String area delivery identification method and device, computer equipment and storage medium

Abstract

The invention relates to the technical field of logistics distribution management, and provides a string area delivery identification method and device, computer equipment and a storage medium, and the method comprises the steps: screening an effective record of which the delivery mode is internal delivery from a waybill table, and removing a whole vehicle type waybill and an internal part type waybill; counting the waybills in the same track from the waybill table according to receiving areas, wherein the receiving area with the most counted votes is a delivery area; waybills of the main delivery district are removed from the waybill table, and whether the remaining waybills are string district problem bills or not is judged; and for each string area problem ticket, according to the sum of the actual distances of the string area problem ticket and the adjacent main delivery district and the linear distance of the adjacent main delivery district corresponding to the string area problem ticket, obtaining a distance increasing value so as to generate an analysis result corresponding to each string area problem ticket, and completing string area delivery identification. The method is suitable for intelligent scheduling and optimization of delivery services in logistics enterprises, and solves the problems of extensive delivery identification and unreasonable resource allocation in the prior art.

Description

Technical Field

[0001] The present application relates to the technical field of logistics distribution management, and in particular to a method, device, computer equipment and storage medium for identifying multi-zone delivery. Background Art

[0002] In the logistics and distribution sector, the problem of out-of-area delivery (i.e., goods being delivered to areas other than the primary area of ​​responsibility) can lead to circuitous delivery routes, inefficient vehicle loading, and increased delivery costs. Efficiently identifying out-of-area delivery tickets and optimizing delivery strategies is a major challenge in logistics scheduling and management. Traditional delivery identification methods typically rely on manual judgment or simple rule-based screening. For example, they perform rough classification based solely on the area to which the delivery address belongs, lacking systematic analysis of multi-dimensional data such as waybill trajectory, vehicle load, and time and distance.

[0003] Existing technologies often rely on a single criterion (such as address ownership) to identify cross-region shipments. A comprehensive identification system that integrates primary delivery zone statistics, vehicle weight utilization, and route distance optimization has yet to be established. This makes it difficult to accurately identify cross-region shipments and generate actionable consolidated delivery recommendations, leading to wasted logistics resources and low delivery efficiency. Therefore, a method that can integrate multi-dimensional data, systematically identify cross-region shipments, and provide optimization strategies is urgently needed to address the crude shipment identification and irrational resource allocation issues inherent in existing technologies. Summary of the Invention

[0004] The present application provides a method, device, computer equipment and storage medium for identifying serial zone delivery, aiming to solve the problems of extensive delivery identification and unreasonable resource allocation in the prior art.

[0005] In a first aspect, the present application provides a method for identifying cross-region delivery, comprising:

[0006] Filter valid records with the delivery method of internal delivery from the waybill table, and exclude full vehicle type waybill and internal parts type waybill;

[0007] From the waybill table, count the waybills of the same track by receiving area. The receiving area with the most tickets is selected as the primary delivery area. If there are identical numbers of tickets, the receiving area with the largest cargo weight is selected as the primary delivery area.

[0008] Remove the waybills from the main delivery area from the waybill table, and determine whether the remaining waybills are cross-area problem tickets. Within the preset departure time range and straight-line distance range corresponding to the cross-area problem tickets, search for waybills with a ratio of vehicle load weight to the preset approved load less than or equal to the preset ratio. Mark these waybills as eligible for combination and generate a combined delivery suggestion.

[0009] For each problem ticket in the string area, the incremental distance value is obtained based on the sum of the actual distance between the problem ticket in the string area and the adjacent main delivery area and the straight-line distance between the problem ticket in the string area and the adjacent main delivery area, so as to generate the analysis results corresponding to each of the problem tickets in the string area and complete the identification of the string area delivery.

[0010] In some embodiments, the method of filtering valid records with internal delivery as the delivery method from the waybill table and eliminating whole vehicle type waybills and internal parts type waybills includes: performing a filtering operation on the waybill data in the waybill table, excluding waybills with the transportation type field marked as whole vehicle transportation or internal parts, and retaining only waybills with the delivery method field marked as internal delivery; for waybills with the same receipt date, delivery person number and delivery vehicle track number, selecting the only record with the earliest delivery check-in time and the status of completed check-in; if the delivery check-in time field is missing, it is replaced by the receipt time field value.

[0011] In some embodiments, the waybills of the same track are counted from the waybill table according to the receiving area, and the receiving area with the largest number of tickets is defined as the main delivery area, including: grouping the waybills according to the delivery vehicle track number, counting and counting the waybills in each group according to the receiving area name field, and generating the number of waybills for each receiving area; determining the receiving area with the highest number of waybills as the main delivery area; if there are multiple receiving areas with the same number of waybills, comparing the sum of the cargo weight fields of each receiving area, and selecting the receiving area with the largest total cargo weight as the main delivery area.

[0012] In some embodiments, the determination of whether the remaining waybills are problem tickets for cross-regional delivery includes: for the waybills remaining after excluding the waybills of the main delivery area, extracting the delivery time field and the departure time field of the corresponding track, and calculating the time difference; if the time difference is greater than 2 hours, and the sum of the actual driving distances from the longitude and latitude of the receiving address of the waybill to the longitude and latitude of the center of the adjacent main delivery area, minus the straight-line distance between the longitude and latitude of the center of the adjacent main delivery area is greater than or equal to 10 kilometers, it is determined to be a problem ticket for cross-regional delivery; if the front and rear main delivery area waybills are missing, the longitude and latitude registered at the delivery point are used as alternative coordinates for distance calculation.

[0013] In some embodiments, the marking is for a combinable ticket and a combined delivery suggestion is generated, including: for each problem ticket in the cross-zone, searching for other waybills with a straight-line distance of no more than 3 kilometers within 2 hours before and after the departure time of the corresponding trajectory; obtaining the preset load value corresponding to the delivery vehicle type, and calculating the ratio of the current vehicle weight to the preset load value; if the ratio is less than or equal to 80%, the qualified waybills are marked as combinable tickets, and the optimal combined delivery route suggestion is generated based on the delivery address of each waybill; among them, the preset load value corresponding to the van is 600 kilograms, the preset load value corresponding to the medium truck is 2 tons, the preset load value corresponding to the 7.6-meter truck is 8 tons, and the preset load value corresponding to the 9.6-meter truck is 10 tons.

[0014] In some embodiments, the analysis results corresponding to each of the string-area problem tickets are generated to complete the string-area delivery identification, including: using the waybill receipt time field as the statistical time caliber, and the delivery point number as the organizational unit, to count the string-area problem tickets, calculate the number of string-area tickets, the proportion of combinable tickets, the average increase value and the month-on-month change rate; the waybill number, increase value, approval mark and optimization suggestions of each string-area problem ticket are associated and archived, and weekly or monthly statistical reports are generated for the dispatching management system to call.

[0015] In some embodiments, the method also includes: constructing a historical cross-region behavior data set and inputting it into a machine learning model for training, where the input features corresponding to the machine learning model include waybill weight, delivery time interval, driving distance difference, and vehicle rated load utilization rate; obtaining a dynamic judgment threshold through machine learning model training, which is used to automatically adjust preset parameters, where the preset parameters include the difference between delivery time and vehicle departure time, distance difference, and vehicle rated load ratio; incrementally updating the machine learning model based on real-time logistics data, optimizing the cross-region delivery identification rules, and improving the accuracy of abnormal behavior identification.

[0016] In a second aspect, the present application provides a device for identifying goods delivered in multiple zones, comprising:

[0017] The waybill removal unit is used to filter out valid records with the delivery method of internal delivery from the waybill table, and remove full vehicle type waybill and internal parts type waybill;

[0018] The waybill statistics unit is used to count waybills of the same track from the waybill table by receiving area, and the receiving area with the most tickets is selected as the primary delivery area; among them, if there are the same number of tickets, the receiving area with the largest cargo weight is selected as the primary delivery area;

[0019] The problem determination unit is used to remove waybills from the main delivery area from the waybill table and determine whether the remaining waybills are cross-area problem tickets. Specifically, within the preset vehicle departure time range and straight-line distance range corresponding to the cross-area problem tickets, the unit searches for waybills with a ratio of vehicle load weight to a preset approved load less than or equal to a preset ratio, marks them as possible to be combined, and generates a combined delivery suggestion;

[0020] The identification completion unit is used to obtain the incremental value for each string-area problem ticket based on the sum of the actual distances between the string-area problem ticket and the adjacent main delivery area and the straight-line distance between the string-area problem ticket and the adjacent main delivery area, so as to generate the analysis results corresponding to each of the string-area problem tickets and complete the string-area delivery identification.

[0021] In a third aspect, the present application further provides a computer device, comprising:

[0022] memory and processor;

[0023] The memory is used to store computer programs;

[0024] The processor is used to execute the computer program and implement the steps of the string zone delivery identification method described in the first aspect above when executing the computer program.

[0025] In a fourth aspect, the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the processor implements the steps of the string zone delivery identification method described in the first aspect above.

[0026] The embodiments of the present application provide a method, device, computer equipment, and storage medium for identifying cross-zone delivery. The method ensures the scientific nature of the main delivery zone determination by performing dual statistics on the number of tickets and cargo weight for the receiving areas of the same track waybills, avoids the one-sidedness caused by a single indicator (such as only the number of tickets), and lays an accurate foundation for subsequent cross-zone identification. Combined with the preset vehicle departure time range, straight-line distance range, and vehicle-load weight ratio, it can not only identify cross-zone tickets with abnormal addresses, but also simultaneously screen out waybills that can be combined for delivery, realizing the integration of "problem identification" and "optimization suggestions" and improving the integration efficiency of delivery resources. By calculating the incremental distance value and generating analysis results, data support is provided for scheduling management, facilitating targeted optimization of delivery routes; at the same time, the suggestion to merge tickets that can be combined can effectively reduce the vehicle's empty driving rate and reduce distribution costs. Through statistical analysis of cross-zone problem tickets (such as the number of cross-zone tickets, the proportion of tickets that can be combined, etc.), a systematic management report is formed to help logistics companies achieve digital and refined control of the delivery process.

[0027] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0029] Figure 1 This is a schematic flow chart of the steps of a method for identifying cross-region delivery provided by an embodiment of the present application;

[0030] Figure 2 This is a structural diagram of a device for identifying goods delivered in a serial area provided in one embodiment of the present application;

[0031] Figure 3This is a schematic block diagram of the structure of a computer device provided in one embodiment of the present application.

[0032] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. DETAILED DESCRIPTION

[0033] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0034] The flowcharts shown in the accompanying drawings are for illustrative purposes only and do not necessarily include all contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may be decomposed, combined, or partially merged, so the actual execution order may vary depending on the actual situation.

[0035] It should be understood that, in order to clearly describe the technical solutions of the embodiments of the present invention, in the embodiments of the present invention, terms such as "first" and "second" are used to distinguish between identical or similar items having substantially the same functions and effects. Those skilled in the art will understand that terms such as "first" and "second" do not limit the quantity or order of execution, and that terms such as "first" and "second" do not necessarily define differences.

[0036] It should be understood that the terms used in this specification are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in this specification and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0037] It will also be understood that the term "and / or" as used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0038] The following describes some embodiments of the present application in detail with reference to the accompanying drawings. In the absence of conflict, the following embodiments and features therein may be combined with each other.

[0039] In the logistics and distribution sector, the problem of out-of-area delivery (i.e., goods being delivered to areas other than the primary area of ​​responsibility) can lead to circuitous delivery routes, inefficient vehicle loading, and increased delivery costs. Efficiently identifying out-of-area delivery tickets and optimizing delivery strategies is a major challenge in logistics scheduling and management. Traditional delivery identification methods typically rely on manual judgment or simple rule-based screening. For example, they perform rough classification based solely on the area to which the delivery address belongs, lacking systematic analysis of multi-dimensional data such as waybill trajectory, vehicle load, and time and distance.

[0040] Existing technologies often rely on a single criterion (such as address ownership) to identify cross-region shipments. A comprehensive identification system that integrates primary delivery zone statistics, vehicle weight utilization, and route distance optimization has yet to be established. This makes it difficult to accurately identify cross-region shipments and generate actionable consolidated delivery recommendations, leading to wasted logistics resources and low delivery efficiency. Therefore, a method that can integrate multi-dimensional data, systematically identify cross-region shipments, and provide optimization strategies is urgently needed to address the crude shipment identification and irrational resource allocation issues inherent in existing technologies.

[0041] To resolve the above issues, please refer to Figure 1 , Figure 1 This is a schematic flow chart of a method for identifying cross-region delivery, provided in one embodiment of the present application. This method can be implemented by a computer device, which can be deployed on a single server or a server cluster. Alternatively, it can be deployed on a handheld terminal, laptop computer, wearable device, or robot.

[0042] It should be noted that the acquisition of any information involved in the provided method complies with relevant regulations and is carried out with the user's consent. It will not infringe on the user's privacy and will not violate relevant laws and regulations.

[0043] Specifically, if Figure 1 As shown, the provided method for identifying cross-region delivery includes steps S101 to S104, which are detailed as follows:

[0044] Step S101: Filter valid records with internal delivery as the delivery method from the waybill table, and remove vehicle type waybills and internal parts type waybills.

[0045] Specifically, filtering valid records from the waybill table that meet the criteria is crucial for eliminating interfering data and focusing on the target waybill for analysis. Specific filtering criteria include: "Internal Delivery" means only tracking waybill operations handled by the company's internal delivery team, excluding third-party delivery or self-pickup scenarios. Excluding full truckload (FTL) waybill operations: FTL deliveries typically utilize dedicated vehicles, and their delivery logic differs from that of individual dispatches, requiring separate processing to avoid interference with regular dispatch analysis. Excluding internal component (e.g., internal documents and materials) means internal components (e.g., documents and materials) do not require external delivery and are not subject to cross-region issues, thus excluding them to reduce data noise. The waybill table fields must include "Delivery Method" (enumeration values: Internal Delivery, Third-Party Delivery, Self-Pickup, etc.) and "Waybill Type" (enumeration values: FTL, LTL, Internal Component, etc.). Filtering can be performed using database SQL statements or data processing scripts (e.g., Python Pandas) to obtain a valid waybill dataset for analysis, containing key information such as the waybill ID, track number, delivery address, cargo weight, departure time, and delivery zone code.

[0046] Step S102. Count the waybills of the same track from the waybill table by receiving area, and the receiving area with the largest number of tickets is the main delivery area; among them, if there are the same number of tickets, the receiving area with the largest cargo weight is selected as the main delivery area.

[0047] Specifically, for waybills with the same delivery track (i.e. the same delivery vehicle or the same planned route), the number of tickets is counted according to the receiving area, and the main delivery area is determined by the "number of tickets first + cargo weight auxiliary" rule, which serves as the basis for judging the cross-area.

[0048] Grouping statistics by track: Waybills on the same track belong to the same route planning range and are therefore combined for analysis. The primary zone determination rules include: the zone with the most tickets is designated as the primary zone; if multiple zones have the same number of tickets, the zone with the largest total cargo weight is selected (cargo weight directly affects vehicle loading efficiency, so routing for heavy cargo zones is prioritized).

[0049] Group by track number and calculate the total number of deliveries and total freight weight by delivery zone within each group. Sort the zones within each group in descending order by number of deliveries. If the number of deliveries is the same, sort by descending freight weight, and select the first zone after sorting as the primary zone. This can be achieved by using SQL grouping and aggregation functions (e.g., GROUP BY track number, delivery zone, COUNT(*) to count deliveries, SUM(freight weight) to count total freight weight). Alternatively, sorting and filtering can be achieved using the groupby and sort_values ​​methods of a data frame (e.g., Pandas). Output a primary delivery zone for each track, associated with all the deliveries associated with that track.

[0050] Step S103. Remove the waybills of the main delivery area from the waybill table, and determine whether the remaining waybills are problem tickets for the cross-area problem; among them, within the preset departure time range and straight-line distance range corresponding to the problem tickets for the cross-area problem, search for waybills whose vehicle weight to preset approved load ratio is less than or equal to the preset ratio, mark them as combinable tickets and generate a combined delivery suggestion.

[0051] Specifically, after removing the main area waybills, the remaining waybills belong to non-main area routes, which are potential cross-area tickets. This process involves screening for combinable tickets by searching for waybills with a low ratio of vehicle load to preset approved load (i.e., load utilization rate) (ratio ≤ a preset value, such as 70%) within the preset time range (e.g., the same departure time) and straight-line distance range (e.g., a reasonable radius around the main area) of the cross-area ticket. This determines whether the vehicle still has loading space and can be combined for dispatch to optimize the route. Merge suggestions can be generated to merge combinable cross-area tickets with the main area route, reducing duplicate dispatches or circuitous transportation.

[0052] For each track, records in the waybill where the "receiving area ≠ main area" are retained and marked as candidate cross-area tickets. Multi-dimensional screening criteria include: Time range: A preset departure time fluctuation range (e.g., ±30 minutes) is set to ensure that candidate tickets and the main area waybill fall within the same delivery period. Straight-line distance: The straight-line distance between the candidate ticket and the center of the main area (or major distribution point within the main area) is calculated using the longitude and latitude of the delivery address. A threshold (e.g., ≤ 10 kilometers) is set to exclude abnormal tickets that are too far away. Load utilization rate: The ratio of the loaded weight to the vehicle's rated load capacity is calculated (vehicle weight / rated load capacity ≤ a preset ratio, such as 0.7) to determine whether there is remaining space on the vehicle. Address latitude and longitude conversion can be used to convert the delivery address into coordinates using an address resolution interface (e.g., the Baidu Maps API). Distance calculation uses the Haversine formula to calculate the spherical distance between two points. Merge suggestion generation can merge eligible cross-area tickets with the main area waybill, adjusting the delivery order to shorten the route. Mark the problem tickets with cross-region issues, associate the waybill groups that can be combined, and generate "merged delivery route suggestions" (such as adding stopover points and adjusting the route order).

[0053] Step S104. For each string-area problem ticket, obtain the incremental distance value based on the sum of the actual distances between the string-area problem ticket and the adjacent main delivery area and the straight-line distance between the string-area problem ticket and the adjacent main delivery area, so as to generate the analysis results corresponding to each of the string-area problem tickets and complete the string-area delivery identification.

[0054] Specifically, by quantifying the increased route length (incremental distance) caused by cross-zone delivery, we assess the impact of cross-zone delivery and provide a basis for optimization strategies. Actual distance includes the actual round-trip distance required from the main zone for a cross-zone ticket delivered separately (or calculated based on the delivery network path). Straight-line distance includes the shortest straight-line distance between the address of a cross-zone ticket and the adjacent main zone (ideally, the optimal distance). The incremental distance = actual distance minus straight-line distance. A larger value indicates more severe circuitous transportation and requires optimization.

[0055] Actual distance is calculated based on path data from the logistics distribution network (such as road connectivity and restricted zones). A path planning algorithm (such as the Dijkstra algorithm) is used to calculate the round-trip distance (or detour distance) from the main area's distribution point to the cross-area ticket address. Straight-line distance is calculated by directly calculating the spherical distance between the cross-area ticket address and the center of the main area. The incremental distance value = actual driving distance minus the straight-line distance. Each cross-area ticket is annotated with the incremental distance value, the associated track, the main area, and a list of available tickets. This information is then combined to create a visual analysis report or data report for dispatchers to use in optimizing delivery sequences or reallocating areas. Actual road distance is obtained by integrating with a GIS geographic information system or a path planning API. Data visualization tools (such as Tableau and Python Matplotlib) are used to display the incremental distance distribution and identify high-impact cross-area issues.

[0056] In some embodiments, the method of filtering valid records with internal delivery as the delivery method from the waybill table and eliminating whole vehicle type waybills and internal parts type waybills includes performing a filtering operation on the waybill data in the waybill table, excluding waybills with the transportation type field marked as whole vehicle transportation or internal parts, and retaining only waybills with the delivery method field marked as internal delivery; for waybills with the same receipt date, delivery person number and delivery vehicle track number, the only record with the earliest delivery check-in time and the status of completed check-in is selected. If the delivery check-in time field is missing, it is replaced by the receipt time field value.

[0057] When filtering valid waybills, in addition to basic filtering criteria, deduplication rules and time field processing logic have been added. The filtering criteria excludes waybills with a transport type of "Full Truckload" or "Internal Parts," retaining only those with a delivery method of "Internal Delivery." The deduplication rules apply to waybills with the same sign-off date, courier ID, and delivery vehicle track ID, retaining only the unique record with the earliest delivery sign-off time and a sign-off completion status. If the "delivery sign-off time" is missing, the "receipt time" is used instead.

[0058] Data filtering is performed using the waybill table fields (transport type, delivery method) to exclude irrelevant waybill records. Deduplication is performed by grouping the filtered waybill records by "Sign-in Date, Delivery Person Number, and Delivery Vehicle Track Number." Within each group, records are sorted in ascending order by "Delivery Sign-in Time," prioritizing records with a "Sign-in Completed" status. If the "Delivery Sign-in Time" field is blank, records are sorted by "Sign-in Time." A window function (such as SQL's ROW_NUMBER()) is used to assign a sequence number to each group of records, retaining records with a sequence number of 1. This results in a unique, valid waybill record after deduplication, ensuring the uniqueness and accuracy of the data for subsequent analysis.

[0059] In some embodiments, the waybills of the same track are counted from the waybill table according to the receiving area, and the receiving area with the largest number of tickets is defined as the main delivery area, including: grouping the waybills according to the delivery vehicle track number, counting and counting the waybills in each group according to the receiving area name field, and generating the number of waybills for each receiving area; determining the receiving area with the highest number of waybills as the main delivery area; if there are multiple receiving areas with the same number of waybills, comparing the sum of the cargo weight fields of each receiving area, and selecting the receiving area with the largest total cargo weight as the main delivery area.

[0060] By defining statistical rules for primary delivery zones, we implement track grouping and two-dimensional sorting (ticket count prioritized, cargo weight secondary): Track grouping groups waybills by "delivery vehicle track number," with each group corresponding to a collection of waybills for the same delivery route. Zone statistics count waybills within each group by "receiving zone name" to generate the number of waybills for each zone. The primary zone is determined by: if a zone has the highest ticket count, it is designated the primary zone; if ticket counts are the same, the total cargo weight of each zone is compared, and the zone with the heaviest cargo weight is selected.

[0061] Grouped statistics utilize the data grouping and aggregation function to count the number of waybills and cargo weight by zone within each track group. Sorting and filtering sorts the statistical results for each track group first by "Number of Waybills" in descending order, then by "Total Cargo Weight" in descending order. The "Receiving Zone Name" of the first record after sorting is used as the primary delivery zone. The unique primary delivery zone corresponding to each track is output as the basis for determining if there are multiple zones.

[0062] In some embodiments, the determination of whether the remaining waybills are problem tickets for cross-regional delivery includes: for the waybills remaining after excluding the waybills of the main delivery area, extracting the delivery time field and the departure time field of the corresponding track, and calculating the time difference; if the time difference is greater than 2 hours, and the sum of the actual driving distances from the longitude and latitude of the receiving address of the waybill to the longitude and latitude of the center of the adjacent main delivery area, minus the straight-line distance between the longitude and latitude of the center of the adjacent main delivery area is greater than or equal to 10 kilometers, it is determined to be a problem ticket for cross-regional delivery; if the front and rear main delivery area waybills are missing, the longitude and latitude registered at the delivery point are used as alternative coordinates for distance calculation.

[0063] By refining the criteria for identifying cross-region problem tickets, dual thresholds for time difference and distance difference are introduced, including: Time condition: If the difference between the delivery time of the waybill and the departure time of the track is greater than 2 hours, it is considered a time mismatch. Distance condition: If the actual driving distance from the delivery address to the center of the adjacent main area minus the straight-line distance between the main area centers is ≥ 10 kilometers, it is considered a severely circuitous route. Coordinate missing processing: If the coordinates of the waybill in the main area are missing, the longitude and latitude registered at the delivery point are used instead.

[0064] The time difference calculation is done by extracting the delivery time of the waybill and the departure time of the track, and calculating the time difference (unit: hours). If the difference is greater than 2 hours, the distance determination is performed.

[0065] Distance calculation includes: Coordinate acquisition: Normal situation: Use the longitude and latitude of the center of the main area (such as the average coordinates of all delivery addresses in the area) and the longitude and latitude of the delivery address of the cross-area ticket. Missing coordinates: Use the fixed longitude and latitude registered by the delivery point (such as the coordinates resolved from the point address). Actual driving distance: Obtain the road distance between two points through the route planning API (such as Gaode Map). Straight-line distance: Use the Haversine formula to calculate the spherical distance. Distance difference = actual driving distance - straight-line distance. If it is ≥10 kilometers, it is determined to be a cross-area problem ticket. Judgment logic: if time difference > 2 hours and distance difference ≥ 10 kilometers: mark it as a cross-area problem ticket.

[0066] In some embodiments, the marking is for a combinable ticket and a combined delivery suggestion is generated, including: for each problem ticket in the cross-zone, searching for other waybills with a straight-line distance of no more than 3 kilometers within 2 hours before and after the departure time of the corresponding trajectory; obtaining the preset load value corresponding to the delivery vehicle type, and calculating the ratio of the current vehicle weight to the preset load value; if the ratio is less than or equal to 80%, the qualified waybills are marked as combinable tickets, and the optimal combined delivery route suggestion is generated based on the delivery address of each waybill; among them, the preset load value corresponding to the van is 600 kilograms, the preset load value corresponding to the medium truck is 2 tons, the preset load value corresponding to the 7.6-meter truck is 8 tons, and the preset load value corresponding to the 9.6-meter truck is 10 tons.

[0067] By defining the filtering criteria and merging strategies for eligible tickets, based on vehicle type's rated load, time window, distance range, and load utilization, we can: Search scope: Waybills within 2 hours before and after the departure time, with a straight-line distance of ≤3 kilometers. Vehicle type rated load: Preset upper limits for different vehicle types (600kg for vans, 2 tons for medium trucks, 8 / 10 tons for large trucks). Load utilization: Merge is determined when the current vehicle load / preset rated load value is ≤80%. Merge suggestion: Generate the optimal delivery route based on the delivery address.

[0068] Conditional search retrieves the preset approved load value based on the waybill's "Vehicle Type" and calculates the ratio of the current load (total loaded cargo weight) to the approved load value. It selects waybill with a departure time of ±2 hours, a straight-line distance ≤3 kilometers, and a load-to-weight ratio ≤0.8. Route optimization uses the TSP (Traveling Salesman Problem) algorithm or the route planning API to generate the optimal delivery sequence based on all delivery address coordinates, minimizing total travel distance. Marking and suggestions add a "Can be combined with tickets" tag to eligible waybill and generate route adjustment suggestions (such as adding a stop sequence).

[0069] In some embodiments, the analysis results corresponding to each of the string-area problem tickets are generated to complete the string-area delivery identification, including: using the waybill receipt time field as the statistical time caliber, and the delivery point number as the organizational unit, to count the string-area problem tickets, calculate the number of string-area tickets, the proportion of combinable tickets, the average increase value and the month-on-month change rate; the waybill number, increase value, approval mark and optimization suggestions of each string-area problem ticket are associated and archived, and weekly or monthly statistical reports are generated for the dispatching management system to call.

[0070] By defining the statistical dimensions and output format of analysis results based on timeframes, organizational units, and quantitative indicators, these include: Statistical dimensions: "Waybill receipt time" is used as the timeframe, and "delivery point number" is used as the organizational unit. Statistical indicators: Number of inter-zone tickets, percentage of combinable tickets (combinable tickets / total number of inter-zone tickets), average incremental distance, and month-over-month change rate (compared to the previous week / month). Data archiving and reporting: Associate waybill numbers, incremental distances, and optimization suggestions to generate periodic reports (weekly / monthly).

[0071] Data aggregation is grouped by "delivery point number" and "receipt time (week / month)" to calculate various metrics. Report generation involves storing the results in a database table and periodically generating PDF / Excel reports using ETL tools or the dispatch system. These reports include trend charts, percentage analysis, and other visualizations. Data archiving records an "approval flag" (a field manually confirmed by the dispatcher) and "optimization suggestions" for each cross-region ticket, creating historical traceability data.

[0072] In some embodiments, the method also includes: constructing a historical cross-region behavior data set and inputting it into a machine learning model for training, where the input features corresponding to the machine learning model include waybill weight, delivery time interval, driving distance difference, and vehicle rated load utilization rate; obtaining a dynamic judgment threshold through machine learning model training, which is used to automatically adjust preset parameters, where the preset parameters include the difference between delivery time and vehicle departure time, distance difference, and vehicle rated load ratio; incrementally updating the machine learning model based on real-time logistics data, optimizing the cross-region delivery identification rules, and improving the accuracy of abnormal behavior identification.

[0073] By introducing a machine learning model, we dynamically adjust preset thresholds to improve recognition accuracy. This includes: Dataset construction: Collecting historical data on cross-region ticketing and labeling whether it is a genuine cross-region ticket. Input features: Waybill weight, delivery interval, travel distance difference, and vehicle load utilization. Model training: Using classification algorithms (such as Random Forest and XGBoost) to train dynamic judgment thresholds. Incremental updates: Continuously optimize the model based on real-time data, automatically adjusting preset parameters such as time, distance, and load.

[0074] Feature engineering extracts four types of features: numerical features: waybill weight (kg), driving distance difference (actual distance - straight-line distance, km), rated load utilization rate (vehicle weight / rated load value); time features: the difference between delivery time and vehicle departure time (hours).

[0075] Model training involves dividing the model into training and test sets and using a supervised learning algorithm to determine whether a ticket is a cross-region problem ticket. Dynamic thresholds are derived from the model's output probability values ​​or feature importance (for example, adjusting the time difference threshold from a fixed 2-hour threshold to a dynamic value that varies with core utilization).

[0076] Threshold updates include regular (e.g., daily) fine-tuning of thresholds based on new data. For example, when the load utilization rate exceeds 70%, the time difference threshold is relaxed to 3 hours; when the load utilization rate is ≤50%, the time difference threshold is tightened to 1.5 hours. Model deployment integrates trained models into the recognition system, which receives logistics data in real time and outputs dynamic parameters, replacing traditional fixed threshold rules.

[0077] We use geographic spatiotemporal data to construct a graph structure, and use the graph convolutional network (GCN) to capture the spatiotemporal correlation characteristics between patches to achieve pattern recognition and prediction of cluster problems.

[0078] Graph structure modeling: Each delivery area is regarded as a graph node. Node features include historical cross-area frequency, average cargo weight, geographic coordinates, etc.; edge features include the historical total number of transport orders between areas, actual driving distance, road travel time, etc.

[0079] Spatiotemporal feature fusion: Based on GCN, the time dimension (such as time sliding window) is introduced to build a spatiotemporal GCN model to learn the evolution law of the clustering behavior of areas in time series.

[0080] Anomaly detection: Identify abnormal cluster patterns (e.g., the connection weight between a certain area and non-main areas is significantly higher than normal) through the node embedding vector output by the model.

[0081] Data preprocessing: Convert area coordinates into graph nodes. Use the Dijkstra algorithm to calculate the real-time road distance between nodes as edge weights. Count the total number of orders carried between areas on a weekly / monthly basis as edge features. Construct a time series feature matrix. Each time slice (e.g., 1 hour) contains dynamic data such as the node's cargo weight distribution and dispatch time.

[0082] The model is constructed using two GCN layers to extract spatial features, combined with the gated recurrent unit (GRU) to process time series, and finally outputs the string area probability value through the fully connected layer.

[0083] Model training and application: Using historical zone-crossing labels as supervisory signals, the model is trained using a cross-entropy loss function to identify whether real-time waybills exhibit abnormal zone-crossing patterns. The output includes a visual report containing zone-related heat maps to assist dispatchers in optimizing long-term zone division strategies.

[0084] In some embodiments, by building a reinforcement learning (RL) agent, the optimal combined delivery decision is dynamically generated based on real-time waybill data, replacing the traditional fixed threshold rules.

[0085] The state space definition includes the current vehicle weight, remaining rated load space, the set of planned path nodes, the time window / distance range of the inter-zone ticket, the vehicle model rated load parameters, etc.

[0086] The action space design includes discrete actions such as "merging current inter-zone tickets," "splitting to subsequent trains," and "adjusting the main zone range." The reward function design uses the reduction in total incremental distance after merging, the improvement in delivery time, and the vehicle load utilization rate as positive rewards, and the increase in route detours as negative rewards.

[0087] Environmental modeling uses a logistics simulation platform to simulate the delivery process, and inputs real-time waybill data and road network data (such as real-time traffic API) as environmental status.

[0088] The agent training uses a deep Q network (DQN) or a policy gradient algorithm (such as PPO) to gradually optimize the merging strategy through offline historical data simulation training.

[0089] Examples of simplified reward functions include: def calculate_reward(distance before merge, distance after merge, load utilization): distance reduction = distance before merge - distance after merge; load reward = 0.5 if 0.7 ≤ load utilization ≤ 0.9 else 0.2; return distance reduction * 0.1 + load reward. The online decision-making application receives new inter-zone ticket data in real time. The agent outputs the optimal action (such as whether to merge and route adjustment plan) based on the current state, and directly connects to the scheduling system through the API for execution.

[0090] In some embodiments, self-supervised learning is used to construct a feature representation of normal delivery patterns based on historical unlabeled waybill data. Unlabeled outliers with out-of-region anomalies are identified through reconstruction error. Self-supervised task design: Pre-training tasks such as "address mask prediction" and "trajectory sequence recovery" are designed to force the model to learn the spatiotemporal correlation features of normal deliveries. Feature reconstruction and anomaly scoring: Waybill features are extracted using an autoencoder (AE) or contrastive learning model (such as SimCLR). The reconstruction error of anomalies is calculated, and if it exceeds a threshold, it is considered an out-of-region anomaly.

[0091] Pre-training data is constructed by performing data augmentation on normal waybill data (such as randomly masking the longitude and latitude of the delivery address and disrupting the order of waybills within the trajectory) to construct self-supervised training samples.

[0092] The model architecture uses a multi-layer Transformer encoder as a feature extractor, and inputs a vector sequence containing features such as waybill time, cargo weight, address coordinates, and track number.

[0093] Anomaly detection involves generating feature representations for new waybill data through a pre-trained model, calculating the cosine distance or reconstruction error from the normal pattern, and setting a dynamic threshold (such as the mean + 3 times the standard deviation) to identify string area anomalies.

[0094] In some embodiments, a heterogeneous "waybill-area-vehicle" graph is constructed, leveraging graph neural networks to predict the potential for merging inter-area tickets with the main area waybill, replacing the traditional single-distance / load rule. Heterogeneous graph modeling: Node types include waybill (including cargo weight and time), area (including coordinates and historical co-load frequency), and vehicle (including vehicle type and rated load value). Edge types include "waybill belongs to area," "vehicle loaded with waybill," and "area adjacency." Graph Attention Mechanism: A graph attention network (GAT) learns the interaction weights between different node types to capture the comprehensive impact of waybill merging on path efficiency (such as time window matching and load complementarity).

[0095] Heterogeneous graph construction includes: using knowledge graph tools (such as Neo4j) to store three types of nodes and their relationships, for example: waybill node attributes: waybill ID, cargo weight, delivery time, address coordinates; area node attributes: area ID, center coordinates, historical average vehicle departure time; vehicle node attributes: vehicle ID, vehicle model, rated load value, and current load.

[0096] The model training uses historical mergeable tickets as positive samples and non-mergeable tickets as negative samples to train the GNN classification model and output the mergeable probability value.

[0097] Application scenarios include: For newly identified cross-zone tickets, the GNN model is used to predict the probability of merging. If the probability is higher than 70%, a merging suggestion is automatically generated; if it is lower than 30%, it is marked as "requires manual review".

[0098] In some embodiments, to address the data sparsity problem in newly developed areas, historical cluster models of mature areas are used for transfer learning to quickly build a localized recognition model.

[0099] The source and target domains are divided into two parts: selecting mature, data-rich regions as the source domain (e.g., Beijing, Shanghai, and Guangzhou) and newer regions as the target domain (e.g., third- and fourth-tier cities). Feature transfer strategies involve retaining the underlying feature extraction layers of the source domain model (e.g., geocoding and temporal feature processing) and fine-tuning only the parameters of the top-level classification layer. Domain adaptation techniques involve using adversarial domain adaptation (e.g., DANN) to reduce the distribution differences between the source and target domains and improve model generalization.

[0100] Source domain model pre-training learns common string region features (such as time difference, distance difference, and load ratio) by training basic models (such as random forest or neural network) on mature regional data.

[0101] Fine-tuning the target domain involves collecting a small amount of annotated data from the new region, freezing the first 90% of the base model's network layers, and updating only the parameters of the last fully connected layer. A domain adversarial loss function is introduced to make it impossible for the model to distinguish between input data from the source and target domains. Localized optimization incorporates specific rules for the new region (such as rural road restrictions and regional delivery time windows) and adds a custom rule layer to the migrated model, achieving hybrid "algorithm + rule" recognition.

[0102] In some embodiments, a causal inference model (such as a structural causal model (SCM)) is used to analyze the root cause of the cross-zone problem, replacing traditional correlation analysis to identify causal relationships such as "vehicle type mismatch" and "unreasonable area division."

[0103] Causal graph construction: Define potential causal variables (such as vehicle type, distance between zones, and departure time), construct a directed acyclic graph (DAG), and annotate the causal relationships between variables (e.g., "undersized vehicle type → insufficient load capacity → forced zone separation"). Intervention analysis: Use "do-calculus" to simulate the impact of intervention measures (such as changing vehicle type or adjusting zone boundaries) on the zone separation rate and quantify the causal effect.

[0104] Variable definition and data preparation include: defining independent variables: vehicle type (X1), distance from area center (X2), and fluctuation in departure time (X3); dependent variable: occurrence of cross-area traffic (Y); control variables: cargo weight distribution (Z1) and road congestion index (Z2).

[0105] Causal graph learning uses the PC algorithm or GES algorithm to learn the causal structure from historical data and combines the knowledge of logistics experts to modify the causal graph.

[0106] Causal effect calculation uses propensity score matching (PSM) to calculate the average treatment effect (ATE) of vehicle type on clustering, for example: ATE = E[Y|do(vehicle type = medium truck)] - E[Y|do(vehicle type = van)]; strategy optimization formulates optimization plans for variables with high causal effects. If "distance from district center > 15 kilometers" is found to be the main reason for clustering, the district redivision process is triggered.

[0107] The embodiments of the present application provide a method, device, computer equipment, and storage medium for identifying cross-zone delivery. The method ensures the scientific nature of the main delivery zone determination by performing dual statistics on the number of tickets and cargo weight for the receiving areas of the same track waybills, avoids the one-sidedness caused by a single indicator (such as only the number of tickets), and lays an accurate foundation for subsequent cross-zone identification. Combined with the preset vehicle departure time range, straight-line distance range, and vehicle-load weight ratio, it can not only identify cross-zone tickets with abnormal addresses, but also simultaneously screen out waybills that can be combined for delivery, realizing the integration of "problem identification" and "optimization suggestions" and improving the integration efficiency of delivery resources. By calculating the incremental distance value and generating analysis results, data support is provided for scheduling management, facilitating targeted optimization of delivery routes; at the same time, the suggestion to merge tickets that can be combined can effectively reduce the vehicle's empty driving rate and reduce distribution costs. Through statistical analysis of cross-zone problem tickets (such as the number of cross-zone tickets, the proportion of tickets that can be combined, etc.), a systematic management report is formed to help logistics companies achieve digital and refined control of the delivery process.

[0108] See also Figure 2 As shown, Figure 2 2 is a schematic diagram of the structure of a device 200 for identifying cross-region delivery of goods provided in an embodiment of the present application. The device 200 is configured to execute the steps of the cross-region delivery identification method described in the above embodiments. The device 200 can be a single server or a server cluster, or it can be a terminal, such as a handheld terminal, a laptop computer, a wearable device, or a robot.

[0109] like Figure 2 As shown, the device 200 for identifying goods delivered in a serial zone includes:

[0110] The waybill removal unit 201 is used to filter out valid records with the delivery method of internal delivery from the waybill table, and remove the whole vehicle type waybill and the internal parts type waybill;

[0111] The waybill statistics unit 202 is used to count the waybills of the same track from the waybill table by receiving area, and the receiving area with the most tickets is selected as the main delivery area; wherein, if there are the same number of tickets, the receiving area with the largest cargo weight is selected as the main delivery area;

[0112] Problem determination unit 203 is used to remove waybills from the main delivery area from the waybill table and determine whether the remaining waybills are inter-area problem tickets. Specifically, within the preset departure time range and straight-line distance range corresponding to the inter-area problem tickets, it searches for waybills with a ratio of vehicle load weight to a preset approved load less than or equal to a preset ratio, marks them as possible to be combined, and generates a combined delivery suggestion;

[0113] The identification completion unit 204 is used to obtain the incremental value for each string-area problem ticket based on the sum of the actual distances between the string-area problem ticket and the adjacent main delivery area and the straight-line distance between the string-area problem ticket and the adjacent main delivery area, so as to generate an analysis result corresponding to each of the string-area problem tickets and complete the string-area delivery identification.

[0114] In some embodiments, the method of filtering valid records with internal delivery as the delivery method from the waybill table and eliminating whole vehicle type waybills and internal parts type waybills includes: performing a filtering operation on the waybill data in the waybill table, excluding waybills with the transportation type field marked as whole vehicle transportation or internal parts, and retaining only waybills with the delivery method field marked as internal delivery; for waybills with the same receipt date, delivery person number and delivery vehicle track number, selecting the only record with the earliest delivery check-in time and the status of completed check-in; if the delivery check-in time field is missing, it is replaced by the receipt time field value.

[0115] In some embodiments, the waybills of the same track are counted from the waybill table according to the receiving area, and the receiving area with the largest number of tickets is defined as the main delivery area, including: grouping the waybills according to the delivery vehicle track number, counting and counting the waybills in each group according to the receiving area name field, and generating the number of waybills for each receiving area; determining the receiving area with the highest number of waybills as the main delivery area; if there are multiple receiving areas with the same number of waybills, comparing the sum of the cargo weight fields of each receiving area, and selecting the receiving area with the largest total cargo weight as the main delivery area.

[0116] In some embodiments, the determination of whether the remaining waybills are problem tickets for cross-regional delivery includes: for the waybills remaining after excluding the waybills of the main delivery area, extracting the delivery time field and the departure time field of the corresponding track, and calculating the time difference; if the time difference is greater than 2 hours, and the sum of the actual driving distances from the longitude and latitude of the receiving address of the waybill to the longitude and latitude of the center of the adjacent main delivery area, minus the straight-line distance between the longitude and latitude of the center of the adjacent main delivery area is greater than or equal to 10 kilometers, it is determined to be a problem ticket for cross-regional delivery; if the front and rear main delivery area waybills are missing, the longitude and latitude registered at the delivery point are used as alternative coordinates for distance calculation.

[0117] In some embodiments, the marking is for a combinable ticket and a combined delivery suggestion is generated, including: for each problem ticket in the cross-zone, searching for other waybills with a straight-line distance of no more than 3 kilometers within 2 hours before and after the departure time of the corresponding trajectory; obtaining the preset load value corresponding to the delivery vehicle type, and calculating the ratio of the current vehicle weight to the preset load value; if the ratio is less than or equal to 80%, the qualified waybills are marked as combinable tickets, and the optimal combined delivery route suggestion is generated based on the delivery address of each waybill; among them, the preset load value corresponding to the van is 600 kilograms, the preset load value corresponding to the medium truck is 2 tons, the preset load value corresponding to the 7.6-meter truck is 8 tons, and the preset load value corresponding to the 9.6-meter truck is 10 tons.

[0118] In some embodiments, the analysis results corresponding to each of the string-area problem tickets are generated to complete the string-area delivery identification, including: using the waybill receipt time field as the statistical time caliber, and the delivery point number as the organizational unit, to count the string-area problem tickets, calculate the number of string-area tickets, the proportion of combinable tickets, the average increase value and the month-on-month change rate; the waybill number, increase value, approval mark and optimization suggestions of each string-area problem ticket are associated and archived, and weekly or monthly statistical reports are generated for the dispatching management system to call.

[0119] In some embodiments, the method also includes: constructing a historical cross-region behavior data set and inputting it into a machine learning model for training, where the input features corresponding to the machine learning model include waybill weight, delivery time interval, driving distance difference, and vehicle rated load utilization rate; obtaining a dynamic judgment threshold through machine learning model training, which is used to automatically adjust preset parameters, where the preset parameters include the difference between delivery time and vehicle departure time, distance difference, and vehicle rated load ratio; incrementally updating the machine learning model based on real-time logistics data, optimizing the cross-region delivery identification rules, and improving the accuracy of abnormal behavior identification.

[0120] It should be noted that, those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described string-zone delivery identification device and each module can refer to the corresponding processes in the string-zone delivery identification method embodiments described in the above-mentioned embodiments, and will not be repeated here.

[0121] The above-mentioned method for identifying cross-region delivery can be implemented in the form of a computer program. The computer program can be used in Figure 2 Run on the device shown.

[0122] See also Figure 3 , Figure 3 1 is a schematic block diagram of the structure of a computer device provided in an embodiment of the present application. The computer device includes a processor, a memory, and a network interface connected via a device bus, wherein the memory may include a storage medium and an internal memory.

[0123] The storage medium can store an operating device and a computer program. The computer program includes program instructions, and when the program instructions are executed, the processor can execute any one of the string area delivery identification methods.

[0124] The processor is used to provide computing and control capabilities and support the operation of the entire computer equipment.

[0125] The internal memory provides an environment for the operation of the computer program in the non-volatile storage medium. When the computer program is executed by the processor, the processor can execute any one of the string area delivery identification methods.

[0126] The network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the terminal to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0127] It should be understood that the processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0128] In one embodiment, the processor is configured to execute a computer program stored in the memory to implement the following steps:

[0129] Filter valid records with the delivery method of internal delivery from the waybill table, and exclude full vehicle type waybill and internal parts type waybill;

[0130] From the waybill table, count the waybills of the same track by receiving area. The receiving area with the most tickets is selected as the primary delivery area. If there are identical numbers of tickets, the receiving area with the largest cargo weight is selected as the primary delivery area.

[0131] Remove the waybills from the main delivery area from the waybill table, and determine whether the remaining waybills are cross-area problem tickets. Within the preset departure time range and straight-line distance range corresponding to the cross-area problem tickets, search for waybills with a ratio of vehicle load weight to the preset approved load less than or equal to the preset ratio. Mark these waybills as eligible for combination and generate a combined delivery suggestion.

[0132] For each problem ticket in the string area, the incremental distance value is obtained based on the sum of the actual distance between the problem ticket in the string area and the adjacent main delivery area and the straight-line distance between the problem ticket in the string area and the adjacent main delivery area, so as to generate the analysis results corresponding to each of the problem tickets in the string area and complete the identification of the string area delivery.

[0133] In some embodiments, the method of filtering valid records with internal delivery as the delivery method from the waybill table and eliminating whole vehicle type waybills and internal parts type waybills includes: performing a filtering operation on the waybill data in the waybill table, excluding waybills with the transportation type field marked as whole vehicle transportation or internal parts, and retaining only waybills with the delivery method field marked as internal delivery; for waybills with the same receipt date, delivery person number and delivery vehicle track number, selecting the only record with the earliest delivery check-in time and the status of completed check-in; if the delivery check-in time field is missing, it is replaced by the receipt time field value.

[0134] In some embodiments, the waybills of the same track are counted from the waybill table according to the receiving area, and the receiving area with the largest number of tickets is defined as the main delivery area, including: grouping the waybills according to the delivery vehicle track number, counting and counting the waybills in each group according to the receiving area name field, and generating the number of waybills for each receiving area; determining the receiving area with the highest number of waybills as the main delivery area; if there are multiple receiving areas with the same number of waybills, comparing the sum of the cargo weight fields of each receiving area, and selecting the receiving area with the largest total cargo weight as the main delivery area.

[0135] In some embodiments, the determination of whether the remaining waybills are problem tickets for cross-regional delivery includes: for the waybills remaining after excluding the waybills of the main delivery area, extracting the delivery time field and the departure time field of the corresponding track, and calculating the time difference; if the time difference is greater than 2 hours, and the sum of the actual driving distances from the longitude and latitude of the receiving address of the waybill to the longitude and latitude of the center of the adjacent main delivery area, minus the straight-line distance between the longitude and latitude of the center of the adjacent main delivery area is greater than or equal to 10 kilometers, it is determined to be a problem ticket for cross-regional delivery; if the front and rear main delivery area waybills are missing, the longitude and latitude registered at the delivery point are used as alternative coordinates for distance calculation.

[0136] In some embodiments, the marking is for a combinable ticket and a combined delivery suggestion is generated, including: for each problem ticket in the cross-zone, searching for other waybills with a straight-line distance of no more than 3 kilometers within 2 hours before and after the departure time of the corresponding trajectory; obtaining the preset load value corresponding to the delivery vehicle type, and calculating the ratio of the current vehicle weight to the preset load value; if the ratio is less than or equal to 80%, the qualified waybills are marked as combinable tickets, and the optimal combined delivery route suggestion is generated based on the delivery address of each waybill; among them, the preset load value corresponding to the van is 600 kilograms, the preset load value corresponding to the medium truck is 2 tons, the preset load value corresponding to the 7.6-meter truck is 8 tons, and the preset load value corresponding to the 9.6-meter truck is 10 tons.

[0137] In some embodiments, the analysis results corresponding to each of the string-area problem tickets are generated to complete the string-area delivery identification, including: using the waybill receipt time field as the statistical time caliber, and the delivery point number as the organizational unit, to count the string-area problem tickets, calculate the number of string-area tickets, the proportion of combinable tickets, the average increase value and the month-on-month change rate; the waybill number, increase value, approval mark and optimization suggestions of each string-area problem ticket are associated and archived, and weekly or monthly statistical reports are generated for the dispatching management system to call.

[0138] In some embodiments, the method also includes: constructing a historical cross-region behavior data set and inputting it into a machine learning model for training, where the input features corresponding to the machine learning model include waybill weight, delivery time interval, driving distance difference, and vehicle rated load utilization rate; obtaining a dynamic judgment threshold through machine learning model training, which is used to automatically adjust preset parameters, where the preset parameters include the difference between delivery time and vehicle departure time, distance difference, and vehicle rated load ratio; incrementally updating the machine learning model based on real-time logistics data, optimizing the cross-region delivery identification rules, and improving the accuracy of abnormal behavior identification.

[0139] A computer-readable storage medium is also provided in an embodiment of the present application, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and the processor executes the program instructions to implement the steps of the string zone delivery identification method provided in the above embodiments of the present application.

[0140] The computer-readable storage medium may be an internal storage unit of the computer device described in the aforementioned embodiment, such as a hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a SmartMedia Card (SMC), a Secure Digital (SD) card, a flash memory card, etc., equipped on the computer device.

[0141] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present application, and such modifications or substitutions should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A method for identifying cross-region delivery, characterized in that: include: Filter valid records with the delivery method of internal delivery from the waybill table, and exclude full vehicle type waybill and internal parts type waybill; From the waybill table, count the waybills of the same track by receiving area. The receiving area with the most tickets is selected as the primary delivery area. If there are identical numbers of tickets, the receiving area with the largest cargo weight is selected as the primary delivery area. Remove the waybills from the main delivery area from the waybill table, and determine whether the remaining waybills are cross-area problem tickets. Within the preset departure time range and straight-line distance range corresponding to the cross-area problem tickets, search for waybills with a ratio of vehicle load weight to the preset approved load less than or equal to the preset ratio. Mark these waybills as eligible for combination and generate a combined delivery suggestion. For each problem ticket in the string area, the incremental distance value is obtained based on the sum of the actual distance between the problem ticket in the string area and the adjacent main delivery area and the straight-line distance between the problem ticket in the string area and the adjacent main delivery area, so as to generate the analysis results corresponding to each of the problem tickets in the string area and complete the identification of the string area delivery.

2. The method according to claim 1, characterized in that The method of filtering valid records with internal delivery as the delivery method from the waybill table and excluding vehicle type waybills and internal parts type waybills includes: Filter the waybill data in the waybill table to exclude waybills with the Transport Type field marked as Full Truckload or Internal Parts, and only retain waybills with the Delivery Method field marked as Internal Delivery. For waybills with the same delivery date, delivery person number, and delivery vehicle track number, select the only record with the earliest delivery arrival time and a status of "Delivery Completed". If the delivery arrival time field is missing, the value of the delivery time field is used instead.

3. The method according to claim 1, characterized in that The waybills of the same track are counted according to the receiving area from the waybill table. The receiving area with the most tickets is the main delivery area, including: Waybills are grouped by delivery vehicle track number, and the waybills in each group are counted by the pickup area name field to generate the number of waybills for each pickup area. The receiving area with the highest number of waybills will be determined as the main delivery area. If there are multiple receiving areas with the same number of waybills, the sum of the cargo weight fields of each receiving area will be compared, and the receiving area with the largest total cargo weight will be selected as the main delivery area.

4. The method according to claim 1, wherein The determination of whether the remaining waybills are problematic tickets for cross-region issues includes: For the remaining waybills after removing the waybills in the main delivery area, extract the delivery time field and the departure time field of the corresponding track and calculate the time difference; If the time difference is greater than 2 hours, and the sum of the actual driving distances from the latitude and longitude of the delivery address to the adjacent main delivery area center latitude and longitude, minus the straight-line distance between the adjacent main delivery area center latitude and longitude, is greater than or equal to 10 kilometers, it is determined to be a cross-region problem ticket; If the waybills for the front and rear main delivery areas are missing, the longitude and latitude registered at the delivery point will be used as alternative coordinates for distance calculation.

5. The method according to claim 1, wherein The ticket marked as combinable and generating a combined delivery suggestion includes: For each cross-zone problem ticket, search for other waybills within a straight-line distance of no more than 3 kilometers within 2 hours before and after the departure time of the corresponding track; Obtain the preset approved load value for the delivery vehicle type and calculate the ratio of the current vehicle weight to the preset approved load value. If the ratio is less than or equal to 80%, mark the eligible waybills as combinable and generate the optimal combined delivery route recommendation based on the delivery address of each waybill. Among them, the preset load value for vans is 600 kilograms, the preset load value for medium trucks is 2 tons, the preset load value for 7.6-meter trucks is 8 tons, and the preset load value for 9.6-meter trucks is 10 tons.

6. The method according to claim 1, characterized in that Generating the analysis results corresponding to each of the problem tickets in the cross-region and completing the cross-region delivery identification includes: Using the waybill receipt time field as the statistical time caliber and the delivery point number as the organizational unit, statistics are collected on problem tickets that cross areas. The number of tickets that cross areas, the proportion of tickets that can be combined, the average incremental value, and the month-on-month change rate are calculated. The waybill number, incremental value, approval mark and optimization suggestions of each problem ticket are associated and archived, and weekly or monthly statistical reports are generated for the dispatching management system to call.

7. The method according to claim 1, characterized in that The method further comprises: Construct a historical data set of cross-region behavior and input it into the machine learning model for training. The input features of the machine learning model include waybill weight, delivery interval, driving distance difference, and vehicle load utilization rate. A dynamic judgment threshold is obtained through machine learning model training, which is used to automatically adjust preset parameters, including the difference between delivery time and vehicle departure time, distance difference, and vehicle load ratio; Incrementally update the machine learning model based on real-time logistics data, optimize the rules for identifying cross-region delivery, and improve the accuracy of identifying abnormal behavior.

8. A device for identifying goods delivered in multiple zones, characterized in that: include: The waybill removal unit is used to filter out valid records with the delivery method of internal delivery from the waybill table, and remove full vehicle type waybill and internal parts type waybill; The waybill statistics unit is used to count waybills of the same track from the waybill table by receiving area, and the receiving area with the most tickets is selected as the primary delivery area; among them, if there are the same number of tickets, the receiving area with the largest cargo weight is selected as the primary delivery area; The problem determination unit is used to remove waybills from the main delivery area from the waybill table and determine whether the remaining waybills are cross-area problem tickets. Specifically, within the preset vehicle departure time range and straight-line distance range corresponding to the cross-area problem tickets, the unit searches for waybills with a ratio of vehicle load weight to a preset approved load less than or equal to a preset ratio, marks them as possible to be combined, and generates a combined delivery suggestion; The identification completion unit is used to obtain the incremental value for each string-area problem ticket based on the sum of the actual distances between the string-area problem ticket and the adjacent main delivery area and the straight-line distance between the string-area problem ticket and the adjacent main delivery area, so as to generate the analysis results corresponding to each of the string-area problem tickets and complete the string-area delivery identification.

9. A computer device, characterized in that: The computer device includes a memory and a processor; The memory is used to store computer programs; The processor is configured to execute the computer program and implement the method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which, when executed by a processor, causes the processor to implement the method according to any one of claims 1 to 7.

Images