Flight passenger volume prediction method, system, electronic device and storage medium
By constructing a transfer matrix for passenger booking increase and decrease change and iterative prediction process, combined with the Markov chain, the problem of poor prediction accuracy of flight passenger volume caused by relying on historical data in the existing technology is solved, and the prediction accuracy and current travel behavior are improved. Considerations.
Patent Information
- Application Number
- CN202410253451.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-03-06
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-03-06
AI Technical Summary
The existing flight passenger volume prediction method relies on a large amount of historical data. When the sample data is insufficient or there is a deviation, the prediction accuracy is poor, and the current passenger travel behavior of the flight to be predicted is ignored.
By obtaining the first-day passenger bookings and the change transfer matrix of passenger bookings on the target flight, the iterative prediction process is used to predict the next day's passenger bookings, and combining the Markov chain to build the change transfer matrix of passenger bookings, considering the take-off before taking off. Passenger booking data for m days predicts the number of passengers on the departure date.
It improves the prediction accuracy of flight passenger volume, reduces dependence on historical travel data, and enhances consideration of current passenger travel behavior.
Smart Images

Figure CN118228860B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of data processing technology, and in particular to a flight passenger volume prediction method, system, electronic equipment and storage medium. Background Art
[0002] International flights refer to flights from a target country to other countries or flights from other countries to the target country. As one of the most important modes of transportation between different countries, international flights play an indispensable role.
[0003] As a hub of aviation network transportation, the understanding and prediction of international flight passenger volume at airports will greatly affect the allocation of airport-related resources and decision-making, and thus affect the travel experience of passengers entering and leaving the country. When predicting flight passenger volume, a large amount of historical data and prior knowledge are usually required to build a prediction model. This prediction model is too dependent on historical data and needs to collect various relevant historical data. When the sample data in the training process is insufficient or biased, the prediction results will be less accurate.
[0004] Therefore, there is an urgent need for a flight passenger volume prediction method, system, electronic device and storage medium to solve the above problems. Summary of the invention
[0005] In view of the problems existing in the prior art, the present invention provides a flight passenger volume prediction method, system, electronic device and storage medium.
[0006] The present invention provides a method for predicting the number of flight passengers, comprising:
[0007] Obtaining the passenger booking volume on the first date corresponding to the target flight to be predicted, wherein the passenger booking volume on the first date is the passenger booking volume of the target flight on a preset date;
[0008] According to the passenger booking volume on the first date and the passenger booking change of the target flight on the day after the preset date, the passenger booking volume on the second date corresponding to the target flight is obtained through an iterative prediction process;
[0009] Determining the passenger volume of the target flight on the departure date according to the passenger booking volume on the second date;
[0010] Among them, the passenger booking change on the day after the preset date is obtained by predicting the passenger booking increase and decrease transfer matrix and the passenger booking change on the first date. The passenger booking change on the first date is the difference in booking volume between the passenger booking volume on the first date and the passenger booking volume on the day before the preset date. The passenger booking volume on the second date is the predicted passenger booking volume of the target flight on the departure date. The passenger booking increase and decrease transfer matrix is constructed by using the historical passenger booking boarding record data of the target flight.
[0011] According to a flight passenger volume prediction method provided by the present invention, the passenger booking increase and decrease change transfer matrix is constructed by the following steps:
[0012] According to the historical passenger booking record data of the target flight, the passenger booking data change amount between adjacent historical dates before the historical take-off of the target flight is obtained;
[0013] According to the passenger booking data change amount, construct a passenger booking amount change set between each adjacent historical date;
[0014] Obtain the corresponding transfer probability of passenger booking change between each adjacent historical date;
[0015] Based on the preset number of days before the historical take-off of the target flight and the Markov chain, the passenger booking increase and decrease change transfer matrix is constructed according to the transfer probability of the passenger booking change corresponding to each adjacent historical date before the historical take-off of the target flight.
[0016] According to a flight passenger volume prediction method provided by the present invention, the step of obtaining the corresponding passenger booking change transfer probability between each adjacent historical date includes:
[0017] According to the number of occurrences of the passenger booking change amount between adjacent historical dates in each of the passenger booking change sets, the corresponding passenger booking change amount transition probability between each adjacent historical date is obtained.
[0018] According to a flight passenger volume prediction method provided by the present invention, the passenger booking volume change set between each adjacent historical date is constructed according to the passenger booking data change, including:
[0019] According to the passenger booking data change amount, obtaining the maximum passenger booking data change amount and the minimum passenger booking data change amount between adjacent historical dates;
[0020] According to the maximum change amount of the passenger booking data and the minimum change amount of the passenger booking data between each adjacent historical date, the corresponding change amount of the passenger booking data is discretized to construct a set of passenger booking amount changes between each adjacent historical date.
[0021] According to a flight passenger volume prediction method provided by the present invention, the passenger booking volume on the second date corresponding to the target flight is obtained through an iterative prediction process according to the passenger booking volume on the first date and the passenger booking change of the target flight on the day after the preset date, including:
[0022] Taking the passenger booking volume of the first date as the target passenger booking volume in the current prediction process, obtaining a current time interval, wherein the current time interval is the time interval between the preset date and the departure date;
[0023] Determine a target passenger booking change amount transfer probability vector from the passenger booking increase / decrease change transfer matrix according to the passenger booking change amount on the first date and the current time interval, wherein the target passenger booking change amount transfer probability vector is a passenger booking change amount transfer probability between the preset date corresponding to the current time interval in the passenger booking increase / decrease change transfer matrix and the next day of the preset date;
[0024] Predicting the passenger booking change amount on the day following the preset date according to the passenger booking change amount on the first date and the target passenger booking change amount transition probability vector;
[0025] Obtaining the number of passenger bookings for the day following the preset date according to the number of passenger bookings for the first date and the change in passenger bookings for the day following the preset date;
[0026] The passenger booking volume on the day following the preset date is used as the target passenger booking volume in the next round of forecasting;
[0027] In each round of prediction, if the date information corresponding to the predicted passenger booking volume is the departure date, the passenger booking volume on the second date corresponding to the target flight is obtained.
[0028] According to a method for predicting flight passenger volume provided by the present invention, the method further includes:
[0029] The passenger booking volume and passenger travel volume of the target flight are collected to obtain the real-time passenger booking volume and the real-time passenger travel volume, wherein the real-time passenger travel volume is the passenger travel volume on the departure date corresponding to the real-time passenger booking volume;
[0030] The passenger booking increase / decrease change transfer matrix is updated according to the real-time passenger booking quantity and the real-time passenger travel quantity to obtain an updated passenger booking increase / decrease change transfer matrix.
[0031] According to a flight passenger volume prediction method provided by the present invention, taking the passenger booking volume of the first date as the target passenger booking volume in the current prediction process includes:
[0032] Determine whether the preset date corresponding to the passenger booking volume on the first date is the date closest to the departure date. If not, use the passenger booking volume corresponding to the date closest to the departure date as the target passenger booking volume in the current prediction process.
[0033] The present invention also provides a flight passenger volume prediction system, comprising:
[0034] A passenger booking volume acquisition module is used to acquire the passenger booking volume on the first date corresponding to the target flight to be predicted, wherein the passenger booking volume on the first date is the passenger booking volume of the target flight on a preset date;
[0035] A passenger booking volume prediction module, configured to obtain the passenger booking volume on the second date corresponding to the target flight through an iterative prediction process according to the passenger booking volume on the first date and the passenger booking change of the target flight on the day after the preset date;
[0036] A passenger travel volume determination module, configured to determine the passenger travel volume of the target flight on the departure date according to the passenger booking volume on the second date;
[0037] Among them, the passenger booking change on the day after the preset date is obtained by predicting the passenger booking increase and decrease transfer matrix and the passenger booking change on the first date. The passenger booking change on the first date is the difference in booking volume between the passenger booking volume on the first date and the passenger booking volume on the day before the preset date. The passenger booking volume on the second date is the predicted passenger booking volume of the target flight on the departure date. The passenger booking increase and decrease transfer matrix is constructed by using the historical passenger booking boarding record data of the target flight.
[0038] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a flight passenger volume prediction method as described above is implemented.
[0039] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the flight passenger volume prediction method as described in any one of the above is implemented.
[0040] The flight passenger volume prediction method, system, electronic device and storage medium provided by the present invention utilize the first-date passenger booking volume of the target flight obtained and the passenger booking change volume of the next day predicted based on the passenger booking increase and decrease change transfer matrix and the first-date passenger booking change volume, through an iterative prediction process, and finally determine the passenger travel volume of the target flight on the departure date based on the predicted passenger booking volume of the target flight on the departure date, thereby greatly improving the prediction accuracy of the flight passenger volume. BRIEF DESCRIPTION OF THE DRAWINGS
[0041] In order to more clearly illustrate the technical solutions in the present invention or 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 invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0042] Figure 1 A schematic diagram of the flow chart of the flight passenger volume prediction method provided by the present invention;
[0043] Figure 2 A schematic diagram of the structure of the flight passenger volume prediction system provided by the present invention;
[0044] Figure 3 This is a schematic structural diagram of an electronic device provided by the present invention. DETAILED DESCRIPTION
[0045] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are 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.
[0046] As the demand for air transport continues to pay attention to passenger volume forecasting, its demand forecasting theories and methods emerge in an endless stream, including regression models, gravity models, time series models, neural networks, gray models, etc. For example, in the prior art, by obtaining the historical passenger volume feature data of the product to be predicted and the date feature data on the day to be predicted, calculating the interval day, and inputting the feature data and the normalized date feature data into the trained passenger volume forecasting model, the passenger volume forecast value with specific time characteristics is obtained.
[0047] In the above method, the prediction model mainly uses the periodicity and trend characteristics of the passenger volume data from the same origin to destination (OD) or airport in the time series to predict the passenger volume of the same OD or airport in the future, which can meet the actual application needs to a certain extent. This prediction model requires a large amount of historical data and prior knowledge, as well as long-term and continuous observation data. Only by training the model can better observation results be obtained. However, the existing flight passenger volume prediction method mainly relies on the statistical characteristics of the passenger volume of past flights on specific routes, and lacks consideration of some attributes of passengers on the predicted day, such as passenger reservation data stored in the system; at the same time, the existing technology requires a lot of time to collect relevant sample data and train the model. When the sample data is insufficient or the training requirements are not met, there are problems with low prediction accuracy and efficiency.
[0048] Passenger booking data can reflect the popularity of a certain flight to a certain extent. After statistical analysis, it is found that the passenger booking data of a specific flight has certain characteristics. The closer to the flight's departure date of 3 to 5 days, the smaller the gap between the booking data and the final actual travel data. On the premise of knowing the passenger booking volume of a certain flight m days before its departure, incorporating the passenger booking data into the prediction of the final number of passengers can greatly improve the accuracy of the prediction.
[0049] Therefore, the present invention predicts and infers the actual number of passengers that will eventually take the flight based on the passenger booking data of the flight to be predicted stored in the airport system m days before departure. Different from the time series analysis and other technologies used in the prior art prediction, the present invention constructs a passenger increment transfer matrix m days before the departure date (i.e., the departure date) based on the statistical characteristics of the booking and final passenger volume, and predicts the final number of passengers in combination with the passenger booking behavior before departure, so as to solve the problem that the prior art relies too much on historical departure data and ignores the current passenger travel behavior of the flight to be predicted, resulting in poor model prediction accuracy.
[0050] Figure 1 A schematic diagram of the flow chart of the flight passenger volume prediction method provided by the present invention, such as Figure 1 As shown, the present invention provides a method for predicting the number of flight passengers, comprising:
[0051] Step 101, obtaining the passenger booking volume on the first date corresponding to the target flight to be predicted, wherein the passenger booking volume on the first date is the passenger booking volume of the target flight on a preset date.
[0052] In the present invention, the passenger booking volume on the first date corresponding to the target flight to be predicted refers to the passenger booking volume that the target flight has on a set date (usually a day before the departure date). In the air transportation industry, airlines usually open the flight booking function in advance so that passengers can book tickets in advance. The booking volume refers to the number of tickets that have been booked by passengers, which can be used to predict future seat sales and flight passenger capacity.
[0053] In the present invention, a specific date (for example, a day before the departure date) can be selected as an observation point to count the passenger bookings of the target flight on that date. At the same time, the passenger bookings of the previous day of the first date (i.e., the preset date) need to be obtained, so as to obtain the difference between the passenger bookings of the first date and the passenger bookings of the previous day of the preset date, so as to be used in the subsequent iteration step to predict the passenger volume of the target flight on the departure date. In one embodiment, based on the current prediction task, the bookings of the date closest to the departure date can be selected as much as possible, so that the subsequent prediction accuracy is higher. For example, there are passenger bookings from 10 days to 8 days before the departure date in the system, therefore, the passenger bookings of 8 days before the departure date can be selected as the passenger bookings of the first date.
[0054] Step 102, according to the passenger booking volume on the first date and the passenger booking change of the target flight on the day after the preset date, the passenger booking volume on the second date corresponding to the target flight is obtained through an iterative prediction process.
[0055] In the present invention, the passenger booking change amount on the next day of the preset date is obtained by combining the passenger booking change amount on the first date with the passenger booking increase / decrease change transfer matrix, which describes the probability of the increase or decrease of passengers from one day to another. Further, the predicted change of booking amount on the next day of the preset date is added to the booking amount on the first date (i.e., the passenger booking amount on the first date) to obtain the booking amount on the next day of the preset date, and the booking amount on the next day of the preset date is used as the observation point in the next prediction process, and the passenger booking amount on the future date corresponding to the target flight is gradually obtained through the iterative prediction process, and the passenger booking amount of the target flight on the departure date is finally predicted.
[0056] Step 103, determining the passenger volume of the target flight on the departure date according to the passenger booking volume on the second date;
[0057] Among them, the passenger booking change on the day after the preset date is obtained by predicting the passenger booking increase and decrease transfer matrix and the passenger booking change on the first date. The passenger booking change on the first date is the difference in booking volume between the passenger booking volume on the first date and the passenger booking volume on the day before the preset date. The passenger booking volume on the second date is the predicted passenger booking volume of the target flight on the departure date. The passenger booking increase and decrease transfer matrix is constructed by using the historical passenger booking boarding record data of the target flight.
[0058] In the present invention, the final number of trips of the target flight is the accumulation of passenger booking behaviors. If the passenger booking change m days before departure is available, the increase or decrease in the passenger booking volume of the next day (i.e., the previous m-1 days) can be inferred through the above steps, and then the passenger booking volume of the previous m-1 days can be obtained. Based on this data, the passenger booking volume of the previous m-2 days can be further inferred, and so on, and the passenger booking volume in the future (i.e., the day of departure) can be finally obtained as the predicted actual passenger trip volume.
[0059] The flight passenger volume prediction method provided by the present invention utilizes the first-date passenger booking volume of the target flight obtained and the passenger booking change volume of the next day predicted based on the passenger booking increase / decrease change transfer matrix and the first-date passenger booking change volume, and through an iterative prediction process, finally determines the passenger travel volume of the target flight on the departure date based on the predicted passenger booking volume of the target flight on the departure date, thereby greatly improving the prediction accuracy of the flight passenger volume.
[0060] Based on the above embodiment, the passenger booking increase and decrease change transfer matrix is constructed by the following steps:
[0061] According to the historical passenger booking record data of the target flight, the passenger booking data change amount between adjacent historical dates before the historical take-off of the target flight is obtained;
[0062] According to the passenger booking data change amount, construct a passenger booking amount change set between each adjacent historical date;
[0063] Obtain the corresponding transfer probability of passenger booking change between each adjacent historical date;
[0064] Based on the preset number of days before the historical take-off of the target flight and the Markov chain, the passenger booking increase and decrease change transfer matrix is constructed according to the transfer probability of the passenger booking change corresponding to each adjacent historical date before the historical take-off of the target flight.
[0065] In the present invention, firstly, the passenger booking record data of the predicted international flight (i.e., the target flight) stored in the airport system is obtained, and the passenger booking record data is the passenger booking volume within m days before the target flight.
[0066] Furthermore, the present invention constructs a transfer matrix of passenger booking increase and decrease from m days before the departure date of the target flight to the next day, and uses a Markov chain to predict short-term passenger booking changes. Specifically, the change in passenger booking data x between m+1 days and m days before the departure date of the target flight on different dates is calculated. m , for the target flight stored in the airport system, the change in passenger booking data for the previous m+1 days and the previous m days x m , using the set X m To express.
[0067] On the basis of the above embodiment, the step of constructing a passenger booking quantity change set between each adjacent historical date according to the passenger booking data change includes:
[0068] According to the passenger booking data change amount, obtaining the maximum passenger booking data change amount and the minimum passenger booking data change amount between adjacent historical dates;
[0069] According to the maximum change amount of the passenger booking data and the minimum change amount of the passenger booking data between each adjacent historical date, the corresponding change amount of the passenger booking data is discretized to construct a set of passenger booking amount changes between each adjacent historical date.
[0070] In the present invention, since the state of the Markov chain is a finite set, before constructing the passenger booking increase and decrease change transfer matrix, the passenger booking data change data in the above embodiment is first discretized, so that the set with continuous values is processed into a finite state set through interval processing. In the present invention, take the set X m The minimum and maximum values in represent the minimum value (i.e. the minimum change in passenger booking data) and maximum value (i.e. the maximum change in passenger booking data) of the booking change m days before departure. The width of each interval is calculated through the minimum and maximum values of the booking change (i.e. the large interval between the minimum and maximum values of the booking change m days before departure is divided), and then the boundaries of the division are determined, so that the value range of the continuous variable is divided into several intervals of equal width.
[0071] In one embodiment, in order to make the discretization process as continuous as possible, the interval interval θ is as small as possible, for example, θ can be set to 3. m After discretization, there are k m Status:
[0072]
[0073] Among them, x m(max) represents the maximum change in passenger booking data between the m+1 days before the target flight’s historical departure and the m days before. m(min) represents the minimum change in passenger booking data between the m+1 days before the target flight and the m days before its departure. Similarly, we can obtain the discretized passenger booking data change set X from the m days before to the m-1 days. m-1 , and finally obtain the set of changes in passenger booking volume between each adjacent historical date before the historical departure of the target flight.
[0074] Furthermore, the transfer probability P of the passenger booking change from m days before the historical departure of the target flight to m-1 days before the historical departure of the target flight is calculated. m In the present invention, the passenger booking increase and decrease change transfer matrix is composed of multiple passenger booking change amount transfer probabilities P m The probability of change in passenger booking transfer P m It can be obtained through statistical methods, that is, obtained by statistics of past flight booking data.
[0075] Based on the above embodiment, the step of obtaining the corresponding passenger booking change transfer probability between adjacent historical dates includes:
[0076] According to the number of occurrences of the passenger booking change amount between adjacent historical dates in each of the passenger booking change sets, the corresponding passenger booking change amount transition probability between each adjacent historical date is obtained.
[0077] In the present invention, by Indicates status The number of times it appears in the past records, that is, the number of times the passenger booking data change from m+1 days before the target flight's historical departure to m days before the departure (which can be referred to as the m-day booking change) appears; Indicates the change in passenger booking data m days before the target flight's historical departure Changes in passenger booking data transferred to the previous m-1 days Therefore, the probability of change in passenger booking transfer is It can be calculated by the following formula:
[0078]
[0079] The key to passenger volume prediction is to ensure the accuracy of the state transition probability. The more complete the historical data, the more accurate the transition probability. In this invention, based on the passenger booking data of the target flight within 3 years, a brief analysis of the historical data shows that the passenger booking behavior will show different characteristics with the number of days approaching the departure date. Therefore, in order to more accurately grasp the changes in passenger bookings under different approaching days, for each day of the target flight within the first m days of historical departure, (15≥m≥1), the corresponding state transition matrix P is constructed. m , thus obtaining the transfer matrix of passenger booking increase and decrease changes:
[0080]
[0081] Furthermore, in the actual flight passenger volume prediction process, based on the number of days between the target flight departure date and different pre-departure dates, the corresponding passenger booking increase and decrease change transfer matrix is used for prediction, and finally the passenger volume of the target flight on the departure date is obtained.
[0082] On the basis of the above embodiment, the passenger booking volume on the first date and the passenger booking change of the target flight on the day after the preset date are obtained through an iterative prediction process, including:
[0083] Taking the passenger booking volume of the first date as the target passenger booking volume in the current prediction process, obtaining a current time interval, wherein the current time interval is the time interval between the preset date and the departure date;
[0084] Determine a target passenger booking change amount transfer probability vector from the passenger booking increase / decrease change transfer matrix according to the passenger booking change amount on the first date and the current time interval, wherein the target passenger booking change amount transfer probability vector is a passenger booking change amount transfer probability between the preset date corresponding to the current time interval in the passenger booking increase / decrease change transfer matrix and the next day of the preset date;
[0085] Predicting the passenger booking change amount on the day following the preset date according to the passenger booking change amount on the first date and the target passenger booking change amount transition probability vector;
[0086] Obtaining the number of passenger bookings for the day following the preset date according to the number of passenger bookings for the first date and the change in passenger bookings for the day following the preset date;
[0087] The passenger booking volume on the day following the preset date is used as the target passenger booking volume in the next round of forecasting;
[0088] In each round of prediction, if the date information corresponding to the predicted passenger booking volume is the departure date, the passenger booking volume on the second date corresponding to the target flight is obtained.
[0089] In the present invention, the interval s between the date to be predicted (i.e., the departure date) and the current date (i.e., the preset date) is first calculated, thereby obtaining the time interval between the preset date and the departure date. Then, based on the interval s, the passenger booking volume of the first date is used as the target passenger booking volume in the current prediction process, and the passenger booking increase and decrease change transfer matrix P constructed based on the above embodiment is obtained. m , m≥s≥1, and the number of passenger reservations stored in the airport system on the current date y s (i.e. passenger bookings on the first date) and the change in passenger bookings in the previous s days x s (i.e., the change in passenger bookings on the first date), determine the corresponding state interval i and the transition probability vector For example, taking the interval s days before departure as an example, the interval s is 10 days, then select m = 10 passenger booking increase and decrease change transfer matrix P m , and based on the change in passenger bookings on the current date x s Determine the corresponding interval (e.g., determine the interval based on the value of the booking change), and then calculate the expected value to obtain the passenger booking change w on the day after the preset date. s :
[0090]
[0091] in, Indicates the change in passenger booking data from s days before the historical departure of the target flight to s-1 days before the historical departure of the target flight.
[0092] Further obtain the passenger booking volume s-1 days before the departure date Then, the passenger booking volume obtained this time before the departure date s-1 day is As the target passenger booking volume in the next round of prediction process, continue to predict the passenger booking change on day s-2, and correspondingly obtain the passenger booking volume on day s-2 before the departure date. Repeat the above steps until the passenger booking volume on the departure date is predicted, thereby obtaining the final estimated passenger volume, that is, the passenger booking volume on the second date corresponding to the target flight.
[0093] Based on the above embodiment, the method further includes:
[0094] The passenger booking volume and passenger travel volume of the target flight are collected to obtain the real-time passenger booking volume and the real-time passenger travel volume, wherein the real-time passenger travel volume is the passenger travel volume on the departure date corresponding to the real-time passenger booking volume;
[0095] The passenger booking increase / decrease change transfer matrix is updated according to the real-time passenger booking quantity and the real-time passenger travel quantity to obtain an updated passenger booking increase / decrease change transfer matrix.
[0096] In the present invention, the subsequent increase or decrease of the real-time passenger booking volume and the real-time passenger travel volume of the target flight needs to be compared with the previous historical data, and a new transition probability is calculated and updated to the corresponding position in the transition matrix. By continuously collecting the real-time passenger booking volume and the real-time passenger travel volume, and updating the passenger booking increase or decrease change transfer matrix, the prediction accuracy can be continuously improved, and the prediction result can be made more accurate.
[0097] Based on the above embodiment, the method of using the passenger booking volume on the first date as the target passenger booking volume in the current prediction process includes:
[0098] Determine whether the preset date corresponding to the passenger booking volume on the first date is the date closest to the departure date. If not, use the passenger booking volume corresponding to the date closest to the departure date as the target passenger booking volume in the current prediction process.
[0099] In the present invention, by calculating the time difference between the first date and the departure date, it is determined whether the first date is the date closest to the departure date. If the preset date corresponding to the passenger booking volume of the first date is not the date closest to the departure date, the date closest to the departure date is found, and the passenger booking volume of this date is stored in the system, and the passenger booking volume corresponding to this date is used as the target passenger booking volume in the current prediction process; if it is determined that the preset date corresponding to the passenger booking volume of the first date is the date closest to the departure date, the passenger booking volume of the first date is directly used as the target passenger booking volume in the current prediction process. The present invention ensures that when making a prediction, the actual passenger booking volume closest to the departure date is used as the target to improve the accuracy and reliability of the prediction, because the actual booking volume closer to the departure date can better reflect the current passenger demand and change trend.
[0100] The flight passenger volume prediction system provided by the present invention is described below. The flight passenger volume prediction system described below and the flight passenger volume prediction method described above can be referenced to each other.
[0101] Figure 2 The schematic diagram of the structure of the flight passenger volume prediction system provided by the present invention is as follows: Figure 2As shown, the present invention provides a flight passenger volume prediction system, including a passenger booking volume acquisition module 201, a passenger booking volume prediction module 202 and a passenger travel volume determination module 203, wherein the passenger booking volume acquisition module 201 is used to obtain the passenger booking volume of the first date corresponding to the target flight to be predicted, wherein the passenger booking volume of the first date is the passenger booking volume of the target flight on a preset date; the passenger booking volume prediction module 202 is used to obtain the passenger booking volume of the second date corresponding to the target flight through an iterative prediction process according to the passenger booking volume of the first date and the passenger booking change of the target flight on the day after the preset date; the passenger travel volume determination module 203 is used to determine the passenger travel volume of the target flight on the departure date according to the passenger booking volume of the second date;
[0102] Among them, the passenger booking change on the day after the preset date is obtained by predicting the passenger booking increase and decrease transfer matrix and the passenger booking change on the first date. The passenger booking change on the first date is the difference in booking volume between the passenger booking volume on the first date and the passenger booking volume on the day before the preset date. The passenger booking volume on the second date is the predicted passenger booking volume of the target flight on the departure date. The passenger booking increase and decrease transfer matrix is constructed by using the historical passenger booking boarding record data of the target flight.
[0103] The flight passenger volume prediction system provided by the present invention utilizes the first-date passenger booking volume of the target flight obtained and the passenger booking change volume of the next day predicted based on the passenger booking increase / decrease change transfer matrix and the first-date passenger booking change volume, and through an iterative prediction process, finally determines the passenger travel volume of the target flight on the departure date based on the predicted passenger booking volume of the target flight on the departure date, thereby greatly improving the prediction accuracy of the flight passenger volume.
[0104] The system provided by the present invention is used to execute the above-mentioned method embodiments. Please refer to the above-mentioned embodiments for the specific processes and detailed contents, which will not be repeated here.
[0105] Figure 3 A schematic diagram of the structure of an electronic device provided by the present invention, such as Figure 3As shown, the electronic device may include: a processor (Processor) 301, a communication interface (Communications Interface) 302, a memory (Memory) 303 and a communication bus 304, wherein the processor 301, the communication interface 302, and the memory 303 communicate with each other through the communication bus 304. The processor 301 can call the logic instructions in the memory 303 to execute the flight passenger volume prediction method, which includes: obtaining the passenger booking volume on the first date corresponding to the target flight to be predicted, wherein the passenger booking volume on the first date is the passenger booking volume of the target flight on the preset date; according to the passenger booking volume on the first date and the passenger booking change of the target flight on the day after the preset date, through an iterative prediction process, obtaining the passenger booking volume on the second date corresponding to the target flight; according to the passenger booking volume on the second date, determining the passenger travel volume of the target flight on the departure date; wherein the passenger booking change on the day after the preset date is obtained by predicting the passenger booking increase and decrease change transfer matrix and the passenger booking change on the first date, the passenger booking change on the first date is the booking volume difference between the passenger booking volume on the first date and the passenger booking volume on the day before the preset date, the passenger booking volume on the second date is the predicted passenger booking volume of the target flight on the departure date, and the passenger booking increase and decrease change transfer matrix is constructed by the historical passenger booking flight record data of the target flight.
[0106] In addition, the logic instructions in the above-mentioned memory 303 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.
[0107] On the other hand, the present invention also provides a computer program product, the computer program product includes a computer program stored on a non-transitory computer-readable storage medium, the computer program includes program instructions, when the program instructions are executed by the computer, the computer can execute the flight passenger volume prediction method provided by the above methods, the method comprising: obtaining the passenger booking volume on the first date corresponding to the target flight to be predicted, wherein the passenger booking volume on the first date is the passenger booking volume of the target flight on a preset date; according to the passenger booking volume on the first date and the passenger booking change of the target flight on the day after the preset date, obtaining the passenger booking volume corresponding to the target flight through an iterative prediction process. the number of passenger bookings on the second date; determining the number of passengers scheduled for the target flight on the departure date based on the number of passenger bookings on the second date; wherein the change in passenger bookings on the day following the preset date is predicted by a passenger booking increase / decrease transfer matrix and a passenger booking change on the first date, the passenger booking change on the first date being the difference in bookings between the passenger bookings on the first date and the passenger bookings on the day before the preset date, the passenger bookings on the second date being the predicted number of passenger bookings on the target flight on the departure date, and the passenger booking increase / decrease transfer matrix being constructed by the historical passenger booking boarding record data of the target flight.
[0108] On the other hand, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it is implemented to execute the flight passenger volume prediction method provided by the above embodiments, the method comprising: obtaining the passenger booking volume of the first date corresponding to the target flight to be predicted, wherein the passenger booking volume of the first date is the passenger booking volume of the target flight on the preset date; according to the passenger booking volume of the first date and the passenger booking change of the target flight on the next day after the preset date, through an iterative prediction process, obtaining the passenger booking volume of the second date corresponding to the target flight; according to the passenger booking volume of the second date, determining the passenger travel volume of the target flight on the departure date; wherein the passenger booking change on the next day after the preset date is obtained by predicting the passenger booking increase and decrease change transfer matrix and the passenger booking change on the first date, the passenger booking change on the first date is the booking volume difference between the passenger booking volume of the first date and the passenger booking volume on the previous day of the preset date, the passenger booking volume of the second date is the predicted passenger booking volume of the target flight on the departure date, and the passenger booking increase and decrease change transfer matrix is constructed by the historical passenger booking flight record data of the target flight.
[0109] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.
[0110] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.
[0111] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention 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 invention.
Claims
1. A method for predicting flight passenger volume, It is characterized in that include: Obtaining the passenger booking volume on the first date corresponding to the target flight to be predicted, wherein the passenger booking volume on the first date is the passenger booking volume of the target flight on a preset date; According to the passenger booking volume on the first date and the passenger booking change of the target flight on the day after the preset date, the passenger booking volume on the second date corresponding to the target flight is obtained through an iterative prediction process; Determining the passenger volume of the target flight on the departure date according to the passenger booking volume on the second date; The passenger booking change amount on the day after the preset date is obtained by predicting the passenger booking increase / decrease change transfer matrix and the passenger booking change amount on the first date, the passenger booking change amount on the first date is the booking amount difference between the passenger booking amount on the first date and the passenger booking amount on the day before the preset date, the passenger booking amount on the second date is the predicted passenger booking amount of the target flight on the departure date, and the passenger booking increase / decrease change transfer matrix is constructed by the historical passenger booking flight record data of the target flight; The passenger booking increase and decrease change transfer matrix is constructed by the following steps: According to the historical passenger booking record data of the target flight, the passenger booking data change amount between adjacent historical dates before the historical take-off of the target flight is obtained; According to the passenger booking data change amount, construct a passenger booking amount change set between each adjacent historical date; Obtain the corresponding transfer probability of passenger booking change between each adjacent historical date; Based on the preset number of days before the historical departure of the target flight and the Markov chain, the passenger booking increase and decrease change transfer matrix is constructed according to the corresponding passenger booking change transfer probability between each adjacent historical date before the historical departure of the target flight; The step of obtaining the passenger booking volume on the second date corresponding to the target flight through an iterative prediction process according to the passenger booking volume on the first date and the passenger booking change of the target flight on the day after the preset date includes: Taking the passenger booking volume of the first date as the target passenger booking volume in the current prediction process, obtaining a current time interval, wherein the current time interval is the time interval between the preset date and the departure date; Determine a target passenger booking change amount transfer probability vector from the passenger booking increase / decrease change transfer matrix according to the passenger booking change amount on the first date and the current time interval, wherein the target passenger booking change amount transfer probability vector is a passenger booking change amount transfer probability between the preset date corresponding to the current time interval in the passenger booking increase / decrease change transfer matrix and the next day of the preset date; Predicting the passenger booking change amount on the day following the preset date according to the passenger booking change amount on the first date and the target passenger booking change amount transition probability vector; Obtaining the number of passenger bookings for the day following the preset date according to the number of passenger bookings for the first date and the change in passenger bookings for the day following the preset date; The passenger booking volume on the day following the preset date is used as the target passenger booking volume in the next round of forecasting; In each round of prediction, if the date information corresponding to the predicted passenger booking volume is the departure date, the passenger booking volume on the second date corresponding to the target flight is obtained.
2. The flight passenger volume prediction method according to claim 1, It is characterized in that The step of obtaining the corresponding passenger booking change transfer probability between each adjacent historical date includes: According to the number of occurrences of the passenger booking change amount between adjacent historical dates in each of the passenger booking change sets, the corresponding passenger booking change amount transition probability between each adjacent historical date is obtained.
3. The flight passenger volume prediction method according to claim 1, It is characterized in that The step of constructing a passenger booking quantity change set between each adjacent historical date according to the passenger booking data change includes: According to the passenger booking data change amount, obtaining the maximum passenger booking data change amount and the minimum passenger booking data change amount between adjacent historical dates; According to the maximum change amount of the passenger booking data and the minimum change amount of the passenger booking data between each adjacent historical date, the corresponding change amount of the passenger booking data is discretized to construct a set of passenger booking amount changes between each adjacent historical date.
4. The flight passenger volume prediction method according to claim 1, It is characterized in that The method further comprises: The passenger booking volume and passenger travel volume of the target flight are collected to obtain the real-time passenger booking volume and the real-time passenger travel volume, wherein the real-time passenger travel volume is the passenger travel volume on the departure date corresponding to the real-time passenger booking volume; The passenger booking increase / decrease change transfer matrix is updated according to the real-time passenger booking quantity and the real-time passenger travel quantity to obtain an updated passenger booking increase / decrease change transfer matrix.
5. The flight passenger volume prediction method according to claim 1, It is characterized in that The step of taking the passenger booking volume on the first date as the target passenger booking volume in the current prediction process includes: Determine whether the preset date corresponding to the passenger booking volume on the first date is the date closest to the departure date. If not, use the passenger booking volume corresponding to the date closest to the departure date as the target passenger booking volume in the current prediction process.
6. A flight passenger volume prediction system, It is characterized in that include: A passenger booking volume acquisition module is used to acquire the passenger booking volume on the first date corresponding to the target flight to be predicted, wherein the passenger booking volume on the first date is the passenger booking volume of the target flight on a preset date; A passenger booking volume prediction module, configured to obtain the passenger booking volume on the second date corresponding to the target flight through an iterative prediction process according to the passenger booking volume on the first date and the passenger booking change of the target flight on the day after the preset date; A passenger travel volume determination module, configured to determine the passenger travel volume of the target flight on the departure date according to the passenger booking volume on the second date; The passenger booking change amount on the day after the preset date is obtained by predicting the passenger booking increase / decrease change transfer matrix and the passenger booking change amount on the first date, the passenger booking change amount on the first date is the booking amount difference between the passenger booking amount on the first date and the passenger booking amount on the day before the preset date, the passenger booking amount on the second date is the predicted passenger booking amount of the target flight on the departure date, and the passenger booking increase / decrease change transfer matrix is constructed by the historical passenger booking flight record data of the target flight; The passenger booking increase and decrease change transfer matrix is constructed by the following steps: According to the historical passenger booking record data of the target flight, the passenger booking data change amount between adjacent historical dates before the historical take-off of the target flight is obtained; According to the passenger booking data change amount, construct a passenger booking amount change set between each adjacent historical date; Obtain the corresponding transfer probability of passenger booking change between each adjacent historical date; Based on the preset number of days before the historical departure of the target flight and the Markov chain, the passenger booking increase and decrease change transfer matrix is constructed according to the corresponding passenger booking change transfer probability between each adjacent historical date before the historical departure of the target flight; The passenger booking volume prediction module is specifically used for: Taking the passenger booking volume of the first date as the target passenger booking volume in the current prediction process, obtaining a current time interval, wherein the current time interval is the time interval between the preset date and the departure date; Determine a target passenger booking change amount transfer probability vector from the passenger booking increase / decrease change transfer matrix according to the passenger booking change amount on the first date and the current time interval, wherein the target passenger booking change amount transfer probability vector is a passenger booking change amount transfer probability between the preset date corresponding to the current time interval in the passenger booking increase / decrease change transfer matrix and the next day of the preset date; Predicting the passenger booking change amount on the day following the preset date according to the passenger booking change amount on the first date and the target passenger booking change amount transition probability vector; Obtaining the number of passenger bookings for the day following the preset date according to the number of passenger bookings for the first date and the change in passenger bookings for the day following the preset date; The passenger booking volume on the day following the preset date is used as the target passenger booking volume in the next round of forecasting; In each round of prediction, if the date information corresponding to the predicted passenger booking volume is the departure date, the passenger booking volume on the second date corresponding to the target flight is obtained.
7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the computer program, the flight passenger volume prediction method according to any one of claims 1 to 5 is implemented.
8. A non-transitory computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the flight passenger volume prediction method according to any one of claims 1 to 5 is implemented.
