Heartbeat type dynamic air ticket pricing method and system based on route search popularity index

Abstract

The invention discloses a heartbeat type dynamic airline ticket pricing method and system based on an airline search popularity index, and the method comprises the steps: collecting the search times of a related airline path by path according to a preset search path at a preset interval time interval, and the preset search path comprises but is not limited to a first path and a second path; obtaining the total number of search times in a preset time period path by path from a historical search database according to preset search paths, and calculating a mean value of the search times of each preset search path; calculating a search popularity index corresponding to each preset search path based on a mean value of search times of each preset search path and the search times, and calculating a weighted search popularity index I based on the search popularity index corresponding to each preset search path and a preset weight; according to the heartbeat type dynamic air ticket pricing method and system based on the airline search popularity index, dynamic heartbeat type air ticket pricing adjustment is realized, so that benefit maximization is ensured according to the real-time sales condition.

Description

Technical Field

[0001] The present invention relates to the technical field of air ticket pricing, and in particular to a heartbeat-type dynamic air ticket pricing method and system based on a route search heat index. Background Art

[0002] In the airline ticketing market, traditional ticket pricing methods are usually based on fixed cost plus or seasonal demand adjustments, and lack the ability to respond to real-time market dynamics. With the development of Internet technology, more and more users search for flight information through online platforms, and these search behaviors reflect the market demand for specific routes. However, the existing pricing system fails to make full use of these real-time data to dynamically adjust ticket prices, resulting in airlines being unable to maximize their profits. Summary of the invention

[0003] In order to solve the technical problems existing in the background technology, the present invention proposes a heartbeat-type dynamic air ticket pricing method and system based on the route search heat index.

[0004] The heartbeat-type dynamic air ticket pricing method based on route search heat index proposed by the present invention comprises the following steps:

[0005] S1. Collecting the number of searches for relevant routes path by path at every preset interval according to a preset search path, wherein the preset search path includes but is not limited to a first path and a second path;

[0006] S2. Obtaining the total number of searches within a preset time period from the historical search database according to the preset search paths one by one, and calculating the average number of searches for each preset search path;

[0007] S3, calculating the search heat index corresponding to each preset search path based on the average of the search times of each preset search path and the search times, and calculating the weighted search heat index I based on the search heat index corresponding to each preset search path and the preset weight;

[0008] S4. Obtain a price adjustment threshold X, and compare the price adjustment threshold X with the weighted search heat index I to obtain a price adjustment direction, which includes price increase and price reduction;

[0009] S5. Obtain a price adjustment sensitivity coefficient k, and calculate a price adjustment range Y based on the price adjustment sensitivity coefficient k, the weighted search heat index I, and the price adjustment threshold X;

[0010] S6. Obtain the basic fare and adjust the fare within a preset price adjustment period based on the basic fare, price adjustment range Y, and price adjustment direction.

[0011] Preferably, the first path specifically collects the number of searches S1 for related routes from the OTA platform at preset intervals, and the second path specifically collects the number of searches S2 for related routes from the airline's official website at preset intervals.

[0012] Preferably, step S2 specifically includes:

[0013] Obtain the total number of searches within a preset time period S from the historical search database according to the first path 总1 , and calculate the mean of the first path search times

[0014] The average number of first path searches The calculation process is as follows:

[0015]

[0016] Among them, S 总1 Obtaining the total number of searches within a preset time period for the first path; nT is the preset time period, T is a single period, and n is the number of periods;

[0017] Obtain the total number of searches S within a preset time period from the historical search database according to the second path 总2 , and calculate the mean of the second path search times

[0018] The average number of second path searches The calculation process is as follows:

[0019]

[0020] Among them, S 总2 The total number of searches within a preset time period is obtained for the second path; nT is the preset time period, T is a single period, and n is the number of periods.

[0021] Preferably, the preset weights include a weight ω1 corresponding to the first path and a weight ω2 corresponding to the second path; step S3 specifically includes:

[0022] When the preset search path is specifically the first path, the corresponding search heat index I1 calculation process is as follows:

[0023]

[0024] Among them, S 总1 Obtaining a total number of searches within a preset time period for the first path; is the mean of the number of searches for the first path;

[0025] When the preset search path is specifically the second path, the corresponding search heat index I2 calculation process is as follows:

[0026]

[0027] Among them, S 总2 Obtaining a total number of searches within a preset time period for the second path; is the mean of the number of second path searches;

[0028] The calculation process of weighted search heat index I is as follows:

[0029] I=ω1I1+ω2I2;

[0030] Among them, I1 is the search heat index corresponding to the first path; I2 is the search heat index corresponding to the second path; ω1 is the weight corresponding to the first path; ω2 is the weight corresponding to the second path.

[0031] Preferably, step S4 specifically includes:

[0032] Obtain a price adjustment threshold X, and compare the price adjustment threshold X with the weighted search popularity index I;

[0033] When the weighted search heat index I is greater than the price adjustment threshold X, the price adjustment direction is specifically a price increase;

[0034] When the weighted search heat index I is less than the price adjustment threshold X, the price adjustment direction is specifically a price reduction.

[0035] Preferably, the calculation process of the price adjustment range Y is as follows:

[0036] Y = k*(IX);

[0037] Among them, k is the price adjustment sensitivity coefficient; I is the weighted search heat index; X is the price adjustment threshold.

[0038] Preferably, step S6 specifically includes:

[0039] Get the basic fare P base , based on the basic fare P base , calculate the adjusted fare P by the price adjustment range Y new ;

[0040] Adjusted fare P new The calculation process is as follows:

[0041] P new =P base +Y;

[0042] Among them, P base is the basic fare; Y is the price adjustment range;

[0043] According to the adjusted fare P within the preset price adjustment period new Carry out corresponding flight ticket sales.

[0044] The heartbeat-type dynamic air ticket pricing system based on route search heat index proposed by the present invention comprises:

[0045] A data collection module, used for collecting the number of searches of related routes path by path at every preset interval according to a preset search path, wherein the preset search path includes but is not limited to a first path and a second path;

[0046] A first processing module, used to obtain the total number of searches within a preset time period from a historical search database according to the preset search paths one by one, and calculate the average of the number of searches for each preset search path;

[0047] The second processing module is used to calculate the search heat index corresponding to each preset search path based on the average of the search times of each preset search path and the search times, and calculate the weighted search heat index I based on the search heat index corresponding to each preset search path and the preset weight;

[0048] The third processing module is used to obtain a price adjustment threshold X, and compare the price adjustment threshold X with the weighted search heat index I to obtain a price adjustment direction, which includes price increase and price reduction;

[0049] The fourth processing module is used to obtain the price adjustment sensitivity coefficient k, and calculate the price adjustment range Y according to the price adjustment sensitivity coefficient k, the weighted search heat index I, and the price adjustment threshold X;

[0050] The price adjustment module is used to obtain the basic fare and adjust the fare within a preset price adjustment period based on the basic fare, price adjustment range Y, and price adjustment direction.

[0051] In the present invention, the proposed heartbeat-type dynamic air ticket pricing method and system based on the route search heat index collects the number of searches of relevant routes path by path at each preset interval period according to the preset search path; obtains the total number of searches within a preset time period from the historical search database according to the preset search path path by path, and calculates the average number of searches for each preset search path; calculates the search heat index corresponding to each preset search path based on the average number of searches for each preset search path and the number of searches, and calculates the weighted search heat index I based on the search heat index corresponding to each preset search path and the preset weight; obtains the price adjustment threshold X, compares the price adjustment threshold X with the weighted search heat index I to obtain the price adjustment direction, which includes price increase and price reduction; obtains the price adjustment sensitivity coefficient k, and calculates the price adjustment range Y according to the price adjustment sensitivity coefficient k, the weighted search heat index I, and the price adjustment threshold X; obtains the basic fare and adjusts the fare within the preset price adjustment period based on the basic fare, the price adjustment range Y, and the price adjustment direction. Dynamic heartbeat-type adjustment of air ticket pricing is achieved to ensure maximum revenue according to real-time sales conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is a schematic diagram of the work flow structure of the heartbeat-type dynamic air ticket pricing method based on the route search heat index proposed by the present invention;

[0053] Figure 2 This is a schematic diagram of the system architecture of the heartbeat-type dynamic air ticket pricing system based on the route search heat index proposed by the present invention. DETAILED DESCRIPTION

[0054] Reference Figure 1 and Figure 2 The present invention proposes a heartbeat-type dynamic air ticket pricing method based on route search heat index, comprising the following steps:

[0055] S1. Collecting the number of searches for related routes path by path at preset intervals according to a preset search path, where the preset search path includes but is not limited to a first path and a second path.

[0056] In this embodiment, the first path is specifically collecting the search times S1 of related routes from the OTA platform at every preset interval, and the second path is specifically collecting the search times S2 of related routes from the airline's official website at every preset interval. The OTA platform is specifically an online travel agency platform.

[0057] In this embodiment, each preset interval period may be specifically every 30 minutes, that is, the number of searches for specific routes by passengers is collected from OTAs and airline websites every 30 minutes.

[0058] S2. Obtain the total number of searches within a preset time period from the historical search database according to the preset search paths one by one, and calculate the average of the number of searches for each preset search path.

[0059] In this embodiment, step S2 specifically includes:

[0060] Obtain the total number of searches within a preset time period S from the historical search database according to the first path 总1 , and calculate the mean of the first path search times

[0061] The average number of first path searches The calculation process is as follows:

[0062]

[0063] Among them, S 总1 Obtaining the total number of searches within a preset time period for the first path; nT is the preset time period, T is a single period, and n is the number of periods;

[0064] Obtain the total number of searches S within a preset time period from the historical search database according to the second path 总2 , and calculate the mean of the second path search times

[0065] The average number of second path searches The calculation process is as follows:

[0066]

[0067] Among them, S 总2 The total number of searches within a preset time period is obtained for the second path; nT is the preset time period, T is a single period, and n is the number of periods.

[0068] S3. Calculate the search heat index corresponding to each preset search path based on the average of the search times of each preset search path and the search times, and calculate the weighted search heat index I based on the search heat index corresponding to each preset search path and the preset weight.

[0069] In this embodiment, the preset weights include a weight ω1 corresponding to the first path and a weight ω2 corresponding to the second path; step S3 specifically includes:

[0070] When the preset search path is specifically the first path, the corresponding search heat index I1 calculation process is as follows:

[0071]

[0072] Among them, S 总1Obtaining a total number of searches within a preset time period for the first path; is the mean of the number of searches for the first path;

[0073] When the preset search path is specifically the second path, the corresponding search heat index I2 calculation process is as follows:

[0074]

[0075] Among them, S 总2 Obtaining a total number of searches within a preset time period for the second path; is the mean of the number of second path searches;

[0076] The calculation process of weighted search heat index I is as follows:

[0077] I=ω1I1+ω2I2;

[0078] Among them, I1 is the search heat index corresponding to the first path; I2 is the search heat index corresponding to the second path; ω1 is the weight corresponding to the first path; ω2 is the weight corresponding to the second path.

[0079] In this embodiment, if the preset weights ω1=0.6 and ω2=0.4, then I=0.6I1+0.4I2.

[0080] S4. Obtain a price adjustment threshold X, and compare the price adjustment threshold X with the weighted search heat index I to obtain a price adjustment direction, which includes a price increase or a price decrease.

[0081] In this embodiment, step S4 specifically includes:

[0082] Obtain a price adjustment threshold X, and compare the price adjustment threshold X with the weighted search popularity index I;

[0083] When the weighted search heat index I is greater than the price adjustment threshold X, the price adjustment direction is specifically a price increase;

[0084] When the weighted search heat index I is less than the price adjustment threshold X, the price adjustment direction is specifically a price reduction.

[0085] In this embodiment, the process of generating the price adjustment threshold X is as follows:

[0086] Calculate the average of all weighted search heat index I in the past 180 days Calculate the standard deviation δ of all weighted search heat index I in the past 180 days. The calculation process of standard deviation δ is as follows:

[0087]

[0088] Among them, μ is the mean of all weighted search heat indexes I.

[0089] The coefficient λ is determined, for example, λ=1.5 to represent that the threshold value X should be higher than the mean value by 1.5 standard deviations.

[0090] The coefficient λ is used to determine the degree of deviation of the price adjustment threshold X from the average value. It can be determined based on market volatility, historical data fluctuation range, and business risk preference. For example, if you want the price adjustment threshold to be relatively conservative, you can choose a smaller value; if you want to respond more actively to market changes, you can choose a larger value.

[0091] The specific calculation process of the price adjustment threshold X is as follows:

[0092] in, is the average value of all weighted search heat indexes I; λ is the coefficient; δ is the standard deviation of all weighted search heat indexes I.

[0093] In this embodiment, the specific process of determining a reasonable price adjustment threshold value X by adjusting the coefficient λ value is as follows:

[0094] Sort all weighted search heat index I data in the past 180 days from small to large, determine the 90% quantile position, and find the 90% quantile value;

[0095] If the 90% quantile location refers to an exact data point in the data set, that value is used directly; if the location falls between two data points, the value of the 90% quantile is determined by linear interpolation.

[0096] If the current value of λ=1.5 is used, according to the formula Calculate the price adjustment threshold X;

[0097] Compare the calculated X value with the 90% quantile value to determine whether X is within the 90% quantile range.

[0098] If X is within the 90% quantile range, it means that λ=1.5 and the corresponding X obtained based on the expert judgment method are both reasonable values. At this time, the finalized λ and X can be implemented in the dynamic pricing system, and its impact on fare adjustment and sales can be continuously monitored.

[0099] If X is not within the 90% quantile range, it means that the λ=1.5 and the corresponding X obtained based on the expert judgment method are unreasonable values. The pricing method is not implemented or cannot be implemented effectively. The λ value needs to be re-determined.

[0100] S5. Obtain a price adjustment sensitivity coefficient k, and calculate a price adjustment range Y according to the price adjustment sensitivity coefficient k, a weighted search heat index I, and a price adjustment threshold X.

[0101] In this embodiment, the calculation process of the price adjustment range Y is as follows:

[0102] Y = k*(IX);

[0103] Among them, k is the price adjustment sensitivity coefficient; I is the weighted search heat index; X is the price adjustment threshold.

[0104] In this embodiment, the process of setting the price adjustment sensitivity coefficient k is as follows:

[0105] Step 1: Dataset construction and preprocessing:

[0106] Construct a data set D = {(I i ,P i )|i=1,...,N}, where I i is the search popularity index of the ith observation, P i is the corresponding ticket price, N is the number of samples;

[0107] Normalize the dataset: and Where μ′ and δ′ are the mean and standard deviation, respectively.

[0108] Step 2: Neural network model establishment and training:

[0109] Establish a neural network model with an input layer, a hidden layer, and an output layer. Assume that the input layer has n nodes, corresponding to n features (for example, the standardized search heat index).

[0110] Train a model to predict ticket prices: Where f is the neural network function and θ is the model parameter.

[0111] Step 3: Determination of price adjustment sensitivity coefficient k:

[0112] The sensitivity of the model to the search heat index is used as an estimate of the value of k. In the simplest linear case, k can be obtained by the model output layer about The derivative of is approximated as: in It is the average value of the normalized search popularity index.

[0113] S6. Obtain the basic fare and adjust the fare within a preset price adjustment period based on the basic fare, price adjustment range Y, and price adjustment direction.

[0114] In this embodiment, step S6 specifically includes:

[0115] Get the basic fare P base , based on the basic fare P base , calculate the adjusted fare P by the price adjustment range Y new ; Adjusted fare Pnew The calculation process is as follows:

[0116] P new =P base +Y;

[0117] Among them, P base is the basic fare; Y is the price adjustment range;

[0118] According to the adjusted fare P within the preset price adjustment period new Carry out corresponding flight ticket sales.

[0119] Reference Figure 1 and Figure 2 The present invention proposes a heartbeat-type dynamic air ticket pricing system based on route search heat index, comprising:

[0120] A data collection module, used for collecting the number of searches of related routes path by path at every preset interval according to a preset search path, wherein the preset search path includes but is not limited to a first path and a second path;

[0121] A first processing module, used to obtain the total number of searches within a preset time period from a historical search database according to the preset search paths one by one, and calculate the average of the number of searches for each preset search path;

[0122] The second processing module is used to calculate the search heat index corresponding to each preset search path based on the average of the search times of each preset search path and the search times, and calculate the weighted search heat index I based on the search heat index corresponding to each preset search path and the preset weight;

[0123] The third processing module is used to obtain a price adjustment threshold X, and compare the price adjustment threshold X with the weighted search heat index I to obtain a price adjustment direction, which includes price increase and price reduction;

[0124] The fourth processing module is used to obtain the price adjustment sensitivity coefficient k, and calculate the price adjustment range Y according to the price adjustment sensitivity coefficient k, the weighted search heat index I, and the price adjustment threshold X;

[0125] The price adjustment module is used to obtain the basic fare and adjust the fare within a preset price adjustment period based on the basic fare, price adjustment range Y, and price adjustment direction.

[0126] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A heartbeat-type dynamic air ticket pricing method based on route search popularity index, characterized in that: The following steps are involved: S1. Collecting the number of searches for relevant routes path by path at every preset interval according to a preset search path, where the preset search path includes but is not limited to a first path and a second path; S2. Obtaining the total number of searches within a preset time period from the historical search database according to the preset search paths one by one, and calculating the average number of searches for each preset search path; S3, calculating the search heat index corresponding to each preset search path based on the average of the search times of each preset search path and the search times, and calculating the weighted search heat index I based on the search heat index corresponding to each preset search path and the preset weight; S4. Obtain a price adjustment threshold X, and compare the price adjustment threshold X with the weighted search heat index I to obtain a price adjustment direction, which includes price increase and price reduction; S5. Obtain a price adjustment sensitivity coefficient k, and calculate a price adjustment range Y based on the price adjustment sensitivity coefficient k, a weighted search heat index I, and a price adjustment threshold value X; S6. Obtain the basic fare and adjust the fare within a preset price adjustment period based on the basic fare, price adjustment range Y, and price adjustment direction.

2. The heartbeat-type dynamic air ticket pricing method based on route search heat index according to claim 1 is characterized in that: The first path specifically collects the number of searches for related routes S1 from the OTA platform at each preset interval, and the second path specifically collects the number of searches for related routes S2 from the airline's official website at each preset interval.

3. The heartbeat-type dynamic air ticket pricing method based on route search heat index according to claim 2 is characterized in that: Step S2 specifically includes: Obtain the total number of searches within a preset time period S from the historical search database according to the first path 总1 , and calculate the mean of the first path search times The average number of first path searches The calculation process is as follows: Among them, S 总1 Obtaining the total number of searches within a preset time period for the first path; nT is the preset time period, T is a single period, and n is the number of periods; Obtain the total number of searches S within a preset time period from the historical search database according to the second path 总2 , and calculate the mean of the second path search times The average number of second path searches The calculation process is as follows: Among them, S 总2 The total number of searches within a preset time period is obtained for the second path; nT is the preset time period, T is a single period, and n is the number of periods.

4. The heartbeat-type dynamic air ticket pricing method based on route search heat index according to claim 3 is characterized in that: The preset weights include the weight ω1 corresponding to the first path and the weight ω2 corresponding to the second path; step S3 specifically includes: When the preset search path is specifically the first path, the corresponding search heat index I1 calculation process is as follows: Among them, S 总1 Obtaining a total number of searches within a preset time period for the first path; is the mean of the number of searches for the first path; When the preset search path is specifically the second path, the corresponding search heat index I2 calculation process is as follows: Among them, S 总2 Obtaining a total number of searches within a preset time period for the second path; is the mean of the number of second path searches; The calculation process of weighted search heat index I is as follows: I=ω1I1+ω2I2; Among them, I1 is the search heat index corresponding to the first path; I2 is the search heat index corresponding to the second path; ω1 is the weight corresponding to the first path; ω2 is the weight corresponding to the second path.

5. The heartbeat-type dynamic air ticket pricing method based on route search heat index according to claim 1 is characterized in that: Step S4 specifically includes: Obtain a price adjustment threshold X, and compare the price adjustment threshold X with the weighted search popularity index I; When the weighted search heat index I is greater than the price adjustment threshold X, the price adjustment direction is specifically a price increase; When the weighted search heat index I is less than the price adjustment threshold X, the price adjustment direction is specifically a price reduction.

6. The heartbeat-type dynamic air ticket pricing method based on route search heat index according to claim 1 is characterized in that: The calculation process of the price adjustment range Y is as follows: Y = k*(IX); Among them, k is the price adjustment sensitivity coefficient; I is the weighted search heat index; X is the price adjustment threshold.

7. The heartbeat-type dynamic air ticket pricing method based on route search heat index according to claim 1 is characterized in that: Step S6 specifically includes: Get the basic fare P base , based on the basic fare P base , calculate the adjusted fare P by the price adjustment range Y new ; Adjusted fare P new The calculation process is as follows: P new =P base +Y; Among them, P base is the basic fare; Y is the price adjustment range; According to the adjusted fare P within the preset price adjustment period new Carry out corresponding flight ticket sales.

8. A heartbeat-type dynamic air ticket pricing system based on route search popularity index, characterized in that: include: A data collection module, used for collecting the number of searches of related routes path by path at every preset interval according to a preset search path, wherein the preset search path includes but is not limited to a first path and a second path; A first processing module, used to obtain the total number of searches within a preset time period from a historical search database according to the preset search paths one by one, and calculate the average of the number of searches for each preset search path; The second processing module is used to calculate the search heat index corresponding to each preset search path based on the average of the search times of each preset search path and the search times, and calculate the weighted search heat index I based on the search heat index corresponding to each preset search path and the preset weight; The third processing module is used to obtain a price adjustment threshold X, and compare the price adjustment threshold X with the weighted search heat index I to obtain a price adjustment direction, which includes price increase and price reduction; The fourth processing module is used to obtain the price adjustment sensitivity coefficient k, and calculate the price adjustment range Y according to the price adjustment sensitivity coefficient k, the weighted search heat index I, and the price adjustment threshold X; The price adjustment module is used to obtain the basic fare and adjust the fare within a preset price adjustment period based on the basic fare, price adjustment range Y, and price adjustment direction.

Images