Goods delivery point calculation method based on three-segment code region refinement and application thereof

Abstract

The invention discloses a delivery point calculation method based on three-segment-code area refinement and application thereof, and relates to the technical field of logistics automatic sorting, and the method comprises the steps: collecting three-segment-code area data of each salesman, and determining a preliminary delivery point division standard of an area in charge of the salesman based on the three-segment-code area data; calculating delivery point identification information corresponding to each delivery address by using an address model algorithm through matching of the delivery address provided by a client and a delivery point geographic fence; when the electronic waybill is generated, the identification information of the delivery points is printed on the electronic waybill according to a calculation result, and a sorter sorts and aggregates parcels according to the identification information of the delivery points on the electronic waybill; according to the method, through an intelligent address analysis model, in combination with historical delivery data and address keyword matching, the matching precision of complex addresses is further improved, and automatic calculation of delivery points is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of logistics automated sorting, and in particular to a delivery point calculation method based on three-segment code area refinement and its application. Background Art

[0002] In recent years, with the rapid development of e-commerce, the demand for the logistics industry has shown explosive growth, especially the surge in express delivery business, making parcel sorting one of the key links in the logistics chain. The traditional sorting method mainly relies on manual identification of address information on the package, which is not only inefficient, but also prone to errors due to human factors, affecting the accuracy and timeliness of delivery. In response to this problem, automated sorting systems have gradually been introduced, among which sorting methods based on technologies such as barcodes and QR codes have been applied to actual operations. However, the application of these technologies still has certain limitations when facing complex geographical areas or huge business volumes during peak hours. Especially in recent years, with the increasing prominence of the "last mile" delivery problem, how to achieve more accurate and efficient sorting at the delivery terminal has become a technical problem that needs to be solved urgently.

[0003] Although the existing sorting technology based on three-segment codes has improved the sorting efficiency to a certain extent, it still has many shortcomings in terms of detailed regional processing and accurate calculation of distribution points. The main manifestations are: First, the existing three-segment code algorithm cannot meet the needs of finer-grained regional division, resulting in the definition of distribution points being too rough and unable to adapt to the precise distribution needs in complex geographical environments; second, due to the high reliance of sales staff on the experience of regional division, the existing system is difficult to precipitate these experience data through algorithms, and thus cannot effectively form an automated regional optimization plan. These shortcomings have a significant impact on the accuracy and efficiency of parcel sorting during the sorting process, especially during peak hours. In addition, the existing three-segment code technology also has certain limitations when dealing with fuzzy address matching, and it is difficult to accurately parse complex address structures through automation technology. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a delivery point calculation method based on three-segment code area refinement and its application to solve the problems of low efficiency and poor accuracy in traditional sorting methods.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a delivery point calculation method based on three-segment code area refinement, which includes collecting three-segment code area data of each salesperson, and determining preliminary delivery point division standards for the salesperson's responsible area based on the three-segment code area data;

[0008] Based on the division criteria, use GIS tools to draw more detailed delivery point geo-fences within each salesperson’s three-segment code area;

[0009] Assign unique identification information to each delivery point and associate the identification information with the corresponding delivery point geofence;

[0010] Using the address model algorithm, the delivery point identification information corresponding to each delivery address is calculated by matching the delivery address provided by the customer with the delivery point geographic fence;

[0011] When an electronic waybill is generated, the identification information of the delivery point is printed on the electronic waybill based on the calculation results, and the sorting personnel sort and aggregate the packages according to the identification information of the delivery point on the electronic waybill.

[0012] As a preferred solution of the delivery point calculation method based on three-segment code area refinement of the present invention, wherein: the three-segment code area data of each salesperson is collected, and the specific steps are:

[0013] The courier management platform calls the application programming interface and database query to extract the three-segment code area data of each salesperson;

[0014] Perform data cleaning and standardization on the extracted three-segment code area data;

[0015] Load historical delivery data and use the point-in-polygon algorithm to verify whether the historical delivery address is within the boundaries of the salesperson's three-segment code area;

[0016] After verification and standardization are completed, the processed salesperson three-segment code area data is stored in the database.

[0017] As a preferred solution of the delivery point calculation method based on three-segment code area refinement of the present invention, wherein: the preliminary delivery point division standard of the salesperson's responsible area is determined based on the three-segment code area data, and the specific steps are:

[0018] Extract the three-segment code area data corresponding to the salesperson from the database;

[0019] Based on the extracted data, analyze the salesperson's historical delivery situation and count the delivery density and delivery volume distribution data in each three-segment code area;

[0020] Based on the delivery density and delivery volume distribution data, the salesperson is responsible for evaluating the area of ​​each three-segment code area, and the number of delivery points is initially set based on the delivery density in the area;

[0021] Estimate the coverage of each delivery point based on the number of delivery points and area;

[0022] According to the business needs of the delivery points, set the delivery density threshold for each delivery point;

[0023] By setting coverage and delivery density thresholds, the workload and service area of ​​each delivery point are clarified, and preliminary delivery point division standards are formed.

[0024] As a preferred solution of the delivery point calculation method based on the refinement of the three-segment code area of ​​the present invention, wherein: according to the division standard, a GIS tool is used to draw a more refined delivery point geographic fence within the three-segment code area of ​​each salesperson, and the specific steps are:

[0025] Import the three-segment code area data and the preliminary divided delivery point information into the GIS tool;

[0026] Based on the three-segment code area covered by each delivery point, use GIS tools to calculate the centroid of the area covered by each delivery point;

[0027] The centroid is taken as the center point of each delivery point, and the Voronoi diagram algorithm is used to automatically generate the centroid-based delivery point geofence.

[0028] As a preferred solution of the delivery point calculation method based on three-segment code area refinement of the present invention, wherein: unique identification information is assigned to each delivery point, and the identification information is associated with the corresponding delivery point geo-fence, the specific steps are:

[0029] Based on GIS tools and historical delivery data, analyze the density and geographical distribution of deliveries within the three-segment code area that each salesperson is responsible for;

[0030] Set coverage for each delivery point based on the delivery quantity limit and administrative divisions;

[0031] Use GIS tools to import the geo-fence information for each delivery point. In the layer manager, bind the geo-fence data to the identifier of the delivery point.

[0032] The unique identifier of each delivery point and the associated information of the geo-fence are stored in the database.

[0033] As a preferred solution of the delivery point calculation method based on three-segment code area refinement of the present invention, wherein: the address model algorithm is used to calculate the delivery point identification information corresponding to each delivery address by matching the delivery address provided by the customer with the delivery point geographic fence. The specific steps are:

[0034] Use natural language processing methods to clean the original address provided by the customer and perform standardized matching through the address dictionary;

[0035] Use geocoding services to convert standardized matching addresses into corresponding longitude and latitude coordinates;

[0036] Load geo-fence data of delivery points from database;

[0037] For each delivery address’s latitude and longitude coordinates, apply the point-in-polygon algorithm to verify whether the customer’s delivery address is within the geographic fence of the delivery point;

[0038] If the customer's delivery address is within the polygon of a delivery point, the identifier of the delivery point is returned, and then the unique identification information of the delivery point is associated with the delivery address to complete the match;

[0039] If the customer address is not within the geo-fence of any delivery point, it will be marked as an exception and manual address verification will be required.

[0040] As a preferred solution of the delivery point calculation method based on three-segment code area refinement of the present invention, wherein: the point-in-polygon algorithm is applied to verify whether the customer delivery address is located within the geographical fence of the delivery point, and the specific steps are:

[0041] Use the ray method to verify the matching of points within the polygon, and set the intersection counter to zero;

[0042] Draw a ray horizontally to the right from the client's latitude and longitude coordinates;

[0043] Traverse each edge of the geo-fence polygon of the delivery point, and determine whether the ray intersects the current edge based on comparing the relative positions of the two endpoints of the polygon edge with respect to the ray;

[0044] If the intersection condition is met, the horizontal coordinate of the intersection point between the ray and the polygon edge is calculated. If the horizontal coordinate of the intersection point is greater than the longitude of the customer address, the intersection counter is increased by one;

[0045] If the counter displays an odd number, the customer address is within the polygon, which means that the customer address is successfully matched to the geofence of the delivery point;

[0046] If it is an even number, it is outside the polygon, which means that the customer's delivery address cannot be matched within the geographic fence of the delivery point and is marked as an exception.

[0047] As a preferred solution of the delivery point calculation method based on three-segment code area refinement described in the present invention, wherein: the sorting personnel sort and aggregate packages according to the delivery point identification information on the electronic waybill, the specific steps are:

[0048] Extract the customer's delivery information and delivery point identification information from the database, and based on the extracted information, load the preset electronic waybill template and fill each field into the corresponding position;

[0049] After the template is filled, an electronic waybill is generated and output to a print format, and the electronic waybill is printed out;

[0050] The sorting staff automatically identifies the corresponding delivery point ID by scanning the barcode on the waybill, and guides the staff to distribute the packages to the corresponding sorting area.

[0051] In a second aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the delivery point calculation method based on three-segment code area refinement as described in the first aspect of the present invention is implemented.

[0052] In a third aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: the computer program, when executed by a processor, implements any step of the method for calculating a delivery point based on three-segment code area refinement as described in the first aspect of the present invention.

[0053] The beneficial effects of the present invention are as follows: the present invention introduces a GIS geographic information system tool to refine the three-segment code area and assigns unique identification information to each refined delivery point, thereby achieving more accurate delivery point calculation and matching. Through this improvement, sorting personnel can quickly classify and aggregate packages according to the delivery point identification information on the electronic waybill, greatly improving the sorting efficiency and reducing sorting errors caused by inaccurate address information; in addition, the present invention further improves the matching accuracy of complex addresses through an intelligent address resolution model, combines historical delivery data and address keyword matching, and realizes automatic calculation of delivery points, effectively solving the problems of inaccurate area division and inaccurate delivery point matching in the prior art, and can also dynamically adjust the delivery point division strategy according to actual business needs to further optimize the sorting process. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0055] Figure 1 This is a flow chart of the method for calculating the delivery point based on the refinement of the three-segment code area in Example 1. DETAILED DESCRIPTION

[0057] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0058] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0059] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0060] Example 1, reference Figure 1 , which is the first embodiment of the present invention, provides a method for calculating a delivery point based on three-segment code area refinement, comprising the following steps:

[0061] S1. Collect the three-segment code area data of each salesperson, and determine the preliminary delivery point division standard for the salesperson's responsible area based on the three-segment code area data.

[0062] Furthermore, the courier management platform calls the application programming interface (API) and database query to extract the three-segment code area data of each salesperson;

[0063] Among them, the three-segment code area data should include the specific area boundaries (polygon coordinates), postal codes and other information that the salesperson is responsible for.

[0064] Database query refers to obtaining the geographical boundary data for which the salesperson is responsible by querying the region table in the database.

[0065] API call means that if the data is stored in an external service, the region data can be obtained through the REST API.

[0066] Perform data cleaning and standardization on the extracted three-segment code area data;

[0067] It should be noted that the core of data cleaning is to remove illegal or incomplete data and ensure data quality. The specific steps are as follows:

[0068] Use regular expressions to check whether the postal code conforms to the standard format (6 digits). Records that do not conform will be marked as anomalies.

[0069] Check whether there are blank or missing postal_code, street_name or region_boundary fields. If so, fill them in or mark them as data to be added.

[0070] By setting a unique constraint on the combination of employee_id and postal_code, ensure that there is no duplicate data. If duplicate records are found, they need to be merged or deleted.

[0071] Standardization processing refers to standardizing geographic boundary data: for each three-segment code area, ensure that the region_boundary data is in a standard polygon coordinate format (such as GeoJSON or WKT format) for subsequent GIS operations; specifically, use a geographic data processing library (such as shapely, geopandas) to convert the format of the original geographic boundary data.

[0072] The collected and cleaned three-segment code data must be verified with the actual salesperson’s delivery area to ensure that the data matches the actual delivery area to avoid omissions or overlapping areas.

[0073] Load historical delivery data and use the Point-in-Polygon (PIP) algorithm to verify whether the historical delivery address is within the boundaries of the salesperson's three-segment code area;

[0074] If the PIP algorithm returns a "yes" result, it means that the longitude and latitude of the delivery address are within the boundaries of the salesperson's three-segment code area. The delivery address is directly associated with the salesperson's three-segment code area, and the result is stored in the database and marked as "verified", indicating that the delivery address has passed the verification of the responsible area and can enter the next step of delivery point calculation;

[0075] If the PIP algorithm returns a "no" result, it means that the longitude and latitude of the delivery address are not within the boundaries of the salesperson's three-segment code area, indicating that the delivery may have been incorrectly assigned to the salesperson. The reallocation algorithm will be called to reallocate the delivery address to the correct salesperson's area. The reallocation algorithm usually finds the closest salesperson's area based on the longitude and latitude information of the address;

[0076] Furthermore, the reallocation algorithm specifically uses the nearest neighbor algorithm to find the salesperson's responsible area closest to the delivery address, and based on the result of the nearest neighbor algorithm, reallocates the delivery to the correct salesperson and updates the corresponding record in the database.

[0077] If there is still a problem with the regional allocation of the delivery after multiple verifications (such as no matching area), the delivery will be recorded in the problem list for manual intervention.

[0078] After verification and standardization are completed, the processed salesperson's three-segment code area data will be stored in the database to ensure that the subsequent delivery point division process can directly reference these data.

[0079] Furthermore, the preliminary delivery point division standard for the salesperson's responsible area is determined based on the three-segment code area data. The specific steps are as follows:

[0080] Extract the three-segment code area data corresponding to the salesperson from the database. The data should include the salesperson ID, postal code, street name and boundary information of the three-segment code area (such as latitude and longitude or area code);

[0081] Based on the extracted data, analyze the salesperson's historical delivery situation and count the delivery density and delivery volume distribution data in each three-segment code area;

[0082] Among them, by analyzing historical delivery data, we can identify areas with large delivery volumes, and these areas will be used as the basis for dividing delivery points.

[0083] It should be noted that on this basis, AI models can be introduced to comprehensively analyze historical delivery data and real-time external data (such as weather, traffic flow, etc.) to predict the delivery volume in the future. Through machine learning algorithms, the delivery density trend of each area can be dynamically evaluated to ensure that the division of delivery points is more in line with future needs.

[0084] Based on the delivery density and delivery volume distribution data, the area of ​​each three-segment code area that the salesperson is responsible for is evaluated, and the number of delivery points is preliminarily set based on the delivery density in the area.

[0085] The delivery density can be defined as:

[0086]

[0087] Through this formula, the delivery density of each three-segment code area can be evaluated, thereby determining the division of delivery points.

[0088] For example: setting up more delivery points in high-density areas to reduce the coverage of each delivery point; reducing the number of delivery points in low-density areas to improve operational efficiency.

[0089] For example, when the delivery density exceeds a certain threshold (such as more than 100 deliveries per square kilometer), additional delivery points can be added to evenly distribute the delivery volume.

[0090] Furthermore, the AI ​​model will automatically adjust the number and coverage of delivery points based on dynamic delivery forecasts. For example, during holidays or promotional events, the system will predict areas with a surge in delivery volume and allocate more delivery points in advance to avoid concentrating delivery pressure in a few areas.

[0091] Estimate the coverage of each delivery point based on the number of delivery points and area;

[0092] The coverage can be estimated by geometric area, or adjusted based on geographical features (such as road density, building distribution, etc.). If the delivery density in a certain area is high, the coverage of the delivery point can be relatively reduced; conversely, in areas with low delivery density, the coverage of the delivery point can be appropriately expanded.

[0093] To further improve delivery efficiency, the AI ​​model will also combine real-time traffic data and road conditions to dynamically adjust the coverage of delivery points to adapt to actual geographical and road conditions. For example, during busy traffic hours, the system will automatically reduce the coverage of delivery points in certain areas to ensure timely delivery.

[0094] According to the business needs of the delivery points, set the delivery density threshold for each delivery point;

[0095] For example, if each delivery point handles an average of 500 parcels per day, the delivery density within the coverage area of ​​a delivery point should be controlled at around 500 parcels per day. This threshold should be adjusted based on business needs and historical data.

[0096] It should be noted that the business needs of delivery point division refer to multiple business factors related to logistics operations, such as delivery time requirements, salesperson workload, delivery peak periods and seasonal changes, customer satisfaction, regional geographical characteristics, delivery types, cost control, special needs of service areas and diversity of service types.

[0097] These requirements will affect the division and optimization of delivery points. At the same time, in the specific implementation process, the number, coverage and workload distribution of delivery points need to be adjusted according to actual business needs. For example, for business scenarios with high delivery timeliness requirements, the coverage of delivery points can be reduced and the number of delivery points can be increased; in scenarios where cost control is prioritized, the coverage of delivery points can be expanded to reduce the redundancy of delivery routes for salesmen.

[0098] Through the AI ​​model, the system will automatically evaluate various business needs and take into account delivery efficiency, customer satisfaction and cost optimization when setting the coverage and workload of delivery points, so that the division of delivery points can dynamically adapt to changes in different business scenarios.

[0099] By setting the coverage and delivery density thresholds, the workload and service area of ​​each delivery point are clarified, forming a preliminary delivery point division standard;

[0100] Specific criteria include: geographic coverage of each delivery point;

[0101] The delivery density threshold for each delivery point;

[0102] The workload of each delivery point (estimated number of deliveries);

[0103] The service area boundaries of the delivery points (visualized as polygonal areas on the map).

[0104] S2. Based on the division criteria, use GIS tools to draw more detailed delivery point geo-fences within each salesperson’s three-segment code area.

[0105] Furthermore, the three-segment code area data (including the boundary information of the area, delivery density, etc.) and the preliminary divided delivery point information are imported into a GIS tool (such as QGIS or ArcGIS);

[0106] Based on the three-segment code area covered by each delivery point, use GIS tools (such as QGIS or ArcGIS) to calculate the centroid of the coverage area of ​​each delivery point;

[0107] It should be noted that after the centroid is generated, the automatically calculated centroid point may not always conform to the actual geographic features or business needs, especially in the following cases:

[0108] Irregular regions: If the shape of the three-segment code region is very irregular, the automatically calculated geometric centroid may be located near the region border or even outside the region in some cases.

[0109] Geographical obstacles: The centroid may fall in some geographical locations that are not suitable as delivery points, such as near obstacles such as rivers, mountains, buildings, etc., and cannot be used as an actual delivery center.

[0110] In summary, after generating the centroid, it is necessary to check the centroid position and manually correct the centroid, as follows:

[0111] Check centroid location: Manually check the location of each centroid point to ensure that it is located in a reasonable geographical location, especially in complex geographical conditions (such as areas with rivers, mountains, or buildings).

[0112] Manually correct the centroid: In some special cases, the centroid can be manually adjusted to make it closer to a practical and easy-to-deliver location. For example, the centroid can be moved to a location close to a main road to make it easier for delivery personnel to operate.

[0113] Taking the centroid as the center point of each delivery point, use the Voronoi diagram (Thiessen polygon) algorithm to automatically generate a delivery point geo-fence based on the centroid point;

[0114] Among them, usually the Voronoi algorithm will divide the surrounding areas according to the location of the centroid of each delivery point, ensuring that each area serves its corresponding centroid nearby;

[0115] It should also be noted that although the Voronoi diagram (Thiessen polygon) can automatically generate coverage areas based on centroids, the generated area boundaries are calculated based on "geometric distance" and may not completely match the actual geographical features. For example:

[0116] Natural obstacles: Natural obstacles such as rivers, lakes, and mountains may cross the polygon boundaries generated by the Voronoi diagram, and these places are likely to not become actual delivery point areas.

[0117] Road network: The boundaries of the Voronoi diagram are geometrically straight lines, but actual delivery routes often rely on road networks, and the Voronoi boundaries cannot reflect these complex geographical factors.

[0118] Administrative boundaries or regional divisions: In some cases, the boundaries of the Voronoi diagram may cross the boundaries of administrative regions (such as cities or neighborhood divisions), which may affect the actual division of delivery operations.

[0119] In summary, after generating the Voronoi diagram, it is necessary to check the centroid position and manually correct the centroid, as follows:

[0120] Load actual geographic layers: Load additional geographic information layers (such as road layers, natural terrain layers, etc.) in the GIS tool, and adjust the boundaries of the Voronoi diagram based on these actual geographic features.

[0121] Manually adjust polygon boundaries: Based on the road network, natural obstacles, and administrative boundaries, manually adjust the polygon boundaries generated by the Voronoi diagram to make them more consistent with actual delivery needs. Factors such as road separation and bridge locations can be prioritized to ensure the actual operability of the delivery point.

[0122] In QGIS, use the Vertex Tool to manually adjust the vertices of the Voronoi polygons so that their boundaries conform to actual geographic features; in ArcGIS, use the Split tool or the Cut Polygons tool in the Editor Toolbar to adjust the boundaries of the polygons to more accurately match geographic barriers and road networks.

[0123] It should be noted that on the basis of generating the centroid and Voronoi diagram, an AI prediction model is introduced to automatically analyze the real-time data of regional traffic flow, road conditions, weather, etc., and dynamically adjust the geographical fence of the delivery point. Through AI model prediction, the system can automatically identify geographical obstacles or traffic problems that may affect delivery efficiency, and adjust the location and coverage of the delivery point accordingly, avoiding manual correction.

[0124] For example, AI models can automatically reduce the coverage of delivery points in areas with heavy traffic to avoid delivery delays due to traffic jams.

[0125] In severe weather conditions (such as heavy rain or snow), the system will automatically increase the number of delivery points in the area to avoid overloading the workload of a single courier.

[0126] At the same time, the AI ​​model can also analyze historical delivery data and future delivery volume forecasts to dynamically adjust the boundaries of delivery points. For example, if it is predicted that the delivery volume in a certain area will surge, the system will automatically re-divide the delivery point coverage area in that area to ensure delivery efficiency during peak periods.

[0127] Furthermore, through the introduction of AI models, the system can automatically combine geographic data (such as road networks, terrain, etc.) and real-time data (such as traffic, weather, etc.) to dynamically adjust the boundaries of the Voronoi diagram. The AI ​​model can identify in real time the geographic or traffic problems that may be encountered during delivery, and automatically optimize the geographic fence of the delivery point to make it more in line with actual delivery needs.

[0128] For example, when certain roads are closed or under construction, the AI ​​model can update traffic information in real time, and the system will automatically adjust the Voronoi polygon boundaries to ensure smooth and efficient delivery routes.

[0129] In complex urban geographical environments (such as commercial areas with tall buildings or dense residential areas), AI models can automatically reduce the coverage area of ​​delivery points and adjust them according to the actual road network to ensure that each delivery point is convenient for the delivery person's work.

[0130] By introducing the AI ​​model, the system can not only automatically generate geographic fences for delivery points, but also dynamically adjust the location and boundaries of delivery points according to real-time conditions, reducing manual intervention and improving delivery efficiency.

[0131] S3. Assign unique identification information to each delivery point and associate the identification information with the corresponding delivery point geo-fence.

[0132] Furthermore, based on GIS tools and historical delivery data, the density and geographical distribution of deliveries within the three-segment code area that each salesperson is responsible for is analyzed;

[0133] Set coverage for each delivery point based on the delivery quantity limit and administrative divisions;

[0134] Among them, the coverage of each delivery point should be reasonably allocated to ensure that the workload of the salesperson is balanced.

[0135] Use a GIS tool to import geofencing information for each delivery point (e.g. from a CSV or database);

[0136] In the layer manager, bind the geo-fence data with the identifier of the delivery point (such as "salesman ID-three-segment code-delivery point serial number") to ensure that each polygon area corresponds to a unique delivery point identifier;

[0137] The unique identifier of each delivery point and the associated information of the geo-fence are stored in the database to ensure that the identifier of each delivery point completely corresponds to its geo-fence information.

[0138] The associated information of geo-fence refers to various metadata and business data related to geo-fence. Its key associated information mainly includes:

[0139] Unique identifier: identifies each delivery point;

[0140] Geofence polygon coordinates: define the coverage area of ​​the delivery point;

[0141] Upper limit of the number of deliveries: determines the maximum processing capacity of the delivery point;

[0142] Salesperson information (optional): Indicate the salesperson in charge of the area;

[0143] Other metadata (such as delivery point name, regional attributes, etc.);

[0144] S4. Using the address model algorithm, the delivery point identification information corresponding to each delivery address is calculated by matching the delivery address provided by the customer with the delivery point geographic fence.

[0145] Furthermore, we use natural language processing (NLP) methods to clean the original addresses provided by customers, remove redundant information (such as punctuation marks, special characters, etc.), and deal with spelling errors or inconsistent formats. Specifically, we use specific technical means such as word segmentation algorithms and regular expression matching to achieve address cleaning and standardization;

[0146] For example, use a word segmentation algorithm (such as Stanford NLP or SpaCy) to automatically segment the address and break the entire text into semantic segments (such as province, city, street, house number, etc.);

[0147] Apply replacement rules to the word segmentation results, such as converting "Rd." to "Road" and "St." to "Street", to improve the standardization of the address.

[0148] Use address dictionaries to standardize matching of administrative divisions, streets, house numbers, etc., and correct common spelling errors or inconsistent address formats;

[0149] Use geocoding services (such as Google Geocoding API, Mapbox, etc.) to convert the standardized address into the corresponding longitude and latitude coordinates. The specific implementation steps are as follows: First, input the cleaned and standardized address string, call the geocoding API interface, and obtain the corresponding longitude and latitude data to handle the fuzzy matching problem;

[0150] Preferably, by introducing natural language processing (NLP) methods, complex address strings can be automatically processed and intelligently cleaned and standardized, which greatly reduces the need for human intervention. Compared with the solutions in the prior art that rely on manual confirmation or simple regular expressions, it has more efficient processing capabilities.

[0151] For multiple results returned (such as fuzzy matching multiple locations), the result that best matches the input is prioritized and sorted based on the address similarity score (similar to the Levenshtein distance algorithm).

[0152] Suppose the input address is string A, and the returned candidate address is B i (where i is the index number of the candidate address), and its similarity score expression is:

[0153]

[0154] Among them, d(A,B i ) is the Levenshtein distance, indicating the difference between string A and B i The edit distance between |A| and |B i | Strings A and B respectively i Length;

[0155] Similarity score S(A,B i ) has a value range of [0,1], where 1 indicates a perfect match and 0 indicates a perfect mismatch.

[0156] Preferably, the introduction of a similarity scoring mechanism can effectively solve the problem of multiple fuzzy matches that may occur during geocoding, ensuring that the result that best matches the customer's input is selected. This process is a dynamic and intelligent selection, rather than simply relying on the first result returned by the API.

[0157] For the obtained latitude and longitude coordinates, further processing is performed to ensure their accuracy, such as:

[0158] Geographic correction: Based on known geographic deviations (such as the deviation between China's GCJ-02 coordinate system and the WGS-84 coordinate system), coordinate conversion is performed to ensure that the longitude and latitude are consistent with the actual geographic location.

[0159] Coordinate accuracy limit: Based on the accuracy requirements of the delivery address (such as community, street, house number, etc.), the latitude and longitude accuracy is limited. Generally, 6 decimal places (such as 123.456789) are used to ensure meter-level accuracy.

[0160] Match the latitude and longitude coordinates with the delivery point geofence, and use the Point-in-Polygon (PIP) algorithm to determine whether the coordinate falls within the geofence of a delivery point;

[0161] Load the geo-fence data of the delivery point from the database, where the geo-fence of each delivery point is a polygon consisting of a set of latitude and longitude coordinate points;

[0162] For each delivery address’s latitude and longitude coordinates, apply the point-in-polygon algorithm to verify whether the customer’s delivery address is within the geographic fence of the delivery point. The specific steps are:

[0163] Use the ray method to verify the matching of points within the polygon, and set the intersection counter to zero;

[0164] Draw a ray horizontally to the right from the client's latitude and longitude coordinates to detect the intersection with the polygon edge;

[0165] Traverse each edge of the geo-fence polygon of the delivery point, and determine whether the ray intersects the current edge based on comparing the relative positions of the two endpoints of the polygon edge with respect to the ray;

[0166] The expression for determining the intersection point of the ray method is:

[0167] Assume that the longitude and latitude of the delivery address are (x0, y0), and the i-th edge of the delivery point polygon is (x i ,y i ) and (x i+1 ,y i+1 ), the horizontal coordinate of the intersection of the ray and the edge is x cos s , the intersection judgment formula is:

[0168]

[0169] Among them, x i and x i+1x0 and x1 are the longitudes of the two endpoints of the polygon side respectively, y0 is the latitude of the delivery address, and y i and y i+1 are the latitudes of the two endpoints of the polygon side respectively,

[0170] Intersection condition: If y i and y i+1 are respectively above and below the ray, that is, (y i > y0 and y i+1 < y0) or (y i < y0 and y i+1 > y0), and the intersection point x cross is greater than the longitude x0 of the delivery address;

[0171] If the intersection condition is met, calculate the abscissa of the intersection point of the ray and the polygon side. If the abscissa of this intersection point is greater than the longitude of the customer address, the intersection point counter is incremented by one;

[0172] If the counter shows an odd number, it means the customer address is inside the polygon, indicating that the customer address has been successfully matched to the geographical fence of the delivery point;

[0173] If it is an even number, it means it is outside the polygon, indicating that the delivery address of the customer fails to match the geographical fence of this delivery point and is marked as abnormal.

[0174] If the delivery address of the customer is inside the polygon of a certain delivery point (inside the geographical fence of the delivery point), return the identifier of this delivery point, and then associate the unique identification information of this delivery point (such as the delivery point code S_ID) with this delivery address to complete the matching for subsequent delivery processing and distribution;

[0175] If the customer address is not within the geographical fence of any delivery point, it is marked as abnormal and requires manual address verification.

[0176] S5. When generating the electronic waybill, print the identification information of the delivery point on the electronic waybill according to the calculation result, and the sorting personnel perform parcel sorting and aggregation according to the delivery point identification information on the electronic waybill.

[0177] Furthermore, extract the customer delivery information and the delivery point identification information matched by the address model algorithm from the database;

[0178] Among them, the delivery point identification information includes the standardized address, longitude and latitude, delivery ID and delivery point ID of the customer, providing a data basis for subsequent electronic waybill generation.

[0179] Based on the extracted delivery information, load the preset electronic waybill template and fill each field (such as customer information, delivery ID, delivery point ID, etc.) into the corresponding position;

[0180] It should be noted that the delivery point ID should be displayed in a prominent position, and a barcode or QR code containing the ID should be generated for subsequent automated sorting. This step ensures that the delivery point identification information is accurately integrated and reserves convenience for subsequent sorting.

[0181] After the template is filled, an electronic waybill is generated and output to a print format, and the electronic waybill is printed out;

[0182] At this point, the delivery note contains all necessary logistics information and the barcode / QR code of the delivery point ID, ensuring that the delivery point information can be quickly read by the scanning device during the sorting process;

[0183] The sorting staff automatically identifies the corresponding delivery point ID by scanning the barcode on the waybill, and guides the staff to distribute the packages to the corresponding sorting area.

[0184] This embodiment also provides a computer device, which is suitable for the case of a delivery point calculation method based on three-segment code area refinement, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the delivery point calculation method based on three-segment code area refinement as proposed in the above embodiment.

[0185] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0186] The present embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the delivery point calculation method based on the refinement of the three-segment code area as proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0187] In summary, the present invention introduces the GIS geographic information system tool to refine the three-segment code area and assigns unique identification information to each refined delivery point, thereby achieving more accurate delivery point calculation and matching. Through this improvement, sorting personnel can quickly classify and aggregate packages according to the delivery point identification information on the electronic waybill, which greatly improves the sorting efficiency and reduces sorting errors caused by inaccurate address information; in addition, the present invention also uses an intelligent address resolution model, combined with historical delivery data and address keyword matching, to further improve the matching accuracy of complex addresses, realizes automatic calculation of delivery points, and effectively solves the problems of inaccurate area division and inaccurate delivery point matching in the prior art. It can also dynamically adjust the delivery point division strategy according to actual business needs to further optimize the sorting process.

[0188] Example 2, referring to Table 1, is the second example of the present invention. To further verify the technical solution of the present invention, experimental simulation data of a delivery point calculation method based on three-segment code area refinement is provided.

[0189] The experiment selected two express delivery business areas in a city, Area A and Area B, which have different delivery densities and geographical distributions. The experiment used this method to evaluate the delivery point division and delivery sorting efficiency in the two areas, and compared it with the delivery point division method based on manual experience.

[0190] Furthermore, in traditional express delivery business, the division of delivery points often depends on the work experience and manual judgment of the salesperson. Specifically, the salesperson subjectively demarcates several delivery points based on the geographical scope and delivery location of the area he is responsible for. The division of these delivery points usually does not have precise geographical fences and lacks support based on data analysis. The salesperson artificially divides the area according to the names of streets and communities, and the coverage and delivery volume of each delivery point may be uneven, resulting in low sorting and delivery efficiency. In addition, the traditional method of parcel sorting process also relies on manual recognition of address information, which is difficult to avoid human errors.

[0191] First, the API is called through the express management platform to obtain the three-segment code area data of the salesperson in area A and area B. Then the data is standardized and cleaned to eliminate duplicates and abnormal data, and the historical delivery data is loaded. The Point-in-Polygon (PIP) algorithm is used to verify whether the historical delivery address is located in the respective three-segment code area to ensure data accuracy. The processed three-segment code area data is stored in the database for subsequent distribution point division.

[0192] Secondly, after extracting the salesperson's three-segment code area data from the database, a preliminary standard for the division of delivery points in each area was formulated based on the historical delivery density and delivery volume distribution. For Area A, historical data shows that the delivery density in this area is high, with an average delivery volume of 500 pieces per square kilometer. The density of Area B is relatively low, with an average delivery volume of 300 pieces per square kilometer. In combination with business needs, Area A is divided into 20 delivery points and Area B is divided into 15 delivery points. The coverage of each delivery point and the delivery density threshold are also optimized and adjusted according to actual conditions. In contrast, under the traditional method, Area A is divided into 15 delivery points and Area B is divided into 12 delivery points. The number and coverage of the delivery points are subjectively adjusted based on the experience of the salesperson.

[0193] Next, by importing the three-segment code area data into the GIS tool, the Voronoi diagram algorithm was used to generate the delivery point geofence for each area. The centroid of each delivery point was used as the center point to generate a centroid-based delivery point fence to ensure that the range covered by each delivery point can reasonably allocate the delivery volume in the area. In traditional methods, the geographical scope of the delivery point is not clearly defined by the GIS tool, resulting in the salesperson relying on business experience to determine whether the package belongs to a certain delivery point area during the sorting process.

[0194] Furthermore, the address model algorithm is used to clean and standardize the customer's delivery address, and the address is converted into longitude and latitude coordinates through the geocoding service. Subsequently, the PIP algorithm is used to verify whether the customer's delivery address is within the corresponding delivery point geofence. If the match is successful, the delivery point identification information is generated and printed on the electronic waybill. In the traditional method, address matching is to determine which delivery point the package should be delivered to by manually identifying the address information, which is prone to misclassification and delays.

[0195] Finally, after the electronic waybill generated by the new method is printed out, the sorting staff automatically identifies the delivery point information by scanning the barcode, and guides the package to be automatically assigned to the corresponding sorting area. Under the traditional method, the sorting staff needs to manually identify the address information of each package and sort it according to the subjective division of the area by the salesperson, which is inefficient and prone to errors. The experiment recorded the parcel sorting efficiency in each area and compared it with the traditional method.

[0196] The details are shown in Table 1 below:

[0197] Table 1 Experimental data comparison table

[0198]

[0199] By comparing the parameters of Area A and Area B under the traditional method based on manual experience and the delivery point calculation method based on the refinement of the three-segment code area, it can be clearly seen that the latter has significant advantages in many key indicators.

[0200] Shortening of sorting time: According to the data in the table, when using the traditional manual experience method in Area A, the average sorting time for each package was 1.8 minutes, while after using the new method, the sorting time was shortened to 1.0 minute. This is mainly due to the fact that the new method uses GIS tools to accurately divide the geographic fences of delivery points, allowing sorting personnel to quickly and accurately assign packages to the corresponding areas. The situation in Area B is similar, with the sorting time reduced from 2.0 minutes to 1.2 minutes. In contrast, under the traditional method, sorting personnel need to rely on manual judgment of the area to which the package belongs, resulting in longer sorting time.

[0201] Reduction in misclassification rate: Under the traditional method, the misclassification rate of Area A was 3.0%, while the new method reduced the misclassification rate to 0.5%. This improvement is mainly due to the new method generating geographic fences and algorithm matching through GIS tools, which greatly reduces the error of human judgment. Similarly, the misclassification rate of Area B dropped from 3.5% to 0.8%. Under the traditional method, the salesperson subjectively divides the delivery points based on experience, which can easily lead to misclassification of packages, especially when the address information is complex or unclear.

[0202] Improvement in delivery delay rate: The new method also performs quite well in terms of delivery delay rate. The delivery delay rate in Area A was reduced from 6.0% under the traditional method to 2.0%; the delay rate in Area B was reduced from 7.0% to 2.5%. This difference is due to the fact that the new method ensures a more balanced delivery workload at each delivery point through precise geographic fence division and data-driven delivery matching, avoiding delivery delays caused by overcrowding at some delivery points. Under the traditional method, due to the lack of detailed division of delivery points, some delivery points may have excessive deliveries, resulting in delays.

[0203] Number of delivery points and distribution of delivery volume: The experiment also shows that the method based on three-segment code area refinement can dynamically adjust the number and coverage of delivery points. The number of delivery points in Area A increased from 15 to 20 based on the experience of the salesperson, while the number of delivery points in Area B increased from 12 to 15. Although the number of delivery points has increased, the average delivery volume of each delivery point has also increased, from 350 pieces / day to 500 pieces / day in Area A, and from 300 pieces / day to 400 pieces / day in Area B. This shows that the new method can more reasonably configure the delivery tasks of each delivery point, avoiding the imbalance problem caused by experience judgment under the traditional method.

[0204] In summary, by comparing the data, it can be clearly seen that the delivery point calculation method based on the three-segment code area refinement is significantly better than the traditional delivery point method based on manual experience in terms of sorting time, misclassification rate, delivery delay rate, etc. This not only improves the overall operational efficiency, but also reduces the misclassification of packages and delivery delays, and improves the quality of express delivery services and customer satisfaction.

[0205] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A delivery point calculation method based on three-segment code area refinement, characterized by: include, Collect the three-segment code area data of each salesperson, and determine the preliminary delivery point division standards for the salesperson's responsible area based on the three-segment code area data; Based on the division criteria, use GIS tools to draw more detailed delivery point geo-fences within each salesperson’s three-segment code area; Assign unique identification information to each delivery point and associate the identification information with the corresponding delivery point geofence; Using the address model algorithm, the delivery point identification information corresponding to each delivery address is calculated by matching the delivery address provided by the customer with the delivery point geographic fence; When an electronic waybill is generated, the identification information of the delivery point is printed on the electronic waybill based on the calculation results, and the sorting personnel sort and aggregate the packages according to the identification information of the delivery point on the electronic waybill.

2. The method for calculating the delivery point based on the three-segment code area refinement according to claim 1, characterized in that: The specific steps of collecting the three-segment code area data of each salesperson are as follows: The courier management platform calls the application programming interface and database query to extract the three-segment code area data of each salesperson; Perform data cleaning and standardization on the extracted three-segment code area data; Load historical delivery data and use the point-in-polygon algorithm to verify whether the historical delivery address is within the boundaries of the salesperson's three-segment code area; After verification and standardization are completed, the processed salesperson three-segment code area data is stored in the database.

3. The method for calculating the delivery point based on three-segment code area refinement as claimed in claim 2, characterized in that: The specific steps of determining the preliminary delivery point division standard for the salesperson's responsible area based on the three-segment code area data are as follows: Extract the three-segment code area data corresponding to the salesperson from the database; Based on the extracted data, analyze the salesperson's historical delivery situation and count the delivery density and delivery volume distribution data in each three-segment code area; Based on the delivery density and delivery volume distribution data, the salesperson is responsible for evaluating the area of ​​each three-segment code area, and the number of delivery points is initially set based on the delivery density in the area; Estimate the coverage of each delivery point based on the number of delivery points and area; According to the business needs of the delivery points, set the delivery density threshold for each delivery point; By setting coverage and delivery density thresholds, the workload and service area of ​​each delivery point are clarified, and preliminary delivery point division standards are formed.

4. The method for calculating the delivery point based on three-segment code area refinement as claimed in claim 3, characterized in that: According to the division criteria, a GIS tool is used to draw a more detailed delivery point geo-fence within the three-segment code area of ​​each salesperson. The specific steps are as follows: Import the three-segment code area data and the preliminary divided delivery point information into the GIS tool; Based on the three-segment code area covered by each delivery point, use GIS tools to calculate the centroid of the area covered by each delivery point; The centroid is taken as the center point of each delivery point, and the Voronoi diagram algorithm is used to automatically generate the centroid-based delivery point geofence.

5. The method for calculating the delivery point based on three-segment code area refinement as claimed in claim 4, characterized in that: The specific steps of allocating unique identification information to each delivery point and associating the identification information with the corresponding delivery point geo-fence are as follows: Based on GIS tools and historical delivery data, analyze the density and geographical distribution of deliveries within the three-segment code area that each salesperson is responsible for; Set coverage for each delivery point based on the delivery quantity limit and administrative divisions; Use GIS tools to import the geo-fence information for each delivery point. In the layer manager, bind the geo-fence data to the identifier of the delivery point. The unique identifier of each delivery point and the associated information of the geo-fence are stored in the database.

6. The method for calculating the delivery point based on three-segment code area refinement as claimed in claim 5, characterized in that: The address model algorithm is used to calculate the delivery point identification information corresponding to each delivery address by matching the delivery address provided by the customer with the delivery point geographic fence. The specific steps are: Use natural language processing methods to clean the original address provided by the customer and perform standardized matching through the address dictionary; Use geocoding services to convert standardized matching addresses into corresponding longitude and latitude coordinates; Load geo-fence data of delivery points from database; For each delivery address’s latitude and longitude coordinates, apply the point-in-polygon algorithm to verify whether the customer’s delivery address is within the geographic fence of the delivery point; If the customer's delivery address is within the polygon of a delivery point, the identifier of the delivery point is returned, and then the unique identification information of the delivery point is associated with the delivery address to complete the match; If the customer address is not within the geo-fence of any delivery point, it will be marked as an exception and manual address verification will be required.

7. The method for calculating the delivery point based on three-segment code area refinement according to claim 6, characterized in that: The point-in-polygon algorithm is applied to verify whether the customer delivery address is located within the geographical fence of the delivery point. The specific steps are: Use the ray method to verify the matching of points within the polygon, and set the intersection counter to zero; Draw a ray horizontally to the right from the client's latitude and longitude coordinates; Traverse each edge of the geo-fence polygon of the delivery point, and determine whether the ray intersects the current edge based on comparing the relative positions of the two endpoints of the polygon edge with respect to the ray; If the intersection condition is met, the horizontal coordinate of the intersection point between the ray and the polygon edge is calculated. If the horizontal coordinate of the intersection point is greater than the longitude of the customer address, the intersection counter is increased by one; If the counter displays an odd number, the customer address is within the polygon, which means that the customer address is successfully matched to the geofence of the delivery point; If it is an even number, it is outside the polygon, which means that the customer's delivery address cannot be matched within the geographic fence of the delivery point and is marked as an exception.

8. The method for calculating the delivery point based on three-segment code area refinement according to claim 7, characterized in that: The sorting personnel sort and aggregate the packages according to the delivery point identification information on the electronic waybill, and the specific steps are as follows: Extract the customer's delivery information and delivery point identification information from the database, and based on the extracted information, load the preset electronic waybill template and fill each field into the corresponding position; After the template is filled, an electronic waybill is generated and output to a print format, and the electronic waybill is printed out; The sorting staff automatically identifies the corresponding delivery point ID by scanning the barcode on the waybill, and guides the staff to distribute the packages to the corresponding sorting area.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the delivery point calculation method based on three-segment code area refinement according to any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the delivery point calculation method based on three-segment code area refinement according to any one of claims 1 to 8 are implemented.

Images