A real-time dynamic allocation method, device and computer equipment for flight space

By adopting a real-time dynamic allocation method based on time granularity in flight cabin allocation, and using segmented discrete time system equations and feedback mechanisms, the problem that static strategies cannot track changes in user demand is solved, and efficient and automated allocation of flight cabins is achieved.

CN112598310BActive Publication Date: 2024-12-03GUANGZHOU YOUCE TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202011598212.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2020-12-29
Publication Date
2024-12-03
Estimated Expiration
2040-12-29

AI Technical Summary

Technical Problem

In the prior art, the static control strategy of flight cabins cannot track the dynamic changes in user needs, resulting in a more one-sided and limited allocation plan.

Method used

Using a real-time dynamic allocation method based on preset time granularity, by obtaining input vectors and cabin adjustment data, using preset coefficient matrix and segmented discrete time system equations, the cabin allocation is adjusted in real time, and a feedback mechanism between the target output and the input vector is established. , dynamically track user needs.

Benefits of technology

Real-time tracking and automatic optimization of flight cabin allocation is realized, ensuring that the output vectors can run along the preset target trajectory to the maximum extent, and improving the allocation efficiency and adaptability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN112598310B_ABST
    Figure CN112598310B_ABST
Patent Text Reader

Abstract

The present invention provides a method, device and computer equipment for real-time dynamic allocation of flight space, wherein the method comprises: based on a preset time granularity, obtaining an input vector and space adjustment data at the kth moment, wherein the time granularity is used to characterize the time interval between two moments; determining the space adjustment data at the (k+1)th moment according to the space adjustment data at the kth moment, the input vector at the kth moment, a preset coefficient matrix and a preset piecewise discrete time system equation; and performing space allocation according to the space adjustment data at the (k+1)th moment. By implementing the present invention, the problem that the generated static space control strategy existing in the related art is relatively one-sided and limited is solved, and the change of the output vector can be tracked in real time, and a real-time feedback operation mechanism between the target output and the input vector is established.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of information processing, and in particular to a real-time dynamic allocation method, device and computer equipment for flight cabin space. Background Art

[0002] Currently, the more effective way to allocate seats on flights is to fully match the potential needs of users with relatively fixed capacity. This method can also be used in other transportation industries such as car rental and railways, or in other service industries such as hotels, cinemas, electricity, telecommunications, and banks, with a wide range of applications.

[0003] When applied to the transportation industry, the classic seat control theory, the EMSR method, can be used to meet user needs. However, in reality, the probability distribution of user needs and demand estimation have great uncertainty. The EMSR model is based on the Littlewood criterion, which assumes that the booking demand arrives at a fixed condition of low first and high later, and calculates the daily booking control quantity based on the fixed condition. In fact, the actual needs of users are not compatible with the EMSR model, and the generated static cabin control strategy cannot track the dynamic changes of users' actual ticket booking needs, resulting in the generated cabin allocation plan being one-sided and limited. Summary of the invention

[0004] In view of this, an embodiment of the present invention provides a real-time dynamic allocation method, device and computer equipment for flight seats to solve the problem that the static seat control strategy generated in the related art cannot track the dynamic changes of user demand, resulting in the generated seat allocation plan being relatively one-sided and limited.

[0005] According to the first aspect, an embodiment of the present invention provides a real-time dynamic allocation method for flight cabin space, comprising: based on a preset time granularity, obtaining an input vector and cabin space adjustment data at the kth moment, the time granularity being used to characterize the time interval between two moments; determining the cabin space adjustment data at the (k+1)th moment based on the cabin space adjustment data at the kth moment, the input vector at the kth moment, a preset coefficient matrix and a preset piecewise discrete time system equation; the preset piecewise discrete time system equation is used to characterize the relationship between the cabin space adjustment data at different moments, and the relationship between the output vector at the kth moment and the cabin space adjustment data; and performing cabin space allocation based on the cabin space adjustment data at the (k+1)th moment.

[0006] In combination with the first aspect, in the first embodiment of the first aspect, the step of obtaining the input vector and cabin adjustment data at the kth moment includes: if the kth moment is the initial moment, the input vector at the kth moment is a preset value, and the cabin adjustment data at the kth moment is the preset cabin adjustment data; or, if the kth moment is not the initial moment, the input vector at the (k-1)th moment is determined according to the input vector at the (k-1)th moment, the error control reference at the (k-1)th moment, and the mean matrix of the input vector at the (k-1)th moment; the cabin adjustment data at the kth moment is determined according to the cabin adjustment data at the (k-1)th moment, the input vector at the (k-1)th moment, and the coefficient matrix.

[0007] In combination with the first embodiment of the first aspect, in the second embodiment of the first aspect, the input vector mean matrix at the (k-1)th moment and the error control reference at the (k-1)th moment are determined by the following process, and the kth moment is not the initial moment: the error at the (k-1)th moment is determined according to the output vector at the (k-1)th moment and the preset target output at the (k-1)th moment; the input vector mean matrix at the (k-1)th moment is calculated and generated according to the error at the (k-1)th moment, the input vector at the kth moment, and the input vector at the (k-1)th moment; the error control reference at the (k-1)th moment is calculated and generated according to the preset first weight matrix, second weight matrix and coefficient matrix.

[0008] In combination with the second implementation of the first aspect, in the third implementation of the first aspect, the output vector at the (k-1)th moment is determined by the following process: a target flight output equation is generated based on the target historical data and the time granularity, and the target flight output equation is used to characterize the relationship between the time granularity and the output vector and the cabin adjustment data; the output vector at the (k-1)th moment is determined based on the target flight output equation and the cabin adjustment data at the (k-1)th moment.

[0009] In combination with the third implementation of the first aspect, in the fourth implementation of the first aspect, the target flight output equation is generated based on the target historical data and the time granularity, including: obtaining initial historical data within a preset time period; removing singular points in the initial historical data to generate target historical data; generating an initial flight output equation based on the time granularity and the target historical data; and performing piecewise approximate fitting processing on the initial flight output equation to generate a target flight output equation.

[0010] In combination with the fourth implementation manner of the first aspect, in the fifth implementation manner of the first aspect, the mean matrix of the input vector at the (k+1)th moment is determined by the following process: the preset piecewise discrete time system equation is differentiated to generate a first difference equation; the error at the kth moment is differentiated to generate a second difference equation; based on the first difference equation and the second difference equation, the error system at the kth moment is calculated and generated; based on the error system at the kth moment, the mean matrix of the input vector at the (k+1)th moment is calculated and generated.

[0011] In combination with the fifth implementation of the first aspect, in the sixth implementation of the first aspect, the input vector at the kth moment is calculated by the following formula:

[0012]

[0013] Among them, u(k) represents the input vector at the kth moment, u(k-1) represents the input vector at the (k-1)th moment, and F(k-1) represents the error control reference at the (k-1)th moment. Represents the input vector mean matrix at the (k-1)th moment;

[0014] The space adjustment data at the kth moment is determined by the following formula:

[0015] x(k)=Ax(k-1)+Bu(k-1),

[0016] Among them, x(k) represents the cabin adjustment data at the kth moment, x(k-1) represents the cabin adjustment data at the (k-1)th moment, and A and B are coefficient matrices.

[0017] In combination with the sixth implementation of the first aspect, in the seventh implementation of the first aspect, the input vector mean matrix at the (k-1)th moment is calculated by the following formula:

[0018] e(k-1)=R(k-1)-y(k-1),

[0019] Δx(k-1)=x(k)-x(k-1),

[0020]

[0021] Among them, e(k-1) represents the error at the (k-1)th moment, R(k-1) represents the preset target output at the (k-1)th moment, and y(k-1) represents the output vector at the (k-1)th moment. Represents the input vector mean matrix at the (k-1)th moment;

[0022] The error control reference at the (k-1)th moment is calculated using the following formula:

[0023]

[0024]

[0025]

[0026] Wherein, F(K-1) represents the error control reference at the (k-1)th moment, H represents the first weight matrix, Q represents the second weight matrix, A and B represent the coefficient matrix, and I m represents the unit matrix, and C represents the ratio of the output vector to the cabin adjustment data.

[0027] In combination with the seventh implementation of the first aspect, in the eighth implementation of the first aspect, the error system is calculated by the following formula:

[0028]

[0029] Δx(k)=x(k+1)-x(k),

[0030] Δe(k)=e(k+1)-e(k),

[0031] Among them, Δx(k)=x(k+1)-x(k) represents the first differential equation, and Δe(k)=e(k+1)-e(k) represents the second differential equation.

[0032] According to the second aspect, an embodiment of the present invention provides a real-time dynamic allocation device for flight cabin space, including: a first acquisition module, used to acquire the input vector and cabin space adjustment data at the kth moment based on a preset time granularity, and the time granularity is used to characterize the time interval between two moments; a determination module, used to determine the cabin space adjustment data at the (k+1)th moment based on the cabin space adjustment data at the kth moment, the input vector at the kth moment, a preset coefficient matrix and a preset piecewise discrete time system equation; the preset piecewise discrete time system equation is used to characterize the relationship between the cabin space adjustment data at different moments, and the relationship between the output vector at the kth moment and the cabin space adjustment data; a cabin space allocation module, used to allocate cabin space according to the cabin space adjustment data at the (k+1)th moment.

[0033] According to the third aspect, an embodiment of the present invention provides a computer device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the steps of the real-time dynamic allocation method of flight slots described in the first aspect or any one embodiment of the first aspect.

[0034] According to the fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for real-time dynamic allocation of flight slots described in the second aspect or any one of the embodiments of the second aspect.

[0035] The technical solution of the present invention has the following advantages:

[0036] 1. A method, device and computer equipment for real-time dynamic allocation of flight space provided by the present invention, wherein the method comprises: obtaining an input vector and space adjustment data at the kth moment based on a preset time granularity, wherein the time granularity is used to characterize the time interval between two moments; determining the space adjustment data at the (k+1)th moment according to the space adjustment data at the kth moment, the input vector at the kth moment, a preset coefficient matrix and a preset piecewise discrete time system equation; the preset piecewise discrete time system equation is used to characterize the relationship between the space adjustment data at different moments, and the relationship between the output vector at the kth moment and the space adjustment data; and performing space allocation according to the space adjustment data at the (k+1)th moment. The seat adjustment data at the kth moment, the input vector at the kth moment, the preset coefficient matrix and the preset piecewise discrete time system equation are combined to determine the seat adjustment data at the (k+1)th moment. The preset piecewise discrete time system equation is used to characterize the relationship between the output vector at the kth moment and the seat adjustment data. This allows real-time tracking of changes in the output vector, and a feedback operation mechanism between the target output and the input vector. When allocating flight seats, even if the output vector deviates from the optimized route, the input can be automatically changed in real time according to the deviated output vector, so that it runs along the trajectory of the preset target output to the greatest extent. Therefore, the target output can be set based on historical seat load data or current seat load data of other flights, and the target output can be dynamically tracked. The input can be continuously changed according to the feedback of the actual output vector, thereby realizing an automated, real-time and efficient feedback optimization control strategy.

[0037] 2. The real-time dynamic allocation method, device and computer equipment of flight space provided by the present invention can select different weight matrices according to specific control requirements to achieve customized control. When the airline is more concerned about the rapid response of the expected occupancy rate, a larger weight is selected; when the airline is more concerned about the smooth transition of the discounted price, a larger weight can be selected; if both are more concerned, the weights are of the same magnitude. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0039] Figure 1 This is a flowchart of a specific example of a method for real-time dynamic allocation of flight space in an embodiment of the present invention;

[0040] Figure 2 It is a flowchart of a specific example of calculating the error control reference amount in the real-time dynamic allocation method of flight space in an embodiment of the present invention;

[0041] Figure 3 It is a flowchart of a specific example of calculating the output vector in the real-time dynamic allocation method of flight space in an embodiment of the present invention;

[0042] Figure 4 It is a flowchart of a specific example of generating a target flight output equation in the real-time dynamic allocation method of flight space in an embodiment of the present invention;

[0043] Figure 5 It is a schematic diagram of initial historical data and target historical data after approximate fitting processing in the real-time dynamic allocation method of flight space in an embodiment of the present invention;

[0044] Figure 6 Schematic diagram of actual output vector and preset target output in the real-time dynamic allocation method of flight space in an embodiment of the present invention;

[0045] Figure 7 Schematic diagram of an error signal in a real-time dynamic allocation method for flight space in an embodiment of the present invention;

[0046] Figure 8 This is a schematic diagram of cabin space adjustment data in a real-time dynamic allocation method for flight cabin space in an embodiment of the present invention;

[0047] Fig. 9 Schematic diagram of input vectors in a real-time dynamic allocation method for flight space in an embodiment of the present invention;

[0048] Fig.10 It is a principle block diagram of a specific example of a real-time dynamic allocation device for flight space in an embodiment of the present invention;

[0049] Fig.11 FIG. 4 is a specific example diagram of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION

[0050] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0052] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, it can also be the internal connection of two components, it can be a wireless connection, or it can be a wired connection. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0053] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0054] In order to solve the problem in the related art that the generated static cabin space control strategy cannot track the dynamic changes of user demand, resulting in the generated cabin space allocation plan being relatively one-sided and limited, when determining that there is a correlation between flight seat sales and flight seat prices, an embodiment of the present invention provides a real-time dynamic allocation method, device and computer equipment for flight cabin space, with the purpose of creating a real-time feedback control mechanism so that the output vector follows the preset target output.

[0055] The embodiment of the present invention provides a method for real-time dynamic allocation of flight space, such as Figure 1 As shown, including:

[0056] Step S11: Based on the preset time granularity, the input vector and cabin adjustment data of the kth moment are obtained, and the time granularity is used to characterize the time interval between two moments. In this embodiment, the preset time granularity is used to characterize the time interval between two moments, and the preset time granularity can be determined according to the needs of the user, that is, the preset time granularity can be determined according to the control frequency that meets the user's expectations. The kth moment is used to characterize the (k-1, k] time period; the k+1th moment is used to characterize the (k, k+1] time period; the time granularity can be the length of the monitored time interval, which can be one hour, half an hour or ten minutes, etc.; the input vector can be an input state value; the cabin adjustment data can be discount data. When the cabin adjustment data is 40, it can be determined that the discount of the flight cabin price at this time is 40%, that is, 40% off; specifically, according to the time granularity at this time, the input vector and cabin adjustment data of the corresponding moment are determined. The kth moment can be the initial moment or any moment, and the change of k can change in real time according to user needs.

[0057] Step S12: Determine the cabin adjustment data at the (k+1)th moment based on the cabin adjustment data at the kth moment, the input vector at the kth moment, the preset coefficient matrix and the preset piecewise discrete time system equation; the preset piecewise discrete time system equation is used to characterize the relationship between the cabin adjustment data at different moments, and the relationship between the output vector at the kth moment and the cabin adjustment data.

[0058] In this embodiment, the cabin adjustment data at the kth moment is the "minimum cabin discount" in the (k-1, k] time period; the preset coefficient matrix can be a matrix of preset dimensions, which can be determined according to the actual flight situation; the preset piecewise discrete time system equation can be a single flight state equation, which can be a relationship function between cabin adjustment data at different moments, state vectors and actual flight occupancy rate constructed based on LQR (linear quadratic regulator), and a control function that optimizes the actual flight occupancy rate based on parameters such as state vectors and cabin adjustment functions at different moments; it is a state equation generated based on the controllability of the flight and the observability of the flight. The preset piecewise discrete time system equation can characterize the correlation between flight adjustment data at different moments and input vectors on the one hand, and can characterize the correlation between the output vector and flight adjustment data at the same moment on the other hand. The above output vector can be the actual occupancy rate data in any time period (k-1, k] time period. That is to say, the cabin discount data at the current moment can be determined based on the cabin discount data at the previous moment, the input vector at the previous moment, the matrix of preset dimensions and the piecewise discrete time system equation.

[0059] Step S13: Allocate seats according to the seat adjustment data at the (k+1)th moment. In this embodiment, adjusting the flight seats according to the calculated seat adjustment data can be achieved by adjusting the discount data of the remaining seats.

[0060] The present invention provides a real-time dynamic allocation method for flight cabin space, comprising: based on a preset time granularity, obtaining an input vector and cabin space adjustment data at the kth moment, wherein the time granularity is used to characterize the time interval between two moments; determining the cabin space adjustment data at the (k+1)th moment according to the cabin space adjustment data at the kth moment, the input vector at the kth moment, a preset coefficient matrix and a preset piecewise discrete time system equation; the preset piecewise discrete time system equation is used to characterize the relationship between the cabin space adjustment data at different moments, and the relationship between the output vector at the kth moment and the cabin space adjustment data; and performing cabin space allocation according to the cabin space adjustment data at the (k+1)th moment. The seat adjustment data at the kth moment, the input vector at the kth moment, the preset coefficient matrix and the preset piecewise discrete time system equation are combined to determine the seat adjustment data at the (k+1)th moment. The preset piecewise discrete time system equation is used to characterize the relationship between the output vector at the kth moment and the seat adjustment data. This allows real-time tracking of changes in the output vector, and establishes a feedback operation mechanism between the target output and the input vector. When allocating flight seats, even if the output vector deviates from the optimized route, the input can be automatically changed in real time according to the deviated output vector to maximize its operation along the trajectory of the preset target output. Therefore, the target output can be set based on historical seat load data or current seat load data of other flights, and the target output can be dynamically tracked. The input is continuously changed according to the feedback of the actual output vector, and an automated real-time feedback optimization control strategy that meets the target output is efficiently generated.

[0061] As an embodiment of the present invention, the specific process of applying the capacity adjustment data to the actual flight capacity control scenario can be: the total number of seats of the actual flight can be 200, that is, capacity = 200; the flight capacity can be divided into multiple classes, for example, the first class, the second class, the third class, the fourth class and the fifth class; the price corresponding to each class can be 1,000 yuan for the first class, 900 yuan for the second class, 800 yuan for the third class, 700 yuan for the fourth class, and 600 yuan for the fifth class. At this time, the first class is full price, the second class is 10% discount, the third class is 20% discount, the fourth class is 30% discount, and the fifth class is 40% discount.

[0062] Specifically, if the output cabin adjustment data is a 30% discount, then n seats can be allocated to the fourth cabin class; for example, the number of seats currently allocated to each cabin class is [30, 50, 20, 10, 2], and when the seats in the higher cabin class are sold out, the seats in the cabin class lower than that can be sold, that is, the nested rule. Therefore, the number of seats currently available for sale in each cabin is [112, 82, 32, 12, 2].

[0063] Specifically, the process of determining the number of initially allocated spaces for each cabin class (i.e., the process of opening cabins) may include: first, determining that the fifth cabin class is initially X, and the number of spaces available for sale in other cabin classes may be as shown in Table 1 below:

[0064] Table 1

[0065] Cabin Class Number of seats allocated First Class μ1+α·σ1 Second Class μ2+α·σ2 Third Class μ3+α·σ3 Fourth Class μ4+α·σ4 Fifth Class X

[0066] Among them, μ1, ..., μ4 represent means, σ1, ..., σ4 represent standard deviations, which can be determined based on historical data of the current time. For example, when the actual flight departs on Monday, data of all Monday flights in the historical time period are obtained to determine the corresponding means and standard deviations of each cabin class.

[0067] α can be determined by the following formula:

[0068]

[0069] Among them, capacity represents the total number of seats on the actual flight.

[0070] As an optional implementation of the present invention, in the above step S11, the process of obtaining the input vector and the cabin adjustment data at the kth moment can be mainly divided into two cases:

[0071] In the first case, if the kth moment is the initial moment, the input vector at the kth moment is a preset value, and the cabin adjustment data at the kth moment is the preset cabin adjustment data; in this embodiment, the initial moment may be when k=0, the input vector at the initial moment may be 0, the cabin adjustment data at the initial moment may be determined according to the actual application scenario (for example, the number of flight seats that have been sold), and the initial discount may be 40% off.

[0072] In the second case, if the kth moment is not the initial moment, the input vector at the kth moment is determined according to the input vector at the (k-1)th moment, the error control reference at the (k-1)th moment, and the input vector mean matrix at the (k-1)th moment; the cabin adjustment data at the kth moment is determined according to the cabin adjustment data at the (k-1)th moment, the input vector at the (k-1)th moment, the preset piecewise discrete time system equation, and the coefficient matrix. In this embodiment, the input vector at the current moment can be determined according to the input vector at the previous moment, the error control reference at the previous moment, and the input vector mean matrix at the previous moment; the cabin adjustment data at the current moment can be determined according to the cabin adjustment data at the previous moment, the input vector at the previous moment, and the preset coefficient matrix; through this step, the input (for example, the input vector at the next moment and the cabin adjustment data) can be continuously changed according to the feedback of the actual output vector, and an automated real-time feedback optimization control strategy that meets the target output can be efficiently generated.

[0073] Exemplarily, the input vector at the kth moment can be calculated by the following formula:

[0074]

[0075] Among them, u(k) represents the input vector at the kth moment, u(k-1) represents the input vector at the (k-1)th moment, and F(k-1) represents the error control reference at the (k-1)th moment. Represents the input vector mean matrix at the (k-1)th moment;

[0076] The preset piecewise discrete time system equation is calculated by the following formula:

[0077]

[0078] The space adjustment data at the kth moment is determined by the following formula:

[0079] x(k)=Ax(k-1)+Bu(k-1),

[0080] Wherein, x(k) represents the space adjustment data at the kth moment, x(k-1) represents the space adjustment data at the (k-1)th moment, A and B are both preset coefficient matrices, and C is a preset dimension matrix, which is used to characterize the ratio of the output vector to the space adjustment data. Specifically, the ratio of the output vector to the space adjustment data can be determined according to the output equation of the target flight;

[0081] x(k)∈R n is the state vector of the above piecewise discrete time system equation, u(k)∈R r , is the input vector of the above piecewise discrete time system equation, y(k)∈R mis the output vector of the above piecewise discrete time system equation;

[0082]

[0083] Among them, z(k) is the actual lowest fare in the time period (k-1,k], and Z is the full-price ticket.

[0084] As an optional embodiment of the present invention, Figure 2 As shown, the input vector mean matrix at the (k-1)th moment and the error control reference at the (k-1)th moment are determined by the following process, provided that the kth moment is not the initial moment:

[0085] Step S21: Determine the error at the (k-1)th moment according to the output vector at the (k-1)th moment and the preset target output at the (k-1)th moment.

[0086] In this embodiment, the output vector at the (k-1)th moment may be the output vector at the previous moment, which is the actual seat occupancy rate in the time period (k-2, k-1); the preset target output is the optimization target, which may be the target seat occupancy rate determined based on historical data (for example, the average seat occupancy rate of the flight, the historical seat occupancy rate of the target flight); the error at the (k-1)th moment may be the difference between the actual seat occupancy rate and the preset target output at the previous moment.

[0087] Specifically, the error at the (k-1)th moment is calculated by the following formula:

[0088] e(k-1)=R(k-1)-y(k-1),

[0089] Among them, e(k-1) represents the error at the (k-1)th moment, R(k-1) represents the preset target output at the (k-1)th moment, and y(k-1) represents the output vector at the (k-1)th moment.

[0090] In another optional embodiment, the output vector at the (k-1)th moment may be the output vector at the previous moment, which is the cabin price within the time period (k-2, k-1); the preset target output is the optimization target, which may be the target cabin price determined based on historical data (for example, the average cabin price of the flight, the historical cabin price of the target flight); the error at the (k-1)th moment may be the difference between the actual cabin price and the preset target cabin price at the previous moment.

[0091] Specifically, the error at the (k-1)th moment is calculated by the following formula:

[0092] e(k-1)=R(k-1)-y(k-1),

[0093] Among them, e(k-1) represents the error at the (k-1)th moment, R(k-1) represents the preset target output at the (k-1)th moment, and y(k-1) represents the output vector at the (k-1)th moment.

[0094] That is to say, the output vector and the preset target output may be any target value set according to user needs, and may be the actual occupancy rate, cabin price, or any other physical quantity related to flight cabin.

[0095] Step S22: Calculate and generate the input vector mean matrix at the (k-1)th moment according to the error at the (k-1)th moment, the input vector at the kth moment, and the input vector at the (k-1)th moment; in this embodiment, the input vector mean matrix at the (k-1)th moment is calculated by the following formula:

[0096]

[0097] Δx(k-1)=x(k)-x(k-1),

[0098] in, represents the input vector mean matrix at the (k-1)th moment, and Δx(k-1) represents the difference between the cabin adjustment data at the kth moment and the cabin adjustment data at the k-1th moment.

[0099] Step S23: Calculate and generate the error control reference at the (k-1)th moment according to the preset first weight matrix, the second weight matrix and the coefficient matrix. In this embodiment, the error control reference at the (k-1)th moment is calculated by the following formula:

[0100]

[0101]

[0102]

[0103]

[0104] Among them, F(K-1) represents the error control reference at the (k-1)th moment, H represents the first weight matrix, Q represents the second weight matrix, A and B represent coefficient matrices, Im represents the unit matrix, and C represents the ratio of the output vector to the cabin adjustment data.

[0105] Wherein, the first weight matrix is ​​an m×m positive definite matrix, the second weight matrix is ​​an r×r positive definite matrix, and m and r are both positive integers;

[0106] The process of determining the first weight matrix and the second weight matrix may include: obtaining a control demand input by a user, and when the control demand is mainly to output a fast response of the attendance rate to a preset target output, e When the control requirement is mainly the smooth transition of the cabin adjustment data, the first weight matrix is ​​set to the preset maximum value; when the control requirement is mainly the output of the seat rate to quickly respond to the preset target output and the smooth transition of the cabin adjustment data, the first weight matrix and Q e Set to the same level.

[0107] As an optional embodiment of the present invention, Figure 3 As shown, the output vector at the (k-1)th moment is determined by the following process:

[0108] Step S31: Generate a target flight output equation based on the target historical data and time granularity, and the target flight output equation is used to characterize the relationship between the time granularity and the output vector and the cabin adjustment data; in this embodiment, the target historical data can be the actual occupancy rate data, flight seat discount data and corresponding time period data of the target flight within fourteen days before the departure date; the target flight output equation can be used to characterize the functional relationship between the ratio of the actual occupancy rate to the flight seat discount and time.

[0109] Step S32: Determine the output vector at the (k-1)th moment according to the output equation of the target flight and the cabin space adjustment data at the (k-1)th moment. In this embodiment, on the one hand, the output vector at the (k-1)th moment can be the actual seat load rate data in the (k-2, k-1] time period, which can be determined according to y(k)=Cx(k), where x(k) represents the cabin space adjustment data at the (k-1)th moment, and C can be determined according to the output equation of the target flight; on the other hand, the output vector at the (k-1)th moment can also be obtained by the following process, where the cabin space adjustment data is applied to the actual application scenario, and the actual seat load rate data in the corresponding time period is extracted according to the actual application scenario, which is the output vector at the corresponding moment.

[0110] As an optional embodiment of the present invention, Figure 4 As shown, the above step S31 generates a target flight output equation according to the target historical data and time granularity, including:

[0111] Step S311: Obtain initial historical data within a preset time period; in this embodiment, the preset time period may be within fourteen days from the departure date of the target flight; the initial historical data may be actual data within fourteen days from the departure date that has not been processed. The relationship between the initial historical data and the ratio of the actual seat rate to the discount data may be as follows: Figure 5 As shown, Figure 5Each point in represents the initial historical data, the horizontal axis k represents each moment (such as a time point), and the vertical axis C represents the ratio of the actual occupancy rate to the discount data.

[0112] Step S312: remove the singular points in the initial historical data and generate target historical data; in this embodiment, the singular points may be data points with large errors; the process of generating the target historical data may be: clean the above-mentioned initial historical data, filter out the singular points in the initial historical data, and generate the target historical data.

[0113] Step S313: Generate an initial flight output equation according to the time granularity and the target historical data; in this embodiment, the generated target historical data is preliminarily simplified to generate an initial piecewise equation divided by each time, that is, the initial flight output equation:

[0114]

[0115]

[0116] C2=-0.01253k+4.235(331≤k≤335)

[0117] Step S314: Perform segmented approximate fitting processing on the initial flight output equation to generate the target flight output equation. In this embodiment, k represents the time granularity. When performing segmented approximate fitting processing, the segmentation standard can be: divided into 5 segments based on the target historical data, namely: 0-7 days (0-168 hours), 8-10 days (168 hours-240 hours), 11-12 days (240 hours-288 hours), 13 days (288 hours-312 hours) and 4 days (312 hours-336 hours);

[0118] Specifically, when the segmentation is performed by the S-type function, the value ranges of k are [0, 168], [168, 240], [240, 288], [288, 312], [288, 312] and [312, 336] respectively.

[0119] For example, when the target historical data is a flight departing on Tuesday, the target historical data can be divided into six segments when performing piecewise approximate fitting processing according to the generated initial flight output equation, and the following piecewise constant function can be calculated:

[0120]

[0121] Specifically, the relationship between the corresponding time after fitting and the ratio of the actual attendance rate to the discount data can be as follows: Figure 5 shown.

[0122] Specifically, c can represent the ratio of the output vector to the cabin adjustment data. For example, the time granularity can be determined according to the cabin allocation period, and then the correlation between c and the time granularity can be determined; the cabin allocation period can represent the change in the passenger's ticket booking time, and the time granularity is used to characterize the time interval between the user's arrival time node; according to historical data, the correlation between the user's ticket booking situation and the time interval to the flight departure node is determined, and the corresponding time granularity is determined according to the correlation; specifically, based on the historical data within the preset time period, when the total number of passengers on the departure day is 100%, more than 90% of the users book tickets two weeks before the departure day (i.e. 14 days before the departure day), and about 60% to 80% of the passengers book tickets one week before the departure day, especially two to three days before departure, the user's user ticket booking situation changes frequently.

[0123] Therefore, when the departure date is closer, the number of user ticket bookings increases, and the time granularity should be adjusted to a smaller value. For example, when only monitoring the changes 14 days before the departure date, when the time interval from the departure date is 14 days to 3 days, the time granularity can be set to one hour, that is, the value range of the moment k can be [0,264], k is a positive integer; when the time interval from the departure date is 3 days to the day of departure, the time granularity can be set to half an hour, and the change process of the moment k can be (264.5,265,265.5,......,336). The time granularity can also be set to ten minutes, five minutes, one minute, etc. according to user needs and actual application scenarios.

[0124] As an optional embodiment of the present invention, the input vector mean matrix at the (k+1)th moment is determined by the following process, including: performing differential processing on the preset piecewise discrete time system equation to generate a first differential equation; performing differential processing on the error at the kth moment to generate a second differential equation; calculating and generating the error system at the kth moment based on the first differential equation and the second differential equation; calculating and generating the input vector mean matrix at the (k+1)th moment based on the error system at the kth moment. In this embodiment, both sides of the state equation in the preset discrete time system may be differentially processed to obtain the first differential equation; and both sides of the calculation equation of the error signal may be differentially processed to obtain the second differential equation. Since differential processing is to calculate the difference between variables at different moments, the error system may be calculated by the following formula:

[0125] x(k+1)=Ax(k)+Bu(k),

[0126] e(k)=R(k)-y(k),

[0127] Δx(k)=x(k+1)-x(k),

[0128] Δe(k)=e(k+1)-e(k),

[0129]

[0130] Where: x(k+1)=Ax(k)+Bu(k) is the state equation in the preset discrete time system;

[0131]

[0132] Specifically, a simulation experiment can be performed based on the real-time dynamic allocation method for flight seats described in the above embodiment. The real-time dynamic allocation method for flight seats described in the above embodiment can be further improved based on the experimental results, and a reasonable number of equation models and weight coefficients in the performance indicators can be selected. For example, the preset target output can be:

[0133]

[0134] At this point, it can be explained that the preset target output is that a total of 222 economy class seats are sold two weeks before departure. Substitute C2 in the piecewise constant function fitted in the above embodiment steps to obtain the control simulation result. The corresponding simulation control result is as follows: Figure 6-Figure 9 As shown, Figure 6 The horizontal axis of the curve shown in FIG. 1 represents time k, the vertical axis represents the output vector, the solid line represents the actual output vector y(k), and the dotted line represents the preset target output R(k). Figure 7 The horizontal axis of the curve shown represents time k, and the vertical axis represents the error signal (The tracking error). The error signal represents the difference between the actual output vector and the preset target output, that is, the tracking error value; Figure 8 The horizontal axis of the curve shown represents time k, and the vertical axis represents the cabin adjustment data x(k); Fig. 9 The horizontal axis of the shown curve represents time k, and the vertical axis represents input vector u(k).

[0135] Through the above control simulation results, the real-time dynamic allocation method of flight slots provided by the embodiment of the present invention can determine that even if the actual seat load rate deviates from the original optimization route during the management of flight slots, the input (such as seat adjustment data and input vector) can be automatically changed in real time according to the state after the deviation, so that the output vector runs along the trajectory of the optimization target to the maximum extent; the target can also be set according to the historical seat load rate data or the current seat load rate data of other flights, so that the actual output vector dynamically tracks the target, and the input is continuously changed according to the feedback of the actual output vector, so as to efficiently generate an automated real-time feedback optimization control strategy that meets the target output situation. The historical seat load rate data can be compared with the data of last week or last month, or last year, and the current seat load rate data of other flights can be compared with the data that has occurred this week or this month.

[0136] The embodiment of the present invention provides a real-time dynamic allocation device for flight space, such as Fig.10 As shown, including:

[0137] The first acquisition module 51 is used to acquire the input vector and cabin adjustment data at the kth moment based on a preset time granularity, where the time granularity is used to characterize the time interval between two moments. For detailed implementation content, please refer to the relevant description of step S11 in the above method embodiment.

[0138] The determination module 52 is used to determine the cabin adjustment data at the (k+1)th moment based on the cabin adjustment data at the kth moment, the input vector at the kth moment, the preset coefficient matrix and the preset piecewise discrete time system equation; the preset piecewise discrete time system equation is used to characterize the relationship between the cabin adjustment data at different moments, and the relationship between the output vector at the kth moment and the cabin adjustment data; for detailed implementation content, please refer to the relevant description of step S12 in the above method embodiment.

[0139] The space allocation module 53 is used to allocate space according to the space adjustment data at the (k+1)th moment. For detailed implementation, please refer to the relevant description of step S13 in the above method embodiment.

[0140] The present invention provides a real-time dynamic allocation device for flight cabin space, comprising: a first acquisition module 51, used to acquire an input vector and cabin space adjustment data at the kth moment based on a preset time granularity, wherein the time granularity is used to characterize the time interval between two moments; a determination module 52, used to determine the cabin space adjustment data at the (k+1)th moment according to the cabin space adjustment data at the kth moment, the input vector at the kth moment, a preset coefficient matrix and a preset piecewise discrete time system equation; the preset piecewise discrete time system equation is used to characterize the relationship between the cabin space adjustment data at different moments, and the relationship between the output vector at the kth moment and the cabin space adjustment data; a cabin space allocation module 53, used to perform cabin space allocation according to the cabin space adjustment data at the (k+1)th moment. The seat adjustment data at the kth moment, the input vector at the kth moment, the preset coefficient matrix and the preset piecewise discrete time system equation are combined to determine the seat adjustment data at the (k+1)th moment. The preset piecewise discrete time system equation is used to characterize the relationship between the output vector at the kth moment and the seat adjustment data. This allows real-time tracking of changes in the output vector, and a feedback operation mechanism between the target output and the input vector. When allocating flight seats, even if the output vector deviates from the optimized route, the input can be automatically changed in real time according to the deviated output vector, so that it runs along the trajectory of the preset target output to the greatest extent. Therefore, the target output can be set based on historical seat load data or current seat load data of other flights, and the target output can be dynamically tracked. The input can be continuously changed according to the feedback of the actual output vector, thereby realizing an automated, real-time and efficient feedback optimization control strategy.

[0141] The embodiment of the present invention also provides a computer device, such as Fig.11 As shown, the computer device may include a processor 61 and a memory 62, wherein the processor 61 and the memory 62 may be connected via a bus 60 or other means. Fig.11 The connection via bus 60 is taken as an example.

[0142] The processor 61 may be a central processing unit (CPU). The processor 61 may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or a combination of the above chips.

[0143] The memory 62 is a non-transient computer-readable storage medium that can be used to store non-transient software programs, non-transient computer executable programs and modules, such as program instructions / modules corresponding to the real-time dynamic allocation method for flight slots in the embodiment of the present invention. The processor 61 executes various functional applications and data processing of the processor by running the non-transient software programs, instructions and modules stored in the memory 62, that is, the real-time dynamic allocation method for flight slots in the above method embodiment is implemented.

[0144] The memory 62 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and applications required for at least one function; the data storage area may store data created by the processor 61, etc. In addition, the memory 62 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage devices. In some embodiments, the memory 62 may optionally include a memory remotely arranged relative to the processor 61, and these remote memories may be connected to the processor 61 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0145] The one or more modules are stored in the memory 62, and when executed by the processor 61, the following is performed: Figure 1 The real-time dynamic allocation method of flight space in the illustrated embodiment.

[0146] For details of the above computer equipment, please refer to Figure 1 The corresponding related descriptions and effects in the illustrated embodiments can be understood and will not be repeated here.

[0147] An embodiment of the present invention further provides a non-transitory computer-readable medium, which stores computer instructions. The computer instructions are used to enable a computer to execute a real-time dynamic allocation method for flight seats as described in any one of the above embodiments, wherein the storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), a flash memory, a hard disk (HDD) or a solid-state drive (SSD), etc.; the storage medium may also include a combination of the above-mentioned types of memories.

[0148] Obviously, the above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the invention.

Claims

1. A real-time dynamic allocation method for flight space, characterized in that: include: Based on a preset time granularity, an input vector and cabin adjustment data at the kth moment are obtained, where the time granularity is used to represent the time interval between two moments; Determine the space adjustment data at the (k+1)th moment according to the space adjustment data at the kth moment, the input vector at the kth moment, the preset coefficient matrix and the preset piecewise discrete time system equation; The preset piecewise discrete time system equation is used to characterize the relationship between the cabin space adjustment data at different times, and the relationship between the output vector at the kth time and the cabin space adjustment data; Performing space allocation according to the space adjustment data at the (k+1)th moment; The step of obtaining the input vector and the cabin adjustment data at the kth moment includes: If the kth moment is the initial moment, the input vector at the kth moment is a preset value, and the space adjustment data at the kth moment is the preset space adjustment data; Or, if the kth moment is not the initial moment, the input vector at the kth moment is determined according to the input vector at the (k-1)th moment, the error control reference at the (k-1)th moment, and the input vector mean matrix at the (k-1)th moment; the space adjustment data at the kth moment is determined according to the space adjustment data at the (k-1)th moment, the input vector at the (k-1)th moment, and the coefficient matrix; The input vector at the kth moment is calculated by the following formula: Among them, u(k) represents the input vector at the kth moment, u(k-1) represents the input vector at the (k-1)th moment, and F(k-1) represents the error control reference at the (k-1)th moment. Represents the input vector mean matrix at the (k-1)th moment; The space adjustment data at the kth moment is determined by the following formula: x(k)=Ax(k-1)+Bu(k-1) where x(k) represents the space adjustment data at the kth moment, x(k-1) represents the space adjustment data at the (k-1)th moment, and A and B are coefficient matrices; The input vector mean matrix at the (k-1)th moment is calculated using the following formula: e(k-1)=R(k-1)-y(k-1), Δx(k-1)=x(k)-x(k-1), Among them, e(k-1) represents the error at the (k-1)th moment, R(k-1) represents the preset target output at the (k-1)th moment, and y(k-1) represents the output vector at the (k-1)th moment. Represents the input vector mean matrix at the (k-1)th moment; The error control reference at the (k-1)th moment is calculated using the following formula: Among them, F(K-1) represents the error control reference at the (k-1)th moment, H represents the first weight matrix, Q represents the second weight matrix, A and B represent coefficient matrices, Im represents the unit matrix, and C represents the ratio of the output vector to the cabin adjustment data.

2. The method according to claim 1, characterized in that The output vector at the (k-1)th moment is determined by the following process: Generate a target flight output equation according to the target historical data and the time granularity, wherein the target flight output equation is used to characterize the relationship between the time granularity and the output vector and the cabin adjustment data; The output vector at the (k-1)th moment is determined according to the target flight output equation and the cabin space adjustment data at the (k-1)th moment.

3. The method according to claim 2, characterized in that The generating of the target flight output equation according to the target historical data and the time granularity includes: Get the initial historical data within the preset time period; removing singular points in the initial historical data to generate target historical data; Generate an initial flight output equation according to the time granularity and the target historical data; The initial flight output equation is subjected to piecewise approximate fitting processing to generate a target flight output equation.

4. A real-time dynamic allocation device for flight space, characterized in that: include: A first acquisition module is used to acquire the input vector and the cabin adjustment data at the kth moment based on a preset time granularity, where the time granularity is used to represent the time interval between two moments; A determination module, used to determine the space adjustment data at the (k+1)th moment according to the space adjustment data at the kth moment, the input vector at the kth moment, a preset coefficient matrix and a preset piecewise discrete time system equation; The preset piecewise discrete time system equation is used to characterize the relationship between the cabin space adjustment data at different times, and the relationship between the output vector at the kth time and the cabin space adjustment data; A space allocation module, used for performing space allocation according to the space adjustment data at the (k+1)th moment; A first acquisition module, configured to, if the kth moment is an initial moment, the input vector at the kth moment is a preset value, and the space adjustment data at the kth moment is the preset space adjustment data; Or, if the kth moment is not the initial moment, the input vector at the kth moment is determined according to the input vector at the (k-1)th moment, the error control reference at the (k-1)th moment, and the input vector mean matrix at the (k-1)th moment; the space adjustment data at the kth moment is determined according to the space adjustment data at the (k-1)th moment, the input vector at the (k-1)th moment, and the coefficient matrix; wherein the input vector at the kth moment is calculated by the following formula: Among them, u(k) represents the input vector at the kth moment, u(k-1) represents the input vector at the (k-1)th moment, and F(k-1) represents the error control reference at the (k-1)th moment. Represents the input vector mean matrix at the (k-1)th moment; The space adjustment data at the kth moment is determined by the following formula: x(k)=Ax(k-1)+Bu(k-1) where x(k) represents the space adjustment data at the kth moment, x(k-1) represents the space adjustment data at the (k-1)th moment, and A and B are coefficient matrices; The input vector mean matrix at the (k-1)th moment is calculated using the following formula: e(k-1)=R(k-1)-y(k-1), Δx(k-1)=x(k)-x(k-1), Among them, e(k-1) represents the error at the (k-1)th moment, R(k-1) represents the preset target output at the (k-1)th moment, and y(k-1) represents the output vector at the (k-1)th moment. Represents the input vector mean matrix at the (k-1)th moment; The error control reference at the (k-1)th moment is calculated using the following formula: Among them, F(K-1) represents the error control reference at the (k-1)th moment, H represents the first weight matrix, Q represents the second weight matrix, A and B represent coefficient matrices, Im represents the unit matrix, and C represents the ratio of the output vector to the cabin adjustment data.

5. A computer device, characterized in that: include: at least one processor; And a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor executes the steps of the real-time dynamic allocation method of flight space according to any one of claims 1-3.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for real-time dynamic allocation of flight space according to any one of claims 1 to 3 are implemented.