Method and device for determining delivery turnover of store, electronic equipment and medium

Abstract

The invention discloses a method and device for determining the delivery turnover of a store, electronic equipment and a medium. The method comprises the steps of obtaining a new store delivery range of a newly opened store and historical order data of the opened store; according to the new store distribution range, determining an overlapped distribution area of the newly opened store and the opened store; according to the historical order data, determining the total order number of the opened stores in the overlapped delivery area and a first customer unit price mean value corresponding to the total order number; through a specified data analysis model, estimating a target order quantity of which the total order quantity is to be shared by a newly opened store; the specified data analysis model is used for determining the probability of selecting a new store by a user in the overlapped distribution area; and according to the target order quantity and the first customer unit price mean value, estimating the target delivery turnover of the newly opened store. Therefore, the number of the target orders shared by the newly-opened store is accurately estimated through the data analysis model, the situation that an average distribution principle ignores the tendency difference of selecting the store by different users is avoided, and the distribution analysis precision of the newly-opened store is improved.

Description

Technical Field

[0001] This application relates to the field of operational analysis technology, and in particular to a method, apparatus, electronic device and medium for determining store delivery revenue. Background Technology

[0002] In existing store operation analysis systems, the distribution of population-rich areas such as residential areas, hospitals, schools, and office buildings around the store is typically used to count the population in the corresponding area and further calculate the number of orders that an existing store can cover within its delivery service radius. Based on this, by analyzing the relationship between delivery revenue and the population covered, a "contribution rate per 10,000 people" indicator is constructed to measure the contribution level of delivery performance per unit of population and to serve as a basis for estimating the delivery revenue of newly opened stores.

[0003] However, this method has several significant drawbacks in practical applications. First, due to differences in consumption capacity and habits across different regions, the contribution rate per 10,000 people for each store may fluctuate significantly. If the contribution rate per 10,000 people of existing stores is used as a benchmark for new store forecasting, the estimated turnover may deviate from the actual situation because regional characteristics are not fully considered.

[0004] Secondly, when assessing the overlap effect of delivery range between newly opened stores and existing stores, existing methods usually use the principle of average allocation to divide the number of orders in the overlapping area, ignoring the differences in the preferences of consumers in different residential areas when choosing stores. Especially for communities in remote locations, the order allocation results may not match the actual consumption behavior, thus affecting the accuracy of diversion estimation.

[0005] Therefore, how to accurately predict the delivery revenue of newly opened stores is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0006] In view of this, one aspect of this application provides a method for determining store delivery revenue, the method comprising: Obtain the delivery range of newly opened stores and historical order data of existing stores; Based on the delivery range of the new store, determine the overlapping delivery areas between the newly opened store and the existing store; Based on the historical order data, determine the total number of orders of the opened stores in the overlapping delivery area and the average first order value corresponding to the total number of orders; By using a specified data analysis model, the target number of orders that will be distributed to the newly opened stores is estimated; the specified data analysis model is used to determine the probability that users in the overlapping delivery area will choose the newly opened stores; Based on the target number of orders and the average first order value, the target delivery revenue of the newly opened store is estimated.

[0007] Optionally, obtain the delivery range for newly opened stores, including: Determine whether there is a target existing store within a circular area formed by the newly opened store as the center and a first preset radius; If it exists, perform the following steps: Obtain the furthest delivery time for each of the target stores that have already opened; and determine the average duration of each of the furthest delivery times; Starting from the newly opened store, the target delivery destination is obtained by traversing each route and taking the average travel time. Based on the latitude and longitude data of the target delivery destination, generate interest surfaces corresponding to each target delivery destination; The interest surfaces are connected to generate a closed area, which serves as the delivery range for the newly opened store.

[0008] Optionally, if there is no target store within the circular area, perform the following steps: Obtain a pre-constructed first mapping relationship; the first mapping relationship is the correspondence between the farthest delivery time and the delivery influencing factor; the delivery influencing factor includes at least one of the city type and population density within the specified area; Determine the delivery influencing factors for the newly opened stores; Based on the mapping relationship and the delivery impact factors of the newly opened stores, determine the target maximum delivery time corresponding to the newly opened stores; Starting from the newly opened store, the system traverses each route and takes the longest delivery time to reach the target destination to obtain the target delivery endpoint; then it proceeds to the step of generating interest surfaces corresponding to each target delivery endpoint based on the latitude and longitude data of the target delivery endpoint, and executes subsequent steps.

[0009] Optionally, determining the overlapping delivery area between the newly opened store and the existing store based on the delivery range of the new store includes: Based on the delivery distance of the old store corresponding to the historical order data, the historical order data is filtered to obtain the target historical order data; the filtering process is to remove data whose old store delivery distance is greater than a threshold, or to remove data with a preset percentage at the end of the ascending sort result of the old store delivery distance; Determine the latitude and longitude data corresponding to points of interest in the target historical order data; The latitude and longitude data are projected to obtain interest surfaces; and the interest surfaces are connected to generate a closed area as the delivery range of the existing stores. The overlapping portion of the delivery range of the new store and the delivery range of the old store shall be defined as the overlapping delivery area.

[0010] Optionally, the specified data analysis model is the Huff model; the step of estimating the target number of orders that will be distributed among the newly opened stores using the specified data analysis model includes: Obtain the new store delivery distance between the newly opened store and the overlapping delivery area, and the preset value of the attenuation coefficient in the Huff model; the attenuation coefficient includes at least the distance attenuation coefficient; the distance attenuation coefficient is used to reflect the degree of influence of distance on the user's choice of store; Determine the new store attractiveness parameters of the newly opened store; the new store attractiveness parameters are used to characterize the degree of influence of the attractiveness factors of the newly opened store on users' store selection; the attractiveness factors include at least one of store area, number of product types and product price; Based on the new store delivery distance, the preset value, and the new store attractiveness parameter, the target probability value that the total number of orders will be shared by the newly opened store is determined through the Huff model; The target order quantity is determined based on the total number of orders and the target probability value.

[0011] Optionally, estimating the target delivery revenue of the newly opened store based on the target order quantity and the average first order value includes: The newly opened store is located within the single-layer delivery area of ​​the new store, excluding the overlapping delivery area. According to preset rules, the new store single-layer capture rate is determined; the new store single-layer capture rate is the probability of delivery orders in the single-layer delivery area; The single-layer delivery revenue of the newly opened store is determined based on the single-layer capture rate of the new store. The product of the target order quantity and the average first order value is used as the delivery revenue of the overlapping area of ​​the newly opened store. The sum of the single-layer delivery revenue and the overlapping area delivery revenue is taken as the target delivery revenue.

[0012] Optionally, determining the new store single-layer capture rate of the newly opened store according to preset rules includes: Determine whether there is a target existing store within a circular area formed by the newly opened store as the center and an initial preset radius; If it exists, the average single-layer capture rate of the existing stores of the target stores shall be used as the single-layer capture rate of the new stores; If not, determine whether the target existing store exists within the preset area where the new store is located; if it exists, execute the step of using the average single-layer capture rate of the existing stores of the target existing stores as the single-layer capture rate of the new store; the circular area is within the preset area.

[0013] Optionally, if there is no target store already opened within the preset area, the following steps are performed; Obtain a pre-constructed second mapping relationship; the second mapping relationship is the correspondence between the unopened store area, the opened store area, and the single-layer capture rate of the opened store area; Determine the target area of ​​existing stores corresponding to the preset area where the new store is located; Based on the second mapping relationship, the single-layer capture rate corresponding to the target already opened store area is used as the single-layer capture rate of the new store.

[0014] Optionally, determining the single-layer delivery revenue of the newly opened store based on the single-layer capture rate of the new store includes: Obtain the permanent resident population of the single-layer delivery area; Determine the average second order value of historical orders from existing stores corresponding to the single-layer capture rate of the new store; The product of the resident population, the single-layer capture rate of the new store, and the average order value is used as the single-layer delivery revenue.

[0015] Another aspect of this application provides an apparatus for determining store delivery revenue, the apparatus comprising: The target acquisition module is used to obtain the delivery range of newly opened stores and historical order data of existing stores; The overlapping area determination module is used to determine the overlapping delivery area between the newly opened store and the existing store based on the delivery range of the new store. The order information determination module is used to determine the total number of orders of the opened stores in the overlapping delivery area and the average first order price corresponding to the total number of orders based on the historical order data. The order forecasting module is used to forecast the target number of orders that will be distributed to the newly opened stores by the total number of orders through a specified data analysis model; the specified data analysis model is used to determine the probability that users in the overlapping delivery area will choose the newly opened stores; The delivery revenue determination module is used to estimate the target delivery revenue of the newly opened store based on the target number of orders and the average first order value.

[0016] Another aspect of this application provides an electronic device including a memory and a processor, wherein the memory stores a computer program executable on the processor, and the processor executes the computer program to implement the steps of the method for determining the store delivery revenue.

[0017] Another aspect of this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for determining the store's delivery revenue.

[0018] The method, apparatus, electronic device, and medium for determining store delivery revenue provided in this application have the following beneficial effects: by accurately predicting the target order quantity to be shared by newly opened stores through a data analysis model, the principle of average allocation avoids ignoring the differences in the preferences of different users for choosing stores, improves the accuracy of traffic diversion analysis for newly opened stores, thereby improving the accuracy of determining the target delivery revenue of newly opened stores, and providing accurate data support for store layout and resource allocation. Attached Figure Description

[0019] Figure 1 A flowchart illustrating a method for determining store delivery revenue provided in an embodiment of this application; Figure 2 This is a schematic diagram of an overlapping delivery area provided in an embodiment of this application; Figure 3 A flowchart illustrating a method for obtaining the delivery range of a newly opened store, provided in an embodiment of this application; Figure 4 A flowchart illustrating a method for determining store delivery revenue according to another embodiment of this application; Figure 5 A schematic diagram of an overlapping delivery area provided for another embodiment of this application; Figure 6 A flowchart illustrating a method for determining store delivery revenue provided in another embodiment of this application; Figure 7 A schematic diagram of the structure of a device for determining store delivery revenue provided in an embodiment of this application; Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0020] The attached diagram is labeled as follows: 70 is the target acquisition module, 71 is the overlapping area determination module, 72 is the order information determination module, 73 is the order estimation module, 74 is the delivery revenue determination module, 80 is the memory, 81 is the processor, 82 is the display screen, 83 is the input / output interface, 84 is the communication interface, 85 is the power supply, 86 is the communication bus, 801 is the computer program, 802 is the operating system, and 803 is the data. Detailed Implementation

[0021] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0022] It should be understood that although the terms first, second, third, etc., may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0023] Figure 1 A flowchart illustrating a method for determining store delivery revenue provided in this application embodiment is shown below. Figure 1 As shown, the method includes: S10: Obtain the delivery range of newly opened stores and historical order data of existing stores; S11: Based on the delivery range of the new store, determine the overlapping delivery areas between the newly opened store and the existing store; In a specific implementation of estimating the delivery revenue of newly opened stores, it is understood that the delivery areas of newly opened stores may overlap with those of existing stores, making accurate estimation of delivery revenue in these overlapping areas crucial. Therefore, in this specific implementation, it is necessary to determine the overlapping delivery areas between newly opened and existing stores.

[0024] Specifically, obtain the delivery range of newly opened stores and the delivery range of existing stores, and determine the overlapping delivery areas based on the delivery ranges of the new and existing stores. Figure 2 This is a schematic diagram of an overlapping delivery area provided in an embodiment of this application. For ease of understanding, it will be described below in conjunction with... Figure 2 Please provide an explanation.

[0025] like Figure 2 As shown, the delivery range of the newly opened store A is represented by the blue rectangle, and the delivery range of the existing store B is represented by the green circle. The overlapping area between the newly opened store A and the existing store B is the overlapping delivery area (i.e., the purple part).

[0026] S12: Based on historical order data, determine the total number of orders from existing stores within the overlapping delivery area and the average first order value corresponding to the total number of orders; S13: Using a specified data analysis model, estimate the target number of orders that will be distributed to the newly opened stores; the specified data analysis model is used to determine the probability that users in the overlapping delivery area will choose the newly opened stores; according to Figure 2 As can be seen, in the specific implementation, there is an overlap in delivery areas between the newly opened stores and the existing stores. In the overlapping delivery areas, some of the orders of the existing stores will be diverted to the newly opened stores. Therefore, when estimating the target delivery revenue of the newly opened stores, it is necessary to determine the target number of orders that the existing stores in the overlapping delivery areas will be diverted to the newly opened stores.

[0027] Specifically, step S10 first obtains historical order data from already opened stores. It should be noted that this historical order data includes, but is not limited to, the location data of the order delivery destination, delivery time, delivery route, and total number of orders. Furthermore, it should be noted that the historical order data can be data within a preset historical time period, such as within a week or a month. This application does not impose any restrictions on this; the appropriate data can be selected based on actual business needs.

[0028] Furthermore, based on historical order data, the total number of orders from existing stores within overlapping delivery areas can be determined, along with the average first order value corresponding to that total. Based on this, it is necessary to determine how many of the total orders will be diverted to newly opened stores. In one optional embodiment, a specified data analysis model can be used to estimate the target number of orders that will be diverted to newly opened stores.

[0029] The specified data analysis model can be used to determine the probability that a user will choose a newly opened store within an overlapping delivery area. After determining the probability of choosing a newly opened store, the product of the total number of orders from existing stores and this probability value is the target order quantity. It should be noted that the specified data analysis model may include, but is not limited to, the Huff model and the competitive interaction model, and this application does not limit it.

[0030] S14: Estimate the target delivery revenue for the new store based on the target number of orders and the average first order value.

[0031] Furthermore, based on the target order data and average first-order value that newly opened stores in overlapping delivery areas will receive, the delivery revenue of these stores can be determined. It is worth noting that, for example... Figure 2 As shown in the specific embodiment, in addition to the overlapping delivery area, the target delivery revenue of the newly opened store also includes the area independently delivered by the newly opened store, namely the blue area in the rectangle. The revenue of this area is used as the single-layer delivery revenue of the newly opened store. The single-layer delivery revenue refers to the revenue generated by the newly opened store's own independently delivered orders.

[0032] It should be noted that the average single-floor delivery revenue of existing stores within a preset nearby area can be used as the single-floor delivery revenue of the new store. This application does not limit the method for determining the single-floor delivery revenue of the new store. Furthermore, after determining the single-floor delivery revenue and the delivery revenue of the overlapping area, the sum of the two is the target delivery revenue of the new store.

[0033] In another alternative embodiment, there may be no existing stores with overlapping delivery areas with the newly opened store, that is, there is no overlapping area delivery revenue. In this case, the single-layer delivery revenue can be used as the target delivery revenue.

[0034] Therefore, the method for determining store delivery revenue provided in this application accurately predicts the target order quantity to be shared by newly opened stores through a data analysis model, avoids ignoring the differences in the preference of different users for stores by the principle of average allocation, improves the accuracy of the traffic diversion analysis of newly opened stores, thereby improving the accuracy of determining the target delivery revenue of newly opened stores and providing accurate data support for store layout and resource allocation.

[0035] Figure 3 The flowchart of a method for obtaining the delivery range of a newly opened store, provided as an embodiment of this application, includes, in one optional embodiment, the method of obtaining the delivery range of a newly opened store as follows: S1 determines whether there is a target existing store within the circular area formed by the new store as the center and the first preset radius; if so, proceed to steps S20 to S23: S20: Obtain the furthest delivery time for each target's existing stores; and determine the average duration of each furthest delivery time. S21: Starting from the newly opened store, traverse all routes and calculate the average travel time to obtain the target delivery destination. S22: Generate interest surfaces corresponding to each target delivery destination based on the latitude and longitude data of the target delivery destination; S23: Connect interest areas to generate a closed area as the delivery range for newly opened stores.

[0036] In specific embodiments, it is understood that accurately determining the delivery range of the new store is crucial for subsequently determining the overlapping delivery area, thereby ensuring the accuracy of the final target revenue forecast for the new store. Therefore, in one optional embodiment, to ensure the accuracy and reasonableness of the delivery range of the new store, the target existing stores are used as a reference standard for determination.

[0037] Specifically, a circle is drawn with the newly opened store as the center and a first preset radius to obtain a circular area. For example, a circular area with a radius of 5 kilometers is formed with the newly opened store as the center. It is then determined whether there is a target existing store within this circular area. If there is, the delivery range of the new store is determined based on the target existing store.

[0038] like Figure 3 As shown, firstly, based on the historical order data of the target's existing stores, the furthest delivery time for all target stores within the circular area is obtained, and the average of all furthest delivery times is calculated. Further, starting from the newly opened store, all drivable paths in the surrounding area are traversed, with each path's travel time being the average of the furthest delivery times for the target's existing stores. It should be noted that the traversed paths are those accessible to non-motorized vehicles.

[0039] After traversing all paths, the latitude and longitude data of each target delivery destination are obtained, and Areas of Interest (AOIs) are generated based on the latitude and longitude data of each destination. Furthermore, all AOIs are stitched together to generate a closed area, which can be used as the delivery range for the newly opened store.

[0040] AOI refers to an area with defined geographical boundaries, representing a functional area or spatial range, not just a single point. Essentially, it's area data—the region corresponding to the target delivery destination, such as a residential community, industrial park, park, or commercial district. Understandably, a closed area generated based on AOI is usually an irregular delivery area.

[0041] It should be noted that this application does not limit the first preset radius of the circular area; it can be set according to different cities or population densities. When the city where the new store is located is a first-tier city, a smaller value should be selected for the first preset radius. That is, the city level is negatively correlated with the first preset radius; the higher the city level, the smaller the first preset radius. Of course, it can also be set based on population density. Specifically, population density is also negatively correlated with the first preset radius; that is, the lower the population density, the larger the first preset radius.

[0042] Therefore, by using nearby existing target stores as a benchmark to determine the delivery range of new stores, a more accurate and realistic delivery range can be obtained. This can ensure the accuracy of subsequent determination of overlapping delivery areas, and thus improve the accuracy of the target delivery revenue forecast for new stores.

[0043] It is understandable that the method for determining store delivery revenue provided in this application must determine the delivery range of the new store. Therefore, based on the above embodiments, as... Figure 3 As shown, if there are no target stores already open within the circular area, proceed to steps S30 to S33: S30: Obtain the pre-built first mapping relationship; the first mapping relationship is the correspondence between the farthest delivery time and the delivery influencing factor; the delivery influencing factor includes at least one of the city type and population density in the specified area; S31: Determine the delivery influencing factors for newly opened stores; S32: Based on the mapping relationship and the delivery impact factors of the newly opened stores, determine the target maximum delivery time for the newly opened stores; First, it should be noted that the target existing stores within the circular area in the above embodiment are different from the existing stores in the overlapping delivery area. It is understandable that when determining the target delivery revenue for a newly opened store, it is necessary to determine the delivery range of the new store. Therefore, it is necessary to obtain a target existing store as a reference to determine the delivery range of the new store. However, existing stores with overlapping delivery areas with the new store may not necessarily exist.

[0044] However, in an alternative embodiment, if there is no target store within the circular area, but it is necessary to obtain a store as a reference, a reference store that can be used as the target store can be obtained according to the pre-constructed first mapping relationship, that is, the time of obtaining the traversal path of the newly opened store.

[0045] The first mapping relationship is the correspondence between the furthest delivery time and the delivery influencing factor, where the delivery influencing factor includes at least one of the city type and population density within the specified area. This can be understood as follows: if there are no existing stores in the area where the newly opened store is located, existing stores in other cities can be used as a benchmark. Table 1 is a schematic table of one such first mapping relationship provided in the embodiments of this application. For ease of understanding, the following explanation will be based on Table 1.

[0046] Table 1 is a schematic diagram of a first type of mapping relationship.

[0047] Based on Table 1, in one optional embodiment, the delivery impact factors of the newly opened store are first determined. For example, based on population density, the farthest delivery time corresponding to the difference between the population density of the area where the newly opened store is located and the population density difference in the first mapping relationship is within a preset difference range can be used as the target farthest delivery time.

[0048] Alternatively, the city type of the newly opened store can be determined, and the furthest delivery time corresponding to a specified area with the same city type can be found in the first mapping relationship as the target furthest delivery time. Of course, population density can also be combined with city type to determine the target furthest delivery time. Specifically, the target furthest delivery time can be the population density difference between cities with the same type and within a preset difference range.

[0049] In another alternative embodiment, the same city can also be divided into different areas, such as Beijing's Tongzhou District and Beijing's Fangshan District. When a new store is opened in Beijing's Fangshan District, the longest delivery time corresponding to the store in Tongzhou District can be used as the longest delivery time for the new store in Fangshan District.

[0050] It should be noted that the table above is only an example implementation. In fact, this application does not limit the storage method and structure of the first mapping relationship. As long as a comparable designated area with similar delivery influencing factors such as city type and population density to the newly opened store can be found based on the first mapping relationship, the furthest delivery time of the comparable area can be obtained.

[0051] S33: Starting from the newly opened store, traverse all routes and find the longest delivery time to the target destination; then proceed to step S22 and execute subsequent steps.

[0052] Furthermore, such as Figure 3 As shown, after obtaining the target maximum delivery time for the newly opened store, similarly, starting from the newly opened store, all drivable paths in the surrounding area are traversed, with the drivable path time being the maximum delivery time. Then, proceed to steps S22 and S23 to generate a closed delivery area for the new store.

[0053] Therefore, based on the method provided in this application embodiment, even if there is no target store within the circular area, a comparable area can be found based on the pre-built first mapping relationship. The duration of the traversal path is determined based on the comparable area, ensuring the accuracy of the new store's delivery range determination, thereby ensuring the accuracy of the new store's target delivery revenue determination.

[0054] In one optional embodiment, determining the overlapping delivery area between the newly opened store and the existing store, based on the new store's delivery range, includes: Based on the delivery distance of the original stores corresponding to the historical order data, the historical order data is filtered to obtain the target historical order data. The filtering process is to remove data whose original store delivery distance is greater than a threshold, or to remove data with a preset percentage at the end of the ascending sorted results of the original store delivery distance. Determine the latitude and longitude data corresponding to points of interest in the target historical order data; The latitude and longitude data are projected to obtain interest surfaces; then the interest surfaces are connected to generate closed areas as the delivery range of existing stores. The overlapping area between the delivery range of the new store and the delivery range of the old store will be designated as the overlapping delivery area.

[0055] Based on the above embodiments, after determining the delivery range of the new store, it is necessary to determine the delivery area of ​​the old store when determining the overlapping delivery area. Specifically, the delivery distance of the old store is extracted from the historical order data, and outliers are removed based on the delivery distance to obtain the target historical order data. Specifically, data with old store delivery distances greater than a threshold are removed, that is, data with excessively long delivery distances are removed. Alternatively, the old store delivery distances are sorted in ascending order, that is, sorted from smallest to largest, and data at the end of the sorted results with a preset percentage (e.g., the last 20%) are removed. Thus, after filtering the historical order data, errors that may occur when reconstructing the latitude and longitude data of the orders can be avoided, thereby reducing the accuracy of determining the overlapping delivery area.

[0056] Similarly, following the same principle as the above embodiments, the latitude and longitude data corresponding to the points of interest in the target historical order data are obtained, and the latitude and longitude data are projected to obtain the AOI. After the AOIs are stitched together, a closed area can be obtained, which can be used as the delivery range of the old store. Therefore, in a specific embodiment, such as Figure 2 As shown, the overlapping area between the delivery range of the old store and the delivery range of the new store is defined as the overlapping delivery area.

[0057] Figure 4 The flowchart illustrates a method for determining store delivery revenue according to another embodiment of this application. In an optional embodiment, the data analysis model is specified as the Huff model. By specifying the data analysis model, the target number of orders that will be distributed to newly opened stores is estimated, including: S40: Obtain the new store delivery distance between the newly opened store and the overlapping delivery area, as well as the preset value of the attenuation coefficient in the Huff model; the attenuation coefficient includes at least the distance attenuation coefficient; the distance attenuation coefficient is used to reflect the degree of influence of distance on the user's choice of store; To facilitate understanding, we will use the Huff model as an example to illustrate the estimation of the target order quantity. The Huff model is a classic consumer behavior modeling method used to predict the distribution of customer flow between competing locations (stores with overlapping delivery areas).

[0058] When using the Huff model, the decay coefficient in the model is crucial for accurately predicting the probability of a new store diverting a target number of orders. It's understandable that factors such as delivery distance, store size, store reputation, and historical order reviews all influence a user's final store choice. To accurately predict the number of orders from existing stores in overlapping delivery areas that will be diverted to a new store, it's necessary to quantify these influencing factors.

[0059] Therefore, in a specific embodiment, the new store delivery distance between the newly opened store and the overlapping delivery area is obtained, along with the preset value of the attenuation coefficient in the Huff model. The new store delivery distance can be the shortest delivery distance from each historical order in the overlapping delivery area to the newly opened store.

[0060] It is understandable that, in specific embodiments, if the number of historical orders in the overlapping delivery area is very large, it will inevitably consume a lot of computing resources, affecting the estimation efficiency. Therefore, in an optional embodiment, the overlapping delivery area can be divided into multiple grid areas of preset size, and the delivery distance to the new store can be determined based on the grid areas. Specifically, the shortest drivable path from the center of each grid area to the newly opened store is used as the delivery distance to the new store, and this distance serves as the delivery distance to the new store for all orders within the grid area.

[0061] The attenuation coefficient may include, but is not limited to, the distance attenuation coefficient. The distance attenuation coefficient reflects the degree to which delivery distance affects a user's choice of store. The greater the distance, the greater the impact, and the higher the value of the attenuation coefficient, the more likely the user is to choose the corresponding store.

[0062] The preset value of the distance attenuation coefficient can be set based on experience or selected according to actual business needs; this application does not impose any limitations on this. It is understood that the accuracy of the attenuation coefficient value determines the accuracy of the target delivery revenue forecast for new stores. Therefore, in one optional embodiment, the target coefficient value of the attenuation coefficient can be determined based on the order allocation pattern reflected by the historical actual order ratio of existing stores.

[0063] Specifically, determining the target value of the attenuation coefficient includes the following steps: Obtain delivery areas and historical order data for existing stores; Based on the delivery area, determine the overlapping delivery areas between the existing stores; Based on historical order data, determine the first target store corresponding to the overlapping delivery area and the actual order ratio within the overlapping delivery area; and determine the second target store that has a target overlapping area with the newly opened store and the total number of orders within the target overlapping area. Based on the order allocation pattern reflected by the actual order ratio, determine the target value of the attenuation coefficient in the specified data analysis model; the specified data analysis model is used to determine the probability of users choosing a store; the attenuation coefficient is used to reflect the degree of influence of selection factors on users' store selection.

[0064] In a specific embodiment, there are overlapping delivery areas among the delivery orders of existing stores. In fact, there are certain historical order distribution patterns in these overlapping areas. Based on these patterns, the influence of factors on user store selection can be analyzed, and the influence of these factors exists in the data analysis model as a decay coefficient. Therefore, to accurately predict the number of delivery orders that newly opened stores will divert from existing stores, the prediction method for delivery orders from newly opened stores provided in this application determines the decay coefficient in a specified data analysis model based on the order distribution patterns reflected by the actual order ratios of existing stores.

[0065] Specifically, in one optional embodiment, the delivery areas and historical order data of existing stores are acquired. These existing stores can be stores within a preset area, such as all existing stores in the entire city, or areas defined by criteria like Chaoyang District in Beijing. This application does not limit the scope of acquired existing stores. However, it should be noted that there are multiple existing stores. It is understood that the more existing stores there are, the more data samples are analyzed, leading to higher accuracy in determining the attenuation coefficient of the subsequent data analysis model. Therefore, the preset area can be selected based on actual business needs. Furthermore, it should be noted that all acquired existing and newly opened stores are of the same type, for example, all are beverage stores of the same brand. Further, based on the delivery areas of existing stores, overlapping delivery areas between existing stores can be determined, meaning that there are historical orders for multiple stores within a certain area. Figure 5 This is a schematic diagram of an overlapping delivery area provided in another embodiment of this application. For ease of understanding, it will be described below in conjunction with... Figure 5 Please provide an explanation.

[0066] For example, such as Figure 5As shown, the opened stores include opened store A and opened store B. The delivery area of ​​opened store A is the blue area + purple area, and the delivery area of ​​opened store B is the green area + purple area. Orders in the blue area are all selected from opened store A, orders in the green area are all selected from opened store B, and orders in the purple area include opened store A and opened store B. That is, the purple area is the overlapping delivery area of ​​opened store A and opened store B.

[0067] It is understandable that overlapping delivery areas include multiple primary target stores; for example, in the above example, existing store A and existing store B are the primary target stores. Based on historical order data, the actual order ratio (i.e., historical order delivery ratio) among the primary target stores in the overlapping delivery area can be determined.

[0068] Simultaneously, a second target store with an overlapping delivery area with the newly opened store is identified, and the total number of orders placed by the second target store within that overlapping delivery area is determined. It's important to note that the first target store refers to stores with overlapping delivery areas among all existing stores, while the second target store only refers to stores with overlapping delivery areas with the newly opened store.

[0069] After determining the actual order proportion of the first target store in the overlapping delivery area, this actual order proportion reflects the influence of factors affecting user store selection in the overlapping delivery area on the final choice. In order to quantify these factors, the data analysis model uses a decay coefficient in its calculations. Therefore, in a specific embodiment, analyzing the order allocation pattern reflected in the actual order proportion allows for the determination of the target value for the decay coefficient in the specified data analysis model.

[0070] In one optional embodiment for determining the actual order ratio, there may be a situation where no existing stores exist within a preset area. Understandably, in this case, there is no order diversion between the newly opened store and the existing stores. However, in another optional embodiment, if there is only one existing store, there is no overlapping delivery area for the existing store, but there may be overlapping delivery areas between the existing store and the newly opened store. In this case, the actual order ratio for the existing store is obtained from a pre-built mapping relationship.

[0071] The mapping relationship is a correspondence between city type, population density, number of existing stores, and the actual order ratio of the first target store. The city type can be a city divided by administrative region, or it can be classified according to rules such as first-tier cities, second-tier cities, etc. This application does not limit this. In a specific embodiment, based on this mapping relationship, a city type and population density can be found that corresponds to the city of the newly opened store. This allows for the analysis of the attenuation coefficient applicable to the current newly opened store scenario based on the actual order ratio of the corresponding city.

[0072] Furthermore, in a specific embodiment, if the total number of orders of the second target store in the target overlapping area has been determined, then if the probability of users in the target overlapping area choosing the newly opened store is determined, the number of target orders that will be diverted from the newly opened store can be determined, that is, the product of the probability and the total number of orders is the target number of orders.

[0073] Therefore, by analyzing the order distribution patterns of all existing stores in overlapping areas, and based on historical order response patterns, the attenuation coefficient influencing user store selection can be determined in a specified data analysis model, thus obtaining an accurate data analysis model suitable for the current scenario. Based on this, and using the determined attenuation coefficient, the data analysis model accurately predicts the number of target orders that newly opened stores will divert from existing stores in the target overlapping area, providing high-precision data support for store layout and resource allocation.

[0074] In one optional embodiment, the target coefficient value of the attenuation coefficient in the specified data analysis model is determined based on the order allocation pattern reflected by the actual order ratio, including: Based on the order allocation pattern reflected by the actual order ratio, determine the target coefficient value for the attenuation coefficient in the specified data analysis model, including: Obtain the preset range of values ​​for the attenuation coefficient; Use any one of the endpoint values ​​in the preset range as the current value of the attenuation coefficient; Based on the current coefficient values, the estimated order proportion of the first target store in the overlapping delivery area is estimated by using a specified data analysis model. Determine whether the difference parameter is within the preset acceptable range; the difference parameter is used to characterize the degree of difference between the actual order ratio and the estimated order ratio. If so, use the current coefficient value as the target coefficient value; If not, according to the preset rules, re-obtain the current coefficient value within the preset value range; and return the step of estimating the estimated order ratio of the first target store in the overlapping delivery area based on the current coefficient value and the specified data analysis model, and execute the subsequent steps.

[0075] In specific embodiments, different attenuation coefficients have a certain range of values ​​for different data analysis scenarios. Therefore, as long as the optimal target coefficient value suitable for the current scenario is determined within the possible range of values ​​for the attenuation coefficient, a high-precision data analysis model can be obtained.

[0076] Therefore, in one optional embodiment, a preset range of values ​​for different attenuation coefficients is obtained. For example, for the distance attenuation coefficient, in spatial economics or geography, the distance attenuation coefficient is used to represent the attenuation effect of distance on interaction, and may be set to 1 (linear attenuation), 2 (inverse square attenuation, similar to the gravitational model in physics), etc. In one optional embodiment, the range of values ​​for the distance attenuation coefficient is [1, 3].

[0077] In order to determine the optimal coefficient value suitable for the current scenario within the preset value range, as an optional embodiment, any endpoint value within the preset value range can be used as the current coefficient value of the attenuation system. For example, in the example above where the preset value range of the distance attenuation coefficient is [1, 3], 1 or 3 can be used as the current coefficient value of the distance attenuation coefficient.

[0078] Furthermore, to determine whether the current coefficient value is the optimal value for the current scenario, based on the current coefficient value, a specified data analysis model is used to estimate the predicted order ratio of the first target store within the overlapping area. It is understood that the first target store has an actual order ratio within the overlapping area; when the difference between the actual order ratio and the predicted order ratio is small, it indicates that the current coefficient value is the target coefficient value suitable for the current scenario.

[0079] Specifically, in one optional embodiment, a difference parameter is determined to characterize the degree of difference between the actual order ratio and the estimated order ratio. When the difference parameter is within a preset acceptable range, it indicates that the current actual order ratio and the estimated order ratio are very close, and the current coefficient value is a suitable value. In this case, the current coefficient value can be used as the target coefficient value. It should be noted that the difference parameter can be calculated in the form of a difference, ratio, etc., and this application does not limit it in this way.

[0080] Of course, if the difference parameter is not within the preset acceptable range, it indicates that the current coefficient value is not the optimal value under the current scenario, and it is necessary to reselect the current coefficient value and recalculate the difference parameter.

[0081] Specifically, according to preset rules, a new current coefficient value is obtained again within a preset value range. The preset rules can be obtained at equal intervals within the preset value range. This application does not limit the method of obtaining the coefficient, as long as it is within the preset value range and is different from the historical value.

[0082] Each time a new current coefficient value is obtained, a new estimated order ratio is calculated using the specified data analysis model to reassess whether the current difference parameter is within an acceptable range. This process is repeated until a target coefficient value that meets the preset acceptable range requirement is obtained.

[0083] Therefore, it can be understood that the obtained target coefficient value is the optimal value for the specified data analysis model in the current scenario. It can be used to estimate the probability that users in the target overlapping area will choose the newly opened store, thereby predicting the number of target orders that will be diverted from the newly opened store.

[0084] In one optional embodiment, based on the current coefficient value, a specified data analysis model is used to estimate the proportion of orders for the first target store within the overlapping delivery area, including: Obtain the existing store attractiveness parameter of the first target store, as well as the existing store delivery distance of historical orders within the overlapping delivery area; the existing store attractiveness parameter is used to characterize the degree of influence of the attractiveness factors of the first target store on the user's store selection; the attractiveness factors include at least one of store area, number of product types and product price; Based on the existing store's attractiveness parameters, delivery distance to the existing store, and target coefficient values, the estimated order ratio is determined using the Huff model.

[0085] In one optional embodiment, based on the existing store's attractiveness parameter, the existing store's delivery distance, and the target coefficient value, the estimated order ratio is determined using the Huff model, including: The overlapping delivery area is divided into multiple grid areas of preset size; the delivery distance to the original store is the shortest delivery distance between the center of the grid area and the first target store; Based on the attractiveness parameters of established stores, the delivery distance of established stores, and the target coefficient values, the Huff model is used to estimate the proportion of sub-predicted orders from the permanent residents in each grid area who choose each first target store. The estimated order ratio is determined based on the sub-estimated order ratio.

[0086] In a specific embodiment, the estimated order ratio is calculated for all historical orders in the overlapping delivery area, and the corresponding original store delivery distance is the actual delivery distance for each historical order. However, in practical applications, the area of ​​the overlapping delivery area may be very large, and the corresponding number of historical orders is enormous. Calculating all historical orders would undoubtedly require significant computing resources.

[0087] A grid is a structured method for organizing, storing, and analyzing spatial data. Grids are typically used to divide continuous space into discrete units, facilitating the modeling and computation of geographical phenomena. A grid is a regular spatial division method, usually composed of a series of equally spaced lines, forming a regular rectangular grid or a honeycomb grid. Each grid cell represents an independent spatial region.

[0088] Therefore, in one optional embodiment, to save computing resources and improve estimation efficiency, the overlapping delivery area is divided into multiple grid areas of preset size, and the estimation is performed on a grid area basis. In this embodiment, the delivery distance to the original store refers to the cycling delivery distance from the center of the grid area to the first target store.

[0089] Based on the above embodiments, as an optional embodiment, determining that the difference parameter is within a preset acceptable range includes: The sum of the proportions of the sub-estimated orders corresponding to the same first target store is taken as the order proportion of the corresponding first target store in the overlapping delivery area; Determine the estimated order proportion based on the ratio of each order's proportion; When the difference between the actual order ratio and the estimated order ratio is within a preset difference range, the difference parameter is determined to be within a preset acceptable range.

[0090] Understandably, the difference parameter is used to characterize the degree of difference between the actual order ratio and the estimated order ratio. In the above embodiment, the estimated order ratio is calculated on a grid area basis. Therefore, when determining the difference parameter, it is also necessary to evaluate it on a grid area basis.

[0091] Specifically, the sum of the estimated sub-order proportions corresponding to the same first target store is taken as the order proportion of that first target store in the overlapping delivery area. For example, if the overlapping delivery area includes 10 grid areas, and for the first target store X, the estimated sub-order proportions of each grid area are P1, P2...P10, then the order proportion of the first target store X in the overlapping delivery area is P1+P2+...+P10.

[0092] Furthermore, the ratio of order proportions among different primary target stores can be used as the estimated order proportion. For example, if the overlapping delivery area includes primary target store X1, primary target store X2, and primary target store X3, and the order proportions corresponding to the three stores are Px1, Px2, and Px3, then the estimated order proportion is Px1 / Px2 / Px3.

[0093] When the difference between the actual order ratio and the estimated order ratio is within a preset range, the difference parameter is considered to be within a preset acceptable range. This indicates that the target coefficient used to determine the attenuation coefficient for the estimated order ratio is an appropriate value. It is important to note that the order in which the actual order ratio is determined for each store must be consistent with the estimated order ratio. For example, in the above example, the ratio should be determined in the order of first target store X1, first target store X2, and first target store X3.

[0094] In another alternative embodiment, determining that the difference parameter is within a preset acceptable range includes: Based on the order ratio of the examples, determine the actual number of orders for each primary target store within the overlapping delivery area; Based on the estimated sub-order ratio and the actual order quantity of the established store, determine the estimated sub-order quantity of the same primary target store in different grid areas; The sum of the estimated sub-orders corresponding to the same first target store is taken as the estimated order quantity for the same first target store. When the difference between the estimated number of orders for each primary target store and the actual number of orders for existing stores is within the preset difference range, the difference parameter is determined to be within the preset acceptable range.

[0095] In a specific embodiment, such as Figure 2 As shown, in the overlapping delivery areas, users in area G1 are more likely to choose the first target store A due to distance, while users in area G2 are more likely to choose the second target store B. Therefore, if the sum of probabilities across the entire overlapping delivery area is used as the basis for the difference parameter evaluation, the final evaluation accuracy will obviously be reduced because the actual selection preferences of each area in the overlapping delivery area are different.

[0096] Therefore, to improve the accuracy of determining the difference parameter—that is, to improve the accuracy of assessing the difference between the actual order ratio and the estimated order ratio—and thus improve the accuracy of the target coefficient value, thereby improving the accuracy of the target order quantity estimation, in one optional embodiment, the actual order quantity of each first target store in the overlapping delivery area can be determined based on the actual order ratio. Simultaneously, by determining the product between the sub-estimated order ratio determined in the above embodiment and the actual order quantity of the existing stores, the estimated sub-order quantity corresponding to each first target store in each grid area can be determined.

[0097] Furthermore, for the same primary target store, the sum of the estimated sub-order quantities within all grid areas is taken as the estimated order quantity. Therefore, when the difference between the estimated order quantity and the actual order quantity of the existing store is within a preset difference range, the difference parameter is determined to be within a preset acceptable range. For ease of understanding, an example will be provided below.

[0098] For example, the preset difference range is 0 to 5. The actual number of orders for the old store in the overlapping delivery area is 50. The actual number of orders for the old store in the overlapping delivery area of ​​the first target store Y1 is 30. The actual number of orders for the old store in the overlapping delivery area of ​​the first target store Y2 is 20. There are a total of 4 grid areas.

[0099] Taking the first target store Y1 as an example, based on the sub-estimated order ratio and the actual number of orders in the old store, the estimated number of sub-orders corresponding to the first target store Y1 in the 4 grid areas are determined to be 8 orders, 4 orders, 10 orders and 5 orders respectively. The estimated number of orders is 27 orders. The difference between the estimated number of orders and the actual number of orders in the old store is 3 orders. Within the preset difference range, it indicates that the current difference parameter is within the preset acceptable range.

[0100] As can be seen, the embodiments of this application calculate from the perspective of the total number of stores' estimated orders, which is closer to the actual selection preferences of users and can improve the accuracy and efficiency of calculation.

[0101] In another optional embodiment, the actual order ratio includes the sub-actual order ratio corresponding to each grid area; the estimated order ratio includes each sub-estimated order ratio. Determine that the difference parameters are within the preset acceptable range, including: Using grid areas as units, determine the sub-difference parameters for the sub-actual order ratio and sub-estimated order ratio corresponding to the same first target store; Determine the target number of sub-difference parameters within a preset acceptable range of grid areas; and determine the ratio of the target number to the total number of grid areas. When the ratio of the number of each first target store is greater than the preset ratio, the difference parameter is determined to be within the preset acceptable range.

[0102] To further improve the accuracy of the determination of the difference parameters, in one optional embodiment, the difference analysis of the order allocation ratio is performed separately on a grid area basis, and then the difference of all grids is analyzed on a total of all stores basis.

[0103] Specifically, taking grid areas as units, for the same primary target store, a sub-difference parameter is determined for the sub-actual order ratio and the sub-estimated order ratio. The sub-difference parameter can be a ratio or a difference, and this application does not limit this. Furthermore, for the same primary target store, the target number of grid areas whose sub-difference parameters are within a preset acceptable range is determined, and the ratio of the target number to the total number of all grid areas is determined.

[0104] On a store-by-store basis, if the target quantity to the total quantity for different primary target stores is greater than a preset ratio, then the difference between the actual order ratio and the estimated order ratio is considered to be within an acceptable range. For ease of understanding, examples will be provided below.

[0105] For example, the preset difference range is 0 to 20%, and the preset ratio is 80%. The overlapping delivery area includes 100 grid areas. When the sub-difference parameter is within 0 to 20%, the difference between the actual sub-order ratio and the estimated sub-order ratio within the corresponding grid area is determined to be within an acceptable range. On a store-by-store basis, for different first target stores, if the number of grids with a sub-difference parameter within 20% is greater than or equal to 80 (i.e., the ratio is greater than 80%), then the corresponding first target store is considered to meet the requirements.

[0106] In this embodiment of the application, it is necessary not only to compare the differences between actual orders and estimated orders in different grid areas to see if they meet the requirements, but also to ultimately determine, on a store-by-store basis, whether the proportion of grids that meet the requirements in each store meets the requirements.

[0107] Therefore, by comparing and calculating from multiple levels and different perspectives, the difference between the actual order ratio and the estimated order ratio can be accurately assessed, thereby obtaining a high-precision target coefficient value for the attenuation coefficient and improving the accuracy of the target order quantity calculation.

[0108] In one optional embodiment, the current coefficient value is re-acquired within a preset value range according to preset rules, including: Starting from any endpoint value, along the direction of another endpoint value, each time the difference parameter is determined to be outside the preset acceptable range, the current coefficient value is reacquired after an interval of the first search step; the difference between endpoint values ​​is an integer multiple of the first search step.

[0109] like Figure 3 As shown, when the difference parameter is not within the preset acceptable range, it is necessary to obtain a new current coefficient value. How to quickly and accurately obtain the target coefficient value is also crucial for increasing the target order quantity.

[0110] In one optional embodiment, new current coefficient values ​​can be obtained according to the rule of equal intervals. For example, for a distance attenuation coefficient with a value of [1, 3], the first search compensation can be 0.5. Each time it is determined that the difference parameter is not within the preset range, a value is taken at intervals of 0.5. That is, the difference parameter is verified in turn for the current coefficient values ​​corresponding to 1, 1.5, 2, 2.5 and 3.

[0111] The difference between endpoint values ​​is an integer multiple of the first search step size, meaning the length of the preset value range is an integer multiple of the first search step size, thus ensuring that values ​​can be uniformly selected within the preset value range. This application does not limit the value of the first search step size; it can be selected according to actual business needs.

[0112] Based on the above embodiments, as an optional embodiment, if the difference parameter is not within the preset acceptable range when searching to another endpoint value with the first search step size, the following steps are performed: The difference parameters corresponding to each first search step are sorted to obtain the sorting results; in the sorting results, the smaller the difference between the actual order ratio and the estimated order ratio, the higher the ranking. The current coefficient values ​​corresponding to the top two positions in the sorting results are used as candidate coefficient values. The numerical range formed by the candidate coefficient values ​​is used as the preset value range. The first search step size is reduced to the second search step size. The step of taking any endpoint value in the preset value range as the current coefficient value of the attenuation coefficient is returned and the subsequent steps are executed. The numerical length of the numerical range is an integer multiple of the second search step size.

[0113] In the initial search step, it may be impossible to obtain the target coefficient value that meets the requirements. In this case, the search step can be reduced to perform a refined search, i.e., a coarse search followed by a refined search. Specifically, if none of the difference parameters are within the preset acceptable range in the initial search step, the difference parameters corresponding to the initial search step are sorted in ascending order, i.e., sorted by the degree of difference between the actual order ratio and the estimated order ratio, gradually increasing. The difference parameters that appear earlier in the sorted results have the smaller degree of difference.

[0114] Furthermore, the current coefficient values ​​corresponding to the top two values ​​in the sorting results are taken as candidate coefficient values, and the numerical range formed between the two candidate coefficient values ​​is taken as a new preset value range. At this time, within the new preset value range, the search step size is reduced to a second search step size, and the iterative search is performed again.

[0115] The numerical length of the range formed by the candidate coefficient values ​​is an integer multiple of the second search step size; that is, the difference between the candidate coefficient values ​​is an integer multiple of the second search step size. For ease of understanding, an example will be provided below.

[0116] For example, for a distance attenuation coefficient with values ​​[1, 3], the first search compensation is 0.5, and the second search compensation is 0.1. At the next search step, the difference parameters for the current coefficient values ​​corresponding to 1, 1.5, 2, 2.5, and 3 are judged. If all current coefficient values ​​are outside the preset acceptable range, the search proceeds at the second search step.

[0117] Specifically, the difference parameters corresponding to the current coefficient values ​​of 1, 1.5, 2, 2.5 and 3 are sorted. If the difference parameters corresponding to 1.5 and 2 are ranked in the first two, then the second search step is used to search at equal intervals in the range [1.5, 2] until the target coefficient value is obtained.

[0118] S41: Determine the new store attractiveness parameters; the new store attractiveness parameters are used to characterize the degree of influence of the attractiveness factors of new stores on users' store selection; the attractiveness factors include at least one of the following: store area, number of product types, and product price; S42: Based on the delivery distance of the new store, preset values, and the attractiveness parameters of the new store, the Huff model is used to determine the target probability value that the total number of orders will be shared by the newly opened stores; S43: Determine the target number of orders based on the total number of orders and the target probability value.

[0119] In a specific embodiment, the store's hardware attributes also influence users' final store selection. These hardware attributes include, but are not limited to, store size, store reputation, store decoration, historical order reviews, product variety and quantity, and product prices. Quantifying these hardware attributes yields a new store attractiveness parameter, which characterizes the degree to which the attractiveness factors of newly opened stores influence users' store selection.

[0120] After determining the new store attractiveness parameters, in one optional embodiment, the target probability value of the total number of orders being shared by the new store can be determined by formula (1): (1) in, For newly opened stores The target probability value distributed within overlapping delivery areas. For newly opened stores New store attractiveness parameters This is the distance attenuation coefficient. The larger the distance, the more significant its impact. For historical orders To the newly opened store The new store's delivery distance, and the new store's delivery distance is the distance covered by the delivery rider for the order. This represents the total number of the first target stores within the overlapping delivery area.

[0121] Based on the above-mentioned new store delivery distance It can be the shortest delivery distance from the destination of historical orders from existing stores in the overlapping delivery area to the newly opened store, or it can be the shortest delivery distance from the center of the grid area to the newly opened store after dividing the overlapping delivery area into multiple grid areas. This application does not limit this.

[0122] Furthermore, by determining the target probability value and the total number of orders from stores already open in the overlapping delivery area, the product of the target probability value and the total number of orders can be used as the target order quantity.

[0123] Therefore, in this embodiment of the application, by quantifying the factors that influence users' store selection, the new store attractiveness parameter is obtained, and the preset value of the distance decay coefficient is determined. The Huff model is used to accurately predict the number of historical orders that will be diverted to new stores in the overlapping delivery area, providing data support for accurately predicting the delivery revenue of newly opened stores.

[0124] In one optional embodiment, the target delivery revenue of the newly opened store is estimated based on the target order quantity and the average first order value, including: Determine the single-layer delivery area within the new store's delivery range, excluding overlapping delivery areas; Based on preset rules, determine the single-layer capture rate of newly opened stores; the single-layer capture rate of new stores is the probability of delivery orders in a single-layer delivery area; The single-layer delivery revenue of a newly opened store is determined based on the single-layer capture rate of the new store. The product of the target number of orders and the average first order value will be used as the delivery revenue for the overlapping area of ​​the newly opened store. The sum of the single-layer delivery revenue and the delivery revenue of the overlapping area is used as the target delivery revenue.

[0125] like Figure 2 As shown in the specific embodiment, when estimating the delivery revenue of a newly opened store, in addition to the delivery revenue generated in the purple overlapping area, the blue area within the rectangle of the newly opened store will also generate delivery revenue. This portion of the delivery revenue is called the single-layer delivery revenue. Therefore, in the specific embodiment, when estimating the target delivery revenue of a newly opened store, it is also necessary to determine the single-layer delivery revenue of the newly opened store.

[0126] Specifically, in one optional embodiment, the area outside the overlapping delivery area within the new store's delivery range is designated as a single-layer delivery area for the new store. All orders within this single-layer delivery area are handled by the new store, meaning they all originate from the new store.

[0127] Furthermore, according to preset rules, the single-layer capture rate of the newly opened store is determined. The single-layer capture rate refers to the probability of delivery orders within a single-layer delivery area. In an optional embodiment, the single-layer capture rate of the new store can be determined based on the single-layer capture rate of existing stores within a preset nearby range. That is, a similar benchmark existing store is found, and the single-layer capture rate of the new store is determined based on the single-layer capture rate of the benchmark existing store. This application does not limit the method for determining the single-layer capture rate of the new store. After determining the single-layer capture rate of the new store, the single-layer delivery revenue of the newly opened store can be determined based on the single-layer capture rate of the new store.

[0128] In the above embodiment, the target number of orders diverted to the overlapping delivery area by the newly opened store was determined, and the product of the target number of orders and the first average order value was used as the delivery revenue of the overlapping area. At this time, the target delivery revenue = single-layer delivery revenue + overlapping area delivery revenue.

[0129] In addition to estimating the target delivery revenue for newly opened stores, it is also possible to estimate the delivery revenue of existing stores in designated delivery areas that overlap with the new stores. Specifically, the delivery revenue of existing stores = the existing store's original revenue - the target delivery revenue.

[0130] Figure 6 This application also provides a flowchart illustrating a method for determining store delivery revenue according to an embodiment. In an optional embodiment, such as... Figure 6 As shown, according to preset rules, the single-layer capture rate of newly opened stores is determined, including: S60: Determine whether there is a target store already opened within the circular area formed by the new store as the center and the initial preset radius; if there is, proceed to step S61; if not, proceed to step S62. S61: Use the average single-layer capture rate of existing stores of the target as the single-layer capture rate of new stores; S62: Determine whether there is a target store already opened within the preset area where the new store is located; if so, proceed to step S61; wherein, the circular area is within the preset area.

[0131] In a specific implementation, accurately determining the single-layer capture rate of newly opened stores is also crucial to the accuracy of the target delivery revenue. After determining the delivery range and overlapping delivery range of the new store, the non-overlapping portion is designated as the single-layer delivery area of ​​the new store, for example... Figure 2 In the diagram, the blue area within the rectangle represents the single-layer delivery area for newly opened store A, while the green area within the circle represents the single-layer delivery area for existing store B. Correspondingly, newly opened store A corresponds to the single-layer capture rate for new stores, and existing store B corresponds to the single-layer capture rate for established stores.

[0132] Based on the above principles, as an optional implementation, since newly opened stores do not have any orders, the estimated single-layer capture rate of new stores can be calculated based on the single-layer capture rate of existing target stores. Specifically, such as... Figure 6 As shown, a circle is drawn with the newly opened store as the center and an initial preset radius to obtain a circular area, and it is determined whether there is a target store already opened within this circular area.

[0133] If such a scenario exists, the average single-layer capture rate of existing stores within the target's existing stores can be used as the single-layer capture rate of newly opened stores. For example, the average single-layer capture rate of existing stores within a circular area with a radius of 8 kilometers centered on the newly opened store can be used as the single-layer capture rate of new stores.

[0134] like Figure 6 As shown, if there is no target store within the circular area, in one optional embodiment, the search range can be expanded, that is, a new circular area can be obtained by drawing a circle with a preset radius of the target to search for the target store. The target store can be obtained by continuously expanding the search range.

[0135] In another optional embodiment, if no target store is found within the circular area, the search can be expanded to a preset area, which includes the circular area. In another optional embodiment, the preset area can be an area defined by the administrative division criteria of the newly opened store. For example, if no target store is found within 8 kilometers, the search can be expanded to the entire county or city. The preset area can be determined based on the city tier (first-tier city, second-tier city, etc.), administrative division criteria (county, city, province, etc.), and population density of the city where the newly opened store is located; this application does not impose any limitations on this.

[0136] In fact, it's understandable that when a newly opened store has a single-layer delivery area, it's essential to calculate the single-layer delivery revenue within that area. Therefore, it's crucial to find a comparable existing store; that is, to identify a target existing store that can serve as a benchmark. Thus, if no suitable store is found within the circular area, the search scope can be expanded until a target existing store is located.

[0137] Based on the above embodiments, as an optional embodiment, such as... Figure 6 As shown, if there is no target store already open within the preset area, proceed to steps S63 to S65: S63: Obtain the pre-built second mapping relationship; the second mapping relationship is the correspondence between the single-layer capture rate of the unopened store area, the opened store area, and the opened store area; S64: Determine the target existing store area corresponding to the preset area where the new store is located; S65: Based on the second mapping relationship, the single-layer capture rate corresponding to the target already opened store area is used as the single-layer capture rate of the new store.

[0138] In the above embodiments, there may not be any target stores in the preset area (i.e., the scope of the search is expanded). For example, some small counties may not have opened any corresponding stores in the entire county. In this case, it is impossible to find the target stores in the county.

[0139] To avoid the aforementioned technical problems, as an optional embodiment, such as Figure 6 As shown, a pre-constructed second mapping relationship is obtained, which stores the correspondence between unopened store areas, opened store areas, and single-layer capture rates of opened store areas. The future store area is benchmarked against the preset area in the above embodiment; for example, if the preset area is a county town, then the unopened store area is that county town area. The opened store area is the benchmark area for the unopened store area. When constructing the second mapping relationship, for example, if city A and city B have similar population densities and the same city level, they can be used as benchmark cities. Table 2 is a schematic table of a second mapping relationship provided in the embodiments of this application. For ease of understanding, the following explanation will be based on Table 2.

[0140] Table 2 is a schematic diagram of a second mapping relationship.

[0141] As shown in Table 2, in specific embodiments, the opened stores can be divided into different areas such as county towns, city areas, cities, and towns, and this application does not limit this. The above embodiments are merely illustrative examples, and this application does not limit the storage format and structure.

[0142] Understandably, based on the second mapping relationship mentioned above, the target area of ​​existing stores for new stores can be determined, thereby determining the corresponding single-layer capture rate.

[0143] Based on the above embodiments, as an optional embodiment, the single-layer delivery revenue of a newly opened store is determined according to the single-layer capture rate of the new store, including: Obtain the resident population of a single-level delivery area; Determine the average second order value of historical orders from existing stores corresponding to the single-layer capture rate of new stores; The product of the resident population, the single-layer capture rate of new stores, and the average second order value is used as the single-layer delivery revenue.

[0144] After determining the capture rate of a new store's single-layer delivery area, the resident population of the delivery area of ​​the new store's single-layer delivery area is obtained, and the average second average order value of the benchmark existing stores is used as the average order value to determine the single-layer delivery revenue. Therefore, the product of the resident population, the capture rate of the new store's single-layer delivery area, and the average second average order value is taken as the single-layer delivery revenue.

[0145] It is worth noting that if there are no existing stores with overlapping delivery areas with the newly opened store, the single-layer delivery revenue of the new store will be directly used as the target delivery revenue.

[0146] In the above embodiments, the method for determining store delivery revenue has been described in detail. This application also provides an embodiment of a device for determining store delivery revenue.

[0147] Figure 7 This is a schematic diagram of a device for determining store delivery revenue provided in an embodiment of this application, as shown below. Figure 7 As shown, the device includes: The target acquisition module 70 is used to acquire the delivery range of newly opened stores and historical order data of existing stores. The overlapping area determination module 71 is used to determine the overlapping delivery areas between newly opened stores and existing stores based on the delivery range of the new stores. The order information determination module 72 is used to determine the total number of orders of the opened stores in the overlapping delivery area and the average first order price corresponding to the total number of orders based on historical order data. The order forecasting module 73 is used to forecast the target number of orders that will be distributed to the newly opened stores by using a specified data analysis model; the specified data analysis model is used to determine the probability that users in the overlapping delivery area will choose the newly opened stores; The delivery revenue determination module 74 is used to estimate the target delivery revenue of newly opened stores based on the target number of orders and the average first order value.

[0148] Furthermore, the device for determining store delivery revenue provided in this application embodiment also includes: The first target store determination module is used to determine whether there is a target existing store within a circular area centered on the newly opened store and with a first preset radius; if so, the following module is called: The first duration determination module is used to obtain the furthest delivery time for each target's opened stores and determine the average duration of each furthest delivery time. The first path traversal module is used to traverse each path starting from the newly opened store and calculate the average travel time to obtain the target delivery destination. The interest surface generation module is used to generate interest surfaces corresponding to each target delivery destination based on the latitude and longitude data of the target delivery destination. The interest-surface splicing module is used to connect interest surfaces to generate a closed area as the delivery range for newly opened stores.

[0149] If no target store is located within the circular area, the following module is invoked: The first mapping relationship acquisition module is used to acquire a pre-built first mapping relationship; the first mapping relationship is the correspondence between the farthest delivery time and the delivery influencing factor; the delivery influencing factor includes at least one of the city type and population density in the specified area. The delivery impact factor determination module is used to determine the delivery impact factors for newly opened stores; The second duration determination module is used to determine the target maximum delivery time for newly opened stores based on the mapping relationship and the delivery impact factors of newly opened stores. The second path traversal module is used to traverse each path starting from the newly opened store, and to obtain the target delivery destination by determining the longest delivery time. It then proceeds to the step of generating interest surfaces corresponding to each target delivery destination based on the latitude and longitude data of the target delivery destination, and executes subsequent steps.

[0150] The order filtering module is used to filter historical order data based on the corresponding delivery distance of the original store, and then obtain the target historical order data. The filtering process is to remove data whose delivery distance to the original store is greater than a threshold, or to remove data with a preset percentage at the end of the ascending sort result of the delivery distance to the original store. The latitude and longitude data determination module is used to determine the latitude and longitude data corresponding to points of interest in the target historical order data; The overlapping delivery area determination module is used to project latitude and longitude data to obtain interest surfaces; connect the interest surfaces to generate a closed area as the delivery range of the existing stores; and define the overlapping part of the delivery range of the new stores and the delivery range of the existing stores as the overlapping delivery area.

[0151] The model parameter acquisition module is used to obtain the new store delivery distance between the newly opened store and the overlapping delivery area, as well as the preset value of the attenuation coefficient in the Huff model; the attenuation coefficient includes at least the distance attenuation coefficient; the distance attenuation coefficient is used to reflect the degree of influence of distance on the user's choice of store; the data analysis model is specified as the Huff model; The attractiveness parameter determination module is used to determine the new store attractiveness parameters of newly opened stores; the new store attractiveness parameters are used to characterize the degree of influence of the attractiveness factors of newly opened stores on users' store selection; the attractiveness factors include at least one of store area, number of product types and product prices; The target probability value determination module is used to determine the target probability value that the total number of orders will be distributed to the newly opened stores based on the delivery distance of the new stores, preset values, and the attractiveness parameters of the new stores, using the Huff model; The target order quantity determination module is used to determine the target order quantity based on the total number of orders and the target probability value.

[0152] The single-layer delivery area determination module is used to determine the single-layer delivery area of ​​a newly opened store within the store's delivery range, excluding overlapping delivery areas. The new store single-layer capture rate determination module is used to determine the new store single-layer capture rate according to preset rules; the new store single-layer capture rate is the probability of delivery orders in a single-layer delivery area; The module for determining single-layer delivery revenue is used to determine the single-layer delivery revenue of a newly opened store based on the single-layer capture rate of the new store. The module for determining the overlapping area delivery revenue is used to take the product of the target order quantity and the average first order value as the overlapping area delivery revenue of the newly opened store. The target delivery revenue determination module is used to calculate the target delivery revenue by summing the delivery revenue of a single layer and the delivery revenue of overlapping areas.

[0153] The second target store determination module is used to determine whether there is a target existing store within a circular area formed by the new store as the center and an initial preset radius. If there is, the average single-layer capture rate of the existing target stores is used as the single-layer capture rate of the new store. If there is no target existing store, it is determined whether there is a target existing store within the preset area where the new store is located. If there is, the step of using the average single-layer capture rate of the existing target stores as the single-layer capture rate of the new store is executed. The circular area is within the preset area.

[0154] If no target store is already open within the preset area, the following module will be invoked; The second mapping relationship acquisition module is used to acquire the pre-built second mapping relationship; the second mapping relationship is the correspondence between the single-layer capture rate of the non-open store area, the opened store area, and the opened store area. The first determination module is used to determine the target existing store area corresponding to the preset area where the new store is located; The second determining module is used to use the single-layer capture rate corresponding to the target already opened store area as the single-layer capture rate of the new store, based on the second mapping relationship.

[0155] The resident population acquisition module is used to acquire the resident population of a single-layer delivery area; The module for determining the average order value is used to determine the second average order value of historical orders from existing stores corresponding to the single-layer capture rate of new stores. The third determining module is used to take the product of the resident population, the single-layer capture rate of new stores, and the average of the second order value as the single-layer delivery revenue.

[0156] Figure 8This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application, such as... Figure 8 As shown, the electronic device includes: a memory 80 for storing computer programs; The processor 81 is used to execute a computer program to implement the steps of the method for determining store delivery revenue as described in the above embodiments.

[0157] The electronic devices provided in this embodiment may include, but are not limited to, tablet computers, laptop computers, or desktop computers.

[0158] The processor 81 may include one or more processing cores, such as a quad-core processor or an octa-core processor. The processor 81 may be implemented using at least one of the following hardware forms: Digital Signal Processor (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 81 may also include a main processor and a coprocessor. The main processor, also known as the Central Processing Unit (CPU), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, the processor 81 may integrate a Graphics Processing Unit (GPU), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, the processor 81 may also include an Artificial Intelligence (AI) processor, which is used to handle computational operations related to machine learning.

[0159] The memory 80 may include one or more computer-readable storage media, which may be non-transitory. The memory 80 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In this embodiment, the memory 80 is used to store at least the following computer program 801, which, after being loaded and executed by the processor 81, is capable of implementing the relevant steps of the method for determining store delivery revenue disclosed in any of the foregoing embodiments. In addition, the resources stored in the memory 80 may also include an operating system 802 and data 803, etc., and the storage method may be temporary storage or permanent storage. The operating system 802 may include Windows, Unix, Linux, etc. The data 803 may include, but is not limited to, the relevant data involved in the method for determining store delivery revenue.

[0160] In some embodiments, the electronic device may further include a display screen 82, an input / output interface 83, a communication interface 84, a power supply 85, and a communication bus 86.

[0161] Those skilled in the art will understand that Figure 8 The structures shown do not constitute a limitation on electronic devices and may include more or fewer components than those shown.

[0162] The electronic device provided in this application includes a memory and a processor. When the processor executes the program stored in the memory, it can implement the method for determining the store delivery revenue in the above embodiments.

[0163] It should be noted that although the operations are depicted in a specific order in the accompanying drawings, this should not be construed as requiring these operations to be performed in the specific order shown or sequentially, or requiring all illustrated operations to be performed to achieve the desired result. In some cases, multitasking and parallel processing may be advantageous. Furthermore, the separation of various system modules and components in the above embodiments should not be construed as requiring such separation in all embodiments, and it should be understood that the described program components and systems can generally be integrated together in a single software product or packaged into multiple software products.

Claims

1. A method for determining a delivery business volume of a store, characterized by, The method comprises: acquiring a new store delivery range of a newly opened store and historical order data of an already opened store; determining an overlapping delivery area of the newly opened store and the already opened store according to the new store delivery range; determining a total number of orders of the already opened store in the overlapping delivery area and a first average single order price corresponding to the total number of orders according to the historical order data; estimating a target order quantity that will be shared by the newly opened store from the total number of orders through a specified data analysis model, the specified data analysis model being used to determine a probability that a user in the overlapping delivery area selects the newly opened store; estimating a target delivery turnover of the newly opened store according to the target order quantity and the first average single order price.

2. The method of claim 1, wherein The method comprises: determining whether there is a target already opened store in a circular area range formed by a first preset radius with the newly opened store as the center; if there is, performing the following steps: acquiring a farthest delivery time length of each target already opened store; and determining a time length average of each farthest delivery time length; traversing each path from the newly opened store as a starting point and driving for the time length average to obtain a target delivery endpoint; generating an interest surface corresponding to each target delivery endpoint based on the latitude and longitude data of the target delivery endpoint; connecting the interest surfaces to generate a closed area as the new store delivery range of the newly opened store.

3. The method of claim 2, wherein the delivery amount of the store is determined based on the number of the delivery orders. if there is no target already opened store in the circular area range, performing the following steps: acquiring a first mapping relationship constructed in advance; the first mapping relationship is a corresponding relationship between a farthest delivery time length and a delivery influence factor; the delivery influence factor includes at least one of a city type and a population density in a specified area; determining a delivery influence factor of the newly opened store; determining a target farthest delivery time length corresponding to the newly opened store according to the mapping relationship and the delivery influence factor of the newly opened store; traversing each path from the newly opened store as a starting point and driving for the target farthest delivery time length to obtain the target delivery endpoint; and performing the subsequent steps after the step of generating an interest surface corresponding to each target delivery endpoint based on the latitude and longitude data of the target delivery endpoint.

4. The method of claim 1, wherein The method comprises: screening the historical order data to obtain target historical order data according to a store delivery distance corresponding to the historical order data; the screening processing is to eliminate data with a store delivery distance greater than a threshold value, or to eliminate data with a last preset proportion in an ascending order sorting result of the store delivery distance; determining latitude and longitude data corresponding to a point of interest in the target historical order data; projecting the latitude and longitude data to obtain an interest surface; and connecting the interest surface to generate a closed area as a store delivery range of the already opened store; taking an overlapping part of the new store delivery range and the store delivery range as the overlapping delivery area.

5. The method of claim 1, wherein The specified data analysis model is a Huff model; the target order quantity to be shared by the new store is estimated by the specified data analysis model, comprising: obtaining a new store delivery distance between the new store and the overlapping delivery area, and a preset value of a decay coefficient in the Huff model; the decay coefficient at least includes a distance decay coefficient; the distance decay coefficient is used to reflect the influence degree of distance on user store selection; determining a new store attraction parameter of the new store; the new store attraction parameter is used to represent the influence degree of the attraction influence factor of the new store on user store selection; the attraction influence factor includes at least one of store area, number of commodity types and commodity price; based on the new store delivery distance, the preset value and the new store attraction parameter, the target probability value of the total order quantity to be shared by the new store is determined by the Huff model; determining the target order quantity according to the total order quantity and the target probability value.

6. The method of claim 5, wherein the delivery amount of the store is determined based on the number of the delivery orders. The target delivery business volume of the new store is estimated according to the target order quantity and the first average single price, comprising: determining a single-layer delivery area of the new store within the new store delivery range except the overlapping delivery area; determining a new store single-layer capture rate of the new store according to a preset rule; the new store single-layer capture rate is an external delivery order probability of the single-layer delivery area; determining a single-layer delivery business volume of the new store according to the new store single-layer capture rate; the product of the target order quantity and the first average single price is taken as the new store delivery business volume in the overlapping area; the sum of the single-layer delivery business volume and the new store delivery business volume in the overlapping area is taken as the target delivery business volume.

7. The method of claim 6, wherein the delivery amount of the store is determined based on the number of the delivery orders. The new store single-layer capture rate of the new store is determined according to a preset rule, comprising: determining whether there is a target opened store in a circular area range formed by the new store as the center and an initial preset radius; if there is, the average of the old store single-layer capture rate of the target opened store is taken as the new store single-layer capture rate; if there is not, it is determined whether there is a target opened store in a preset area where the new store is located; if there is, the step of taking the average of the old store single-layer capture rate of the target opened store as the new store single-layer capture rate is executed; the circular area range is within the preset area.

8. The method of claim 7, wherein the delivery amount of the store is determined based on the number of the delivery orders. if there is no target opened store in the preset area, the following steps are executed; obtaining a second mapping relationship constructed in advance; the second mapping relationship is a corresponding relationship among unopened store areas, opened store areas and single-layer capture rates of the opened store areas; determining a target opened store area corresponding to the preset area where the new store is located; based on the second mapping relationship, the single-layer capture rate corresponding to the target opened store area is taken as the new store single-layer capture rate.

9. The method of claim 8, wherein the delivery amount of the store is determined based on the number of the delivery orders. The single-layer delivery business volume of the new store is determined according to the new store single-layer capture rate, comprising: obtaining the permanent population of the single-layer delivery area; determining a second average single price of historical orders of the opened store corresponding to the new single-layer capture rate of the new store; multiplying the permanent population, the new single-layer capture rate, and the second average single price to obtain a single-layer delivery turnover of the new store.

10. A device for determining store delivery revenue, characterized in that, The device comprises: a target acquisition module configured to acquire a new store delivery range of a new opened store and historical order data of an opened store; an overlapping area determination module configured to determine an overlapping delivery area of the new opened store and the opened store according to the new store delivery range; an order information determination module configured to determine a total number of orders of the opened store in the overlapping delivery area and a first average single price corresponding to the total number of orders according to the historical order data; an order estimation module configured to estimate a target order quantity to be shared by the new opened store from the total number of orders by using a specified data analysis model, wherein the specified data analysis model is used to determine a probability of a user in the overlapping delivery area selecting the new opened store; a delivery turnover determination module configured to estimate a target delivery turnover of the new opened store according to the target order quantity and the first average single price.

11. An electronic device comprising a memory and a processor, said memory having stored thereon a computer program operable to run on said processor, characterized in that, The processor executes the computer program to realize the steps of the method for determining the delivery turnover of the store according to any one of claims 1 to 9.

12. A computer readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to realize the steps of the method for determining the delivery turnover of the store according to any one of claims 1 to 9.

Images