Store planning methods, media, computer equipment and software products

Abstract

A store planning method, medium, computer equipment, and program product determine candidate geographic grids containing the target access points within a pre-defined time period based on the location information of the target access points. These candidate geographic grids are then assigned to target stores, thereby expanding the target store's business district outside the pre-defined time period. The target access points dynamically change as the location of users demanding instant delivery services changes at different times, and the candidate geographic grids and target business districts determined based on this dynamic information also change dynamically. By using geographic grids as spatial index units and mapping dynamically changing target access points to the geographic grids, the order redistribution problem, which originally required processing across the entire geographic space, is simplified to local processing only within the candidate geographic grids and their neighboring grids. This reduces computational complexity and minimizes the computational cost of redundant calculations.

Description

Technical Field

[0001] This application relates to the field of Internet technology, and in particular to a store planning method, medium, computer equipment, and program product. Background Technology

[0002] Currently, many stores (such as restaurants, supermarkets, and convenience stores) provide instant delivery services to users through service platforms. In this service model, to achieve efficient order fulfillment and delivery scheduling, a fixed geographical service area, called a trade area, is usually defined for each store. When a user is within a store's trade area, that store can provide service; if the user is outside the trade area, the store cannot provide service. In related technologies, the trade areas defined for stores are usually fixed. Specifically, once a store's trade area is set, it generally remains unchanged throughout all business hours. However, in actual operation, factors such as order volume, spatial distribution density of orders, delivery capacity, and delivery efficiency vary significantly across different time periods. Fixed trade area definitions make it difficult for service platforms to dynamically match these changes, requiring cross-regional reallocation of orders outside the trade area, thus increasing the computational distance and computational cost of route planning. Summary of the Invention

[0003] This application provides a store planning method, medium, computer equipment, and program product to overcome the technical problems of low utilization rate of delivery capacity and high order fulfillment costs in related technologies.

[0004] In a first aspect, embodiments of this application provide a store planning method, the method comprising: Obtain the location information of the target access point within a preset time period; the target access point is the location of the user when accessing the service platform that provides instant delivery service, and the target access point is outside the original business district of multiple first stores; the multiple first stores are in operation during the preset time period; the original business district is the business district outside the preset time period. Based on the location information of the target access point, the target access point is mapped to multiple pre-divided geographic grids to obtain a candidate geographic grid including the target access point; Determine the distance between the candidate geographic grid and the plurality of first stores, determine the target store among the plurality of first stores based on the distance between the candidate geographic grid and the plurality of first stores, and assign at least a portion of the subgrids in the candidate geographic grid to the target store; The subgrids assigned to the target store in the candidate geographic grid and the original business district of the target store are merged to plan the target business district of the target store within the preset time period.

[0005] In some embodiments, the number of candidate geographic grids is greater than 1; the method further includes: Obtain the number of the target access points in multiple candidate geographic grids; A dynamic threshold is determined based on a preset quantile of the number of target access points in the plurality of candidate geographic grids; Candidate geographic grids whose number of target access points is less than the dynamic threshold are filtered.

[0006] In some embodiments, determining a target store among the plurality of first stores based on the distance between the candidate geographic grid and the plurality of first stores, and assigning at least a portion of the sub-grids in the candidate geographic grid to the target store, includes: From the plurality of first stores, determine a number of candidate stores whose distance from the candidate geographic grid is less than a preset distance; Obtain the first score of the candidate stores, and the first score of any candidate store is inversely correlated with the distance of the candidate store to the candidate geographic grid. If the number of candidate stores whose first score meets the first preset condition is 1, the candidate store whose first score meets the first preset condition is determined as the target store, and all subgrids in the candidate geographic grid are assigned to the candidate store whose first score meets the first preset condition.

[0007] In some embodiments, determining a target store among the plurality of first stores based on the distance between the candidate geographic grid and the plurality of first stores, and assigning at least a portion of the subgrids in the candidate geographic grid to the target store, further includes: If the number of candidate stores whose first score meets the first preset condition is greater than 1, perform the following operations for each subgrid in the candidate geographic grid: Obtain the second score of each candidate store whose first score satisfies the first preset condition, and the second score of any candidate store is inversely correlated with the distance of the candidate store to the sub-grid; If the number of candidate stores whose second score meets the second preset condition is 1, the candidate store whose second score meets the second preset condition is determined as the target store, and the subgrid is assigned to the candidate store whose second score meets the second preset condition.

[0008] In some embodiments, determining a target store among the plurality of first stores based on the distance between the candidate geographic grid and the plurality of first stores, and assigning at least a portion of the subgrids in the candidate geographic grid to the target store, further includes: If the number of candidate stores whose second score meets the second preset condition is greater than 1, determine the distance between each candidate store whose second score meets the second preset condition and the subgrid; Candidate stores whose distance from the subgrid meets the preset distance condition are identified as target stores, and the subgrid is assigned to the candidate stores whose distance from the subgrid meets the preset distance condition.

[0009] In some embodiments, merging the sub-grids assigned to the target store in the candidate geographic grid and the original business district of the target store to plan the target business district of the target store within the preset time period includes: If the subgrid assigned to the target store in the candidate geographic grid is not connected to the original business district of the target store, determine the shortest path between the subgrid assigned to the target store in the candidate geographic grid and the original business district of the target store; The subgrids assigned to the target store in the candidate geographic grid, the original business district of the target store, and the subgrids on the shortest path are merged to plan the target business district of the target store within the preset time period.

[0010] In some embodiments, the preset time period includes multiple sub-time periods, and the target store is open during some of the multiple sub-time periods; the method further includes: For any one of the multiple sub-time periods in which the target store is not in operation, perform the following operations: The instant delivery service is obtained as follows: the historical order conversion rate of the instant delivery service in the sub-time period, the target number of target access points in the target business district in the sub-time period, and the proportion of the sub-time period to the total business hours of the target store. The estimated sales revenue of the target store in the sub-period is determined based on the historical order conversion rate, the target quantity, and the ratio. If the estimated sales amount is greater than the preset limit, the operating hours of the target store will be extended to include the sub-period.

[0011] In some embodiments, the method further includes: Obtain target access points outside the target business district within the preset time period, and based on the location information of the obtained target access points, cluster the obtained target access points to obtain several clusters; The clusters are mapped to the pre-divided geographic grids to obtain the clusters to which the geographic grids belong; Divide geographical grids belonging to the same cluster into several regions to be assigned; Identify multiple second stores that are not in operation during the preset time period; For each area to be allocated, a second store is allocated to the area based on the distance between the area to be allocated and each second store and the closing time of each second store. The business hours of the second stores allocated to the area to be allocated are extended to include the preset time period. The original business district of the second stores allocated to the area to be allocated and the area to be allocated are merged to plan the target business district of the second stores allocated to the area to be allocated.

[0012] In some embodiments, the clustering of the acquired target access points based on their location information yields several clusters, including: Based on the location information of the target access points, several initial clusters are obtained. The distance between any two target access points in the same initial cluster is less than a preset distance threshold, and the number of target access points in the initial cluster is greater than a preset number threshold. The following operations are performed iteratively: for each candidate target access point in any initial cluster, obtain target access points from target access points outside the target business district that are less than a preset distance threshold from the candidate target access point, and add the target access points obtained from the target access points outside the target business district to the initial cluster to which the candidate target access point belongs.

[0013] In some embodiments, dividing geographic grids belonging to the same cluster into several regions to be assigned includes: Multiple random points are identified from geographic grids belonging to the same cluster, and these random points are clustered to obtain multiple sub-clusters; Obtain the centroids of each sub-cluster and construct a von Lono diagram based on the obtained centroids. The von Lono diagram divides the geographic grid belonging to the same cluster into multiple regions to be assigned. The distance from any point in each region to the centroid of that region is less than the distance to the centroids of other regions to be assigned.

[0014] In some embodiments, allocating second stores to the area to be allocated based on the distance between the area to be allocated and each second store, and the closing time of each second store, includes: For any given second store, obtain the score of the second store. The score of the second store is inversely correlated with the distance from the area to be assigned to the second store, positively correlated with the on-time rate of the second store, and positively correlated with the closing time of the second store. The area to be assigned is allocated to the second store whose score meets the third preset condition.

[0015] Secondly, embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the methods described in any embodiment of this application.

[0016] Thirdly, embodiments of this application provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method described in any embodiment of this application.

[0017] Fourthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the methods described in any embodiment of this application.

[0018] In this embodiment, candidate geographic grids containing the target access point are determined based on the location information of the target access point within a preset time period. These candidate geographic grids are then assigned to the target store, thereby expanding the store's business district outside the preset time period. The target access point dynamically changes as the location of users demanding instant delivery services changes at different times. The candidate geographic grids and target business districts determined based on this dynamic information also change dynamically. By using geographic grids as spatial index units and mapping dynamically changing target access points to the geographic grids, the order redistribution problem, which originally required processing across the entire geographic space, is simplified to local processing only within the candidate geographic grids and their neighboring grids. This reduces computational complexity and minimizes the computational cost of redundant calculations.

[0019] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this application. Attached Figure Description

[0020] The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate embodiments consistent with this application and, together with the description, serve to explain the technical solutions of this application.

[0021] Figure 1A and Figure 1B This is a schematic diagram illustrating an application scenario of an embodiment of this application.

[0022] Figure 2 This is a flowchart of the business district division method according to an embodiment of this application.

[0023] Figure 3 This is a schematic diagram illustrating the process of assigning a first store to a candidate geographic grid according to an embodiment of this application.

[0024] Figure 4 This is a schematic diagram of a candidate geographic grid and its subgrids according to an embodiment of this application.

[0025] Figure 5 This is a schematic diagram illustrating the process of determining the affiliation of candidate geographic grids in an embodiment of this application.

[0026] Figure 6 This is a schematic diagram of the candidate geographic grid and its subgrids belonging to the stores in an embodiment of this application.

[0027] Figure 7 This is a schematic diagram showing that the original business district and the candidate geographic grid are not connected in an embodiment of this application.

[0028] Figure 8 This is a schematic diagram illustrating the process of filling the original business district and candidate geographic grid in an embodiment of this application.

[0029] Figure 9 This is a flowchart of the process of opening a new store according to an embodiment of this application.

[0030] Figure 10 This is a schematic diagram of a geographic grid belonging to the same cluster, as described in an embodiment of this application.

[0031] Figure 11 This is a schematic diagram of a patchwork of geographic grids in a cluster according to an embodiment of this application.

[0032] Figure 12 This is a schematic diagram of the cluster division method in an embodiment of this application.

[0033] Figure 13 This is a schematic diagram of a clustering method according to another embodiment of this application.

[0034] Figure 14 This is a schematic diagram of the Voronoi diagram in an embodiment of this application.

[0035] Figure 15 This is a general flowchart of an embodiment of this application.

[0036] Figure 16 This is a block diagram of a business district division device according to an embodiment of this application.

[0037] Figure 17 This is a schematic diagram of a computer device according to an embodiment of this application. Detailed Implementation

[0038] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0039] 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. Additionally, the term “at least one” herein means any combination of at least two of any one or more of a plurality.

[0040] 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."

[0041] To enable those skilled in the art to better understand the technical solutions in the embodiments of this application, and to make the above-mentioned objectives, features and advantages of the embodiments of this application more apparent and understandable, the technical solutions in the embodiments of this application will be further described in detail below with reference to the accompanying drawings.

[0042] Currently, many stores (such as restaurants, fresh food supermarkets, and convenience stores) rely on service platforms to provide users with instant delivery services. To achieve effective order fulfillment and delivery scheduling, service platforms typically assign a geographical service area, generally referred to as a "trade area," to each store. Only when a user is located within a store's trade area can that store accept the order and provide delivery service; if the user's location is outside the trade area, the service platform will not assign the order to that store.

[0043] like Figure 1A As shown, the elliptical area represents the business district 400 of a certain store. Business district 400 includes store 100 and several delivery capacity units 200. A certain residential community 300 is located within business district 400; therefore, store 100 can fulfill instant delivery services for users within community 300. Specifically, as... Figure 1BAs shown, after a user in community 300 submits an order through user terminal 101, the order information is sent to service platform 102 for processing and recording. Service platform 102 then distributes the order information to the corresponding store terminal 103, where store staff can prepare goods (such as food preparation) based on the order information. Additionally, service platform 102 can also send the order information to delivery terminal 104, where delivery personnel will execute the delivery task. However, another community 500 is outside the business district 400; therefore, store 100 does not fulfill the instant delivery service for users in community 500. It is understood that the application scenario and system architecture shown in the figure are merely illustrative examples and are not intended to limit this application.

[0044] In related technologies, the business district boundaries of a store are often fixed. Specifically, once a store's business district is determined, it usually remains fixed throughout its entire operating period and will not be adjusted with changes in time or order volume. For example, the business district of a restaurant may always be set as a circular area with a radius of 3 kilometers centered on the store, and this range remains consistent regardless of weekday lunch peaks, weekend evenings, or off-peak hours.

[0045] However, in actual operation, factors such as order volume, geographical distribution density of orders, delivery capacity, and road traffic conditions can all change at different times. Fixed business district divisions struggle to flexibly and precisely schedule and optimize delivery capacity based on these dynamic changes. For example, if orders are concentrated in business districts at midday and dispersed to residential areas in the evening, fixed business districts cannot flexibly adapt to the spatial shift in service demand, easily leading to localized capacity shortages or underutilization. To adapt to these changes, orders outside of business districts need to be redistributed across regions, increasing the computational distance and overhead for route planning.

[0046] Based on this, this application provides a store planning method. This method determines candidate geographic grids containing the target access point from multiple pre-divided geographic grids based on the location information of the target access point within a preset time period. These candidate geographic grids are then assigned to target stores, thereby expanding the business district of the target store outside the preset time period. Since the target access point is the location of a user accessing a service platform providing instant delivery services, this location has significant spatiotemporal dynamics, changing dynamically with the location of users demanding instant delivery services at different times. Therefore, the candidate geographic grids determined based on this dynamic information can also change dynamically, allowing the determined target business district to also change dynamically with the location of users demanding instant delivery services. The dynamic business district determined in this way can flexibly adjust the service range according to real-time demand distribution, ensuring that delivery capacity deployment always aligns with high-demand areas. This reduces the situation of idle or insufficient capacity caused by fixed boundaries, thereby improving the utilization efficiency of delivery capacity. Furthermore, by employing a geographic grid as the spatial index unit, dynamically changing target access points are mapped onto the geographic grid. This simplifies the order reallocation problem, which originally required processing across the entire geographic space, to local processing only within the candidate geographic grid and its neighboring grids. This reduces computational complexity and minimizes the computational cost of redundant calculations. The specific details of the embodiments of this application are illustrated below with reference to the accompanying drawings.

[0047] like Figure 2 The diagram shown is a flowchart of a store planning method according to an embodiment of this application. The store planning method includes: Step S11: Obtain the location information of the target access point within the preset time period; the target access point is the location of the user when accessing the service platform that provides instant delivery service, and the target access point is outside the original business district of multiple first stores; multiple first stores are in operation during the preset time period; the original business district is the business district outside the preset time period. Step S12: Based on the location information of the target access point, map the target access point to multiple pre-divided geographic grids to obtain candidate geographic grids including the target access point; determine the distance between the candidate geographic grids and multiple first stores, determine the target store among the multiple first stores based on the distance between the candidate geographic grids and multiple first stores, and assign at least some sub-grids in the candidate geographic grids to the target store. Step S13: Merge the subgrids assigned to the target store in the candidate geographic grid with the original business district of the target store to plan the target business district of the target store within the preset time period.

[0048] The method in the embodiments of this application can be derived from... Figure 1BThe service platform executes the process. In step S11, the service platform can obtain the location information of the target access point within a preset time period.

[0049] The preset time period can be a period with lower order volume and / or higher fulfillment efficiency. In other words, the preset time period has lower order volume and / or higher fulfillment efficiency compared to periods outside the preset time period. For example, the preset time period could be nighttime, such as 10:00 PM to 6:00 AM the next day, which can be set according to actual needs. Periods outside the preset time period are daytime, such as 6:00 AM to 10:00 PM of the same day. Because the preset time period has lower order volume compared to periods outside the preset time period, delivery capacity cannot easily aggregate orders from adjacent areas for bulk delivery, resulting in a reduced order volume completed per unit time and lower delivery capacity utilization. Therefore, it is necessary to expand the store's catchment area. Furthermore, during the preset time period, due to smoother road traffic, stable store inventory preparation speed, and shorter order waiting times, the overall order fulfillment efficiency is higher than other periods. This means that delivery capacity completes delivery tasks in less time. Therefore, the service platform can plan longer delivery routes for delivery capacity, thus allowing the service platform to plan a larger catchment area for the store.

[0050] The target access point refers to the location of a user when accessing a service platform providing instant delivery services, and this target access point is outside the original business district of multiple primary stores. These multiple primary stores are operational during a preset time period, specifically, they may be operational during some or all sub-segments of that preset time period. For example, assuming the preset time period is from 22:00 to 6:00 the next day, the operating hours of the multiple primary stores could be from 6:00 to 22:30, from 7:00 to 23:00, or 24 hours. The operating hours of any two primary stores can be the same or different. The original business district of any one of the multiple primary stores can be a fixed-range business district. For example, the area centered on the store with a fixed radius can be defined as the store's original business district; or, the area that can be delivered within a preset delivery time, such as 30 minutes, can be defined as the primary store's original business district. Furthermore, other methods can be used to determine the original business district of a primary store, which will not be listed here.

[0051] Users can access the service platform through various methods, including applications (APPs), mini-programs, or H5 pages. When a user opens an APP, mini-program, or H5 page, if the user is outside the original business district of the aforementioned primary stores, then the user's current location is the target access point. Figure 1A For example, in Figure 1AIn this scenario, we assume that the first store includes store 100, community 300 is within the original business district of store 100, while community 500 is outside the original business district of store 100. Therefore, the location of community 500 is the target access point, while the location of community 300 is not the target access point.

[0052] When a user visits a target access point, the user's terminal can collect the location information of that access point. Specifically, when a user opens a mini-program, the mini-program can request location access permissions from the user via a pop-up window through a standard interface provided by the operating system (such as iOS or Android). After the user authorizes, the mini-program can call the user's terminal's built-in positioning module to collect the current location information, which is the location information of the target access point. In addition, the location information of each target access point can also be associated with time information, which can be the time when the location information of the target access point was collected, also obtained by the user terminal. By associating location information with time information, it is easy to determine whether the target access point was generated within a preset time period. After the user terminal collects the location information and associated time information of the target access point, it can report the location information and associated time information of the target access point to the service platform.

[0053] This embodiment uses the location information of target access points within a preset time period as the basis for subsequent expansion of the original business district. This has the following advantages: First, the target access point is the location of the user when accessing the service platform. Users typically assume a demand for instant delivery services when accessing the platform; therefore, the location information of the target access point accurately reflects the distribution of locations where there is demand for instant delivery services. Second, the location of the target access point changes with the user's location. For example, in the food delivery industry, an office worker typically orders food at their workplace during lunchtime (11:00-13:00) and at home during the evening (around 22:00). The location of the target access point accurately reflects the changing demand for instant delivery services. Therefore, the location information of the target access point provides reliable demand data for subsequent business district expansion, ensuring that the expanded business district matches the distribution of user demand.

[0054] In step S12, the service platform can map the target access point to multiple pre-divided geographic grids based on the location information of the target access point, thereby obtaining candidate geographic grids including the target access point. These geographic grids can be obtained by dividing geographic space. For example, the geographic space surface can be divided into polygons such as squares, hexagons, or triangles, or into irregular polygons based on geographic features such as rivers or roads, or administrative boundaries such as administrative divisions. Each polygon is a geographic grid. Different geographic grids can have the same shape and / or different sizes. In some embodiments, the geographic grid can be an H3 grid, which is a hexagonal grid under the H3 spatial indexing system. The H3 grid is obtained by recursively hierarchically partitioning the geographic space: starting from the lowest resolution level 0 grid, each parent grid is recursively divided into multiple sub-grids according to fixed rules, thus forming multiple levels from coarse to fine. The geographic grid in this application can be a grid of a specified level in the H3 grid, such as the level 8 grid.

[0055] In some embodiments, the service platform can obtain the coordinates of the target access point and call a coordinate transformation function provided by the H3 spatial index library, such as geoToH3, to map the coordinates to a pre-defined H3 grid at a specified level, thus obtaining a grid code. The grid corresponding to this grid code is the geographic grid containing the target access point, and the service platform can identify this grid as a candidate geographic grid that includes the target access point.

[0056] In some embodiments, if the number of candidate geographic grids is greater than 1, after determining the candidate geographic grids including the target access point, the number of target access points in multiple candidate geographic grids can be obtained, and geographic grids with fewer than a preset number of target access points can be filtered. In this way, candidate geographic grids with low demand for on-demand delivery services can be filtered out, thereby concentrating limited delivery capacity on candidate geographic grids with high demand, shortening the average delivery distance and time, and improving the timeliness and stability of order fulfillment. The preset number can be a static threshold or a dynamic threshold. Optionally, the preset number is a dynamic threshold determined based on a preset quantile of the number of target access points. For example, if the preset quantile is the 80th percentile, the 80th percentile of the number of target access points refers to the number of candidate geographic grids where the number of target access points is less than or equal to 80%. That is, filtering candidate geographic grids with fewer than the 80th percentile of the number of target access points means retaining the top 20% of candidate geographic grids by the number of target access points and filtering out the rest. This method of setting dynamic thresholds can automatically adjust the screening criteria based on the real-time distribution of overall demand. During periods with a high volume of orders, the threshold is automatically raised to focus on core high-density areas, while during periods with weaker overall demand, the threshold is lowered to expand the effective coverage area, thereby ensuring that delivery capacity is always prioritized for areas with relatively high demand.

[0057] After determining the candidate geographic grid including the target access point, several candidate stores whose distance to the candidate geographic grid is less than a preset distance can be selected from a plurality of first stores, and a first score for each candidate store is determined based on the aforementioned distance. The first score of any candidate store is inversely correlated with the distance from that candidate store to the candidate geographic grid. In some embodiments, for any candidate store i and candidate geographic grid j, the distance between candidate store i and candidate geographic grid j can be obtained, normalized based on the side length of candidate geographic grid j, and then rounded up. The first score of candidate store i is determined based on the rounded result. In a specific example, the first score can be denoted as: (Formula 1) in, This represents the first score of candidate store i determined for candidate geographic grid j. Indicates the centroid of candidate geographic grid j. Represents spherical distance. Let represent the side length of candidate geographic grid j, and [] denotes the round-up operation.

[0058] In some cases, there may only be one candidate store with the highest first score. In this case, the candidate store whose first score meets the first preset condition (e.g., the highest first score) can be directly identified as the target store, and the candidate geographic grid can be directly assigned to the candidate store with the highest first score. For example... Figure 3 As shown, assume there are three candidate geographic grids, denoted as grid G1, grid G2, and grid G3, and two candidate stores, denoted as store A and store B. Target stores can be assigned to grids G1, G2, and G3 respectively. Taking grid G2 as an example, the distance d1 from store A to grid G2 and the distance d2 from store B to grid G2 can be determined. Store A's first score is determined based on distance d1, and store B's first score is determined based on distance d2. If store A's first score is higher than store B's first score, then store A is determined as the target store for grid G2, and grid G2 is assigned to store A.

[0059] In other cases, there may be multiple candidate stores whose first score meets the first preset condition. In this case, the candidate geographic grid can be further divided into multiple sub-grids, and the target store assigned to each sub-grid can be determined. For example, if the candidate geographic grid is an H3 grid at level 8, it can be further divided into multiple sub-grids, each of which is an H3 grid at level 9. Figure 4 A schematic diagram is shown of a candidate geographic grid 401 and multiple sub-grids 4011 within that candidate geographic grid 401. (See diagram for reference.) Figure 4 As shown, candidate geographic grid 401 is a hexagonal grid, as indicated by the solid hexagons in the figure. Candidate geographic grid 401 can be further divided into seven smaller hexagonal grids, as indicated by the dashed hexagons in the figure. Each smaller hexagonal grid is a sub-grid 4011. It is understood that the division of candidate geographic grid 401 and sub-grid 4011 shown in the figure is merely illustrative. In other examples, candidate geographic grids and sub-grids can also have other shapes, the number of sub-grids is not limited to seven, and sub-grids can partially overlap.

[0060] Specifically, the second scores of each candidate store whose first score satisfies the first preset condition can be obtained, and the candidate stores whose second scores satisfy the second preset condition can be determined. The second preset condition could be, for example, having the highest second score. The second score of any candidate store is inversely correlated with the distance from that candidate store to the sub-grid. For any candidate store i and sub-grid k, the distance between candidate store i and sub-grid k can be obtained, normalized based on the side length of sub-grid k, and then rounded up. The second score of candidate store i is determined based on the rounded result. The specific method for determining the second score can refer to the method for determining the first score in the aforementioned embodiments, and will not be repeated here.

[0061] When determining the second score of each candidate store, two scenarios still exist. One scenario is that only one candidate store meets the second preset condition in terms of second score. In this case, the candidate store whose second score meets the second preset condition can be directly identified as the target store, and subgrid k can be assigned to it. The other scenario is that multiple candidate stores meet the second preset condition in terms of second score. In this case, the distance between each candidate store and subgrid k can be determined, and the store that meets the preset distance condition can be identified as the target store, and subgrid k can be assigned to the identified target store. The preset distance condition could be, for example, being the closest to subgrid k.

[0062] Figure 5 The process for determining candidate geographic grid affiliations is illustrated. Figure 6 The diagram shows the area assigned to store A. In the process of determining ownership, the store is first assigned using candidate geographic grid 501 as the granularity. If a unique store cannot be determined, the candidate geographic grid is then divided into sub-grids 502, and the assigned store is determined at the sub-grid granularity. Finally, the area assigned to store A is region 503, enclosed by the dashed line in the diagram.

[0063] The above process employs a two-step iterative approach to determine target stores. In the initial determination, target stores are identified based on the candidate geographic grid as a whole. Since the candidate geographic grid is relatively large, target stores can be quickly identified for most of them. When multiple candidate stores have the same first score, it indicates that the expected effects (such as delivery efficiency and delivery costs) of assigning the candidate geographic area to these multiple candidate stores are relatively similar. In this case, to further determine the superior store among these multiple candidate stores, a second iteration is performed, that is, store allocation is performed at the sub-grid level. This allows each sub-grid to be assigned a more suitable target store, improving allocation accuracy. Furthermore, when multiple candidate stores have the same second score, the improvement in delivery efficiency from continuing with finer-grained iterations is limited. Therefore, to improve allocation efficiency, further iterations are discontinued, and target stores are directly assigned to sub-grids based on distance.

[0064] In step S13, the candidate geographic grids and the original business district of the target store can be merged to obtain the target business district of the target store within a preset time period. In other words, the target business district planned for the target store includes not only the target store's original business district but also the sub-grids assigned to the target store in the candidate geographic grids. This method expands the original business district of the target store. For any preset time period, the target business district of the target store within that preset time period can be determined based on the above method, thus making the target store's business district no longer fixed but dynamically variable based on the order demand volume within the preset time period.

[0065] In some embodiments, such as Figure 7 As shown, the candidate geographic grid 603 and the original business district 604 of the target store may not be connected. This would cause delivery capacity to run empty between unconnected areas, significantly increasing invalid mileage and time costs. Simultaneously, the service platform would struggle to efficiently merge orders and plan global routes for unconnected business districts, leading to decreased delivery capacity utilization. Therefore, if the candidate geographic grid and the original business district of the target store are not connected, grid filling can be performed between the candidate geographic grid and the original business district of the target store. The filled grid, the candidate geographic grid, and the original business district of the target store are then merged to obtain the target business district of the target store within a preset time period. In some embodiments, constraints can be determined based on the area of ​​the filled grid (e.g., minimizing the area of ​​the filled grid is a constraint), and grid filling can be performed between the candidate geographic grid and the original business district based on these constraints.

[0066] Specifically, the shortest path between the sub-grids assigned to the target store in the candidate geographic grid and the target store's original business district can be determined. The sub-grids along this shortest path are then designated as fill grids. This process merges the sub-grids assigned to the target store in the candidate geographic grid, the target store's original business district, and the sub-grids along the shortest path to plan the target store's target business district within a preset time period. Since the number of sub-grids along the shortest path is minimal, the area of ​​the fill grid is also minimized when filling the candidate geographic grid and the target store's original business district based on these sub-grids.

[0067] like Figure 8 As shown, assuming the target store is store A, the shortest path 601 from the candidate geographic grid to the original business district boundary can be determined, and the geographic grid 602 on this shortest path 601 is designated as the filling grid for filling the candidate geographic grid 603 assigned to store A with the original business district 604 of store A. This method minimizes the number of filling grids, thereby reducing the area of ​​invalid regions in the target business district and reducing delivery difficulties, excessively long distances, and scattered orders caused by excessive filling areas.

[0068] After planning the target business district for the target store within a preset time period, the time period to which the current time belongs can be determined within each operating cycle, and the business district of the target store at the current time can be dynamically determined based on the time period to which the current time belongs. Specifically, if the current time belongs to the target time period, the business district of the target store at the current time is determined as the target business district; if the current time belongs to a time period other than the target time period, the business district of the target store at the current time is determined as the original business district. Unlike related technologies that determine the business district of the target store as a fixed business district in all time periods, i.e., the original business district, this embodiment dynamically determines the business district of the target store at the current time based on the time period to which the current time belongs. This allows for dynamic adjustment of the target store's business district range according to actual needs, thereby matching its business district range with the actual needs of the current time period.

[0069] In some embodiments, store planning includes not only planning the store's business district but also planning the store's operating hours. Specifically, after determining the target business district of the target store within a preset time period, it is also possible to determine each sub-time period in which the target store is not operating within the multiple sub-time periods included in the preset time period, and perform the following operations for any one of the determined sub-time periods: obtain the historical order conversion rate of the instant delivery service in that sub-time period, the target number of target access points within the target business district in that sub-time period, and the proportion of that sub-time period to the total operating hours of the target store; and determine the estimated sales revenue of the target store in that sub-time period based on the historical order conversion rate, target number, and proportion. In some embodiments, the estimated sales revenue can be denoted as: (Formula 2) Where S represents the estimated sales revenue, and n represents the target quantity mentioned above. The order conversion rate for a sub-period can be determined by the ratio of the number of users who actually placed orders during that sub-period to the number of users who accessed the service platform during that sub-period. This indicates the number of orders for that sub-period. This indicates the total number of orders received by the service platform. Indicates the duration of a sub-period. This indicates the average price of the order.

[0070] If the estimated sales revenue exceeds the preset limit, the target store's operating hours can be extended to include that sub-period. This embodiment identifies sub-periods from the preset timeframes that would normally be off-peak for the target store, and accurately determines whether there is potential user demand during those sub-periods by calculating the estimated sales revenue. If so, the target store's operating hours are extended to include that sub-period, allowing the store's operating hours to flexibly adapt to the actual needs of surrounding users, thus realizing a shift from fixed operating hours to dynamically adjusted operating hours on demand.

[0071] In some embodiments, there are multiple sub-periods in the preset time period where the target store is not in operation. The estimated sales revenue of each sub-period can be determined sequentially according to its chronological order. If so, the next sub-period is retrieved, and the step of determining whether the estimated sales revenue of that sub-period exceeds the preset amount is returned. The above determination process ends after the determination of the last sub-period is completed, or after identifying a sub-period with estimated sales revenue less than or equal to the preset amount.

[0072] After identifying the target business district for the target stores, there may still be target access points located outside the target business district during the preset time periods. The target stores may not be able to provide instant delivery services to these target access points. Therefore, stores not operating during the preset time periods can be designated as "second stores." Suitable stores can be selected from these second stores to provide instant delivery services to the areas of these target access points, and the operating hours of the selected stores can be extended to at least a portion of the preset time periods. This process is called opening new stores.

[0073] Specifically, such as Figure 9 As shown, the process of opening a new store includes the following steps: Step S21: Obtain target access points outside the target business district within a preset time period, and based on the location information of the obtained target access points, cluster the obtained target access points to obtain several clusters; Step S22: Map the obtained clusters to multiple pre-divided geographic grids to obtain the clusters to which the multiple geographic grids belong; Step S23: Divide the geographic grids belonging to the same cluster into several regions to be assigned; Step S24: Identify multiple second stores that are not in operation during the preset time period; Step S25: For each area to be allocated, based on the distance between the area to be allocated and each second store and the closing time of each second store, allocate a second store to the area to be allocated, extend the business hours of the second store allocated to the area to include the preset time period, and merge the original business district of the second store allocated to the area to be allocated and the area to be allocated, so as to plan the target business district of the second store allocated to the area to be allocated.

[0074] In step S21, clustering the acquired target access points can be achieved using clustering algorithms such as DBSCAN. Through clustering, sets of target access points with significant spatial proximity or density correlation can be identified, thereby transforming discrete target access points into spatial groups with inherent structure.

[0075] In some embodiments, the above clustering process can be implemented as follows: First, several initial clusters are obtained based on the location information of the acquired target access points. The distance between any two target access points in the same initial cluster is less than a preset distance threshold, such as 700 meters, and the number of target access points in the initial cluster is greater than a preset number threshold, such as 5. Then, the following operations are iteratively performed: For each candidate target access point in any initial cluster, target access points whose distance to the candidate target access point is less than the preset distance threshold are obtained from target access points outside the target business district, and these target access points are added to the initial cluster to which the candidate target access point belongs. For example, assuming the initial cluster includes five target access points P1, P2, P3, P4, and P5, target access points whose distance to each of the target access points P1, P2, P3, P4, and P5 is less than the preset distance threshold can be obtained respectively. For example, if target access point P6 is found to be less than the preset distance threshold to P1, P6 can be added to the initial cluster, and further target access points whose distance to P6 is less than the preset distance threshold can be obtained, and so on. If no target access point less than a preset distance threshold can be found in the initial cluster, the iteration stops. Based on this chain reaction, the above method aggregates target access points into several clusters. Furthermore, after clustering the acquired target access points to obtain several clusters, clusters with fewer than a first minimum number of target access points can be filtered out.

[0076] In step S22, the clusters obtained in step S21 can be mapped to multiple pre-divided geographic grids to obtain the clusters to which these geographic grids belong. Each of these geographic grids covers a certain geographic area. Based on the location information of the target access point within the cluster, the cluster can be mapped to a geographic grid. In this way, geographic grids belonging to the same cluster can be identified. For example... Figure 10 The diagram illustrates geographic grids belonging to the same cluster in some embodiments. In some embodiments, geographic grids belonging to the same cluster with a number of target access points less than a second minimum number can be filtered out. The second minimum number is less than a first minimum number. In some embodiments, geographic grid filtering can be performed through multiple iterations. In each iteration, geographic grids located on the boundary (i.e., the outermost layer) with a number of target access points less than the second minimum number are filtered out. If the number of target access points in the outermost geographic grid of the same cluster is greater than or equal to the second minimum number, the iteration stops.

[0077] In practice, it often happens that a certain cluster contains a large number of geographic grids, which aggregate into a large connected region, such as... Figure 11As shown. Therefore, in step S23, it is necessary to segment the geographic grids within the same cluster to obtain several regions to be assigned. In some embodiments, such as Figure 12 As shown, an outer geographic grid can be randomly selected, and its surrounding geographic grids can be continuously acquired until the number of geographic grids reaches the required quantity, ultimately ensuring that the area of ​​each area to be allocated after division is approximately equal.

[0078] In other embodiments, multiple random points can be determined from a geographic grid belonging to the same cluster, and these random points can be clustered to obtain multiple sub-clusters. The k-means clustering algorithm can be used to cluster the random points. The number of sub-clusters, u, can be determined based on the maximum area of ​​the business district. In some embodiments, the number of sub-clusters, u, can be denoted as: (Formula 3) in, This represents the total area of ​​geographic grids belonging to the same cluster. Represents the maximum area of ​​the business district, and [] indicates rounding up. Assume the number of determined sub-clusters u is equal to 3, and the sub-clusters are denoted as R1, R2, and R3 respectively, as follows... Figure 13 As shown, the centroids of subclusters R1, R2, and R3 can be determined respectively, denoted as Q1, Q2, and Q3.

[0079] Then, the centroids of each sub-cluster are obtained, and a von Rooi diagram is constructed based on these centroids. This von Rooi diagram divides the geographic grid belonging to the same cluster into multiple regions to be assigned. Within each region, the distance from any point to the centroid of that region is less than the distance to the centroids of other regions to be assigned. Figure 14 As shown.

[0080] In steps S24 and S25, multiple second stores that are not in operation during a preset time period can be identified, and second stores can be assigned to each area to be assigned.

[0081] Specifically, for any given second store, a score can be obtained for that second store. This score is inversely correlated with the distance from the assigned area to the second store, and is related to the second store's on-time performance. Positively correlated with, and with, the closing time of the second store. Positive correlation. In some embodiments, the aforementioned distance may include navigation distance. and / or straight-line distance Navigation distance affects delivery punctuality; therefore, the greater the navigation distance, the lower the score. A greater straight-line distance indicates a higher probability of spatial disconnection between the second store and the assigned area; therefore, a greater straight-line distance results in a lower score. Punctuality reflects the second store's order-handling capacity; therefore, a higher punctuality rate leads to a higher score. A later closing time reduces the cost of extending the second store's operating hours to the preset time period; therefore, a later closing time results in a higher score. Taking a distance that includes both navigation distance and straight-line distance as an example, the final formula for calculating the second store's score is: (Formula 4) in, This indicates the score of the second store. , , and These represent the weights of navigation distance, straight-line distance, punctuality rate, and closing time, respectively. and These represent the farthest navigation distance and the farthest straight-line distance, respectively.

[0082] After determining the scores of each second store, the areas to be allocated can be assigned to the second stores whose total scores meet a third preset condition, such as having the highest total score.

[0083] Figure 15 The overall flow of an embodiment of this application is illustrated. This embodiment uses the example of a preset time period being the nighttime period and the time period outside the preset time period being the daytime period. This embodiment considers the following three aspects: For the primary store operating at night, can its catchment area be expanded? Generally, traffic is smoother and order flow is more dispersed at night than during the day, meaning the original daytime catchment area is insufficient to meet nighttime demand. Therefore, it is necessary to expand the catchment area beyond the original daytime catchment area to obtain the target catchment area for the nighttime period. For the first store that is open during the nighttime hours, can its operating hours be extended? This can be done by dividing the time into sub-periods and determining whether the store's estimated sales can reach a predetermined target, such as the profit margin. If so, extending its operating hours can be considered. These sub-periods can be divided into half-hour or one-hour periods. For a second store that is not open during the nighttime hours, could it be opened during the nighttime hours while maintaining profitability? If there is still a significant demand for orders during the nighttime hours, then extending the operating hours of the second store, which is not open during the nighttime hours, could be considered.

[0084] Based on the above considerations, this application embodiment plans nighttime business districts for all stores within a certain geographical area, such as a city. The main process is as follows: Preprocessing When a user opens the app or mini-program, if their current location is not within the original business district, their location will be recorded and referred to as the target access point. Assuming the geographic grid is an H3 grid, for ease of processing, the target access point is mapped to H3 grids of similar area based on the H3 spatial index, thus obtaining an H3 grid including the target access point, i.e., the candidate geographic grid in the aforementioned embodiment, and determining the number of target access points in each candidate geographic grid. Due to the significant long-tail effect in the statistical distribution, the 80th percentile of the number of target access points is used as a dynamic threshold, i.e., selecting the top 20% of candidate geographic grids after sorting the number of target access points from high to low, and filtering out the remaining candidate geographic grids.

[0085] Business District Expansion First, calculate the distance matrix between the selected candidate geographic grids and the first store, and select the first store with a distance less than the threshold x for each candidate geographic grid as the candidate store.

[0086] Due to the large number of candidate geographic grids, a recursive logic is adopted here, progressing from coarse to fine based on the grid's specifications. If the previous iteration can determine the affiliation between a candidate geographic grid and a candidate store, all subgrids within that grid are assigned to that candidate store. However, if the previous iteration cannot determine the affiliation, the candidate geographic grid is broken down into more granular subgrids in the next iteration, and the candidate store assigned to each subgrid is determined. If the next iteration also fails to determine the affiliation, the subgrid is assigned to the nearest candidate store.

[0087] Taking a candidate geographic grid as an example where the candidate geographic grid is an 8-level H3 grid and the subgrid is a 9-level H3 grid, the process of determining the candidate geographic grid to which the subgrids belong in the candidate geographic grid is as follows: First, in the first iteration, an 8-level H3 grid containing the target access points is generated, and the first score of each 8-level H3 grid and each candidate store is calculated sequentially. The calculation method is as shown in Formula 1 above.

[0088] At this point, the candidate store to which the level 8 H3 grid belongs will be determined based on the highest first score. Generally, the level 8 H3 grid will be assigned to the candidate store with the highest first score. However, due to the rounding function, it's possible that the level 8 H3 grid has the highest first score in two or more candidate stores; this is called a one-to-many situation. If the level 8 H3 grid has the highest first score in only one candidate store, this is called a one-to-one situation.

[0089] When there is no one-to-many situation, the algorithm process ends because all H3 grids now have a unique owner.

[0090] When there is a one-to-many situation, the algorithm will proceed to the next round.

[0091] In the second round, addressing the one-to-many situation from the previous round, a level 9 H3 grid is first generated from the one-to-one level 8 H3 grid. This generated level 9 H3 grid still belongs to the candidate store to which the one-to-one level 8 H3 grid belonged. Next, for the one-to-many H3 grid, it is also decomposed into level 9 H3 grids, and their affiliation is determined. The specific method is as follows: calculate the second score of the level 9 H3 grid and the multiple candidate stores with the highest first score, using formula 1. Because of the rounding function, it's still possible for the level 9 H3 grid to have the highest second score among two or more candidate stores, i.e., a one-to-many situation.

[0092] If there is still a one-to-many situation in the 9th level H3 grid, the attribution will be determined directly based on the shortest distance from the 9th level H3 grid to the candidate store with the highest score. This method is called algorithm fallback.

[0093] Once all target stores belonging to candidate geographic grids have been confirmed, there may be instances where candidate geographic grids are not connected to the original business districts of the target stores, making it impossible to merge the original business districts and candidate geographic grids. In this case, the shortest path between the candidate geographic grids and the original business districts of the target stores can be found, and all sub-grids along this shortest path can be used as fill grids to fill the blank areas between the candidate geographic grids and the original business districts.

[0094] Next, the original business district, the filled grid, and the candidate geographic grid are merged, and operations such as spatial merging, boundary trimming, and smoothing are performed to obtain the target business district of the target store, i.e., the business district expansion result.

[0095] Business Hours Planning Here, the number of orders for any sub-period of the nighttime period will be estimated based on the target access point and the order conversion rate of that sub-period, as well as the proportion of orders in that sub-period to the total number of orders. Combined with data such as the average price of the orders, the estimated sales revenue will be calculated, as shown in Formula 2.

[0096] If the estimated sales amount exceeds the preset limit, the operating hours of the target store can be extended to that sub-period.

[0097] New store planning First, target access points outside the target business district are clustered (DBSCAN) with a minimum neighborhood radius of 700 meters and a minimum number of cluster points of 5. This means that each core point has at least 5 target access points within 700 meters. Based on this chain reaction, target access points can be clustered into several groups. Then, clusters with fewer than the first minimum number of target access points are filtered out.

[0098] Next, the target access points and their clusters are mapped onto multiple geographic grids, resulting in the clusters to which these grids belong. The H3 grids are then merged, and geographic grids belonging to the same cluster are filtered out if the number of target access points is less than the second minimum. This filtering process may require multiple rounds, with each round continuously filtering from the outermost layer.

[0099] In practice, the following problem may arise: A large number of geographic grids belonging to a certain cluster aggregate into a large connected region. In this case, it's necessary to segment these geographic grids into sub-clusters, and how to segment them into individual sub-clusters is a challenge. One approach is to randomly select an outer geographic grid and continuously acquire its surrounding geographic grids until the number of geographic grids in the sub-clusters reaches a certain requirement, ultimately ensuring that all sub-clusters have approximately equal areas. This segmentation method may result in excessively long and narrow business districts, significantly impacting delivery capacity. Therefore, a segmentation method is needed that can both reasonably segment the area and ensure that the areas of the sub-clusters are approximately equal. The solution is as follows: Multiple random points are generated within a cluster; the number of random points can be determined based on the number of clusters. For these random points, k-means clustering is performed to obtain multiple sub-clusters. The centroid of each sub-cluster is obtained, and a von Runeau diagram is constructed based on the obtained centroids. The von Runeau diagram is used to divide the geographic grid belonging to the same cluster into multiple areas to be assigned. The number of sub-clusters can be determined by the maximum area of ​​the business district and rounded up, as shown in Formula 3. Here, three sub-clusters are used as an example.

[0100] Finally, a second store is assigned to each of the regions to be allocated.

[0101] When there are two or more second stores, the following indicators are mainly considered in determining the sub-cluster affiliation: The navigation distance between the second store and the area to be assigned is considered the score; the greater the navigation distance, the lower the score. This will reduce the on-time delivery rate during actual delivery. The greater the straight-line distance between the second store and the area to be assigned, the lower the score. This indicates a higher probability that the two stores are spatially disconnected. The on-time rate of the second store is a key metric; the higher the on-time rate, the higher the score, indicating greater capacity to handle demand. The later the closing time of the second store, the higher the score.

[0102] Then, Formula 4 can be used to calculate the score of each store, and the area to be allocated can be assigned to the second store with the highest score to obtain the new store planning results.

[0103] In some embodiments, the results of new store planning and business district expansion can be merged to generate a planning report, which can record the original business district, target business district and business hours of each store.

[0104] This application also provides a store planning device, see [link to relevant documentation] Figure 16 The device includes: The acquisition module 201 is used to acquire the location information of the target access point within a preset time period; the target access point is the location of the user when accessing the service platform that provides instant delivery services, and the target access point is outside the original business district of multiple first stores; the multiple first stores are in operation during the preset time period; the original business district is the business district outside the preset time period. The allocation module 202 is used to map the target access point to a plurality of pre-divided geographic grids according to the location information of the target access point, to obtain a candidate geographic grid including the target access point; determine the distance between the candidate geographic grid and the plurality of first stores; determine the target store among the plurality of first stores based on the distance between the candidate geographic grid and the plurality of first stores; and allocate at least a portion of the sub-grids in the candidate geographic grid to the target store. The merging module 203 is used to merge the sub-grids assigned to the target store in the candidate geographic grid and the original business district of the target store, so as to plan the target business district of the target store within the preset time period.

[0105] The apparatus provided in this application has functions or includes modules that can be used to perform the methods described in the above method embodiments. The specific implementation can be referred to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0106] This application also provides a computer device, which includes at least a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the methods described in any of the foregoing embodiments.

[0107] Figure 17 This illustration shows a more specific hardware structure diagram of a computer device provided in an embodiment of this application. The device may include: a processor 301, a memory 302, an input / output interface 303, a communication interface 304, and a bus 305. The processor 301, memory 302, input / output interface 303, and communication interface 304 are interconnected internally via the bus 305.

[0108] The processor 301 can be implemented using a general-purpose central processing unit (CPU), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The processor 301 may also include a graphics card, such as an Nvidia Titan X graphics card or a 1080Ti graphics card.

[0109] The memory 302 can be implemented in the form of read-only memory (ROM), random access memory (RAM), static storage device, dynamic storage device, etc. The memory 302 can store the operating system and other applications. When the technical solutions provided in the embodiments of this application are implemented by tools or firmware, the relevant program code is stored in the memory 302 and is called and executed by the processor 301.

[0110] Input / output interface 303 is used to connect input / output modules to realize information input and output. Input / output modules can be configured as components in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touch screens, microphones, various sensors, etc., and output devices may include displays, speakers, vibrators, indicator lights, etc.

[0111] The communication interface 304 is used to connect a communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0112] Bus 305 includes a pathway for transmitting information between various components of the device, such as processor 301, memory 302, input / output interface 303, and communication interface 304.

[0113] It should be noted that although the above-described device only shows the processor 301, memory 302, input / output interface 303, communication interface 304, and bus 305, in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the embodiments of this application, and does not necessarily include all the components shown in the figures.

[0114] This application provides a computer program product, including a computer program that, when executed by a processor, implements the methods described in any embodiment of this application.

[0115] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the methods described in any of the foregoing embodiments.

[0116] Computer-readable media include both permanent and non-permanent, removable and non-removable media that can store information by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by computer devices. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0117] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments. The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separate. When implementing the embodiments of this application, the functions of each module can be implemented in one or more tools and / or hardware. Alternatively, some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0118] The above description is only a specific implementation of the embodiments of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the embodiments of this application, and these improvements and modifications should also be considered as the protection scope of the embodiments of this application.

Claims

1. A store planning method, characterized in that, The method includes: Obtain the location information of the target access point within a preset time period; the target access point is the location of the user when accessing the service platform that provides instant delivery service, and the target access point is outside the original business district of multiple first stores; the multiple first stores are in operation during the preset time period; the original business district is the business district outside the preset time period. Based on the location information of the target access point, the target access point is mapped to a plurality of pre-divided geographic grids to obtain a candidate geographic grid including the target access point; the distance between the candidate geographic grid and the plurality of first stores is determined, the target store among the plurality of first stores is determined based on the distance between the candidate geographic grid and the plurality of first stores, and at least a portion of the subgrids in the candidate geographic grid are assigned to the target store. The subgrids assigned to the target store in the candidate geographic grid and the original business district of the target store are merged to plan the target business district of the target store within the preset time period.

2. The method according to claim 1, characterized in that, The number of candidate geographic grids is greater than 1; the method further includes: Obtain the number of the target access points in multiple candidate geographic grids; A dynamic threshold is determined based on a preset quantile of the number of target access points in the plurality of candidate geographic grids; Candidate geographic grids whose number of target access points is less than the dynamic threshold are filtered.

3. The method according to claim 1, characterized in that, The step of determining the target store among the plurality of first stores based on the distance between the candidate geographic grids and the plurality of first stores, and assigning at least a portion of the sub-grids in the candidate geographic grids to the target store, includes: From the plurality of first stores, determine a number of candidate stores whose distance from the candidate geographic grid is less than a preset distance; Obtain the first score of the candidate stores, and the first score of any candidate store is inversely correlated with the distance of the candidate store to the candidate geographic grid. If the number of candidate stores whose first score meets the first preset condition is 1, the candidate store whose first score meets the first preset condition is determined as the target store, and all subgrids in the candidate geographic grid are assigned to the candidate store whose first score meets the first preset condition.

4. The method according to claim 3, characterized in that, The step of determining the target store among the plurality of first stores based on the distance between the candidate geographic grid and the plurality of first stores, and assigning at least a portion of the sub-grids in the candidate geographic grid to the target store, further includes: If the number of candidate stores whose first score meets the first preset condition is greater than 1, perform the following operations for each subgrid in the candidate geographic grid: Obtain the second score of each candidate store whose first score satisfies the first preset condition, and the second score of any candidate store is inversely correlated with the distance of the candidate store to the sub-grid; If the number of candidate stores whose second score meets the second preset condition is 1, the candidate store whose second score meets the second preset condition is determined as the target store, and the subgrid is assigned to the candidate store whose second score meets the second preset condition.

5. The method according to claim 4, characterized in that, The step of determining the target store among the plurality of first stores based on the distance between the candidate geographic grid and the plurality of first stores, and assigning at least a portion of the sub-grids in the candidate geographic grid to the target store, further includes: If the number of candidate stores whose second score meets the second preset condition is greater than 1, determine the distance between each candidate store whose second score meets the second preset condition and the subgrid; Candidate stores whose distance from the subgrid meets the preset distance condition are identified as target stores, and the subgrid is assigned to the candidate stores whose distance from the subgrid meets the preset distance condition.

6. The method according to claim 1, characterized in that, The step of merging the sub-grids assigned to the target store in the candidate geographic grid and the original business district of the target store to plan the target business district of the target store within the preset time period includes: If the subgrid assigned to the target store in the candidate geographic grid is not connected to the original business district of the target store, determine the shortest path between the subgrid assigned to the target store in the candidate geographic grid and the original business district of the target store; The subgrids assigned to the target store in the candidate geographic grid, the original business district of the target store, and the subgrids on the shortest path are merged to plan the target business district of the target store within the preset time period.

7. The method according to claim 1, characterized in that, The preset time period includes multiple sub-time periods, and the target store is open during some of these sub-time periods; the method further includes: For any one of the multiple sub-time periods in which the target store is not in operation, perform the following operations: The instant delivery service is obtained as follows: the historical order conversion rate of the instant delivery service in the sub-time period, the target number of target access points in the target business district in the sub-time period, and the proportion of the sub-time period to the total business hours of the target store. The estimated sales revenue of the target store in the sub-period is determined based on the historical order conversion rate, the target quantity, and the ratio. If the estimated sales amount is greater than the preset limit, the operating hours of the target store will be extended to include the sub-period.

8. The method according to claim 1, characterized in that, The method further includes: Obtain target access points outside the target business district within the preset time period, and based on the location information of the obtained target access points, cluster the obtained target access points to obtain several clusters; The clusters are mapped to the pre-divided geographic grids to obtain the clusters to which the geographic grids belong; Divide geographical grids belonging to the same cluster into several regions to be assigned; Identify multiple second stores that are not in operation during the preset time period; For each area to be allocated, a second store is allocated to the area based on the distance between the area to be allocated and each second store and the closing time of each second store. The business hours of the second stores allocated to the area to be allocated are extended to include the preset time period. The original business district of the second stores allocated to the area to be allocated and the area to be allocated are merged to plan the target business district of the second stores allocated to the area to be allocated.

9. The method according to claim 8, characterized in that, Based on the acquired location information of the target access points, the acquired target access points are clustered to obtain several clusters, including: Based on the location information of the target access points, several initial clusters are obtained. The distance between any two target access points in the same initial cluster is less than a preset distance threshold, and the number of target access points in the initial cluster is greater than a preset number threshold. The following operations are performed iteratively: for each candidate target access point in any initial cluster, obtain target access points from target access points outside the target business district that are less than a preset distance threshold from the candidate target access point, and add the target access points obtained from the target access points outside the target business district to the initial cluster to which the candidate target access point belongs.

10. The method according to claim 8, characterized in that, The process of dividing geographic grids belonging to the same cluster into several regions to be assigned includes: Multiple random points are identified from geographic grids belonging to the same cluster, and these random points are clustered to obtain multiple sub-clusters; Obtain the centroids of each sub-cluster and construct a von Lono diagram based on the obtained centroids. The von Lono diagram divides the geographic grid belonging to the same cluster into multiple regions to be assigned. The distance from any point in each region to the centroid of that region is less than the distance to the centroids of other regions to be assigned.

11. The method according to claim 8, characterized in that, The process of allocating second stores to the area to be allocated based on the distance between the area to be allocated and each second store, and the closing time of each second store, includes: For any given second store, obtain the score of the second store. The score of the second store is inversely correlated with the distance from the area to be assigned to the second store, positively correlated with the on-time rate of the second store, and positively correlated with the closing time of the second store. The area to be assigned is allocated to the second store whose score meets the third preset condition.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the method described in any one of claims 1 to 11.

13. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method according to any one of claims 1 to 11.

14. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the method described in any one of claims 1 to 11.

Images