Passenger flow prediction method, device and equipment based on multi-level business district penetration model

Abstract

The invention relates to a passenger flow prediction method based on a multi-level business district penetration model, and the method comprises a business district basic calculation unit which is used for generating various parameters related to the business district passenger flow in historical years according to the data summarization of the population and passenger flow in the business district in historical years, and a business district prediction analysis unit which is used for predicting the passenger flow in the business district according to the various parameters of the business district passenger flow. And the passenger flow prediction value generation unit is used for generating a passenger flow prediction value of the # imgabs0 # month of the prediction year, and the business district calibration feedback unit is used for generating an error index according to an error between the passenger flow prediction value and an actual value. According to the invention, passenger flow prediction can be carried out based on the population penetration difference and the monthly offset of each level of business district, and the prediction accuracy is improved. Through an error index generated by a predicted passenger flow value and an actual passenger flow value, index comparison covering multiple scenes is realized, and a prediction model can be evaluated more accurately and a decision can be guided.

Description

Technical Field

[0001] The present application relates to the technical field of passenger flow prediction, and in particular to a passenger flow prediction method, device and equipment based on a multi-level business district penetration model. Background Art

[0002] Shopping district passenger flow forecasting technology is of great significance for resource allocation, marketing strategy formulation, and operational efficiency improvement. However, existing forecasting methods still have the following key limitations: Traditional models often ignore the differences in penetration across different business districts. They typically treat business districts as homogeneous, failing to distinguish between the significant differences in population density, spending power, and reach of core, secondary, and fringe business districts. This leads to significant deviations between predicted results and actual customer flow distribution patterns. For example, the customer flow attenuation effect of fringe business districts and the agglomeration effect of core business districts require differentiated modeling.

[0003] Lack of consideration for deviations between forecast results and actual conditions: Existing technologies often rely on static historical averages and fail to incorporate dynamic exogenous variables (such as holidays, climate, and seasonal promotions) for adjustment. This makes it difficult to accurately reflect true cyclical fluctuations. If the model fails to identify the deviation caused by certain variables and fails to dynamically adjust the forecast baseline in a timely manner, significant forecast bias may occur, affecting the accuracy of operational decisions and resource allocation.

[0004] The above problems restrict the practicality and real-time performance of the prediction results. There is an urgent need for a passenger flow prediction technology that integrates multi-level business district characteristics, dynamic weight adjustment, and fine-grained time series analysis. Summary of the Invention

[0005] The present application provides a customer flow prediction method based on a multi-level business district penetration model, which is characterized by including: The business district basic calculation unit is used to generate various parameters related to the business district passenger flow in historical years based on the data of the business district population and passenger flow in historical years; The business district forecast analysis unit is used to generate the forecast year and the first Monthly passenger flow forecast; The business district calibration feedback unit is used to generate an error indicator based on the error between the predicted customer flow value and the actual value.

[0006] Optionally, the business district basic calculation unit includes: Based on the annual population and monthly passenger flow of each level of business district in the historical year, the total passenger flow and total population of the business district in the historical year are generated; Generate an average annual number of visits to the business district by the population of the business district in the historical year based on the total passenger flow and total population of the business district in the historical year; Generate the business district penetration rate of each level of business district in the historical year based on the average annual number of visits to the business district by the business district population in the historical year, the weight of each level of business district and the total population; According to historical years Monthly passenger flow and historical year average passenger flow, generate historical year Monthly offset factor for month.

[0007] Optionally, the business district prediction and analysis unit includes: Based on the monthly deviation coefficient of the historical year, the historical error mean of the historical year, the business district penetration rate of each level of business district in the historical year and the annual population of each level of business district in the forecast year, the monthly passenger flow of each level of business district in the forecast year is generated.

[0008] Optionally, the business district calibration feedback unit includes: Generate the mean square error based on the average of the squared errors between the predicted value and the actual value; Generate the mean absolute error based on the absolute value of the error between the predicted value and the actual value; Generates the mean absolute percentage error based on the average percentage of the predicted errors relative to the actual values.

[0009] Optionally, the customer flow prediction method based on the multi-level business district penetration model is characterized by: The penetration rate of each level of business district in the historical years is as follows: The weight of the business district is , The population of the business district in the historical year is , the average annual number of times the business district population visits the business district in historical years is , The business district belongs to the core business district, secondary business district and marginal business district, and the business district penetration rate of each level of business district in historical years The formula is: ; The historical year Monthly deviation coefficient of the month, set the historical year The monthly passenger flow is , the average annual passenger flow in historical years is , then the historical year Monthly deviation factor for the month The formula is: .

[0010] Optionally, the customer flow prediction method based on the multi-level business district penetration model is characterized by: Set historical year Monthly deviation factor for the month , The population of the business district in the forecast year , the penetration rate of business districts at all levels in historical years , historical year Monthly historical error mean , If the business district belongs to the core business district, secondary business district and marginal business district, the first Monthly passenger flow The formula is: .

[0011] Optionally, the customer flow prediction method based on the multi-level business district penetration model is characterized by: Set up the first The actual monthly passenger flow is , No. The monthly passenger flow forecast is , the number of samples of the predicted data is ; The mean square error MSE formula is: ; The mean absolute error MAE formula is: ; The absolute percentage error MAPE formula is: .

[0012] The present application also provides a passenger flow prediction device based on a multi-level business district penetration model, characterized in that the device comprises: The business district basic calculation module is used to generate various parameters related to the historical business district passenger flow based on the data of the historical business district population and passenger flow; The business district forecast analysis module is used to generate the forecast year and the first Monthly passenger flow forecast; The business district calibration feedback module is used to generate an error indicator based on the error between the predicted customer flow value and the actual value.

[0013] Optionally, the business district basic calculation module includes: Business district data collection module, used to collect and organize business district demographics and customer flow; The commercial facility data collection module is used to collect and organize the work calendar, configuration information and turnover of commercial facilities.

[0014] The present application also provides an electronic device, characterized by comprising: A memory and a processor, wherein the memory stores a computer program run by the processor, and the processor runs any step of the passenger flow prediction method based on the multi-level business district penetration model in the computer program.

[0015] The beneficial effects of this application include: improving passenger flow forecasting accuracy by using population penetration differences and monthly offsets across various business districts. By generating error indicators between predicted and actual passenger flow values, we can compare indicators across multiple scenarios, enabling more accurate evaluation of forecasting models and guiding decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required in the embodiments or the description of the prior art. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0017] Figure 1 A flowchart showing a passenger flow prediction method based on a multi-level business district penetration model disclosed in this application is shown; Figure 2 A schematic diagram of each level of business districts is shown in a passenger flow prediction method based on a multi-level business district penetration model disclosed in this application. DETAILED DESCRIPTION

[0018] Various exemplary embodiments, features, and aspects of the present application will be described in detail below with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise indicated.

[0019] The terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature specified as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, "plurality" means two or more, unless otherwise specifically defined.

[0020] The word “exemplary” is used exclusively herein to mean “serving as an example, example, or illustration.” Any embodiment described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments.

[0021] In addition, numerous specific details are provided in the detailed description below to better illustrate the present application. Those skilled in the art will appreciate that the present application can be practiced without certain specific details. In some instances, methods, means, components, and circuits well known to those skilled in the art are not described in detail in order to highlight the main purpose of the present application.

[0022] This application is a passenger flow prediction method based on a multi-level business district penetration model. In this application, passenger flow predictions for each month of each level of business district are made based on the passenger flow data of commercial facilities in historical years, the business district corresponding to the commercial facility area and the set business district range size (core business district, secondary business district, marginal business district), and the error index between the predicted value and the actual value is calculated for dynamic correction.

[0023] Example 1 like Figure 1 FIG. 1 is a flowchart of a method for predicting customer flow based on a multi-level business district penetration model according to an embodiment of the present application, which specifically includes the following contents: S100, a business district basic calculation unit, is used to generate various parameters related to the business district passenger flow in historical years based on the data aggregation of the business district population and passenger flow in historical years.

[0024] Specifically, by collecting data from business districts in historical years, the total population and total passenger flow of each level of business districts in historical years are generated, so as to obtain the average annual number of times the population in the business district visits the business district in historical years, and the penetration rate of each level of business districts in historical years is generated based on the weight of the business district. In addition, the monthly deviation coefficient of a month is generated based on the ratio of the passenger flow in a certain month in the historical year to the average annual passenger flow of that year, providing data support for the passenger flow forecast of the business district.

[0025] S200, a business district prediction analysis unit is used to generate a forecast year and a forecast year according to various parameters of the business district passenger flow. Monthly passenger flow forecast.

[0026] Specifically, the monthly deviation coefficients of the historical years generated in S100 and the penetration rates of the business districts at each level are used to generate the passenger flow of each business district at a certain month in the forecast year based on the population of each business district at each level in the forecast year.

[0027] S300, a business district calibration feedback unit, is used to generate an error index according to the error between the passenger flow prediction value and the actual value.

[0028] Specifically, the error between the passenger flow of each level of business district in a certain month in the predicted year generated in S200 and the actual passenger flow of each level of business district in that month is calculated to generate a residual value to verify the accuracy of the prediction and dynamically correct the prediction.

[0029] To summarize, this application improves the accuracy of passenger flow prediction by adding data such as penetration rate, business district weight value, and average annual number of times in the passenger flow prediction calculation, and can dynamically correct the error between the predicted value and the actual value within the prediction time period to adapt to changes in the business district.

[0030] As an optional implementation scheme of the present application, optionally, in step S100, the business district basic calculation unit is used to generate various parameters related to the historical business district passenger flow based on the data aggregation of the historical business district population and passenger flow, including: S101, based on the annual population and monthly passenger flow of each level of business district in the historical year, generate the total passenger flow and total population of the business district in the historical year.

[0031] like Figure 2 As shown in the figure, each level of business district refers to the radiation range of a commercial facility, which is centered on its location and extends along a certain direction and distance to attract customers. The radiation range is divided into core business district, secondary business district, and marginal business district. Among them: Core business district: refers to an area with a radius of X kilometers from a commercial facility; Secondary business district: refers to the area with a radius of Y kilometers from a commercial facility (excluding the core business district); Marginal business district: refers to an area with a radius of Z kilometers from a commercial facility (excluding core business districts and secondary business districts); The radiation range of each level of business district must meet the following requirements: Z>Y>X.

[0032] The annual population refers to the total annual population of business districts at all levels, including permanent residents and floating population.

[0033] The monthly passenger flow refers to the total number of people passing through a commercial facility each month.

[0034] Specifically, the business district passenger flow calculation unit may further calculate the total population of the business district in historical years: in, Indicates business district The total population of the business district in 2018, Indicates business district Year The total population of the business district, Represents the year, Represents a specific year.

[0035] This formula calculates It can be used to evaluate the overall service scope or market potential of a business district. For example, it can be used to compare the total population change trend of the same business district in different years (e.g. and ), analyze whether the attractiveness of the business district has increased, and calculate the coverage rate of the business district in combination with the total population of the city. As a denominator, it can be used to calculate the population share of each level of business district in the business district, so that when allocating marketing budgets or commercial facilities, the investment weight can be adjusted according to the share of each level of business district.

[0036] The above calculation of the total population of the business district in historical years is the cornerstone of the business district population coverage analysis, and provides key input for subsequent penetration rate, customer flow and other models.

[0037] Specifically, the business district passenger flow calculation unit calculates the total passenger flow of the business district in the historical year, which is the sum of the passenger flow of each month in the historical year: in, Indicates business district The total passenger flow in the year, Indicates business district Passenger flow in month i of year, The value range is 1 to 12 (indicates each month), Represents the year, Represents a specific year.

[0038] Specifically, the formula compares different years The data can be used to obtain the annual growth rate of the business district to analyze the growth or decline of the overall attractiveness of the business district. It can identify seasonal patterns (such as peak months such as Spring Festival and summer vacation). For example, if If it's significantly higher than other months, it could be due to summer promotions or peak travel season. Finally, annual customer traffic can be used to allocate budgets for each business district and optimize monthly resource allocation.

[0039] The above calculation of the historical annual total passenger flow of the business district is used for trend analysis to evaluate long-term business vitality, optimize resource allocation through seasonal management, and the generated results are used to support subsequent forecasts and decisions.

[0040] S102: Generate an average annual number of visits to the business district by the population of the business district in the historical years based on the total passenger flow and total population of the business district in the historical years.

[0041] Specifically, the business district passenger flow calculation unit obtains the number of times an average person in the business district visits the business district each year based on the quotient of the total passenger flow of the business district in a certain year and the total population of the business district: in, Indicates business district The average number of times the business district population visits the business district in 2018, Indicates business district Annual total passenger flow in the business district, Indicates business district Total population of the business district in the year, Represents the year, Represents a specific year.

[0042] Specifically, this formula assesses a business district's attractiveness by calculating the average annual number of visits by its residents. A higher annual average number of visits indicates a stronger stickiness or frequency of consumption. The number of visits across different years can also be compared to analyze user trends. Resource allocation is adjusted based on the average annual number of visits to different business districts. For example, high-volume districts could be better served by strengthening service capabilities (such as more rest areas and cashier counters), while low-volume districts could be better served by increasing marketing incentives (such as membership benefits and promotions).

[0043] At the same time, the formula can also calculate the average annual number of visits by level, for example: in, express business district The average number of times the business district population visits the business district in 2018, express business district Annual total passenger flow in the business district, Indicates business district Year The total population of the business district, Belongs to the core business district, secondary business district and marginal business district.

[0044] In this way, the differences in user loyalty at each level can be obtained in order to optimize resource allocation.

[0045] The average annual number of visits to the business district by the population in the above historical years directly reflects the frequency of users’ visits and is one of the core indicators for measuring the vitality of the business district.

[0046] S103, generating the business district penetration rate of each level of business district in the historical years based on the average annual number of visits of the business district population in the historical years, the weight of each level of business district and the total population.

[0047] The penetration rate of each level of business district is used to measure the attractiveness and coverage of a commercial facility in each level of business district. Core business district penetration rate: the proportion of core business district population attracted by a commercial facility to the total customer flow during a certain period of time; Secondary business district penetration rate: the proportion of the secondary business district population attracted by a commercial facility to the total customer flow during a certain period of time; Permeability rate of marginal business districts: the proportion of the population of marginal business districts attracted by a commercial facility to the total customer flow within a certain period of time.

[0048] Specifically, the business district penetration rate calculation unit generates the business district penetration rate of each level of business district in the historical year according to the business district weight, the business district population and the annual average number of times: in, represents the penetration rate of the business district, Represents the weight of the business district, Indicates business district The population of the business district in the year Indicates business district The average number of times the business district population visits the business district in 2018, Represents the year, Represents a specific year.

[0049] Specifically, the numerator of the formula is ,express Weighted population of the business district (weight population ), reflecting the importance and coverage of the business district. The denominator is , represents the weighted population sum of all business districts and is used for normalization to make the penetration rates of business districts at different levels comparable. As a correction for access frequency, it reflects the direct impact of user activity on penetration rate. for The independent penetration rate of a business district is determined by the relative weight of the population and the frequency of visits to the business district. At the same time, after calculating the penetration rate of all business districts, the formula Generate the comprehensive penetration rate of the integrated business district.

[0050] The penetration rates of various business districts at various levels in the above historical years are primarily used to accurately identify high-value areas and dynamically optimize resource allocation. For example, high-penetration business districts can reduce subsidies and improve the customer experience, while low-penetration business districts need to increase traffic-driving activities (such as transportation connections and community marketing). Penetration rates can also be compared with those of similar business districts to identify strengths and weaknesses. Monitoring long-term trends in annual penetration rates can also be used to assess the lifecycle of a business district to support decision-making and implementation.

[0051] S104, according to the historical year Monthly passenger flow and historical year average passenger flow, generate historical year Monthly offset factor for month.

[0052] The monthly deviation coefficient refers to the degree of deviation of the passenger flow in the current month relative to a certain base month or the monthly average.

[0053] Specifically, the monthly deviation coefficient is: in For the Monthly deviation coefficient for month, for Year Monthly passenger flow, for The average annual passenger flow in the year.

[0054] Specifically, this formula can quantify the passenger flow fluctuations in different months to reflect seasonal patterns, such as when When the number of passengers in a month is higher than the annual average (such as during peak seasons such as Spring Festival and summer vacation), When the passenger flow of the month is lower than the annual average level (such as the rainy season, winter and other off-seasons), When the passenger flow of the month is on par with the annual average, users can use this as a basis to calculate the peak month ( ) or low month ( ) to dynamically allocate resources regarding inventory, costs, etc. and to conduct long-term analysis By analyzing the annual changes and historical changes over the same period, users can adjust their business strategies in a timely manner.

[0055] The above monthly deviation coefficient reveals patterns by quantifying passenger flow fluctuations, assisting decision-making from manpower allocation to marketing investment, and can be updated annually to achieve dynamic optimization and adapt to market changes.

[0056] As an optional implementation scheme of the present application, optionally, in step S200, the business district prediction analysis unit is used to generate the predicted year-first according to various parameters of the business district passenger flow. Monthly passenger flow forecast, including: S201, generating monthly passenger flow for each level of business district in the forecast year based on the monthly deviation coefficient of the historical year, the historical error mean of the historical year, the business district penetration rate of each level of business district in the historical year, and the annual population of each level of business district in the forecast year.

[0057] Specifically, the first level of each level of business district in the forecast year The monthly passenger flow formula is: in, Indicates the first level of each level of business district in the forecast year Monthly passenger flow, Representative forecast year Business district population, Indicates the penetration rate of business districts at all levels in historical years, For historical years Monthly deviation coefficient of the month, historical year Monthly historical error mean .

[0058] Specifically, the numerator in this formula For the population at each business district level Its penetration rate The sum of the products of represents the theoretical annual total passenger flow in the forecast year, and the denominator The annual passenger flow is evenly distributed to the monthly average base to eliminate the impact of the total scale. and the denominator Together they constitute the weighted annual passenger flow base. The monthly average base is adjusted to the forecast value to reflect monthly characteristics (such as the Spring Festival peak season and the rainy season low season). This formula is suitable for medium- and long-term passenger flow forecasting and budget allocation.

[0059] The first level of each business district in the above forecast years The monthly passenger flow formula predicts future passenger flow based on historical penetration rates and deviation patterns.

[0060] S202, assigning business district weights based on the business district types of core business district, secondary business district, and marginal business district, and obtaining the total weights of the business districts in historical years according to the weights of the business districts and the historical population.

[0061] Specifically, the business district penetration rate calculation unit divides each level of business district into three layers according to the hierarchical structure. The first-level core business district can cover 60-80% of customers and has a high penetration rate, so it is assigned a weight of 3; the second-level business district can cover 15-25% of customers and has a medium penetration rate, so it is assigned a weight of 2; the third-level marginal business district can cover 5-15% of customers and has a low penetration rate, so it is assigned a weight of 1. The business district range weight is generated by summing up the product of the weight of each level of business district and the population: in, Indicates business district The sum of the weights of the years, , , They are the weights of the core business district, secondary business district and marginal business district respectively. , , Business districts The population of the core business district, secondary business district and marginal business district in a year, Represents the year, Represents a specific year.

[0062] The sum of the weights of the above business districts in different years can be obtained by comparing different years. changes, evaluate the overall development trend of the business district, and combine the population change trend of the business district to predict the future development potential of the business district.

[0063] As an optional implementation scheme of the present application, optionally, in step S300, the business district calibration feedback unit is configured to generate an error indicator based on the error between the passenger flow prediction value and the actual value, including: S301: Generate a mean square error based on the average of the square errors between the predicted value and the actual value.

[0064] Specifically, the mean square error (MSE) is generated based on the average of the squared errors between the predicted value and the actual value. The unit is consistent with the passenger flow, and the smaller the value, the more accurate the prediction.

[0065] in, For the The actual passenger flow of the month, For the Monthly predicted passenger flow, is the number of samples for prediction data.

[0066] The mean square error (MSE) is more sensitive to outliers (large errors) due to the square amplification effect. It is suitable for optimizing parameters during model training (such as the gradient descent algorithm) or scenarios where extreme errors need to be strictly controlled.

[0067] S302: Generate a mean absolute error based on the absolute value of the error between the predicted value and the actual value.

[0068] Specifically, the mean absolute error (MAE) is generated based on the absolute value of the error between the predicted value and the actual value. The smaller the value, the lower the average error of the prediction.

[0069] in, For the The actual passenger flow of the month, For the Monthly predicted passenger flow, is the number of samples for prediction data.

[0070] The linear penalty of the mean absolute error (MAE) ensures that all errors are treated with equal weights and are not excessively affected by extreme values. It is suitable for scenarios with uniform error distribution (such as passenger flow in normal months).

[0071] S303: Generate a mean absolute percentage error based on the average percentage of the predicted error to the actual value.

[0072] Specifically, the mean absolute percentage error (MAPE) is generated based on the average percentage of the prediction error to the actual value to measure the relative error of the model. The smaller the value, the more reliable the prediction.

[0073] in, For the The actual passenger flow of the month, For the Monthly predicted passenger flow, is the number of samples for prediction data.

[0074] The relative error of the mean absolute percentage error (MAPE) is expressed in percentage form, which can eliminate the impact of data scale and is suitable for comparing data sets of different sizes (such as comparing core business districts and peripheral business districts).

[0075] Example 2 Based on the same principle as the above method, a passenger flow prediction device based on a multi-level business district penetration model is also proposed. A Drools-based rule engine efficiency optimization device 100 of an embodiment of the present disclosure includes: 110, a business district basic calculation module, for generating various parameters related to the business district passenger flow in historical years based on the data aggregation of the business district population and passenger flow in historical years; 120, a business district forecast analysis module, for generating a forecast year and a forecast year according to various parameters of the business district passenger flow. Monthly passenger flow forecast; 130, a business district calibration feedback module, is used to generate an error indicator according to the error between the passenger flow prediction value and the actual value.

[0076] As an optional implementation scheme of the present application, optionally, the business district basic calculation module 110 includes: 111, the business district data collection module, is used to collect and organize business district demographics and customer flow; 112. Commercial facility data collection module, used to collect and organize the work calendar, configuration information and turnover of commercial facilities.

[0077] Obviously, those skilled in the art should understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program. The program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned control methods. The modules or steps of the present invention can be implemented by a general-purpose computing device. They can be concentrated on a single computing device or distributed on a network composed of multiple computing devices. Alternatively, they can be implemented by program codes executable by the computing device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. In this way, the present invention is not limited to any specific combination of hardware and software.

[0078] Those skilled in the art will appreciate that all or part of the processes in the above-described embodiments can be implemented by instructing the relevant hardware through a computer program. The program can be stored in a computer-readable storage medium. When executed, the program can include the processes of the above-described control method embodiments. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), a flash memory, a hard disk drive (HDD), or a solid-state drive (SSD). The storage medium can also include a combination of the above-mentioned types of memory.

[0079] Example 3 The present application provides an electronic device, including: a memory and a processor.

[0080] Wherein, the processor is configured to implement the passenger flow prediction method based on the multi-level business district penetration model described in Example 1 when executing the executable instructions.

[0081] The electronic device of the embodiment of the present disclosure includes a processor and a memory for storing processor-executable instructions, wherein the processor is configured to implement any of the aforementioned passenger flow prediction methods based on a multi-level business district penetration model when executing the executable instructions.

[0082] It should be noted that the number of processors can be one or more. Furthermore, the electronic device in the embodiments of the present disclosure may also include an input device and an output device. The processor, memory, input device, and output device may be connected via a bus or other means, which are not specifically limited herein.

[0083] The memory, as a computer-readable storage medium for the customer flow prediction method based on a multi-level commercial district penetration model, can be used to store software programs, computer executable programs, and various modules, such as the program corresponding to the customer flow prediction method based on a multi-level commercial district penetration model in the embodiments of the present disclosure. The processor executes the software programs stored in the memory to perform various functional applications and data processing of the electronic device.

[0084] The input device can be used to receive input numbers or signals. The signals can be key signals related to user settings and function control of the device / terminal / server. The output device can include a display device such as a display screen.

[0085] The embodiments of the present application have been described above. The above description is illustrative and not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is selected to best explain the principles of the embodiments, their practical applications, or improvements to the technology in the market, or to enable other persons skilled in the art to understand the embodiments disclosed herein.

Claims

1. A passenger flow prediction method based on a multi-level business district penetration model, characterized in that: include: The business district basic calculation unit is used to generate various parameters related to the business district passenger flow in historical years based on the data of the business district population and passenger flow in historical years; The business district forecast analysis unit is used to generate the forecast year and the first Monthly passenger flow forecast; The business district calibration feedback unit is used to generate an error indicator based on the error between the predicted customer flow value and the actual value.

2. The customer flow prediction method based on the multi-level business district penetration model according to claim 1, characterized in that: The business district basic calculation unit includes: Based on the annual population and monthly passenger flow of each level of business district in the historical year, the total passenger flow and total population of the business district in the historical year are generated; Generate an average annual number of visits to the business district by the population of the business district in the historical year based on the total passenger flow and total population of the business district in the historical year; Generate the business district penetration rate of each level of business district in the historical year based on the average annual number of visits to the business district by the business district population in the historical year, the weight of each level of business district and the total population; According to historical years Monthly passenger flow and historical year average passenger flow, generate historical year Monthly offset factor for month.

3. The customer flow prediction method based on the multi-level business district penetration model according to claim 1, characterized in that: The business district prediction and analysis unit includes: Based on the monthly deviation coefficient of the historical year, the historical error mean of the historical year, the business district penetration rate of each level of business district in the historical year and the annual population of each level of business district in the forecast year, the monthly passenger flow of each level of business district in the forecast year is generated.

4. The customer flow prediction method based on the multi-level business district penetration model according to claim 1, characterized in that: The business district calibration feedback unit includes: Generate the mean square error based on the average of the squared errors between the predicted value and the actual value; Generate the mean absolute error based on the absolute value of the error between the predicted value and the actual value; Generates the mean absolute percentage error based on the average percentage of the predicted errors relative to the actual values.

5. The customer flow prediction method based on the multi-level business district penetration model according to claim 2, characterized in that: The penetration rate of each level of business district in the historical years is as follows: The weight of the business district is , The population of the business district in the historical year is , the average annual number of times the business district population visits the business district in historical years is , The business district belongs to the core business district, secondary business district and marginal business district, and the business district penetration rate of each level of business district in historical years The formula is: ； The historical year Monthly deviation coefficient of the month, set the historical year The monthly passenger flow is , the average annual passenger flow in historical years is , then the historical year Monthly deviation factor for the month The formula is: 。 6. The customer flow prediction method based on the multi-level business district penetration model according to claim 3, characterized in that: Set historical year Monthly deviation factor for the month , The population of the business district in the forecast year , the penetration rate of business districts at all levels in historical years , historical year Monthly historical error mean , If the business district belongs to the core business district, secondary business district and marginal business district, the first Monthly passenger flow The formula is: 。 7. The customer flow prediction method based on the multi-level business district penetration model according to claim 4, characterized in that: Set up the first The actual monthly passenger flow is , No. The monthly passenger flow forecast is , the number of samples of the predicted data is ; The mean square error MSE formula is: ； The mean absolute error MAE formula is: ； The absolute percentage error MAPE formula is: 。 8. A passenger flow prediction device based on a multi-level business district penetration model, characterized in that: The device comprises: The business district basic calculation module is used to generate various parameters related to the historical business district passenger flow based on the data of the historical business district population and passenger flow; The business district forecast analysis module is used to generate the forecast year and the first Monthly passenger flow forecast; The business district calibration feedback module is used to generate an error indicator based on the error between the predicted customer flow value and the actual value.

9. The passenger flow prediction system based on the multi-level business district penetration model according to claim 8 is characterized in that: The business district basic calculation module includes: Business district data collection module, used to collect and organize business district demographics and customer flow; The commercial facility data collection module is used to collect and organize the work calendar, configuration information and turnover of commercial facilities.

10. An electronic device, characterized in that: include: A memory and a processor, wherein the memory stores a computer program run by the processor, and the processor runs the computer program to implement the steps of the passenger flow prediction method based on the multi-level business district penetration model as described in any one of claims 1 to 7.

Images