Flight reservation willingness payment prediction method and device, equipment and medium

Abstract

The invention provides a flight seat reservation willingness payment prediction method and device, equipment and a medium, and the method comprises the steps: firstly obtaining and sorting historical seat reservation data of a target flight segment, the historical seat reservation data comprising passenger data of all shared flights of the target flight segment; designing a log-likelihood function about price sensitivity based on an MNL customer selection model and historical seat reservation data, and performing maximum likelihood estimation calculation based on a result calculated by the log-likelihood function; for a given open shipping space set and pricing, predicting a purchase probability of a customer for each shipping space; therefore, according to the method for estimating and predicting the customer willing payment of the flight reservation seat provided by the scheme, the influence of the flight price on the reservation seat quantity can be considered, the value is provided for the subsequent flight pricing decision of an airline company by predicting the reservation seat demand, and the method is simple and effective.

Description

Technical Field

[0001] The present invention belongs to the technical field of information data processing, and in particular relates to a method, device, equipment and medium for predicting willingness to pay for flight bookings. Background Art

[0002] Air transport refers to a mode of transportation that uses aircraft such as jet planes or propeller planes (referred to as aircraft) as the main means of transport to transport passengers, cargo or mail in the sky. Usually, people need to build a series of facilities such as terminal buildings, runways, navigation equipment, communication equipment, security inspection procedures, etc. at the airport in order to take a civil aircraft or use aviation services. The emergence of airplanes has changed people's travel methods and greatly promoted the economic and social development of mankind, making the entire earth connected into a big family, known as the "global village". At present, for airlines, in order to maximize profits, they need to reasonably set ticket prices. How to effectively estimate passengers' willingness to pay for flights is the main basis for airlines to set prices.

[0003] Estimating the demand of airline booking customers is the basis for airlines to optimize flight pricing and increase flight revenue. Currently, most of the methods used are based on independent estimation and prediction of the booking demand for each flight cabin. There are mainly the following problems: first, the impact of flight price factors on the number of bookings is not taken into account; second, the forecast results have low auxiliary value for subsequent pricing decisions; and finally, the forecast results are poorly adaptable to changes in the market environment. Summary of the invention

[0004] In view of the above problems, the present disclosure proposes a method, device, equipment and medium for predicting willingness to pay for flight bookings, which can effectively consider the impact of flight prices on the number of bookings, characterize customer booking demand in the form of customer purchase probability under a given arbitrary flight seat set and seat pricing, thereby predicting booking demand and providing value for the airline's subsequent flight pricing decisions.

[0005] In order to achieve the above objectives, the present disclosure provides the following technical solutions:

[0006] A method for predicting willingness to pay for flight bookings, comprising:

[0007] Acquire and organize historical booking data of the target flight segment, wherein the historical booking data includes boarding data of all co-flying flights of the target flight segment;

[0008] Based on the MNL customer choice model and historical booking data, a log-likelihood function of price sensitivity is designed to obtain parameters α and β.

[0009] Based on the results calculated by the log-likelihood function, the maximum likelihood estimation calculation is performed for the parameters α and β;

[0010] Given a set of open cabins and pricing, predict the probability of a customer purchasing each cabin.

[0011] Optionally, the acquiring and arranging the historical booking data of the target flight segment includes:

[0012] Get historical booking data of airlines;

[0013] Determining the historical booking data within a preset time range as the historical booking data of the target flight segment;

[0014] Organize all open cabins and corresponding prices for the target flight segment and enter the data.

[0015] Optionally, the acquisition and organization of the historical booking data of the target flight segment also includes related data, and the related data is the cabin space and cabin price of each historically open flight during the booking period.

[0016] Optionally, the log-likelihood function defines a parameter X jt The specified variable is 1 to indicate that cabin j is selected, and 0 to indicate that cabin j is not selected.

[0017] Optionally, the log-likelihood function is designed based on the freight rate search system. Since parameter α is the basic utility vector of each space and parameter β is the other attribute utility coefficient vector of each space, the parameters α, β and X are used to calculate the log-likelihood function. jt So as to determine the corresponding cabin price, { p j | j∈S}, thus calculating the main parameters of customer willingness to pay on the target segment of this method.

[0018] Optionally, the maximum likelihood estimation optimizes the calculation of parameters α and β, and the calculation result is used as the parameter of the customer's willingness to pay for the target flight segment. Parameter θ is determined based on a formula and parameter θ is used to predict the probability of future customers purchasing space.

[0019] Optionally, based on the parameters α, β, β0 obtained after optimization calculation, obtain a given open space set S and the corresponding space price { p j | j∈S} and obtain the probability of future customers purchasing any cabin through the probability formula.

[0020] A device for estimating and predicting the willingness to pay of flight booking customers, comprising:

[0021] An acquisition module, used to acquire the historical booking data of the airline, wherein the historical booking data includes the boarding data of all co-flying flights of the target flight segment;

[0022] A determination module, used to determine the historical booking data of a preset time as the historical booking data of the target flight segment;

[0023] The calculation module is used to calculate the probability of future customers purchasing space based on a probability formula.

[0024] Based on the same inventive concept, another aspect of the embodiment of the present disclosure further provides an electronic device, comprising: a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; the memory stores a computer program;

[0025] The processor implements a method for predicting willingness to pay for flight booking when executing the program stored in the memory.

[0026] Based on the same inventive concept, the embodiment of the present disclosure further provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements a method for predicting willingness to pay for flight bookings.

[0027] The above one or at least one technical solution in the embodiments of the present disclosure has at least the following technical effects: Compared with the prior art, the present disclosure provides a prediction method, device, equipment and medium for flight booking willingness to pay. The prediction method for flight booking willingness to pay first obtains and organizes the historical booking data of the target flight segment, and the historical booking data includes the boarding data of all co-flying flights of the target flight segment; based on the MNL customer selection model and the historical booking data, a log-likelihood function about price sensitivity is designed, and the maximum likelihood estimation calculation is performed based on the result calculated by the log-likelihood function; for a given open class set and pricing, the customer's purchase probability for each class is predicted. It can be seen that the estimation and prediction method provided by this solution can effectively consider the impact of flight prices on the number of bookings, and characterize the customer's booking demand in the form of the customer's purchase probability under a given arbitrary flight class set and class pricing, thereby predicting the specific results of the booking demand under different market environments, and providing value for the airline's subsequent flight pricing decisions.

[0028] Other features and advantages of the present disclosure will be described in the following description, and partly become apparent from the description, or be understood by implementing the present disclosure. The purpose and other advantages of the present disclosure can be realized and obtained by the structures indicated in the description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0030] Figure 1 A flowchart of a method for predicting willingness to pay for flight booking disclosed in the present invention;

[0031] Figure 2 A schematic diagram of an electronic device disclosed in the present invention. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0033] It should be noted that the terms "first", "second" etc. in the present application are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged in appropriate circumstances, so that the embodiments of the present application described here. In the present application, the orientation or positional relationship indicated by the terms "upper", "lower", "left", "right", "front", "back", "top", "bottom", "inside", "outside", "middle", "vertical", "horizontal", "lateral", "longitudinal" etc. are based on the orientation or positional relationship shown in the accompanying drawings.

[0034] As mentioned in the background technology, in order to maximize profits, it is necessary to reasonably set air ticket prices. How to effectively estimate passengers' willingness to pay for flights is the main basis for airlines to set prices. Traditional booking demand estimation and prediction generally removes restrictions on the booking quantity trends of different DCP flights, and then predicts the future booking quantity based on the changes in the booking quantity curve after removal of restrictions. The quantitative prediction model is divided into three categories: time series model, causal model and artificial intelligence model. Most of the above methods are based on the independent estimation and prediction of the booking demand for each flight cabin, and there are some problems that have not been solved:

[0035] First, the impact of flight prices on the number of bookings is not taken into account. The number of bookings is the result of the demand for bookings under the flight price and the set of available flights. Simply predicting the number of future bookings based on the number of historical bookings ignores the impact of flight prices and the flights opened by competitors on the number of bookings.

[0036] Second, the prediction results of traditional methods have low auxiliary value for subsequent pricing decisions. The prediction results of the number of bookings only provide the trend of the number of bookings in the future. Airlines can only estimate the scale of future demand through this result, but cannot specifically quantify the changes in the number of bookings with different pricing.

[0037] Third, the forecast results are poorly adaptable to changes in the market environment. The forecast of the number of bookings is generally based on static historical data. When the aviation market changes, such as changes in the number of shared flights or flight pricing, the forecast results are difficult to adjust and reflect in a timely manner.

[0038] In addition, demand estimation is mainly based on the assumption that product demand is independent of each other. Under this assumption, the demand flows between different products do not affect each other, thus ignoring the substitutability between products. In order to better characterize the diversity of customer demand and the richness of decision-making scenarios, this paper proposes an estimation and prediction method for flight booking customer willingness to pay.

[0039] (Example 1)

[0040] In view of these problems, this embodiment provides a method for estimating and predicting the willingness to pay of flight booking customers, including:

[0041] S11: Acquire historical booking data of the target flight segment, wherein the historical booking data includes boarding data of all co-flying flights of the target flight segment;

[0042] S12: sorting historical booking data of the target flight segment;

[0043] S2: Based on the MNL customer choice model and historical booking data, design the log-likelihood function of price sensitivity and obtain its parameters;

[0044] S3: Based on the results calculated by the log-likelihood function, the maximum likelihood estimation is performed on the parameters;

[0045] S4: For a given set of open cabins and pricing, predict the customer’s purchase probability for each cabin.

[0046] Among them, S11 is to obtain and organize the historical booking data of the target flight segment, including: obtaining its historical booking related data, the passenger boarding data of all co-flying flights of the flight segment, the flight space conditions and space prices historically opened for each DCP during the booking period; further, for the selection of departure date data, it is necessary to consider factors such as holidays, tourist seasons, and conference events in the cities at the beginning and end of the flight segment; for the data sets of these special departure dates, they can be performed separately.

[0047] S12: sorting out the historical booking data of the target flight segment of the airline; determining the historical booking data within a preset time range as the historical booking data of the target flight segment; sorting out all open classes and corresponding prices of the target flight segment, and entering the data; sorting out all open classes and corresponding prices of the node where the booking record occurs, and sorting out the historical booking data according to each booking record, including PNR, AV, ticket sales data and other data. After sorting, the main information of each booking record should at least include the booking class, booking price, booking time, and all other classes and prices open at the booking time; after sorting, the data is saved and entered into the database.

[0048] The historical booking data obtained includes different prices and numbers of passengers in different classes.

[0049] In addition, the above steps also include: verification to ensure the completeness, accuracy and timeliness of the data.

[0050] After obtaining the specific flights for sale, key information can be further extracted: price, airline type, departure time, refund and change policy, mileage points and other attribute information.

[0051] S2: Based on the MNL (MultinomialLogit) customer choice model, a log-likelihood function of price sensitivity is designed for the freight rate search system.

[0052] In the fare search results, the utility of the prices of the same flights for users is β0p jt , which is the product of price sensitivity and price. Given an open slot set S and the corresponding slot price {p j | j∈S}, the probability model of cabin selection is the utility of the selected cabin price to the customer divided by the sum of the utility of all cabin prices to the customer

[0053]

[0054] Define indicator variable I tj A value of 1 indicates that cabin j is selected, and a value of 0 indicates that cabin j is not selected.

[0055]

[0056] α represents the basic utility vector of each cabin;

[0057] β represents the utility coefficient vector of other attributes of each space;

[0058] β0 represents price sensitivity;

[0059] S t represents the set of flight seats available on the booking record t;

[0060] X jt A vector representing other attribute values ​​(such as departure time, refund and change) of cabin j in booking record t;

[0061] I tj Indicates whether the class purchased in booking record t is class j, which is a Boolean variable;

[0062] p jt represents the price of cabin j at the time of historical record t;

[0063] In this formula, t represents a historical booking record, including the set of available seats S when the booking occurs. t , buy a certain space j∈S t The indicator function I tj , and the basic utility vector α and price sensitivity β0 of each class. Based on the historical booking data set T, the data of each historical booking record t∈T is substituted into the above formula to complete the preparation for the optimization calculation of the subsequent step S4. Among them, α, β, β0 are unknown parameters, which are also the estimation targets of this method. The following are the main fields: carrier, flight number, departure airport, landing airport, departure date, ticket purchase class, ticket purchase date, ticket purchase time, departure time, flight segment identification, kilometers, published price (yuan), whether the remaining class 1 is open, whether the remaining class 2 is open, whether the remaining class 3 is open, etc.

[0064] S3: Perform maximum likelihood estimation of the log-likelihood function.

[0065] Based on the preparation results of S2, in this step, the maximum likelihood estimation method will be used to optimize the unknown parameters α, β and store the calculation results. The calculation results will be used as the main parameters of the customer's willingness to pay on this flight segment and used to predict the probability of customers purchasing space in the future.

[0066]

[0067] S4: Given a set of open cabins and pricing, predict the probability of a customer purchasing each cabin.

[0068] Based on the parameter estimation results α, β, β0 of S3, for customers arriving in the future, among all the shared flight results in the fare search, for a given open space set S and the corresponding space price {p j |j∈S}, according to the probability formula in step S3, the probability of a future customer purchasing any cabin j, j∈S in the shopping results can be calculated.

[0069] The present invention provides a method for estimating the willingness to pay of flight booking customers. Based on the impact of historical flight prices and cabin availability on the number of bookings, the maximum likelihood estimation algorithm is used to estimate the price sensitivity of customers through the data set of cabin availability and specific prices faced by each customer booking behavior in history, thereby more accurately and scientifically describing the customer's booking demand; and being able to adapt to the changing civil aviation transportation market environment: the estimation result of this method can predict the probability of a customer purchasing each cabin under any open flight set and cabin price, so as to be able to respond immediately to the changing aviation market, and at the same time effectively provide rich and detailed information input for pricing decisions, thereby providing airlines with more scientific and quantitative pricing guidance.

[0070] It should be noted that the basic principle of the model in the above method is: MNL (Muller-Normalized Linear Model, MLN): the main effect size is equal to the product of the benchmark envelope and the air carrier price sensitivity index.

[0071] P(x; i; c): represents the probability that the customer chooses the i-th cabin, x represents the customer's characteristics, c represents the cabin structure; α: intercept term; β: slope term; x: customer characteristic indicators, such as an array of seat location, departure time and other indicators; w: weighted vector; e: error term; λ: weight vector; similarly, p, q, r...u represent different compensation coefficients, u represents the cabin to which the user belongs; p: cardinality; t: time; q: days; r: hours; Let θ = (αβwλe)∈R be the parameter to be estimated, Θ is the matrix form of a normal vector, then the objective function is expressed as follows: In the above log-likelihood function, the dependent variable is the observed variable, and the explanatory variables include not only the decision variables (cabin structure variables), but also some variables that can affect the customer's ability to pay and the relationship between the variables.

[0072] Therefore, explanatory variables can be divided into two categories: decision variables and state variables.

[0073] State variables are variables that determine consumers' product choice behavior. There are two main categories: (1) exogenous state variables - these are variables that reflect the external environment or the behavior of others; (2) endogenous state variables - these are variables that reflect the individual's own characteristics.

[0074] Endogenous state variables can be divided into two types: (a) those that can be explained as functions of exogenous variables; and (b) residual endogenous state variables - those that remain after the part that can be explained indirectly by the exogenous state variables.

[0075] (c) Exogenous explanatory variables are variables controlled by the decision makers themselves. Obviously, the analysis of dependent variables is very important in decision science. Accurate measurement of dependent variables can improve the accuracy and statistical efficiency of predictions, and has an important impact on the effectiveness and stability of decisions.

[0076] The obtained historical booking data includes multiple different passenger types.

[0077] The log-likelihood function is designed based on a fare search system, including: defining an indicator variable as 1 to indicate that cabin j is selected and as 0 to indicate that cabin j is not selected; defining parameters as a basic utility vector for each cabin, parameters as other attribute utility coefficient vectors for each cabin, and price sensitivity; and calculating the main parameters of customer willingness to pay on the target segment of the method based on a formula.

[0078] Among them, the log-likelihood function is a generalized linear model, whose main feature is that it can handle nonlinear regression problems; and the maximum likelihood estimation is a common and effective method for parameter estimation of statistical data.

[0079] The steps for using maximum likelihood estimation are generally as follows: (1) setting the sample space Ω; (2) setting the probability density function L(θ) of the random variable; (3) setting the value range of the parameter θ to be estimated; (4) generating samples based on the data; (5) estimating the value of θ based on the samples; in the present application, step (5) is the maximum likelihood estimation calculation process derived from the above formula.

[0080] In addition, (6) determines whether the convergence theorem is satisfied. If so, the parameter estimate is obtained; otherwise, new samples are generated.

[0081] The maximum likelihood estimation optimizes the calculation of parameters and parameters, and the calculation result is used as the parameter of the customer's willingness to pay for the target flight segment, which is used to predict the probability of future customers purchasing space.

[0082] The steps further include: based on the parameters obtained after the optimization calculation, obtaining a given open cabin set S and the corresponding cabin prices and obtaining the probability of a future customer purchasing any cabin through a probability formula.

[0083] In the log-likelihood function, the log probability is written as follows: The basic idea of ​​the maximum likelihood estimation method is to regard the possibility of various situations occurring in the sample (i.e., the value of the random variable) as the cause of the occurrence of the sample, and then use the frequency of occurrence of the sample (i.e., the number of sampling times) as the probability of occurrence of each cause to infer the possibility and parameters of each cause.

[0084] Based on the results obtained from the log-likelihood function, the maximum likelihood estimation is calculated for the parameters and parameters.

[0085] Different constraints will be given for different businesses. For example, if the total sales of a certain cabin does not exceed 70%, there will be a constraint of not selling out the cabin.

[0086] The partial derivative of the parameter is finally solved as a binary function using Newton's algorithm (using the gradient steepest descent method, the closer the starting point is to the target value, the faster it is).

[0087] Given a set of open cabins and pricing, predict the probability of a customer purchasing each cabin.

[0088] The parameters obtained through the above steps are used to calculate the probability of future customers' willingness to pay for different cabins.

[0089] Therefore, personalized pricing strategies can be implemented according to the characteristics of customers, which can increase the speed of cabin sales and improve customer satisfaction.

[0090] (Example 2)

[0091] This embodiment provides a device for estimating and predicting the willingness to pay of flight booking customers, including:

[0092] An acquisition module, used to acquire the historical booking data of the airline, wherein the historical booking data includes the boarding data of all co-flying flights of the target flight segment;

[0093] A determination module, used to determine the historical booking data of a preset time as the historical booking data of the target flight segment;

[0094] The calculation module is used to calculate the probability of future customers purchasing space based on a probability formula.

[0095] (Example 3)

[0096] Furthermore, a specific example of a specific target segment in embodiment 1 of the present disclosure during actual operation.

[0097] In this specific case, the target segment is CA001, and the goal is to estimate and predict the ticket purchasing intention of passengers on this target segment.

[0098] First, historical booking data is retrieved from the airline database, and the passenger boarding data of all co-operating flights in the target segment are obtained and organized based on the records of passengers taking actual carrier flights from January 1, 2019 to December 31, 2019; then, according to the settings of the MNL customer choice model, a log-likelihood function suitable for this estimation problem is designed: Next, the maximum likelihood estimation method is used to perform maximum likelihood estimation calculations on parameters and parameters, and the log-likelihood function is obtained and further used to calculate the purchase probability of customers on the predicted day according to the different prices of each cabin class, thereby obtaining the probability of customer purchase intention in different cabin classes.

[0099] At the same time, the system can also provide ticket prices for multiple classes and the corresponding number of remaining seats based on the travel history of passengers on the same route during the same period in the past, so that airlines can formulate ticket prices and provide guidance to passengers on seat selection.

[0100] Based on the same inventive concept, the present disclosure also provides an electronic device 161, see Figure 2 , including a processor 164, a communication interface 165, a memory 162 and a communication bus, wherein the processor 164, the communication interface 165 and the memory 162 communicate with each other through the communication bus;

[0101] A memory 162 storing a computer program 163;

[0102] The processor 164 implements the track circuit direction switching method when executing the program stored in the memory 162.

[0103] The above communication bus may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus may be divided into an address bus, a data bus, a control bus, etc.

[0104] The communication interface 165 is used for communication between the electronic device 161 and other devices.

[0105] The memory 162 may include a random access memory 162 (RAM), or may include a non-volatile memory 162 (non-volatile memory), such as at least one disk memory 162. Optionally, the memory 162 may also be at least one storage device located away from the processor 164.

[0106] The above-mentioned processor 164 can be a general-purpose processor 164, including a central processing unit 164 (CPU), a network processor 164 (NP), etc.; it can also be a digital signal processor 164 (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.

[0107] Based on the same inventive concept, the embodiment of the present disclosure further provides a computer-readable storage medium storing a computer program 163 , which implements the method for predicting willingness to pay for flight booking when executed by a processor 164 .

[0108] The computer-readable storage medium may be included in the device / apparatus described in the above embodiment; or it may exist independently without being assembled into the device / apparatus. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, a method for predicting willingness to pay for flight booking according to an embodiment of the present disclosure is implemented.

[0109] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for predicting willingness to pay for flight booking, characterized in that: include: Acquire and organize historical booking data of the target flight segment, wherein the historical booking data includes boarding data of all co-flying flights of the target flight segment; Obtain parameters α and β, and design a log-likelihood function about price sensitivity based on historical booking data; Based on the results calculated by the log-likelihood function, the maximum likelihood estimation calculation is performed for the parameters α and β; Given a set of open cabins and pricing, predict the probability of a customer purchasing each cabin.

2. The method according to claim 1, characterized in that: The obtaining and collating of historical booking data of the target flight segment includes: Get historical booking data of airlines; Determining the historical booking data within a preset time range as the historical booking data of the target flight segment; Organize all open cabins and corresponding prices for the target flight segment and enter the data.

3. The method according to claim 2, characterized in that The acquisition and arrangement of the historical booking data of the target flight segment also includes related data, and the related data is the cabin space and cabin price of each historically open flight during the booking period.

4. The method according to claim 1, characterized in that: The log-likelihood function is defined by the parameter X jt The specified variable is 1 to indicate that cabin j is selected, and 0 to indicate that cabin j is not selected.

5. The method according to claim 4, characterized in that The log-likelihood function is designed based on the freight rate search system, where the parameters α, β and X are used to calculate the log-likelihood function. jt Determine the cabin price, i.e. p j |j∈S}, the main parameters of the customer's willingness to pay on the target flight segment are determined based on the seat price.

6. The method according to claim 5, characterized in that The method further comprises: performing optimization calculation on the parameter α and the parameter β by using maximum likelihood estimation, and using the calculation result as the parameter of the customer's willingness to pay for the target flight segment, and determining the parameter θ and using the parameter θ to predict the probability of future customers purchasing space.

7. The method according to claim 6, characterized in that The method further includes: based on the parameters α, β, β0 obtained after optimization calculation; obtaining a given open space set S and a corresponding space price {p j |j∈S} and obtain the probability of future customers purchasing any cabin through the probability formula.

8. A device for estimating and predicting the willingness to pay of flight booking customers, characterized in that: include: An acquisition module, used to acquire the historical booking data of the airline, wherein the historical booking data includes the boarding data of all co-flying flights of the target flight segment; A determination module, used to determine the historical booking data of a preset time as the historical booking data of the target flight segment; The calculation module is used to calculate the probability of future customers purchasing space based on a probability formula.

9. An electronic device, characterized in that: include: A processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; the memory stores a computer program; The processor, when executing the program stored in the memory, implements the method for predicting the willingness to pay for flight booking as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a processor, the method for predicting the willingness to pay for flight booking according to any one of claims 1 to 7 is implemented.

Images