Passenger selection model method and system based on passenger preference, and electronic equipment

Abstract

The invention discloses a passenger selection model method and system based on passenger preferences and electronic equipment, and belongs to the technical field of airline income management. The method mainly comprises the following steps: acquiring historical big data of passenger flow volume of a certain journey in a typical week period, wherein the historical big data comprises a journey starting place, a destination, first, second and third flights (takeoff and landing airports, takeoff and landing time and models), ticket prices, shipping spaces and passenger numbers; obtaining passenger flow volume ratio curve data of all time from the starting place to the destination in a typical week; dynamic parameters such as deviation days and deviation minutes and static parameters such as flight hours, models, service types and passenger preference attributes in the travel attributes are cleaned and sorted; establishing a passenger utility function by using a linear combination of the dynamic parameter and the static parameter of the travel; a new logit model and a passenger selection model are constructed by using a travel utility function and a passenger flow volume ratio at a typical week time point, the flight market share and the number of passengers on a seat are accurately predicted, and the defects of difficulty in data collection, high cost and poor accuracy in a questionnaire mode are overcome.

Description

[0001] This invention is funded by the Civil Aviation Joint Fund of the National Natural Science Foundation of China (U2233214) and the Civil Aviation Safety Capacity Building Project (Civil Aviation Administration

[2024] No. 9). Technical Field

[0002] This application relates to the technical field of airline revenue management, and particularly to a passenger choice model method, system, and electronic device based on passenger preferences. Background Art

[0003] It is very important for airlines to predict flight market share and passenger load, and it is also one of the difficult tasks for airlines. Due to the complexity of the route network and the diversity of passenger travel behaviors, it is extremely difficult for airlines to predict passenger load. Therefore, to design a reasonable flight schedule, it is necessary to first study passenger travel behaviors, and establishing a passenger choice itinerary model is the top priority. The decision-making process of passenger choice of itinerary is affected by various factors, mainly including ticket price, time, airline brand and service, aircraft type, itinerary transfer, competition, comfort and convenience, etc. To master passenger travel patterns, airlines often use a subjective passenger boarding questionnaire in practice, and use statistical data to reflect passengers' stated preferences. However, the disadvantages are obvious, such as difficult data collection, high cost, and poor accuracy. This patent establishes a passenger choice itinerary model, adopts a nested logit model with stratified travel time, and constructs a nested logit model using key attributes that affect passenger travel, including dynamic and static attributes, so as to reflect the revealed preferences of passengers' choice of itinerary through historical big data, and can scientifically reveal its patterns and characteristics, thereby accurately predicting flight market share and passenger load. Summary of the Invention

[0004] Embodiments of this application provide a passenger choice model method, system, and electronic device based on passenger preferences to solve the problems existing in related technologies.

[0005] In the first aspect, this application proposes a passenger choice model method based on passenger preferences, which is characterized by including:

[0006] Step S1, select a typical week in a flight season, and obtain historical big data of passenger flow during the typical week for a certain itinerary, including the origin O, destination D, the first flight (departure and arrival airports, departure and arrival times, aircraft type), the second flight (departure and arrival airports, departure and arrival times, aircraft type), the third flight (departure and arrival airports, departure and arrival times, aircraft type), ticket price including cabin class and number of passengers;

[0007] Step S2, obtain the data of the passenger flow proportion curve of TOW (Time of Week) at the origin O and destination D in the typical week;

[0008] Step S3: Clean and sort the dynamic parameters such as deviation days and deviation minutes, and the static parameters such as flight hours, aircraft type, service type (direct flight, transfer, and code sharing), and passenger preference attributes (departure airport, direct flight rate, airline, alliance, and fare) in the attributes of the passenger's selected itinerary.

[0009] Step S4: Establish a passenger utility function using a linear combination of the dynamic and static parameters of the itinerary.

[0010] Step S5: Use the itinerary utility function and the proportion of passenger flow at the corresponding time point of TOW to construct a nested logit model to calculate the market occupancy rate of the itinerary, thereby constructing a passenger itinerary selection model.

[0011] In a second aspect, an embodiment of the present application provides a passenger selection model system based on passenger preferences, including:

[0012] An itinerary acquisition module for selecting a typical week in a flight season and obtaining historical data on passenger flow during the typical week for a certain passenger itinerary, including the itinerary origin O, destination D, the first flight (departure and arrival airports, departure and arrival times, aircraft type), the second flight (departure and arrival airports, departure and arrival times, aircraft type), the third flight (departure and arrival airports, departure and arrival times, aircraft type), and the fare including the cabin class and the number of passengers.

[0013] A TOW acquisition module for obtaining the curve data of the proportion of passenger flow of TOW (Time of Week) at the origin O and destination D of the typical week.

[0014] A parameter cleaning module for cleaning and sorting the dynamic parameters such as deviation days and deviation minutes, and the static parameters such as flight hours, aircraft type, service type (direct flight, transfer, and code sharing), and passenger preference attributes (departure airport, direct flight rate, airline, alliance, and fare) in the attributes of the passenger's selected itinerary.

[0015] A utility function construction module for establishing a passenger utility function using a linear combination of the dynamic and static parameters of the itinerary.

[0016] A passenger selection model module for using the itinerary utility function and the proportion of passenger flow at the corresponding time point of TOW to construct a nested logit model to calculate the market occupancy rate of the itinerary, thereby constructing a passenger itinerary selection model.

[0017] In a third aspect, an embodiment of the present application provides an electronic device, including a memory, a processor, and a computer program stored on the memory. The processor implements the method described in any one of the above when executing the computer program.

[0018] Compared with the prior art, the present application has the following advantages:

[0019] According to the embodiments of the present application, the decision-making process for passengers to select a travel itinerary is affected by various factors, mainly including ticket price, time, airline brand and service, aircraft type, itinerary transfer, competition, comfort and convenience, etc. Through historical data analysis, this patent establishes a passenger travel itinerary selection model, adopting a nested logit model with stratified travel time, and constructs the nested logit model using the key attributes affecting passenger travel, including dynamic attributes and static attributes, so as to reveal the laws and characteristics of passengers' travel itinerary selection through historical big data, thereby accurately predicting the flight market share and the number of passengers on board, and overcoming the disadvantages of difficult data collection, high cost, and poor accuracy in revealing statistical laws by using the method of passenger boarding questionnaires.

[0020] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the following specifically gives the specific implementation manners of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In the drawings, unless otherwise specified, the same reference numerals throughout the several views refer to the same or like parts or elements. These drawings are not necessarily drawn to scale. It should be understood that these drawings only depict some embodiments according to the present application and should not be regarded as limiting the scope of the present application.

[0022] Figure 1 is a flowchart showing a method for a passenger selection model based on passenger preferences according to an embodiment of the present application;

[0023] Figure 2 is a typical weekly time curve TOW of North China NCN - West China WCN according to an embodiment of the present application;

[0024] Figure 3 is a nested Nested Logit model diagram nested by departure time according to an embodiment of the present application;

[0025] Figure 4 is a structural diagram of a passenger selection model system based on passenger preferences according to an embodiment of the present application; and

[0026] Figure 5 is a block diagram of an electronic device for a passenger selection model based on passenger preferences according to an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0027] In the following text, only some exemplary embodiments are briefly described. As those skilled in the art can recognize, the described embodiments can be modified in various different ways without departing from the concept or scope of the present application. Therefore, the drawings and the description are considered to be exemplary in nature and not restrictive.

[0028] To facilitate the understanding of the technical solutions of the embodiments of the present application, the related technologies of the embodiments of the present application are described below. The following related technologies can be arbitrarily combined with the technical solutions of the embodiments of the present application as optional solutions, and they all fall within the protection scope of the embodiments of the present application.

[0029] An embodiment of the present application provides a passenger selection model method 100 based on passenger preferences. The passenger selection model method 100 based on passenger preferences can be executed on a computer device, especially on a computing device of an airline, or can be executed in the cloud through cloud computing. No matter which computer entity it is executed on, it can be shared among all airlines through uploading and sharing. The following will refer to Figure 1 to describe a passenger selection model method 100 based on passenger preferences of the present application. Figure 1 is a flowchart showing the passenger selection model method based on passenger preferences according to an embodiment of the present application. As Figure 1 shown, the passenger selection model method 100 based on passenger preferences includes the following steps S1 to S5. The following will describe steps S1 to S5 in combination with specific embodiments.

[0030] First, enter step S1. In step S1, a typical week of a flight season is selected, and big data on historical passenger flow during the typical week for a certain itinerary is obtained, including the origin O, destination D, the first flight (departure and arrival airports, departure and arrival times, aircraft type), the second flight (departure and arrival airports, departure and arrival times, aircraft type), the third flight (departure and arrival airports, departure and arrival times, aircraft type), ticket price, class of service, and the number of passengers.

[0031] Furthermore, the typical week is the week that is closest to the average number of passengers per week in a flight season of an airline. The typical week is selected, and its flight and passenger data are used as historical data for training the parameters of the passenger selection model. A passenger itinerary is segmented when the interval between two itineraries exceeds 12 hours, and the origin O and destination D of the itinerary, i.e., the itinerary O&D, constitute a market.

[0032] The annual flight schedules of airlines reflect the characteristics of weather conditions and tourism demands in the four seasons of spring, summer, autumn, and winter. For example, the increase in ski tourism and winter vacations in winter and spring, and the peak seasons for business activities and vacations in summer and autumn throughout the year. Airlines will reasonably arrange flight schedules, aircraft maintenance plans, and crew flight, training, and vacations according to the characteristics of the four seasons. Airlines generally follow the flight season division rules of the International Civil Aviation Organization (ICAO). The year is divided into two flight seasons: winter-spring and summer-autumn. The switching times between winter-spring and summer-autumn are fixed. The switching time for the winter-spring flight schedule is from the last Sunday in October of each year to the last Saturday in March of the following year, with a duration of 5 months, generally the off-season of the whole year; the switching time for summer-autumn is from the last Sunday in March of each year to the last Saturday in October of the following year, lasting for 7 months, generally the peak season of the whole year, which is the key profit-making period for airlines. Airlines formulate flight schedules for the two seasons to meet the travel needs of passengers and optimize the allocation of airline resources.

[0033] Due to the actual operating characteristics of airlines and the travel characteristics of passengers throughout the year, airlines regularly release flight plans or flight schedules. Flight schedules generally reflect the weekly periodicity of flight operations. Therefore, the long-term plans of airlines generally feature weekly plan cycles. The plans and decisions of airlines are based on weekly flight and passenger data. Therefore, it is necessary to reasonably select representative typical weeks, and the passenger data of which has the characteristics of representing flight seasons.

[0034] The historical travel data of passengers can reflect the preferences of passengers' travel choices, and it is also data that airlines can easily obtain for themselves. However, it is very difficult to obtain all the data of global airlines. Generally, data from global MIDT and GDS are purchased for airlines' decision-making.

[0035] The data sources of this patent are the post-departure data PDD, revenue settlement data RA, flight plan data, OAG foreign airline flight plan data, MIDT, GDS, etc. of this airline.

[0036] Next, enter step S2. In step S2, obtain the data of the passenger flow proportion curve of TOW (Time of Week) on the origin O and destination D of the typical week;

[0037] Furthermore, according to the airline flight schedule, count the percentage of all passenger flows from Sunday to Saturday in the O&D market of the typical week, and smooth the demand at discrete time points (24 * 7 = 168) into continuous time demand TOW curve, which is also the preferred departure time of passengers, also known as the weekly time curve, indicating the probability that passengers choose to depart at a specific time within a week.

[0038] In an embodiment, the itinerary O&D data is the summary of passengers of all global airlines in this O&D market and the proportion at 168 whole points, such as Figure 2As shown, the horizontal axis represents the whole-hour time points of a week \(t = 0, 1, 2,\cdots, 167\), and the vertical axis represents the passenger flow proportion \(p(0), p(1), p(2),\cdots, p(167)\), which respectively represent the passenger distribution probabilities at 0:00 on Sunday, 01:00 on Sunday, 02:00 on Sunday,\(\cdots\), 23:00 on Saturday. It satisfies The TOW passenger flow proportion curve reflects the actual distribution of the passenger flow in a typical week. When calculating the market share, using the weekly time curve as the weight of the departure time and predicting the share of the itinerary or market based on this is a scientific and feasible method.

[0039] Next, enter step S3. In step S3, clean and sort the dynamic parameters such as the deviation days and deviation minutes and the static parameters such as flight hours, aircraft types, service types (non-stop, connecting, and code sharing), and passenger preference attributes (departure airport, non-stop rate, airline, alliance, and fare) in the attributes of the passengers' selected itineraries.

[0040] Furthermore, among the passenger preference attributes, the itinerary attributes that provide many different choices for passengers, such as departure time, departure date, arrival time, and arrival date, are defined as dynamic attributes, which are determined by the passenger flow proportion of the TOW curve. The linear combination of the dynamic attribute parameters is the passenger dynamic utility function, and the specific attributes include but are not limited to the day deviation DD (Displacement Day) and the minute deviation DM (Displacement Minute).

[0041] Furthermore, the invariant attributes related to the flight are defined as passenger static attributes, and the linear combination of the static attribute parameters is the passenger static utility function. The specific attributes include but are not limited to the itinerary flight hours Elasped Time, aircraft type AircraftType, service type Service Type (non-stop, connection, and code sharing), and passenger preference attributes Preference (departure airport OPP, non-stop rate nonstop ratio, airline airline, alliance alliance, and fare relatedprice).

[0042] In one embodiment, the dynamic attribute parameters are attributes that depend on the Time of Week (TOW). These attributes are classified as Dynamic Utility (DU) items. Airlines cannot offer flights with departure or arrival times that all passengers want, resulting in time deviations, including the number of days off (DD) and the number of minutes off (DM). Passengers will preferentially choose their preferred flights with smaller deviation degrees among all alternative flights. These dynamic attributes directly affect passengers' decisions to choose itineraries. This dynamic parameter is more obvious for time-sensitive business and commercial passengers. Due to business and commercial activity arrangements, passengers must arrive at the destination city before a certain time, and the dynamic attributes play a major role.

[0043] Furthermore, the dynamic attribute of the number of days off (DD, Displacement Day) measures the number of days between the passenger's preferred travel day and the actual flight day.

[0044] The dynamic attribute of the number of minutes off (DM, Displacement Minute) measures the number of hours between the passenger's preferred travel time and the actual flight time.

[0045] In one embodiment, the factors that affect passengers' choice of itinerary also include those itinerary attributes that remain unchanged in the model and are independent of the TOW time t. Therefore, they are classified as Static Utility (SU) items. Specific static attributes include, but are not limited to, the elapsed flight hours of the itinerary, service type (non-stop, connection, and code share), equipment type, passenger preference attributes (origin airport OPP, non-stop ratio, airline, alliance, and fare related price).

[0046] Furthermore, the static attribute of the elapsed flight hours of the itinerary is the additional time of a itinerary compared to a non-stop itinerary with the same origin and destination, including flight time and any layover time (the flight time of a non-stop itinerary is defined as zero; the flight time of a connecting flight is defined as the difference between the fastest connecting service in the market and the current service).

[0047] The static attribute of service type is the passenger's preference for a specific service type, including non-stop, thru, online connection, interline; codeshare non-stop, codeshare online connection, codeshare interline.

[0048] The static attribute Equipment Type represents the passenger's preference for Turbo Prop, Regional Jet, Large Regional Jet, Narrow Body, Wide Body, Extra Wide Body, and Supersonic aircraft.

[0049] Optionally, for a transit via Beijing PEK from Chongqing CKG to New York JFK, the itinerary CKG-PEK-JFK is selected as CA4141 (CKG 14:10-PEK 16:30), aircraft type 738, distance 1466 km, and CA981 (PEK 20:15-JFK 11:45), aircraft type B747, distance 10991 km. Assuming that β for aircraft type B738 738 = 1 and β for aircraft type B747 747 = 2, then the aircraft type for the itinerary CKG-PEK-JFK is more accurate than the value of 2 obtained by the method of selecting the aircraft type based on the longer flight distance.

[0050] The static attribute Nonstop Ratio represents the possible share advantage of an airline in a certain market, which is measured based on the frequency of a large number of nonstop flights in the market.

[0051] Optionally, Nonstop Ratio = (number of flights operated by the airline between OD) / (total number of nonstop flights between OD). Generally, a larger proportion of nonstop flights will result in more market share. Therefore, the static attribute of a single seat strengthens the impact of the nonstop feature on the market.

[0052] The static attribute Change of Airport allows penalties for itineraries with transfers between airports. No passenger wants to change airports during a transfer. For example, if transferring in Beijing from Capital Airport PEK to Daxing Airport PKX, such an itinerary will be abandoned by passengers and a certain penalty value will be given.

[0053] The static attribute Change of Gauge allows penalties for changing aircraft types. Especially when changing from a large aircraft to a small aircraft, it has some impact on passengers' choices, and a certain penalty value will be given to such an itinerary.

[0054] The static attribute Origin Point Presence indicates that an airline may have a share advantage in a specific market due to having a large number of seats departing from the origin station.

[0055] The additional stop of the static attribute is the impact of adding stopover sites during the journey on passengers' preferences. It is set as the number of stopover sites minus 1, indicating that passengers do not like journeys with many stopover sites.

[0056] The static attribute relative fare will directly affect passengers' decisions to choose this journey, especially for leisure travelers who are sensitive to air ticket prices. If the air ticket price of the journey is lower than the industry average level (equal to 1), it will receive a reward. Similarly, if the relative fare ratio is greater than 1, the route will be penalized.

[0057] The static attribute MetroAirport Preference is the degree to which one airport is more popular than another in a city region with multiple airports. For example, in the Beijing-Tianjin-Hebei urban agglomeration, Daxing Airport will be popular among passengers in the Beijing-Tianjin-Hebei region.

[0058] The static attribute Airline Preference represents passengers' specific preferences for the quality and service experience of airlines. Airline Preference takes into account intangible influences such as the frequent flyer programs of each airline, arrival and departure on-time rates, and the catering and service quality of airlines. The airline preference for connecting flights is close to that of the main carrier, i.e., the airline with the longest flight segment.

[0059] The static attribute Alliance Preference indicates that passengers will prefer alliance journeys over other similar journeys.

[0060] These static attributes reflect passengers' preferences for journeys. Generally speaking, the relative fare is more sensitive to leisure travelers, and the dynamic attributes are more sensitive to business travelers who are sensitive to time. This passenger choice model can also well verify this view.

[0061] Next, enter step S4. In step S4, a passenger utility function is established using the linear combination of journey dynamic parameters and static parameters.

[0062] The passenger utility function is a linear combination of passenger preference attribute parameters, which include the dynamic utility part DU and the static utility part SU.

[0063] Furthermore, the dynamic utility calculation formula is DU = β DD X DD + β DM X DM .

[0064] Furthermore, the static utility is SU = β elap X elap + β service_type Xservice_type +β type X type +...+β alliance X alliance 。

[0065] The mathematical expression of the passenger utility function is abstracted as U i = DU + SU = β 1 X 1 +…+β n X n 。

[0066] Next, enter step S5. In step S5, a nested logit model is constructed using the trip utility function and the proportion of passenger flow at the corresponding time point of TOW to calculate the trip market share Market Share, thereby constructing a passenger trip selection model.

[0067] Use the nested logit function as the activation function to map the passenger utility function value to the percentage of trip share ms(i), as Figure 3 shown. The nested relationship of the market share analysis model is as Figure 3 , divided by the departure time of the trip. Starting from 0:00 on Sunday of a typical week, with a time interval of 1 hour until 23:00 on Saturday, there are a total of 24 * 7 = 168 time points. The proportion of passenger flow p(t) at each point is statistically obtained through historical data. The specific mathematical expression for calculating the market share of a certain trip i among k trips is While satisfying the requirements of market share integrity, proving the rationality of the passenger selection model proposed in this patent. The mathematical expression of the utility function is U i,t = β 1 X 1t +…+β n X nt , where p(t) is the proportion of passenger flow at time t of the TOW of the trip. Among them, X 1t ,..., X nt may be dynamic attributes related to time t or static attributes unrelated to time t. That is to say, the passenger trip market share is determined by the passenger preference attributes.

[0068] Furthermore, the nested logit model is nested by time, dividing a week into 168 discrete points, i.e., discrete integer time points t = 0, 1,..., 167.

[0069] Corresponding to the application scenario and method of the method provided in the embodiment of the present application, an embodiment of the present application also provides a passenger selection model system based on passenger preferences, and this device is deployed on a computer device. The following will refer toFigure 4 Describe the passenger selection model system based on passenger preferences of the present application. Figure 4 It is a block diagram showing the structure of the passenger selection model system based on passenger preferences according to an embodiment of the present application. As Figure 4 shown, the passenger selection model system 400 based on passenger preferences may include: a trip acquisition module 401, a TOW acquisition module 402, a parameter cleaning module 403, a utility function construction module 404, and a passenger selection model module 405.

[0070] The trip acquisition module 401 is used to select a typical week in a flight season and obtain historical data on passenger flow during the typical week for a certain passenger trip, including the origin O, destination D, the first flight (departure and arrival airports, departure and arrival times, aircraft type), the second flight (departure and arrival airports, departure and arrival times, aircraft type), the third flight (departure and arrival airports, departure and arrival times, aircraft type), and the fare including the cabin class and the number of passengers;

[0071] The TOW acquisition module 402 is used to obtain the data of the passenger flow proportion curve of TOW (Time of Week) on the origin O and destination D in the typical week;

[0072] The parameter cleaning module 403 is used to clean and sort the dynamic parameters such as the number of days of deviation and the number of minutes of deviation, and the static parameters such as flight hours, aircraft type, service type (direct flight, transfer, and code sharing), and passenger preference attributes (departure airport, direct flight rate, airline, alliance, and fare) in the attributes of the passenger-selected trip;

[0073] The utility function construction module 404 uses a linear combination of trip dynamic parameters and static parameters to establish a passenger utility function;

[0074] The passenger selection model module 405 is used to construct a nested logit model by using the trip utility function and the passenger flow proportion at the corresponding time point of TOW to calculate the market occupancy rate of the trip, thereby constructing a passenger selection trip model.

[0075] For the functions of each module in each device of the embodiment of the present application, reference may be made to the corresponding descriptions in the above method, and they have corresponding beneficial effects, which will not be elaborated here.

[0076] Figure 5 It is a block diagram showing an electronic device according to an embodiment of the present application. As Figure 5 shown, the electronic device includes: a memory 501 and a processor 502. The memory 501 stores a computer program that can run on the processor 502. When the processor 502 executes the computer program, the method in the above embodiment is implemented. The number of the memory 501 and the processor 502 may be one or more.

[0077] The electronic device further includes: a communication interface 503, configured to communicate with external devices and perform data interaction and transmission.

[0078] If the memory 501, the processor 502, and the communication interface 503 are implemented independently, the memory 501, the processor 502, and the communication interface 503 can be interconnected through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of simplicity of representation, Figure 5 only a thick line is used to represent it in the figure, but it does not mean that there is only one bus or one type of bus.

[0079] Optionally, in specific implementation, if the memory 501, the processor 502, and the communication interface 503 are integrated on a single chip, the memory 501, the processor 502, and the communication interface 503 can communicate with each other through an internal interface.

[0080] It should be understood that the above-mentioned processor can be a Central Processing Unit (CPU), or other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc. It is worth noting that the processor can be a processor that supports the Advanced RISC Machines (ARM) architecture.

[0081] Further, optionally, the above-mentioned memory may include a read-only memory and a random access memory. The memory may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may include a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may include a random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of RAM are available. For example, static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), sync link dynamic random access memory (SLDRAM), and direct rambus random access memory (DR RAM).

[0082] In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions according to the present application are generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium.

[0083] In the description of this specification, the descriptions referring to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of this application. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0084] In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of such features. In the description of this application, "a plurality of" means two or more unless otherwise specifically defined.

[0085] Any process or method described in the flowchart or otherwise described herein can be understood to represent a module, segment, or portion of code including one or more executable instructions for implementing a specific logical function or process. And the scope of the preferred embodiments of this application includes additional implementations, where the functions can be executed in a substantially simultaneous manner or in a reverse order according to the functions involved, rather than in the order shown or discussed.

[0086] The logic and / or steps described in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing a logical function and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device), or in connection with these instruction execution systems, apparatus, or devices.

[0087] It should be understood that each part of this application can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. All or part of the steps of the method in the above embodiments can be completed by a program instructing relevant hardware, and this program can be stored in a computer-readable storage medium. When this program is executed, it includes one or a combination of the steps of the method embodiment.

[0088] In addition, in each embodiment of the present application, each functional module may be integrated into one processing module, may exist separately as individual physical modules, or two or more modules may be integrated into one module. The above-mentioned integrated modules may be implemented in the form of hardware or in the form of software functional modules. If the above-mentioned integrated modules are implemented in the form of software functional modules and sold or used as independent products, they may also be stored in a computer-readable storage medium. This storage medium may be a read-only memory, a magnetic disk, an optical disc, or the like.

[0089] As described above, only the exemplary embodiments of the present application are provided, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope recorded in the present application can easily think of various changes or substitutions, and these should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A passenger selection model method based on passenger preferences, the characteristics of which include: Step S1, select a typical week of a flight season, and obtain historical big data of passenger flow of a certain trip during the typical week, including the trip origin O, destination D, first flight (take-off and landing airports, take-off and landing times, aircraft type), second flight (take-off and landing airports, take-off and landing times, aircraft type), third flight (take-off and landing airports, take-off and landing times, aircraft type), ticket prices including cabin space and number of passengers; Step S2, obtaining TOW (Time of Week) passenger flow ratio curve data at the origin O and the destination D in a typical week; Step S3, cleaning and sorting the dynamic parameters such as deviation days and deviation minutes and static parameters such as flight hours, aircraft type, service type (direct flight, transfer and code sharing), and passenger preference attributes (origin airport, direct flight rate, airline, alliance and fare) in the attributes of the passenger's selected itinerary; Step S4, establishing a passenger utility function using a linear combination of the dynamic parameters and static parameters of the trip; Step S5, constructing a nested logit model using the itinerary utility function and the passenger flow ratio at the corresponding time point of TOW, constructing a passenger itinerary selection model, and calculating the itinerary market occupancy rate.

2. A passenger selection model method based on passenger preferences according to claim 1, characterized in that: The typical week is the week closest to the average number of passengers per week in a season of the airline. The typical week is selected, and its flight and passenger data are used as historical data for training the parameters of the passenger selection model.

3. According to the passenger selection model method based on passenger preferences as described in claim 1, its characteristics are: The one-week model TOW curve includes: According to the airline flight plan, the percentage of all passenger traffic in the O&D market in a typical week from 0:00 on Sunday to 23:00 on Saturday is calculated, and the demand at discrete time points is smoothed into continuous time demand.

4. According to claim 1, a passenger selection model method based on passenger preferences is characterized by: The passenger utility function includes: The passenger utility function is a linear combination of passenger preference attribute parameters. The passenger preference attribute parameters n include dynamic utility parameter part and static utility parameter part. The mathematical expression is U i =DU+SU=β1X1+…+β n X n。 5. The passenger selection model method based on passenger preferences according to claim 1, characterized in that: The nestedlogit model specifically includes: The nested logit function is used as the activation function to map the passenger utility function value to the trip share percentage ms(i), and its specific mathematical expression is: Satisfy at the same time Market share integrity requirement, the mathematical expression of the utility function is U i,t =β1X 1t +…+β n X nt , p(t) is the passenger flow ratio of TOW of the trip at time t, where X 1t ,...,X nt It may be a dynamic attribute, related to TOW time t, or it may be a static attribute, unrelated to TOW time t, that is, the passenger trip market share is determined by the passenger preference attributes.

6. A passenger selection model method based on passenger preferences according to claim 4, characterized in that: The dynamic utility parameters include: The itinerary attributes that provide passengers with many different choices include departure time, departure date, arrival time and arrival date, which are defined as dynamic attributes and are determined by the passenger flow ratio of the TOW curve. The linear combination of dynamic attribute parameters is the passenger dynamic utility function. The specific attributes include but are not limited to day deviation DD (Displacement Day) and minute deviation DM (Displacement Minute).

7. A passenger selection model method based on passenger preferences according to claim 4, characterized in that: The static utility parameters specifically include: The invariant attributes related to flights are defined as passenger static attributes. The linear combination of static attribute parameters is the passenger static utility function. The main attributes include but are not limited to Elasped Time, Aircraft Type, Service Type (nonstop, connection and code share), and passenger preference attributes (origin airport OPP, nonstop ratio, airline, alliance and related price).

8. A passenger selection model method based on passenger preferences according to any one of claims 1 and 5, characterized in that: The nested logit model specifically includes: The nested logit model is nested by time, dividing a week into 168 discrete points, namely, discrete hourly times t=0, 1, ..., 167.

9. A passenger selection model system based on passenger preferences, the characteristics of which include: The itinerary acquisition module is used to select a typical week of a flight season and obtain the historical passenger flow data of a passenger itinerary during the typical week, including the itinerary origin O, destination D, first flight (take-off and landing airports, take-off and landing times, aircraft type), second flight (take-off and landing airports, take-off and landing times, aircraft type), third flight (take-off and landing airports, take-off and landing times, aircraft type), ticket prices including cabin space and number of passengers; A TOW acquisition module is used to obtain TOW (Time of Week) passenger flow ratio curve data at the origin O and destination D in a typical week; Parameter cleaning module, used to clean and organize the dynamic parameters such as deviation days and deviation minutes in the attributes of passengers' selected itineraries, as well as static parameters such as flight hours, aircraft type, service type (direct flight, transfer and code sharing), and passenger preference attributes (origin airport, direct flight rate, airline, alliance and fare); The utility function building module uses the linear combination of the dynamic and static parameters of the trip to build the passenger utility function; The passenger selection model module is used to construct a nested logit model using the itinerary utility function and the proportion of passenger flow at the corresponding time point of TOW, to construct a passenger itinerary selection model, and to calculate the market occupancy rate of the itinerary.

10. An electronic device comprising a memory, a processor and a computer program stored in the memory, wherein the processor implements the method according to any one of claims 1 to 8 when executing the computer program.

Images