flight plan data is processed based on the task instructions to obtain neighborhood data, and the neighborhood data is then transmitted to the third and fourth intelligent agents; and The third and fourth agents are configured to determine an adjustment scheme based on the task instructions, the neighborhood data, and the rules for flight adjustments. The fourth agent is configured to generate candidate adjustment schemes based on the task instructions, the neighborhood data, and the rules for flight adjustment, and then provide the candidate adjustment schemes to the third agent. The third agent is configured to, in response to receiving the candidate adjustment scheme, filter the candidate adjustment scheme based on the task instruction to determine an adjustment scheme and output the determined adjustment scheme.

16. An electronic device comprising: processor; A memory storing computer-executable instructions, which, when executed by the processor, cause the processing to perform the flight adjustment method according to any one of claims 1 to 14.

17. A computer storage medium having stored thereon computer-executable instructions, which, when executed by a processor, cause the processor to perform the flight adjustment method according to any one of claims 1 to 14.

18. A computer program product comprising instructions that, when executed by a processor, implement the flight adjustment method according to any one of claims 1 to 14.

Images