Method and system for calculating flight passenger overflow and electronic equipment

Abstract

The invention discloses a flight passenger overflow calculation method and system and electronic equipment, and belongs to the technical field of flight route network planning. Comprising the following steps: collecting historical travel data of a specified flight of a passenger, and calculating the sum of demands of all travels on a flight segment through a market share model, i.e., the total flight demand; for the leg, selecting a demand distribution model with high goodness of fit according to the size range of demand variation coefficients; based on flight historical data, the passenger load factor when the flight is closed is regressed by using a machine learning method to obtain the effective capacity of the flight; comparing the adjusted demand with the effective capacity of the airplane to calculate the expected passenger capacity; according to the method, passenger overflow is scientifically predicted based on the real-time demand fluctuation characteristics by using a demand model method with better goodness of fit, the travel experience of passengers is improved, an airline network of airlines is reasonably planned, and the defects that manual prediction and judgment cannot be carried out and experiencing cannot be carried out are overcome.

Description

Technical Field

[0001] The present invention relates to the field of transportation, and particularly to a method, system and electronic device for calculating flight passenger spillover, which are used to simulate and predict the spillover behavior of passengers among different transportation modes, and can be applied to aspects such as ticket distribution and capacity planning of transportation systems such as aviation and railway. Background Art

[0002] The route network planning of airlines is a very important basic work for airlines to improve market competitiveness. Facing the complex route network of airlines, it is a major problem to maximize the satisfaction of passengers' travel needs by using limited air traffic rights, aircraft and crew resources. Among them, passenger spillover management is one of the inevitable problems to meet passengers' travel needs. Sometimes, when a flight cannot meet the needs of all passengers, passengers will spill over to other alternative transportation modes or flights. Airlines need to adopt effective means to meet passengers' needs, expand the airlines' revenue and achieve a win-win situation.

[0003] China has become the second largest civil aviation market after the United States. Every year, 700 million person-times of passengers in China take civil aviation aircraft for business or leisure travel. China has world-class HUB airports such as PEK (PKX), SHA (PVG), CAN or HKG, etc. In Northeast Asia, there are ICN in South Korea and TYO in Japan. At the same time, the aviation market is highly competitive. Chinese passengers have diverse travel choices. When Chinese passengers fly to Europe and America, there are hundreds of travel combinations to meet passengers' needs. For those who cannot fly directly, some transfer at PEK or PVG, and some transfer in Japan or South Korea. In such an O&D market, there will be some flights that are the preferred flights for air travel. Sometimes, the seats on the flights are not enough and it is impossible to meet the travel requirements of all passengers. How to reasonably divert passengers to other flights is a problem that airlines need to consider in their planning and decision-making. The route network planning and flight spillover management of airlines are extremely important and must be considered in the prediction and management of passenger flow in the route network. Sometimes, when the capacity is insufficient, such as there are bottlenecks, passenger spillover will occur. Sometimes, there are empty seats on the aircraft, resulting in waste and loss of revenue.

[0004] The present invention provides a method, system and electronic device for calculating flight passenger spillover, which can more accurately simulate and predict the spillover behavior of passengers among different transportation resources, and provide a more scientific decision-making basis for the operation and management of transportation enterprises. Summary of the Invention

[0005] The embodiments of the present application provide a method and system for calculating flight passenger spillover to solve the deficiencies of existing models in related technologies. When transportation enterprises conduct capacity allocation and ticket sales, it is difficult to accurately predict the spillover behavior of passengers, resulting in empty seats on some flights while other flights are in short supply, which affects the effective utilization of transportation resources and the economic benefits of airlines.

[0006] In a first aspect, the present application proposes a method for calculating flight passenger spillover, including:

[0007] Collect historical travel data of passengers flying through a designated flight segment, including information such as travel purpose, departure time, arrival time, ticket price, origin flight, transfer flight, and aircraft type, and calculate the sum of the demands of all trips on the said flight segment, that is, the total flight demand, through a market share model;

[0008] For the said flight segment, compare the change in current demand and previous demand to obtain the demand coefficient of variation, and select a demand distribution model with high goodness of fit according to the range of the demand coefficient of variation;

[0009] Based on flight historical data, use machine learning methods to regress the load factor at flight closure and calibrate it to obtain the effective capacity of the flight;

[0010] Compare the adjusted demand with the effective capacity of the aircraft to calculate the expected passenger load. If the demand is less than the effective capacity, the expected passenger load is equal to the demand. If the demand is greater than the effective capacity, the expected passenger load is equal to the effective capacity;

[0011] Step S5, calculate the flight spillover number and flight passenger flow according to different demand distribution models.

[0012] In a second aspect, an embodiment of the present application provides a system for calculating flight passenger spillover, including:

[0013] A flight demand module, used to collect historical travel data of passengers flying through a designated flight segment, including information such as travel purpose, departure time, arrival time, ticket price, origin flight, transfer flight, and aircraft type, and calculate the sum of the demands of all trips on the said flight segment, that is, the total flight demand D, through a market share model.

[0014] A demand model module, used to compare the change in current demand and previous demand for the said flight segment to obtain the demand coefficient of variation CV, and select a demand distribution model with high goodness of fit according to the range of the demand coefficient of variation.

[0015] A flight capacity module, used to regress the load factor at flight closure LFCF based on flight historical data by machine learning methods and calibrate it to obtain the effective capacity C of the flight eff .

[0016] An expected passenger flow module, used to compare the adjusted demand with the effective capacity of the aircraft to calculate the expected passenger load. If the demand is less than the effective capacity, the expected passenger load is equal to the demand. If the demand is greater than the effective capacity, the expected passenger load is equal to the effective capacity.

[0017] A flight spillover module, used to calculate the flight spillover number (spill) and flight passenger flow (traffic) according to different demand distribution models.

[0018] 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. When the processor executes the computer program, the method described in any one of the above is implemented.

[0019] Compared with the prior art, the present application has the following advantages:

[0020] According to the embodiments of the present application, in flight management practice of airlines, due to reasons such as transport capacity, fare, seat control, and sales policies, when the market is bad, it is easy to cause weak demand, and there are many empty seats on flights, resulting in wasted seats. When the market is in the peak season, the resources of airlines are insufficient, causing passengers not to be able to buy tickets and resulting in passenger overflow. Supply-demand matching is an important issue that airlines need to face in flight management. This patent mainly uses historical flight and passenger data based on real-time demand fluctuation characteristics and uses model methods to scientifically predict how many passengers are transferred to other companies due to the airline's route network restrictions and thus generate overflow. The overflow passengers are scientifically and reasonably calculated using a goodness-of-fit demand model, so as to accurately predict the overflow of all flights in the route network, overcome the shortcomings of manual judgment and prediction empiricism, use the method of this patent to scientifically reveal statistical laws, optimize resource allocation, and transportation enterprises can reasonably adjust the allocation of transport capacity and ticket sales strategies, improve the utilization rate of transport resources, and reduce the situation of wasted seats and supply falling short of demand; overcome the shortcomings of difficult, costly, and inaccurate manual data collection, reduce the overflow prediction error by about 25% compared with the traditional EMSR-b method, and through numerical operation optimization, the calculation time for a single flight segment is <0.5 seconds, improving the operation speed; by better meeting the needs of passengers, improving the travel experience of passengers, enhancing passenger satisfaction, and strengthening passenger loyalty to airlines.

[0021] 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 description. 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

[0022] In the drawings, unless otherwise specified, the same reference numerals throughout the multiple drawings denote the same or similar components 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.

[0023] Figure 1 is a flowchart showing a method for calculating flight passenger overflow according to an embodiment of the present application;

[0024] Figure 2It is a flowchart of a goodness-of-fit model for selection demand distribution showing an embodiment of the present application;

[0025] Figure 3 It is a flowchart of flight spillover of a two-stage exponential distribution mixture model showing an embodiment of the present application;

[0026] Figure 4 It is a flowchart of flight spillover of a demand gamma distribution model showing an embodiment of the present application;

[0027] Figure 5 It is a flowchart of flight spillover of a demand normal distribution model showing an embodiment of the present application;

[0028] Figure 6 It is a system structure diagram of a calculation of flight passenger spillover showing an embodiment of the present application; and

[0029] Figure 7 It is a block diagram of an electronic device for calculating flight passenger spillover showing an embodiment of the present application. Detailed implementation manners

[0030] In the following, only some exemplary embodiments are simply 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.

[0031] 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. The detailed implementation manners are as follows:

[0032] An embodiment of the present application provides a method 100 for calculating flight passenger spillover. The method 100 for calculating flight passenger spillover 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 method 100 for calculating flight passenger spillover of the present application. Figure 1 It is a flowchart of a method for calculating flight passenger spillover showing an embodiment of the present application. As Figure 1As shown in the figure, the method 100 for calculating flight passenger spillover includes the following steps S1 to S5, and steps S1 to S5 will be described below in combination with specific embodiments. Symbol definitions: D = pDemand (current demand), D_prev = pPrevDmd (previous demand), C = pCapacity (capacity), LF = pLfcf (load factor), CV_leg = pLegCV (coefficient of variation of demand).

[0033] First, enter step S1. In step S1, collect the historical travel data of passengers flying through a designated flight segment, including information such as travel purpose, departure time, arrival time, ticket price, origin flight, transfer flight, and aircraft type, and calculate the sum of the demands of all trips on the flight segment, that is, the total flight demand, through the market share model.

[0034] First, look at the function input variable parameters pCapacity, pDemand, pPrevDmd, pLegCV, pLfcf.

[0035] The total flight demand D = pDemand is the sum of the trip demands of the flight segment, and the trip demand is calculated by the market share Logit model.

[0036] In the Logit model, passenger attributes comprehensively consider factors such as passengers' travel purpose (business, leisure, etc.), travel time preference, ticket price sensitivity, transfer willingness, etc., and construct a passenger demand feature vector.

[0037] Use the logit model to map the passenger utility function value to the trip share percentage ms(i). Divide 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 168 time points. Calculate the passenger flow proportion p(t) at each point through historical data. Calculate the market share of a certain trip i among k trips. The specific mathematical expression is The mathematical expression of the utility function is U i,t = β 1 X 1t +…+ β n X nt , p(t) is the proportion of passenger flow at time t of the TOW of the trip, where X 1t ,..., X nt are passenger preference attributes. The passenger trip market share is determined by the passenger preference attributes.

[0038] The travel of air passengers shows certain regularity throughout the year with the change of seasons. Generally, winter and spring are off-seasons, while summer and autumn are peak seasons. Major holidays are also peak passenger flow periods. In particular, July to September is the peak period when students, tourism, and business trips overlap. Airlines' flight schedules also follow the same pattern as the market. Flight schedules generally reflect a weekly cycle of flight operations. Airlines' planning and decision-making are based on weekly flight and passenger data.

[0039] Exemplarily, take flight CA855 from PEK to LHR on August 9, 2023, as an illustration. There are 564 passenger itineraries passing through CA855. Using the logit model, the passenger utility function values are mapped to the percentage share of itinerary ms(i). The total demand for the 564 itineraries including flight CA855 is 311, as shown in Table 1. The first 19 itineraries with more demand are selected. The demand size is directly related to dynamic attributes such as departure time, and also related to static attributes such as non-stop flights, transfers, number of transfers, relative fare, and aircraft type, as well as the TOW of the passenger's departure airport. In the embodiment, there is only a non-stop itinerary from PEK to LHR, and the other 18 are itineraries departing from China, New Zealand, Australia, Japan, and South Korea, etc. According to the Logit model and the corresponding historical itinerary TOW curve data, the market share Market_share and itinerary passenger demand are obtained as shown in Table 1. Among them, for itinerary 1, from PEK to LHR, flight CA855, the demand dmd = 43.35 is the highest; for itinerary 2, from Chengdu Tianfu TFU to LHR, connecting with CA855 via CA1418, the demand dmd = 21.72; for itinerary 3, from Auckland AKL to LHR, connecting with CA855 via CA784 the previous day, the demand dmd = 12.09.

[0040] Table 1: Itineraries passing through CA855, as well as demand, spillover, and recapture

[0041]

[0042] Next, enter step S2. In step S2, for the flight segment, compare the change in current demand and previous demand to obtain the demand coefficient of variation, and select a demand distribution model with a high goodness of fit according to the range of the demand coefficient of variation.

[0043] Calculate CV as (pDemand - pPrevDmd) / pDemand × pLegCV. The coefficient of variation CV of the flight is the standard deviation of demand divided by the mean, reflecting the intensity of fluctuations. When CV is greater than 1, the demand distribution model used is the two-stage mixed exponential model with a better goodness of fit; when CV is not greater than 1 and greater than 0.6, the gamma distribution is used to simulate demand fluctuations; when CV ≤ 0.6, the normal distribution is used to simulate demand fluctuations.

[0044] The Spill model consists of the effective capacity when the aircraft is full, the average load factor LFCF when passengers spill over, and the probability of air passengers. The passenger spillover value can be calculated using the passenger spillover Spill model. Among them, the effective capacity = the average load factor LFCF when passengers spill over * the total number of seats on the aircraft.

[0045] Due to the good market, the flight load factor is relatively high and the supply falls short of demand. According to the CV calculation formula:

[0046]

[0047] In one embodiment, as Figure 2 shown, according to the change range of CV, three demand models with good goodness of fit are adopted in three cases.

[0048] (1) If CV > 1, it indicates that the passenger demand changes violently, usually occurring in first-class demand. The common normal distribution model cannot be used, and a two-stage exponential mixture model with better goodness of fit is adopted, as Figure 3 shown.

[0049] The two-stage exponential mixture model is a mixture of two negative exponentials. The first stage has a rate of γ 1 , and the second stage has a rate of γ 2 . Thus, the probability density function of the two-stage exponential mixture model is

[0050] where

[0051] The mean of the demand can be obtained from historical flight demand the standard deviation of the demand And the two-stage exponential mixture model has

[0052]

[0053] Generally, a can be verified through data, and take a = 0.1.

[0054] In this way, the parameters of the two-stage exponential mixture model can be explicitly obtained:

[0055]

[0056] The flight spillover is

[0057] where

[0058] Generally, initially take the default values CV = 0.05 and a = 0.1.

[0059] (2) If 0.6 < CV ≤ 1, it indicates that the passenger demand changes greatly. Generally, this occurs in the economy class demand during the off-peak season of the market. The common normal distribution model cannot be used either. Instead, a gamma distribution model with better goodness of fit is adopted to fit the passenger demand, as Figure 4 shown.

[0060] The probability density function PDF(x) of the gamma distribution is

[0061]

[0062] The parameters of the gamma distribution can be calculated

[0063]

[0064] The flight spill is

[0065]

[0066] Among them, the cumulative density function CDF(x) of the gamma distribution is

[0067] (3) If CV ≤ 0.6, it indicates that the passenger demand changes normally. Generally, this occurs in the economy class demand during the peak season of the market. A normal distribution model with good goodness of fit is adopted to fit the passenger demand, as Figure 5 shown.

[0068] The probability density function f(x) of the normal distribution and the probability density function PDF(x) of the standard normal distribution

[0069]

[0070] The cumulative density function CDF(x) of the standard normal distribution is

[0071]

[0072] The flight spill is

[0073] spill = σPDF(k) - (C - μ)(1 - CDF(k))

[0074] Among them, PDF is the probability density function of the standard normal distribution, and CDF is the cumulative density function of the standard normal distribution.

[0075] Next, enter step S3. In step S3, based on the historical flight data, the load factor at flight closure LFCF is regressed using machine learning methods and calibrated to obtain the effective capacity C of the flight eff .

[0076] The load factor at flight closure (LFCF) of an aircraft. Due to a certain percentage of no-shows among passengers, there will be a few empty seats on the aircraft. At this time, the aircraft's passenger capacity is the effective capacity of the aircraft. The maximum number of seats in the aircraft cabin configuration is C, and the average load factor at passenger spillover (LFCF, generally defined as 0.95). The effective capacity of the aircraft is C eff = min(C, C * LFCF)

[0077] The load factor at flight closure (LFCF) is a comprehensive indicator that incorporates revenue management strategies such as overbooking or dynamic pricing and affects the final available capacity. It can be dynamically adjusted through real-time data collection, machine learning prediction models, dynamic threshold adjustment mechanisms, closed-loop feedback systems, and calibration strategies to adjust LFCF

[0078] Generally speaking, when the load factor at flight closure is less than 1.0, the number of overflow passengers is relatively large. In fact, the load factor at flight closure will never be equal to 1.0 because of the uncertainties in the input data (boarding rate, cancellation rate, no-shows, and improper connections) used to determine the optimal overbooking level, resulting in uncertainties in the overbooking process. There are different estimation methods for the load factor of closed flights, with different levels of refinement in the estimation. When a monthly system-wide correction factor is required, one method is as follows

[0079] Step 1: Calculate the average load factor of closed flights based on the reading date (time point before departure) and booking class of all flights in a specific historical month

[0080] Step 2: Estimate the percentage of overflow passengers classified by reading date and booking class in the same historical period

[0081] Step 3: Calculate the weighted average of the load factors of the closed flights in Step 1 above, and then weight it according to the percentage of overflow passengers in Step 2 above to determine the comprehensive value of the load factor of closed flights

[0082] Illustratively, in an embodiment, the total demand for 584 itineraries containing flight CA855 is summarized in Table 1 as dmd(CA855) = D = 311 people, but the aircraft type of CA855 is A359, and the total number of seats on the aircraft is C = pCapacity = 312. However, due to reasons such as passenger no-shows, the load factor generally cannot be 1. Generally, LFCF = 0.95, and the effective capacity = average load factor at passenger spillover LFCF * total number of seats on the aircraft = 296. There are probably overflow passengers on this flight

[0083] Next, enter step S4. In step S4, compare the adjusted demand with the effective capacity of the aircraft to calculate the expected passenger load. If the demand is less than the effective capacity, the expected passenger load is equal to the demand. If the demand is greater than the effective capacity, the expected passenger load is equal to the effective capacity.

[0084] In one embodiment, as shown in Table 1, for flight 855 with aircraft type 359, the maximum capacity of the aircraft is 312 and the effective capacity is 296 (see the dmd column in Table 1). The demand D = 311. Since the demand is greater than the effective capacity of the aircraft, the flight spill is 311 - 296 = 15.

[0085] The passenger flow of the overbooked flight CA855, traffic = demand - spill = 296.

[0086] Next, enter step S5. In step S5, calculate the number of flight spills and the flight passenger flow according to different demand distribution models.

[0087] The calculation method of the flight class overbooking probability model is as follows:

[0088]

[0089] Where f(x) is the demand probability density function and D is the flight demand.

[0090] For the flight passenger flow traffic, there are the following operations:

[0091] E(traffic) = E(Demand) - E(spill).

[0092] Regardless of whether the flight demand is normally distributed, gamma distributed or two-stage mixed exponential distributed, the calculation of the flight class overbooking value can be carried out by Monte Carlo simulation for 10,000 iterations to reduce the calculation error.

[0093] In one embodiment, for example, since there are relatively few high-end business passengers in the airline's cabin layout, in the airline's cabin layout, economy class accounts for 80%. The demand of economy class passengers generally shows a normal distribution pattern. Here, in detail, in the case of CV < 0.6, as Figure 5 shown in the process, numerically calculate the flight spill.

[0094] The first condition is to check whether pDemand is not equal to 0. If so, calculate CV as (pDemand - pPrevDmd) / pDemand × pLegCV. Otherwise, CV is equal to pLegCV.

[0095] Next, determine whether pDemand is equal to pPrevDmd or CV is less than a very small value. If so, calculate legSpill2 as pDemand - pLfcf × pCapacity, and if the result is less than or equal to 0, return 0; otherwise, return legSpill2.

[0096] When in the market stable state or low - volatility condition, when D = D_prev or CV < 10 -6 Spill volume = max(D - (LF × C), 0)

[0097] If pDemand is between 0.5 times and 2 times, enter another branch, especially when pCapacity is greater than 10. At this time, calculate the nominal load factor nomload (i.e., demand divided by capacity), and the standard deviation stdev is CV multiplied by nomload. Then the standardized variable tmpval is (pLfcf - nomload) / stdev, and use polynomial approximation to calculate the cumulative distribution function (CDF) and probability density function (PDF) of the normal distribution. Here, a fourth - degree polynomial is used to approximate the CDF, and the result is adjusted according to the sign of tmpval.

[0098] When 0.5×LF×C ≤ D ≤ 2×LF×C, the calculation of legSpill can be decomposed into the following piece - wise mathematical formula:

[0099] Sub - condition 1 (when C > 10):

[0100] Nominal load factor μ = D / C

[0101] Standard deviation σ = CV×μ

[0102] Standardized value z = (LF - μ) / σ CDF(z)=1 - 0.5×[1 + 0.196854z + 0.115194z 2 +0.000344z 3 +0.019527z 4 -4

[0103]

[0104] Spill volume = spill = σPDF(z)-(C - μ)(1 - CDF(z))

[0105] Sub - condition 2 (when C ≤ 10):

[0106] Spill volume = max(D - (LF × C), 0)

[0107] ​If pDemand is greater than twice pLfcf × pCapacity, directly return the same difference. When D > 2 × LF × C: Overflow volume = D - (LF × C)

[0108] Final correction is performed in all cases:

[0109] Flight overflow volume = max(overflow volume, 0).

[0110] In one embodiment, during the actual calculation process, integral calculation may be encountered. Since there is no explicit expression, only the Monte Carlo simulation method can be used. Whether the flight demand is normally distributed, gamma distributed, or two-stage mixed exponential distributed, the Monte Carlo simulation can be used for 10,000 iterations to calculate the flight class overflow value, reducing the calculation error.

[0111] Corresponding to the application scenario and method of the method provided by the embodiments of the present application, an embodiment of the present application also provides a flight passenger overflow calculation system, which is deployed on a computer device. The following will refer to Figure 6 to describe a flight passenger overflow calculation system of the present application. Figure 6 is a structural block diagram showing a flight passenger overflow calculation system according to an embodiment of the present application. As Figure 6 shown, a flight passenger overflow calculation system 600 may include: a flight demand module 601, a demand model module 602, a flight capacity module 603, an expected passenger flow module 604, and a flight overflow module 605.

[0112] The flight demand module 601 is used to collect historical travel data of passengers flying through a specified flight segment, including information such as travel purpose, departure time, arrival time, ticket price, origin flight, transfer flight, and aircraft type, and calculate the sum of the demands of all trips on the flight segment, that is, the total flight demand D, through the market share model.

[0113] The demand model module 602 is used to compare the current demand and the previous demand changes for the flight segment to obtain the demand variation coefficient CV, and select a demand distribution model with high goodness of fit according to the range of the demand variation coefficient.

[0114] The flight capacity module 603 is used to regress the load factor at flight closure LFCF based on flight historical data using a machine learning method and perform calibration to obtain the effective capacity C of the flight eff .

[0115] The expected passenger flow module 604 is used to compare the adjusted demand with the effective capacity of the aircraft to calculate the expected passenger load. If the demand is less than the effective capacity, the expected passenger load is equal to the demand. If the demand is greater than the effective capacity, the expected passenger load is equal to the effective capacity.

[0116] The flight spill module 605 is used to calculate the number of flight spills and flight traffic according to different demand distribution models.

[0117] For the functions of the modules in each system of the embodiments of the present application, reference may be made to the corresponding descriptions in the above methods, and they have the corresponding beneficial effects, which will not be elaborated here.

[0118] Figure 7 It is a block diagram showing an electronic device according to an embodiment of the present application. As Figure 7 shown, the electronic device includes: a memory 701 and a processor 702. A computer program that can run on the processor 702 is stored in the memory 701. When the processor 702 executes the computer program, the methods in the above embodiments are implemented. The number of the memory 701 and the processor 702 can be one or more.

[0119] The electronic device further includes: a communication interface 703, which is used to communicate with external devices and perform data interaction and transmission.

[0120] If the memory 701, the processor 702, and the communication interface 703 are implemented independently, the memory 701, the processor 702, and the communication interface 703 can be connected to each other through a bus and complete communication 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, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of representation, Figure 7 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.

[0121] Optionally, in specific implementation, if the memory 701, the processor 702, and the communication interface 703 are integrated on a chip, the memory 701, the processor 702, and the communication interface 703 can complete communication with each other through an internal interface.

[0122] 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.

[0123] Further, optionally, the above-mentioned memory can include a read-only memory and a random access memory. The memory can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. Among them, the non-volatile memory can include a Read-Only Memory (ROM), a Programmable ROM (PROM), an Erasable PROM (EPROM), an Electrically EPROM (EEPROM), or a flash memory. The volatile memory can 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 RAM (SRAM), Dynamic Random Access Memory (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDR SDRAM), Enhanced SDRAM (ESDRAM), Sync link DRAM (SLDRAM), and Direct Rambus RAM (DR RAM).

[0124] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can 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 can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium.

[0125] In the description of this specification, the descriptions with reference to the terms "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 the present 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.

[0126] 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" can explicitly or implicitly include at least one of the features. In the description of the present application, "a plurality" means two or more unless otherwise specifically defined.

[0127] 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 the present application includes additional implementations, where the functions can be executed in a substantially simultaneous manner or in the reverse order according to the functions involved, rather than in the order shown or discussed.

[0128] 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, system, 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 conjunction with these instruction execution systems, apparatuses, or devices.

[0129] It should be understood that each part of the present 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 instructing relevant hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiments.

[0130] In addition, in each embodiment of the present application, each functional module can be integrated into a processing module, or each module can exist physically alone, or two or more modules can be integrated into one module. The above integrated modules can be implemented in the form of hardware or in the form of software functional modules. When the above integrated modules are implemented in the form of software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. The storage medium can be a read-only memory, a disk, an optical disc, etc.

[0131] 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 should be subject to the protection scope of the claims.

Claims

1. A method for calculating flight passenger overflow, characterized in that the steps include: Step S1, collecting historical travel data of passengers flying through a specified flight segment, including travel purpose, departure time, arrival time, fare, starting flight, connecting flight, and aircraft type, and calculating the sum of the demand for all itineraries on the flight segment, i.e., the total flight demand, through a market share model; Step S2, for the flight segment, compare the changes of current demand and previous demand to obtain the demand variation coefficient, and select a demand distribution model with high goodness of fit according to the range of the demand variation coefficient; Step S3, regressing the passenger load factor when the flight is closed using a machine learning method based on the flight history data, and performing calibration to obtain the effective capacity of the flight; Step S4, comparing the adjusted demand with the effective capacity of the aircraft to calculate the expected passenger capacity, if the demand is less than the effective capacity, the expected passenger capacity is equal to the demand, if the demand is greater than the effective capacity, the expected passenger capacity is equal to the effective capacity; Step S5, calculating the number of flight overflow passengers and flight passenger flow according to different demand distribution models.

2. The method according to claim 1, characterized in that The total flight demand includes: The total flight demand D is the sum of the itinerary demands containing the flight segments, and the itinerary demand is calculated by the market share Logit model.

3. The method according to claim 1, characterized in that The demand distribution model includes: The flight coefficient of variation CV is the standard deviation of demand divided by the mean, reflecting the intensity of fluctuation. When CV is greater than 1, the two-stage mixed exponential model with better demand model fitting is adopted; when CV is not greater than 1 and greater than 0.6, the gamma distribution is used to simulate demand fluctuations; when CV≤0.6, the normal distribution is used to simulate demand fluctuations.

4. The method according to claim 1, characterized in that: The effective capacity calculation specifically includes: The LFCF is the occupancy rate of the aircraft when the flight is closed. The number of passengers after the flight is closed is the effective capacity of the flight. The maximum number of seats in the aircraft cabin configuration is C. The average occupancy rate LFCF when there is overflow of passengers (generally defined as 0.95) is the effective capacity of the aircraft. eff =min(C,C×LFCF) 5. The method according to claim 1, characterized in that The flight-level overflow probability model includes: Where f(x) is the demand probability density function, D is the flight demand.

6. The method according to claim 2, characterized in that The logit model specifically includes: The passenger attributes in the Logit model comprehensively consider factors such as the passenger’s travel purpose (business, leisure, etc.), travel time preference, fare sensitivity, and transfer willingness, which construct the passenger demand feature vector.

7. The method according to claim 4, characterized in that The flight closed load factor (LFCF) includes: Flight closed load factor (LFCF) is a comprehensive indicator that combines revenue management strategies, such as overbooking or dynamic pricing, to affect the final effective capacity. LFCF can be dynamically adjusted through real-time data collection, machine learning prediction models, dynamic threshold adjustment mechanisms, closed-loop feedback systems, and calibration strategies.

8. The method according to claims 1, 3 and 5, characterized in that The calculation of flight-level overflow numbers includes: Regardless of whether the flight demand is normally distributed, gamma distributed, or two-stage mixed exponentially distributed, the numerical calculation of flight-level overflow can be performed using Monte Carlo simulation with 10,000 iterations to reduce the calculation error.

9. A system for calculating flight passenger overflow, characterized in that: The system comprises: The flight demand module is used to collect the historical travel data of passengers flying through a specified flight segment, including travel purpose, departure time, arrival time, fare, starting flight, connecting flight and aircraft type, and calculate the sum of the demand for all itineraries on the segment, i.e. the total flight demand D, through the market share model. The demand model module is used to compare the changes in current demand and previous demand for the flight segment to obtain the demand variation coefficient CV, and select a demand distribution model with high fitting goodness of fit based on the size range of the demand variation coefficient. The flight capacity module is used to regress the flight closure load factor LFCF based on historical flight data using machine learning methods and calibrate it to obtain the effective capacity C of the flight. eff . The expected passenger flow module is used to compare the adjusted demand with the effective capacity of the aircraft to calculate the expected passenger capacity. If the demand is less than the effective capacity, the expected passenger capacity is equal to the demand. If the demand is greater than the effective capacity, the expected passenger capacity is equal to the effective capacity. The flight overflow module is used to calculate the flight overflow number (spill) and flight passenger flow (traffic) according to different demand distribution models.

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