Customer waybill straightening method and device, computer equipment and storage medium

Abstract

The invention relates to the technical field of logistics information processing, and provides a customer waybill straightening method and device, computer equipment and a storage medium, and the method comprises the steps: obtaining to-be-straightened waybill data of customer waybill data according to the dimensions of a sending company and a sending point; excluding waybill data of which the allocation flow-through times are less than or equal to 1 from the waybill data to be straightened, and obtaining attribution allocation information corresponding to each piece of waybill data to be straightened; generating a distribution configuration cargo flow table according to the plurality of pieces of attribution distribution information; extracting to-be-straightened waybill data of which the customer type is a sending company or a payment company and the pass weight is greater than or equal to a preset weight from the distribution configuration cargo flow table, and obtaining corresponding unstraightened cargo quantity information and to-be-straightened cargo quantity information; and generating analysis result information according to the to-be-straightened waybill data, the unstraightened goods quantity information and the to-be-straightened goods quantity information, and finishing straightening of the customer waybill data. The method is suitable for transportation route optimization and allocation resource allocation in a logistics transportation network.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of logistics information processing, and in particular to a customer waybill straightening method and device, computer equipment and a storage medium. BACKGROUND

[0002] In logistics transportation management, the straightening processing of customer waybills (i.e. reducing unnecessary distribution links and improving transportation efficiency by optimizing transportation routes) is a key link to reduce transportation costs and shorten transportation time. In the prior art, the following problems exist in the processing of customer waybills:

[0003] 1. Extensive data preprocessing: the traditional method does not perform accurate filtering for specific transportation modes (such as whole vehicle transportation, direct flight / air-to-air transportation), resulting in a large amount of invalid data in subsequent analysis, and lacking a unified data format and field mapping standard, poor data compatibility;

[0004] 2. Blind distribution flow configuration: the selection rules for the number of waybill distribution flows are not clear, which cannot effectively exclude low-value distribution data, and the determination of the attribution distribution information relies on manual experience, without forming a systematic attribution rule based on actual flow logs, resulting in insufficient scientificity of the distribution of goods flow direction table;

[0005] 3. Lack of quantitative basis for straightening strategy: the existing scheme does not combine customer types (such as sending companies and payment companies) and weight thresholds for weight analysis, which cannot accurately assess the difference between unstraightened weight and straightened weight, resulting in a lack of data support for the priority ranking and effect evaluation of straightening operation, making it difficult to achieve optimal allocation of transportation resources.

[0006] Therefore, there is an urgent need for a method to solve at least one of the above problems. SUMMARY

[0007] The present application provides a customer waybill straightening method, device, computer equipment and storage medium, which aims to solve the problems of extensive data preprocessing, blind distribution flow configuration and lack of quantitative basis for straightening strategy in the prior art.

[0008] In a first aspect, the present application provides a customer waybill straightening method, comprising:

[0009] performing general pre-filtering on the obtained customer waybill data, excluding waybills with service mode of whole vehicle transportation and transportation mode of direct flight and air-to-air, and unifying the data format and field mapping of the customer waybill data; obtaining waybill data to be straightened according to the dimensions of sending company and sending point;

[0010] Excluding the waybill data with the number of distribution flow-throughs less than or equal to 1 from the to-be-straightened waybill data, obtaining the corresponding home distribution information of each of the to-be-straightened waybill data; and generating a distribution-configured cargo flow direction table according to the plurality of home distribution information.

[0011] Extracting the to-be-straightened waybill data with the customer type being a sending company or a payment company and the weight being greater than or equal to a preset weight from the distribution-configured cargo flow direction table, and obtaining the corresponding un-straightened cargo volume information and the should-straightened cargo volume information;

[0012] Generating analysis result information according to the to-be-straightened waybill data, the un-straightened cargo volume information, and the should-straightened cargo volume information, and completing the straightening of the customer waybill data according to the analysis result information.

[0013] In some embodiments, the obtained customer waybill data is subjected to general pre-filtering, the waybills with the service mode being whole vehicle transportation are excluded, and the waybills with the transportation mode being direct flight and air-to-air are excluded, the data format and field mapping of the customer waybill data are unified, including: analyzing the service mode field and the transportation mode field in the customer waybill data, matching the value of the service mode field with a preset whole vehicle transportation service identifier, and matching the value of the transportation mode field with a preset direct flight transportation mode identifier and an air-to-air transportation mode identifier; logically deleting or marking the filtered state of the waybill data with the service mode matching the whole vehicle transportation identifier or the transportation mode matching the direct flight or air-to-air identifier; performing field standardization processing on the unfiltered waybill data, including mapping data items with the same meaning but different field names in different customer systems to a unified standard field, converting unstructured data into a structured data format, and filling in default values or marking abnormal data for missing fields.

[0014] In some embodiments, the exclusion of the waybill data with the number of distribution flow-throughs less than or equal to 1 from the to-be-straightened waybill data and the obtaining of the corresponding home distribution information of each of the to-be-straightened waybill data include: extracting the distribution node flow log of the waybill for the to-be-straightened waybill data filtered according to the sending company and the sending point, counting the number of distribution center nodes actually passed by each waybill in the transportation process as the number of distribution flow-throughs; retaining the waybill data with the number of distribution flow-throughs greater than 1 and excluding the waybill data with the number of distribution flow-throughs less than or equal to 1; and for the retained waybill data, determining the home distribution center corresponding to the waybill as the home distribution information according to the last actual operation distribution center node information in the distribution node flow log and combining a preset distribution center home rule, the distribution center home rule including home matching according to the geographical area, the operation subject, or the network level of the distribution node.

[0015] In some embodiments, the obtaining of the corresponding non-straightened cargo volume information and the should-straightened cargo volume information comprises: screening the shipment data of the shipment configuration cargo flow table with the customer type field being the sender company or the payment company, and extracting the cargo weight field in the shipment; comparing the cargo weight with a preset weight threshold, and retaining the shipment data with the cargo weight greater than or equal to the preset weight threshold; for the retained shipment data, calculating the total weight of the cargo in the shipment data that is not marked as straightened as the non-straightened cargo volume information, and calculating the total weight of the cargo in all shipments that meet the customer type and weight conditions as the should-straightened cargo volume information, wherein the preset weight threshold is pre-configured according to the straightening cost and efficiency parameters of the transportation network.

[0016] In some embodiments, the generating of the shipment configuration cargo flow table according to the multiple home distribution information comprises: grouping the to-be-straightened shipment data with the same home distribution information, and aggregating and statistically analyzing the transportation destination, transportation line and transportation time limit of each group of shipment data; generating a cargo flow record starting from the home distribution center and ending at the transportation destination according to the corresponding aggregation results of the aggregation and statistical analysis, wherein each flow record contains the number of shipments, the cargo weight distribution and the configuration information of the commonly used transportation line type of the flow; and performing deduplication and standardization processing on the flow record to form a shipment configuration cargo flow table containing the distribution center code, the destination code, the transportation line rule and the cargo type adaptation condition, wherein the shipment configuration cargo flow table is used to guide the cargo sorting and transportation line planning of the distribution center.

[0017] In some embodiments, the generating of the analysis result information according to the to-be-straightened shipment data, the non-straightened cargo volume information and the should-straightened cargo volume information comprises: establishing an analysis data model, and associating the sender company information, the sender point department information and the home distribution information in the to-be-straightened shipment data with the non-straightened cargo volume information and the should-straightened cargo volume information; calculating the ratio of the non-straightened cargo volume to the should-straightened cargo volume as a straightening rate index, and analyzing the differences in the straightening rate of different sender companies, sender point departments or home distribution centers; combining the transportation cost data and the distribution center processing efficiency data to generate a cargo volume straightening priority ranking in each dimension, and forming analysis result information containing straightening potential evaluation, cost-benefit analysis and operation suggestions, wherein the analysis result information is stored in the form of a structured data table or a visual report.

[0018] In some embodiments, the straightening of the customer shipping order data according to the analysis result information comprises: generating a straightening operation instruction for a high-priority shipping company or shipping point department based on the straightening priority ranking in the analysis result information, the operation instruction comprising adjusting the goods sorting rules of the distribution center to match the optimal route in the distribution configuration goods flow table, and sending a configuration request for a straightened transport route to a transport scheduling system; marking the shipping order data that has performed the straightening operation as straightened, and recording the execution time, operation subject and transport route change information of the straightening operation; periodically verifying the straightened shipping order data, comparing the actual transport time limit and transport cost with the expected indicators in the analysis result information, and forming a straightening effect evaluation report for optimizing subsequent straightening strategies.

[0019] In a second aspect, the present application provides a customer shipping order straightening device, comprising:

[0020] A pre-filtering unit is configured to perform general pre-filtering on the obtained customer shipping order data, exclude shipping orders with a service mode of whole vehicle transportation, and exclude shipping orders with a transport mode of direct flight and air-to-air, unify the data format and field mapping of the customer shipping order data, and obtain to-be-straightened shipping order data according to the shipping company and shipping point department dimensions.

[0021] A flow generation unit is configured to exclude shipping order data with a distribution flow frequency less than or equal to 1 from the to-be-straightened shipping order data, obtain the corresponding belonging distribution information of each to-be-straightened shipping order data, and generate a distribution configuration goods flow table according to a plurality of the belonging distribution information.

[0022] An information acquisition unit is configured to extract to-be-straightened shipping order data with a customer type of a shipping company or a payment company and a weight greater than or equal to a preset weight from the distribution configuration goods flow table, and obtain corresponding un-straightened goods volume information and should-straightened goods volume information.

[0023] A straightening completion unit is configured to generate analysis result information according to the to-be-straightened shipping order data, the un-straightened goods volume information and the should-straightened goods volume information, and complete the straightening of the customer shipping order data according to the analysis result information.

[0024] In a third aspect, the present application further provides a computer device, comprising:

[0025] a memory and a processor;

[0026] the memory is configured to store a computer program;

[0027] the processor is configured to execute the computer program and implement the steps of the customer shipping order straightening method of the first aspect when the computer program is executed.

[0028] 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 customer waybill straightening method according to the first aspect.

[0029] The customer waybill straightening method and device, the computer device and the storage medium provided by the embodiments of the present application can exclude invalid waybills through general pre-filtering and unify the data format, thereby improving the accuracy and efficiency of subsequent analysis and avoiding invalid data interference. Based on the sorting frequency screening and the home sorting rule matching, it is ensured that the goods distribution configuration goods flow table only contains waybill data with actual sorting value, thereby providing a scientific basis for the sorting and line planning of the sorting center. In combination with the customer type and the weight threshold, the non-straightened goods volume and the straightened goods volume are counted, the analysis result information is generated through the data model, the priority sorting and the effect evaluation of the straightening operation are realized, the resource utilization efficiency of the transportation network is significantly improved, the transportation cost is reduced, and the goods turnover time is shortened.

[0030] It should be understood that the foregoing general description and the following detailed description are only exemplary and explanatory, and are not limiting to the present application. BRIEF DESCRIPTION OF DRAWINGS

[0031] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0032] Figure 1 is a step schematic flow chart of a customer waybill straightening method provided by an embodiment of the present application;

[0033] Figure 2 is a structural schematic diagram of a customer waybill straightening device provided by an embodiment of the present application;

[0034] Figure 3 is a structural schematic block diagram of a computer device provided by an embodiment of the present application.

[0035] It should be understood that the foregoing general description and the following detailed description are only exemplary and explanatory, and are not limiting to the present application. DETAILED DESCRIPTION

[0036] With reference to the drawings, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only some of the embodiments of the present application, but not all of them. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of the present application.

[0037] The flowcharts shown in the drawings are only illustrative, and do not necessarily include all contents and operations / steps, nor are they necessarily executed in the order described. For example, some operations / steps can be further decomposed, combined or partially merged, so that the actual execution order can be changed according to actual conditions.

[0038] It should be understood that, in order to clearly describe the technical solutions of the embodiments of the present application, in the embodiments of the present application, the terms "first", "second", etc. are used to distinguish the same items or similar items with basically the same functions and effects. Those skilled in the art can understand that the terms "first", "second", etc. do not limit the quantity and execution order, and the terms "first", "second", etc. also do not necessarily mean that they are different.

[0039] It should be understood that the terms used in the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the present application specification and the appended claims, unless otherwise clear from the context, the singular forms "a", "an" and "the" are intended to include the plural forms.

[0040] It should also be understood that the term "and / or" used in the present application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes these combinations.

[0041] Some embodiments of the present application will be described in detail below with reference to the accompanying drawings. The following embodiments and features in the embodiments can be combined with each other without conflict.

[0042] In logistics transportation management, the straightening processing of customer waybills (i.e. reducing unnecessary distribution links and improving transportation efficiency by optimizing transportation routes) is a key link to reduce transportation costs and shorten transportation time. In the prior art, the following problems exist in customer waybill processing:

[0043] Data preprocessing is extensive: the traditional method does not accurately filter for specific transportation modes (such as whole vehicle transportation, direct flight / air-to-air transportation), resulting in a large amount of invalid data in subsequent analysis, and lacking a unified data format and field mapping standard, poor data compatibility;

[0044] The allocation flow direction configuration is blind: the screening rules for the number of allocation flow-through of the waybill are not clear, which cannot effectively exclude low-value allocation data, and the determination of the attribution allocation information relies on manual experience, without forming a systematic attribution rule based on actual flow log, resulting in insufficient scientificity of the goods flow direction table configured by the allocation.

[0045] The straightening strategy lacks quantitative basis: the existing scheme does not analyze the freight volume in combination with the customer type (such as the sender company and the payment company) and the weight threshold of the goods, which cannot accurately assess the difference between the non-straightened freight volume and the straightened freight volume, resulting in a lack of data support for the priority ranking and effect evaluation of the straightening operation, and it is difficult to achieve optimal allocation of transportation resources.

[0046] Although there are some waybill data processing or route optimization methods in the prior art, none of them form a complete systematic method of "multi-dimensional waybill filtering-allocation flow-through number screening-attribution allocation information matching-goods flow direction table generation-freight volume difference analysis", especially lacking precise filtering for specific transportation modes, attribution analysis based on actual allocation node flow log, and quantitative evaluation combined with customer type and weight threshold. Therefore, the prior art does not give technical inspiration for efficiently straightening customer waybills through the above multi-step cooperation, and an improved method is needed that can improve data processing accuracy, scientifically configure allocation flow direction, and quantify straightening strategies.

[0047] To solve the above problems, please refer to Figure 1 , Figure 1 is a schematic flowchart of a customer waybill straightening method provided by an embodiment of the present application. The customer waybill straightening method can be implemented by a computer device, which can be deployed on a single server or a server cluster. It can also be deployed on a handheld terminal, a notebook computer, a wearable device, or a robot, etc.

[0048] It should be noted that the acquisition of any information involved in the provided method is in accordance with relevant regulations and with the consent of the user, and does not infringe on the user's privacy or violate relevant laws and regulations.

[0049] Specifically, as shown in Figure 1 , the provided customer waybill straightening method includes steps S101 to S104, which are described in detail as follows:

[0050] Step S101. Perform general pre-filtering on the acquired customer waybill data, exclude waybills with service mode of truckload transportation, and exclude waybills with transportation mode of direct flight and air-to-air, unify the data format and field mapping of the customer waybill data; obtain the to-be-straightened waybill data according to the sender company and sender point dimensions.

[0051] Specifically, for the natural "direct" attribute of specific transportation modes (whole vehicle transportation, direct flight / air-to-air transportation) in logistics transportation, through data cleaning and format unification, accurate screening of shipping order data that need to be straightened and optimized is carried out, and the objects to be processed are divided according to business dimensions (mail sending company, mail sending point department), so as to provide high-quality data set for subsequent distribution analysis.

[0052] Transportation mode accurate filtering: define filtering rules: exclude whole vehicle transportation shipping orders through the "service mode" label (such as "whole vehicle transportation" and "less than full load transportation") in the shipping order field; exclude direct flight and air-to-air transportation shipping orders through the "transportation mode" label (such as "direct flight", "air-to-air" and "transit transportation"), because there is no distribution optimization space for such transportation. Through the use of regular expressions or database query statements (such as SQL WHERE conditions), batch screening of original shipping order data is carried out to generate a preliminary valid data set.

[0053] Data format and field mapping unification is achieved by establishing a standardized data model: defining unified field specifications (such as mail sending address, mail receiving address, cargo weight, distribution node, etc.), mapping fields of different data sources (such as ERP system, third-party platform) (for example, "shipping company" is uniformly mapped to "mail sending company", and "originating point" is mapped to "mail sending point department"). Through ETL (extraction-transformation-loading) tools or programming scripts (Python / PySpark), field cleaning, type conversion (such as converting string weight to numerical type) are carried out to eliminate data heterogeneity.

[0054] Dimension extraction of straightened shipping orders includes: dimension division: taking "mail sending company" (customer subject) and "mail sending point department" (shipping point) as grouping dimensions to ensure that subsequent analysis focuses on batch shipping order optimization of the same customer or point. Grouping and aggregation of the cleaned data set according to mail sending company + mail sending point department are carried out to extract shipping order data with non-direct transportation mode (need to be distributed) in each group to generate "straightened shipping order data set

[0055] Step S102. In the straightened shipping order data, exclude shipping order data with distribution flow-through frequency less than or equal to 1, obtain the corresponding attribution distribution information of each said straightened shipping order data, and generate a goods distribution configuration goods flow direction table according to a plurality of said attribution distribution information.

[0056] Specifically, through the quantitative screening of distribution flow-through frequency, low-value distribution data (too few flow-through frequencies, small optimization space) is excluded, and systematic attribution rules are established based on actual distribution node flow log to automatically determine the attribution distribution node of the shipping order, and finally a scientific goods flow direction table is generated to provide data support for distribution configuration.

[0057] The distribution flow frequency screening passes through the definition of screening rules, extracts the "distribution node number" (i.e. the number of distribution centers that the consignment passes through from sending out to signing) in the consignment flow log, and excludes consignments with flow frequency ≤ 1 (because such consignments are close to direct delivery, and the straightening space is limited). Associate consignment data with flow logs (through the consignment number primary key), calculate the distribution node count of each consignment (such as the number of "distribution center operation records" in the log), and filter the data using WHERE distribution flow frequency > 1.

[0058] The automatic matching of home distribution information includes: establishing a home rule model: based on historical flow logs, through machine learning (such as decision tree) or rule engine, defining the judgment logic of "home distribution center" (for example: the origin of the consignment belongs to a distribution center A, and more than 80% of the same point consignments are the first distribution node A, then default to A). For each consignment to be straightened, according to the sending point, historical distribution records, geographic fence (distribution center service range) and other data, through database association query or algorithm matching, automatically fill in the "home distribution center" field, replacing manual experience judgment.

[0059] Generating a distribution configuration freight flow table includes: data integration: structurally organizing the screened consignment data (including home distribution information) according to the flow chain of "sending point → home distribution center → target distribution center → receiving address", generating a freight flow table containing distribution node levels and transportation paths. Use a pivot table or a graph database (such as Neo4j) to build a distribution node relationship graph, visualize the freight flow, and provide intuitive basis for subsequent route optimization.

[0060] Step S103. Extracting consignment data with customer type as a sending company or a payment company and weight greater than or equal to a preset weight in the distribution configuration freight flow table, obtaining corresponding unstraightened freight information and should be straightened freight information.

[0061] Specifically, combining customer attributes (sending company / payment company) and freight weight threshold, accurately positioning high-value optimization objects, and providing priority ranking basis for straightening strategies by quantifying the difference between "unstraightened freight" (actual distribution freight) and "should be straightened freight" (theoretical direct delivery freight).

[0062] Customer type and weight threshold screening extracts shipments with customer type as "sender company" (actual shipper) or "payment company" (cost bearer), as these customers are usually more sensitive to transportation cost and time efficiency. According to transportation tool load limit (such as maximum load of trunk truck, air freight single limit) and cost-benefit analysis, a preset weight threshold (such as ≥ 300 kg) is set to screen "heavy cargo" shipments that need to be optimized. Data filtering is performed through WHERE customer type IN ('sender company', 'payment company') AND cargo weight ≥ preset threshold.

[0063] Unstraightened cargo volume information includes the actual distribution cargo volume (such as total weight, number of pieces) of the screened shipments, i.e. the total volume of cargo that needs to go through multiple distribution links under the current process. The should-be-straightened cargo volume information includes the theoretical cargo volume (consistent with the unstraightened cargo volume, the difference lies in the number of distribution links) that these shipments can be directly transported through straightening (reducing distribution links), which is used for comparative analysis of optimization potential. Aggregated calculation (SUM(cargo weight), COUNT(shipment quantity)) is performed on the screened shipment data to generate "unstraightened total weight", "should-be-straightened total weight" and other indicators.

[0064] Step S104. Generating analysis result information according to the to-be-straightened shipment data, unstraightened cargo volume information and should-be-straightened cargo volume information, and completing straightening of the customer shipment data according to the analysis result information.

[0065] By integrating to-be-straightened shipment data and cargo volume difference indicators, analysis results containing priority ranking and expected effect are generated to guide transportation line adjustment and achieve distribution link simplification and optimal resource allocation.

[0066] Analysis result information generation includes core indicators such as "number of to-be-straightened shipments", "difference between unstraightened cargo volume and should-be-straightened cargo volume", "estimated time efficiency reduction time" (calculated according to the number of distribution links reduction), and "cost saving estimate" (based on distribution link cost model). Priority ranking generates a priority list according to "cargo volume difference × customer value coefficient" (customer value can be calculated by weighting historical order volume, profit contribution, etc.), which clearly indicates that shipments of high cargo volume and high value customers should be optimized first. Structured analysis reports are generated through data reporting tools (such as Tableau) or custom algorithms to support visual display. Shipment straightening execution includes line optimization: according to the cargo flow table, direct lines are designed for high-priority shipments (such as skipping intermediate redundant distribution centers), and distribution routing rules are adjusted (such as configuring "prohibit access to certain distribution center" in WMS system). By recording the actual number of distribution links, transportation time efficiency and cost data after straightening, and comparing them with the analysis results, a closed-loop optimization is formed (such as periodically updating the preset weight threshold and attribution distribution rules). The optimized routing instructions are synchronized to the transportation management system (TMS) through the API interface to trigger line scheduling update.

[0067] In some embodiments, the acquired customer waybill data is subjected to general pre-filtering, excluding waybills with service mode of full truckload transportation and excluding waybills with transportation mode of direct flight and air-to-air, and the data format and field mapping of the unified customer waybill data are standardized, including: parsing the service mode field and the transportation mode field in the customer waybill data, matching the value of the service mode field with the preset full truckload transportation service identifier, and matching the value of the transportation mode field with the preset direct flight transportation mode identifier and air-to-air transportation mode identifier; logically deleting or marking the filtered state for waybill data matching the full truckload transportation identifier for service mode or matching the direct flight or air-to-air identifier for transportation mode; performing field standardization processing on unfiltered waybill data, including mapping data items with the same meaning but different field names in different customer systems to a unified standard field, converting unstructured data to a structured data format, and filling in default values for missing fields or marking abnormal data.

[0068] By parsing the service mode and transportation mode identifiers in the waybill field, the transportation types that do not need to be straightened (full truckload, direct flight / air-to-air) are accurately filtered, and the remaining data is subjected to field standardization processing, solving the data compatibility problem and providing a unified data basis for subsequent analysis.

[0069] Field parsing and identifier matching parse the [service mode field] (such as "transportation type" "service type") and [transportation mode field] (such as "transportation mode" "transportation path") in the waybill data, and extract the field values (such as "full truckload transportation" "less-than-truckload transportation" "direct flight" "air-to-air").

[0070] Establishing a preset identifier dictionary includes: full truckload transportation identifier, including keywords such as "full truckload" "full truckload transportation" "truckload"; direct flight transportation mode identifier, including keywords such as "direct flight" "direct air transport" "no transfer"; air-to-air transportation mode identifier, including keywords such as "air-to-air" "air direct" "airport direct delivery". Use string matching algorithm (such as regular expression, fuzzy matching) to compare the field values and identify waybills that meet the filtering conditions.

[0071] Data filtering processing performs "logical deletion" (marked as "no need to straighten" state, not physically deleted, for ease of auditing) or "marking filtering state" (adding "filtering reason" label, such as "full truckload transportation" "direct flight mode") on waybills matching the filtering identifier.

[0072] Field standardization processing includes: field mapping: establish cross-system field mapping table (such as mapping "shipping enterprise" of A system and "shipper name" of B system to standard field "sender company"), replace field name in batches through ETL tool or code script. Structured conversion parses ambiguous address by splitting unstructured data (such as address text) into structured fields (province / city / district / street) using natural language processing (NLP). Missing value processing fills in default values (such as "unknown") or marks as "data exception" for waybills with missing key fields (such as weight and sender point department), and subsequent manual verification.

[0073] In some embodiments, the step of excluding waybill data with a distribution flow-through frequency less than or equal to 1 from the to-be-straightened waybill data, and obtaining the belonging distribution information corresponding to each of the to-be-straightened waybill data, includes: for the to-be-straightened waybill data filtered according to the sender company and the sender point department, extracting the distribution node flow log of the waybill, and counting the number of distribution center nodes actually passed by each waybill in the transportation process as the distribution flow-through frequency; retaining the waybill data with a distribution flow-through frequency greater than 1, and excluding the waybill data with a distribution flow-through frequency less than or equal to 1; and for the retained waybill data, determining the belonging distribution center corresponding to the waybill as the belonging distribution information according to the last actually operated distribution center node information in the distribution node flow log and in combination with a preset distribution center belonging rule, the belonging rule including belonging matching according to the geographical area, the operation subject or the network level of the distribution node.

[0074] The number of actual distribution times of the waybill is counted through the distribution node flow log, low-value waybills (flow-through frequency ≤1) are excluded, and the belonging distribution center is determined based on the last operated node and the preset rule, replacing manual experience judgment, and the accuracy of the distribution information is improved.

[0075] The distribution flow-through frequency statistics include: associating the waybill data with the distribution node flow log (through the waybill number primary key), and extracting all records with the operation type "distribution center processing" in the log. The number of distribution center nodes of each waybill (i.e. the number of different distribution center codes in the log) is counted as the "distribution flow-through frequency".

[0076] The waybill screening includes: retaining the waybills with a distribution flow-through frequency greater than 1 (there are redundant distribution links, and there is room for straightening optimization), and excluding the waybills with a frequency less than or equal to 1 (close to direct, low optimization value).

[0077] The belonging distribution information is determined by extracting the "last actually operated distribution center node" (i.e. the distribution center closest to the consignee address) in the distribution node flow log of the retained waybill.

[0078] The application of the preset attribution rule to match the attribution distribution center includes: a geographical area rule: according to the service area of the distribution center (such as the East China distribution center covering Jiangsu, Zhejiang and Shanghai), the distribution center corresponding to the geographical area of the sending point department is matched; an operation subject rule: according to the company to which the distribution center belongs (such as a self-operated distribution center and a franchised distribution center), the logistics network attribution of the waybill is matched; and a network level rule: a trunk distribution center (rather than an end distribution center) is preferentially matched as an attribution node to ensure the coverage of the core distribution link.

[0079] In some embodiments, the obtaining of the corresponding non-straightened cargo quantity information and the should-straightened cargo quantity information comprises: screening the waybill data with the customer type field being a sending company or a payment company from the cargo distribution configuration table, and extracting the cargo weight field in the waybill; comparing the cargo weight with a preset weight threshold, and retaining the waybill data with the cargo weight being greater than or equal to the preset weight threshold; for the retained waybill data, the total weight of the cargo in the waybill that is not marked as having been straightened is taken as the non-straightened cargo quantity information, and the total weight of the cargo in all the waybills that meet the customer type and weight conditions is taken as the should-straightened cargo quantity information, and the preset weight threshold is pre-configured according to the straightening cost and efficiency parameters of the transportation network.

[0080] In combination with the customer type (sending / paying company) and the cargo weight threshold, high-value optimization objects are screened, the total weight of the waybills that meet the conditions is counted, the difference between the “non-straightened cargo quantity” (current distribution cargo quantity) and the “should-straightened cargo quantity” (theoretically straight-through cargo quantity) is quantified, and data support is provided for strategy making.

[0081] The customer type and weight screening screen the waybills with the [customer type field] being “sending company” or “payment company” from the cargo distribution configuration table (such customers are more sensitive to cost and time efficiency). The [cargo weight field] of the waybill is extracted, compared with a preset weight threshold (such as 300 kg, configured based on the lower limit of the load of the trunk transportation vehicle, the critical point of air transportation cost and other parameters), and the waybill with the weight being greater than or equal to the threshold is retained (heavy cargo distribution cost is high, and the straightening benefit is more significant).

[0082] The cargo quantity information calculation includes: non-straightened cargo quantity information: the total weight of the cargo in the waybill that is not marked as “straightened” (i.e. the total weight of the heavy cargo that still needs to go through multiple distributions); and should-straightened cargo quantity information: the total weight of the cargo in all the waybills that meet the customer type and weight conditions (whether straightened or not, representing the potential cargo quantity that can be optimized in theory). The weight summation of the screened waybill data is performed through the SQL aggregation function (SUM) or the data framework (such as Pandas), and the “straightened” state mark (such as filtering through the Boolean field “is_straightened”).

[0083] In some embodiments, the generating the goods flow table according to the plurality of home distribution information comprises: grouping the to-be-straightened shipping order data with the same home distribution information, and aggregating the transportation destination, transportation line and transportation time limit requirement of each group of shipping order data; generating a goods flow record starting from the home distribution center and ending at the transportation destination according to the aggregation result corresponding to the aggregation, each flow record containing the number of shipping orders, the weight distribution of goods and the configuration information of the commonly used transportation line type of the flow; and performing deduplication and standardization processing on the flow record to form a goods flow table for goods distribution and transportation line planning, which contains the distribution center code, the destination code, the transportation line rule and the goods type adaptation condition.

[0084] By grouping the shipping orders of the same home distribution center, aggregating the transportation destination, line, time limit and other information, a standardized goods flow table is generated to clearly define the optimal line configuration from the distribution center to the destination, providing rule guidance for sorting and scheduling.

[0085] Grouping and aggregation statistics group the to-be-straightened shipping order data according to the [home distribution center] (each group corresponds to the outgoing shipping order of a distribution center). The following information is aggregated for each group of shipping orders: transportation destination (province / city / district level of the recipient address, or custom area code); commonly used transportation line (such as “Distribution Center A → trunk transportation → Destination Distribution Center B”); time limit requirement (the latest delivery time specified by the customer, transportation time limit historical statistics value); number of shipping orders, weight distribution of goods (such as weight interval statistics proportion).

[0086] Flow record generation generates a flow record for each [home distribution center → destination] combination, which contains: basic information: distribution center code, destination code; traffic information: number of shipping orders, average weight, weight distribution interval; line information: commonly used transportation line type (highway trunk, railway, air transportation), line cost and time limit parameter.

[0087] Deduplication and standardization deduplicate the flow records according to [distribution center code + destination code], merge the statistical data of the same flow (such as average time limit, total shipping volume); standardize the field format (such as unify the line type naming as “trunk highway” “regional distribution”), and finally form a structured goods flow table containing the distribution center code, the destination code, the transportation line rule and the goods type adaptation condition (such as weight ≤ 500 kg suitable for highway trunk), which is stored in the database or configuration file.

[0088] In some embodiments, the generating analysis result information according to the to-be-straightened shipping order data, the non-straightened freight volume information and the freight volume information that should be straightened includes: establishing an analysis data model, associating the sender company information, the sender point department information and the home distribution center information in the to-be-straightened shipping order data with the non-straightened freight volume information and the freight volume information that should be straightened; calculating the proportion of the non-straightened freight volume to the freight volume that should be straightened as a straightening rate index, analyzing the straightening rate differences of different sender companies, sender point departments or home distribution centers; combining the transportation cost data and the distribution center processing efficiency data to generate a freight volume straightening priority ranking in each dimension, forming analysis result information including straightening potential evaluation, cost-benefit analysis and operation suggestions, and the analysis result information is stored in the form of a structured data table or a visual report.

[0089] By associating multi-dimensional data to establish an analysis model, calculating a straightening rate index, analyzing the straightening potential of different dimensions (sender company, point department, distribution center), combining cost and efficiency data to generate a priority ranking, and forming an analysis report that can guide operations.

[0090] The data model is established by constructing a multi-dimensional analysis table and associating the following data: to-be-straightened shipping order basic information: sender company, sender point department, home distribution center; freight volume difference index: non-straightened freight volume, freight volume that should be straightened, straightening rate (non-straightened freight volume / freight volume that should be straightened x 100%); additional attributes: customer historical order volume, distribution center processing cost (yuan / kg), average transportation time (hours).

[0091] The straightening rate analysis and priority calculation are grouped according to dimensions such as sender company, sender point department and home distribution center, the straightening rate of each group is calculated, and the high-potential group with low straightening rate (i.e. high proportion of non-straightened freight volume) is identified. A weight factor is introduced to calculate the priority: priority = (freight volume that should be straightened x customer value coefficient) + (1 - straightening rate) x distribution center cost saving coefficient; wherein the customer value coefficient is assigned according to historical order amount and cooperation period, and the distribution center cost saving coefficient is calculated according to the number of distribution times reduced after straightening x single distribution cost.

[0092] The analysis result output includes: generating a structured table containing group, straightening rate, freight volume that should be straightened, expected cost saving, priority level and other fields; visual report: showing the straightening rate differences of different groups through a column chart, displaying the straightening potential distribution of distribution centers through a heat map, supporting interactive filtering and drilling analysis; storage method: storing the results in a data warehouse or generating a PDF report for reference by the business department.

[0093] In some embodiments, the straightening of the customer shipment data according to the analysis result information comprises: generating straightening operation instructions for high-priority shipment companies or shipment point departments based on the straightening priority ranking in the analysis result information, the operation instructions including adjusting the cargo sorting rules of the distribution center to match the optimal route in the distribution configuration cargo flow table, and sending a configuration request for straightening the transportation route to the transportation scheduling system; marking the shipment data that executes the straightening operation as straightened, and recording the execution time, operation subject, and transportation route change information of the straightening operation; periodically verifying the straightened shipment data, comparing the actual transportation time limit and transportation cost with the expected indicators in the analysis result information, and forming a straightening effect evaluation report for optimizing subsequent straightening strategies.

[0094] Based on the analysis result, generate straightening operation instructions, adjust the distribution rules and transportation routes, mark the operation records, and evaluate the effect through backtracking verification to form a strategy optimization closed loop.

[0095] The straightening operation instruction generation includes: according to the priority ranking, generating operation instructions for high-priority groups (such as shipment companies with a straightening rate <30% and a shipment volume >10 tons that should be straightened): distribution rule adjustment by configuring sorting rules in WMS (warehouse management system), such as “shipment from shipment point department X, belonging to distribution center Y, destination Z, prohibited from entering intermediate distribution center M”; line configuration request sends line optimization instructions to TMS (transportation management system) through API, specifying direct routes (such as “distribution center Y→destination Z” dry line transportation route).

[0096] Operation records and state marking mark the shipment that executes the straightening operation as “straightened”, record key information: operation time, operation subject (system automatically triggered or manually confirmed); route change details (original distribution path and new direct path); expected indicators in the analysis result (such as expected time limit reduction of 2 hours, cost saving of 5%). Backtracking verification and effect evaluation: periodically (such as every week / month), extract the actual transportation data (distribution times, time limit, cost) of the straightened shipment, and compare it with the expected indicators in the analysis result: time limit evaluation: actual transportation time and expected shortened time; cost evaluation: actual transportation cost and expected cost saving (based on the reduction of distribution times); generate an effect evaluation report, summarize the optimization effect, identify abnormal cases (such as the time limit of a certain route is not up to standard after straightening, analyze whether it is caused by the processing efficiency bottleneck of the distribution center), and adjust the preset weight threshold, attribution rules, etc. parameters to optimize subsequent strategies.

[0097] In some embodiments, by introducing natural language processing (NLP) technology to address the limitations of traditional rule matching in parsing ambiguous or non-standard fields, semantic understanding of unstructured text fields such as service mode and transportation mode is achieved, improving the robustness of filtering rules and automatically completing missing key information (such as implied transportation mode).

[0098] Domain corpus construction involves collecting service mode and transportation mode description texts (such as "whole vehicle package" "air direct flight without transfer" "air-to-air transfer") from historical shipment data, labeling standard categories (whole vehicle / non-whole vehicle, direct flight / non-direct flight), and constructing a logistics domain-specific corpus. Fine-tune pre-trained models such as BERT to train a "transportation mode classification model" that supports multi-label classification (such as identifying "air-to-air" and "direct flight" labels simultaneously).

[0099] Intelligent parsing and completion includes NLP parsing of free text fields in the original shipment (such as "notes" and "special requirements"), extracting implied transportation mode information (e.g., from "please do not transfer, direct to destination" to infer "direct flight mode"). For shipments with missing key fields (such as no explicit service mode), complete the missing key fields by context semantic inference (e.g., combine the distance between the sender and receiver addresses and the weight of the goods to determine whether it is a whole vehicle transportation).

[0100] Dynamic rule optimization includes cross-validation of classification results from the model and traditional rule matching results, automatic labeling of samples with classification accuracy below a threshold (e.g., 90%), and reverse optimization of the corpus, forming a closed loop of "data labeling-model training-rule updating".

[0101] In some embodiments, by constructing a distribution node relationship graph, using graph neural networks (GNN) to model the complex relationships between distribution centers, sender point departments, and geographic regions, and breaking through the limitations of traditional rule engines, dynamic attribution rules are automatically learned to adapt to changes in network structure (such as the activation of new distribution centers and adjustments to regional service ranges).

[0102] Distribution network graph construction includes node definition: distribution center (attributes: coordinates, service range, processing capacity), sender point department (attributes: coordinates, historical distribution preference), and geographic region (province / city / district). Edge definition: service relationship between distribution centers and sender point departments (historical attribution records), trunk transportation connection between distribution centers, and coverage relationship between geographic regions and distribution centers.

[0103] The GNN model training includes: input: sender point department coordinates, historical distribution node flow log (forming path sequence). Output: predict the optimal attribution distribution center of the shipping order corresponding to the sender point department (considering the current distribution center load, transportation distance, time limit constraint). Training data: the real attribution distribution center of the historical shipping order as the label, using graph convolution network (GCN) or graph attention network (GAT) to learn the dependence between nodes.

[0104] The real-time attribution matching includes: for a new shipping order, input the sender point department information, and output the attribution distribution center in real time through the trained GNN model, replacing the static matching of the traditional rule, which is especially suitable for dynamic adjustment of newly opened regions or abnormal distribution paths.

[0105] In some embodiments, by modeling the straightening strategy as a sequence decision problem, the priority ranking is dynamically optimized using reinforcement learning (RL), and the real-time transportation resource state (such as distribution center congestion, trunk vehicle empty rate), customer dynamic demand (such as time limit urgent order proportion) are comprehensively considered to realize the real-time optimal configuration of transportation resources.

[0106] The state space definition includes: environment state: distribution center processing load (current backlog / processing capacity), real-time transportation line rate of trunk line, customer historical straightening response rate (such as the complaint rate of a customer after straightening), available transportation tool capacity in the current period.

[0107] The action space definition includes: available actions: performing straightening operation on different priority groups of shipping orders (grouped according to customer type / weight threshold), adjusting straightening strategy parameters (such as temporarily increasing the weight threshold in a certain region to match the vehicle load).

[0108] The reward function design includes: positive reward: actual cost saving after straightening, time limit shortening time, distribution center processing efficiency improvement amplitude; negative reward: line conflict cost caused by straightening (such as penalty for temporary adjustment of line), reputation loss caused by customer complaints.

[0109] The strategy optimization trains the model by using deep reinforcement learning algorithm (such as DQN, PPO), inputs real-time state, and outputs optimal priority ranking strategy; regularly updates the reward function weight according to the historical straightening effect data (such as backtracking verification result) to adapt to the changes of business scenarios (such as higher time limit weight than cost during e-commerce promotion period).

[0110] In some embodiments, by introducing a spatio-temporal sequence model (such as LSTM+graph network), combining historical straightening data (distribution times, time limit, cost) and geographical space features (distribution center location, transportation line distance), the expected effect of different straightening strategies is predicted, solving the lag problem of traditional methods relying on historical average data.

[0111] The spatio-temporal feature engineering includes: time features: waybill creation time (hours / working days / quarters), historical same period straightening time limit data; spatial features: distance between the sending point and the home distribution center, trunk transportation time length distribution between the distribution center and the destination.

[0112] The prediction model construction includes: input: customer type, cargo weight, current distribution path (node sequence) of the waybill to be straightened; output: predicted reduction in distribution times after straightening, time limit shortening time length, cost saving amount; the model architecture adopts LSTM to process time series dependence, combines Graph Embedding to encode the distribution network structure, and constructs an end-to-end spatio-temporal prediction model.

[0113] The prediction result application includes: in step S103, the predicted "expected cost saving" is used to replace the traditional estimation based on historical average, to provide more accurate quantitative basis for priority ranking; in the backtracking verification of step S104, the prediction effect is compared with the actual effect, the prediction accuracy is continuously optimized through model updating (transfer learning), and an intelligent closed loop of "prediction-execution-feedback" is formed.

[0114] In some embodiments, by overcoming the limitations of traditional grouping relying on artificial preset dimensions, multi-dimensional clustering of transportation destinations, cargo weights, and time limit requirements is performed using clustering algorithms such as DBSCAN and K-means, to automatically discover the implied distribution flow patterns and generate dynamic and fine-grained cargo flow tables.

[0115] Multi-dimensional feature clustering includes: feature selection includes: destination latitude and longitude (spatial distance), cargo weight interval, customer required delivery time limit window, historical transportation route preference (such as preference for air transport). The clustering algorithm performs unsupervised clustering on the retained waybill data to be straightened, and identifies waybill groups with similar transportation characteristics (such as "short-distance heavy cargo group" and "long-distance time limit sensitive group").

[0116] Flow pattern mining analyzes the optimal distribution path for each clustering group: for example, the short-distance heavy cargo group is suitable for bypassing regional distribution centers and directly reaching through the trunk line; the long-distance time limit sensitive group is suitable for preferentially matching air transport nodes.

[0117] Generating dynamic flow rules includes: automatically adjusting the cargo allocation according to the clustering results (such as adding the rule of "time limit sensitive cargo preferentially allocated to direct flight route"), replacing the traditional fixed route configuration.

[0118] Dynamic updating of the flow table detects changes in flow patterns (such as changes in weight distribution caused by seasonal cargo) by periodically (such as every week) re-clustering based on the latest waybill data, automatically updates the cargo allocation cargo flow table, and adapts to business fluctuations.

[0119] In some embodiments, the problem of low efficiency of traditional manual investigation is solved by identifying inefficient or ineffective straightening operations from the backtracking verification data using anomaly detection algorithms such as Isolation Forest, One-Class SVM, etc., and accurately locating strategy vulnerabilities (such as load imbalance of distribution centers caused by excessive straightening).

[0120] Abnormal index definition includes: core indicators: the number of distribution after straightening does not decrease (abnormal type: ineffective straightening), time limit increases by more than threshold (such as +2 hours), cost increases (abnormal type: negative benefit straightening).

[0121] Abnormal detection model deployment includes: training data includes: actual effect data of historical straightening operations (distribution frequency change, time limit change rate, cost change rate), and known abnormal samples are marked (such as manually confirmed inefficient cases). The model selection combines supervised (such as random forest classification) and unsupervised (such as isolation forest) algorithms to detect never-before-seen abnormal patterns (such as a distribution center that explodes due to straightening).

[0122] Closed-loop optimization triggers automatically mark detected abnormal operations, associate distribution paths, customer types, etc., and generate optimization suggestions (such as restoring the original distribution path for the customer or adjusting the weight threshold); periodically summarize abnormal cases and optimize input parameters of the straightening strategy (such as correcting the geographical area division error in the attribution distribution rule).

[0123] In some embodiments, a scalable customer straightening identification algorithm model is built, which can: cover three major business scenarios (unstraightened by sender, unstraightened by attribution distribution, unstraightened by background planning); automatically determine whether the unstraightened freight volume in different distance intervals meets the business threshold; based on the path matching mechanism of multi-data source fusion, verify the consistency of the shipping path and the planning path; unified periodic statistical results are output according to the sender time and sender point department dimensions. The general pre-filtering module excludes shipments with "service mode = whole vehicle" transportation; excludes shipments with transportation mode = direct flight, air-to-air; unified data format and field mapping to ensure that subsequent modules make judgments under the same data. According to the "sender company + sender point department" dimension, the same recipient customer's shipments that meet any of the following conditions in a week are counted: sender point department - recipient point department distance <400km and unstraightened freight volume ≥6T; sender point department - recipient point department distance ∈ [400km, 1000km] and unstraightened freight volume ≥9T; sender point department - recipient point department distance ≥1000km and unstraightened freight volume ≥13T. Recipient customer consistency determination: ① the recipient address is exactly the same; ② or any of the recipient mobile phone / signature mobile phone / recipient company fields is exactly the same, and the recipient address is within 1km (grid code is the same). Unstraightened freight volume determination: the cumulative value of [site freight] unloading records exists.

[0124] After excluding the delivery of the delivery frequency ≤1, compare the unloading delivery of the delivery with the arrival delivery or the route belonging delivery of [distribution information]; through the line binding relationship of [route background] trunk type = point branch line / secondary branch line, identify the delivery node that should belong to; if there are multiple return point deliveries, select the delivery with the most "return point dispatch department" as the belonging; set the straightening freight volume threshold (6T, 9T, 13T) according to the distance interval to determine. Extract the straightening delivery that meets the conditions (customer type = mailing company or payment company, weight ≥4T) from the [distribution configuration-freight flow] table; calculate the straightening freight volume: match "flow direction final origin + associated final origin", and the final origin organization is consistent, weight ≥2T; if the straightening freight volume / the straightening freight volume ≥30% in any 3 days of the last 7 days, it is determined that the "background planning is not straightened"; the straightening freight volume is defined as: the first unloading site ≠ the final origin and ≠ the delivery organization.

[0125] Statistical time: mailing time; organization dimension: mailing point department; cycle: sliding one-week time window; output fields include: problem type, straightening days, average weight per day, straightening freight volume, analyst information, etc.; support association with analysis records to form a closed-loop management.

[0126] Compared with the existing path analysis method, the present application has the following significant advantages: multi-scene coverage: the unified model framework can simultaneously identify three main "straightening" scenes, reducing the maintenance cost caused by scattered business rules. Fine threshold control: set the freight volume threshold according to different distance intervals, the determination standard is scientific and extensible. Multi-data source fusion: automatically integrate multiple system data such as delivery, route, site, distribution configuration, etc., to realize cross-system path consistency verification. Periodic statistics and attribution: unified according to the mailing time and mailing point department dimension for periodic statistics, facilitating performance evaluation and operation analysis. Scalability and maintainability: modular design can flexibly add / adjust business rules to meet the needs of different business stages.

[0127] In summary, the customer straightening algorithm model of the present application can greatly improve the path abnormality recognition efficiency and accuracy, providing strong data support for logistics enterprises to optimize transportation paths, reduce operating costs and improve customer satisfaction, and has high industrial application value and popularization prospect.

[0128] Please refer to Figure 2 as shown, Figure 2FIG. 1 is a structural schematic diagram of a customer waybill straightening device 200 provided by an embodiment of the present application. The customer waybill straightening device 200 is used to execute the steps of the customer waybill straightening method shown in each of the embodiments described above. The customer waybill straightening device 200 can be a single server or a server cluster, or the customer waybill straightening device 200 can be a terminal, which can be a handheld terminal, a notebook computer, a wearable device, a robot, or the like.

[0129] As shown in FIG. 1, the customer waybill straightening device 200 includes: Figure 2

[0130] a pre-filtering unit 201, configured to perform general pre-filtering on the obtained customer waybill data, exclude waybill data whose service mode is whole vehicle transportation, and exclude waybill data whose transportation mode is direct flight and air-to-air, and unify the data format and field mapping of the customer waybill data; and obtain, according to the customer waybill data, waybill data to be straightened according to the dimension of the sending company and the sending point.

[0131] a flow direction generating unit 202, configured to exclude, from the waybill data to be straightened, waybill data whose distribution flow frequency is less than or equal to 1, and obtain, for each of the waybill data to be straightened, attribution distribution information corresponding to the waybill data; and generate a distribution-configured goods flow direction table according to a plurality of the attribution distribution information.

[0132] an information obtaining unit 203, configured to extract, from the distribution-configured goods flow direction table, waybill data to be straightened whose customer type is a sending company or a payment company and whose weight is greater than or equal to a preset weight, and obtain, for the waybill data to be straightened, corresponding unstraightened goods quantity information and straightened goods quantity information.

[0133] a straightening completion unit 204, configured to generate analysis result information according to the waybill data to be straightened, the unstraightened goods quantity information, and the straightened goods quantity information, and complete straightening of the customer waybill data according to the analysis result information.

[0134] In some embodiments, the general pre-filtering on the obtained customer waybill data, the exclusion of waybill data whose service mode is whole vehicle transportation, and the exclusion of waybill data whose transportation mode is direct flight and air-to-air, and the unification of the data format and field mapping of the customer waybill data include: parsing a service mode field and a transportation mode field in the customer waybill data, matching a value of the service mode field with a preset whole vehicle transportation service identifier, and matching a value of the transportation mode field with a preset direct flight transportation mode identifier and a preset air-to-air transportation mode identifier; performing logical deletion or marking a filtering state for waybill data whose service mode matches the whole vehicle transportation identifier or whose transportation mode matches the direct flight or air-to-air identifier; and performing field standardization processing on unfiltered waybill data, including mapping data items with the same meaning but different field names in different customer systems to a unified standard field, converting unstructured data into a structured data format, and filling in default values for missing fields or marking abnormal data.​

[0135] In some embodiments, the excluding, from the to-be-straightened waybill data, waybill data with a number of distribution center nodes passed through less than or equal to 1, obtaining the corresponding home distribution center information of each of the to-be-straightened waybill data, comprises: for the to-be-straightened waybill data filtered according to the sender company and the sender point, extracting the distribution node transfer log of the waybill, and counting the number of distribution center nodes actually passed through by each waybill in the transportation process as the number of distribution center nodes passed through; retaining the waybill data with the number of distribution center nodes passed through greater than 1, and excluding the waybill data with the number of distribution center nodes passed through less than or equal to 1; for the retained waybill data, determining the home distribution center corresponding to the waybill as the home distribution center information according to the information of the last actually operated distribution center node in the distribution node transfer log and combining a preset distribution center home rule, wherein the distribution center home rule comprises home matching according to the geographical area, the operation subject or the network level of the distribution node.

[0136] In some embodiments, the obtaining the corresponding un-straightened freight volume information and the straightened freight volume information comprises: filtering, from the freight configuration freight flow table, the waybill data with the customer type field being the sender company or the payment company, and extracting the freight weight field in the waybill; comparing the freight weight with a preset weight threshold, and retaining the waybill data with the freight weight greater than or equal to the preset weight threshold; for the retained waybill data, counting the total freight weight of the waybills not marked as straightened to obtain the un-straightened freight volume information, and counting the total freight weight of all the waybills meeting the customer type and weight conditions to obtain the straightened freight volume information, wherein the preset weight threshold is preconfigured according to the straightening cost and efficiency parameters of the transportation network.

[0137] In some embodiments, the generating the freight configuration freight flow table according to the plurality of home distribution center information comprises: grouping the to-be-straightened waybill data with the same home distribution center information, and aggregating and counting the transportation destination, the transportation line and the transportation time limit of each group of waybill data; generating, according to the aggregation result corresponding to the aggregation and counting, a freight flow record with the home distribution center as the starting point and the transportation destination as the end point, wherein each flow record contains the number of waybills, the freight weight distribution and the configuration information of the commonly used transportation line type of the flow; and performing deduplication and standardization processing on the flow record to form a freight configuration freight flow table containing the distribution center code, the destination code, the transportation line rule and the freight type adaptation condition, wherein the freight configuration freight flow table is used to guide the freight sorting and the transportation line planning of the distribution center.

[0138] In some embodiments, the generating analysis result information according to the to-be-straightened shipping order data, the non-straightened freight volume information and the freight volume information that should be straightened comprises: establishing an analysis data model, associating the sender company information, the sender point department information and the home distribution center information in the to-be-straightened shipping order data with the non-straightened freight volume information and the freight volume information that should be straightened; calculating the proportion of the non-straightened freight volume to the freight volume that should be straightened as a straightening rate index, and analyzing the straightening rate differences of different sender companies, sender point departments or home distribution centers; combining the transportation cost data and the distribution center processing efficiency data to generate a freight volume straightening priority ranking in each dimension, forming analysis result information including straightening potential evaluation, cost-benefit analysis and operation suggestions, and the analysis result information is stored in the form of a structured data table or a visual report.

[0139] In some embodiments, the completing straightening of the customer shipping order data according to the analysis result information comprises: generating a straightening operation instruction for a high-priority sender company or sender point department based on the straightening priority ranking in the analysis result information, the operation instruction comprising adjusting the freight sorting rules of the distribution center to match the optimal line in the freight configuration freight flow table, and sending a configuration request for a straightened transportation line to a transportation scheduling system; marking the shipping order data that executes the straightening operation as straightened, and recording the execution time, the operation subject and the transportation line change information of the straightening operation; periodically verifying the straightened shipping order data, comparing the actual transportation time limit and transportation cost with the expected index in the analysis result information, and forming a straightening effect evaluation report for optimizing subsequent straightening strategies.

[0140] It should be noted that, for the convenience and brevity of description, the specific working processes of the customer shipping order straightening device and each module described above can refer to the corresponding processes in the customer shipping order straightening method embodiments described above, and will not be described here.

[0141] The customer shipping order straightening method described above can be implemented in the form of a computer program, which can run on the device as shown in Figure 2 .

[0142] Please refer to Figure 3 , Figure 3 is a structural schematic block diagram of a computer device provided by the embodiment of the present application. The computer device comprises a processor, a memory and a network interface connected through a device bus, wherein the memory can comprise a storage medium and an internal memory.

[0143] The storage medium can store an operating device and a computer program. The computer program comprises program instructions, which, when executed, can cause the processor to execute any kind of customer shipping order straightening method.

[0144] The processor is configured to provide computing and control capabilities to support the operation of the entire computer device.

[0145] The internal memory provides an environment for the operation of a computer program in a non-volatile storage medium, which, when executed by the processor, can cause the processor to perform any one of the customer waybill straightening methods.

[0146] The network interface is configured to perform network communication, such as sending assigned tasks, etc. Figure 3 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the terminal to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0147] It should be understood that the processor can be a central processing unit (CPU), and the processor can also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.

[0148] In one embodiment, the processor is configured to run a computer program stored in the memory to perform the following steps:

[0149] The obtained customer waybill data is subjected to general pre-filtering to exclude waybills whose service mode is whole vehicle transportation and whose transportation mode is direct flight and air-to-air, and to unify the data format and field mapping of the customer waybill data; the customer waybill data is classified according to the dimensions of the sending company and the sending point, and the data of the waybills to be straightened is obtained;

[0150] In the waybill data to be straightened, the waybill data whose distribution flow frequency is less than or equal to 1 is excluded, and the corresponding attribution distribution information of each waybill data to be straightened is obtained; and a distribution and configuration of goods flow direction table is generated according to the attribution distribution information.

[0151] In the distribution and configuration of goods flow direction table, the waybill data to be straightened whose customer type is a sending company or a payment company and whose weight is greater than or equal to a preset weight is extracted, and the corresponding non-straightened goods volume information and the straightened goods volume information are obtained.

[0152] According to the to-be-straightened waybill data, the non-straightened freight volume information and the should-straightened freight volume information, analysis result information is generated, and the straightening of the customer waybill data is completed according to the analysis result information.

[0153] In some embodiments, the obtained customer waybill data is subjected to general pre-filtering, waybills with a service mode of full truckload transportation are excluded, and waybills with a transportation mode of direct flight and air-to-air are excluded, the data format and field mapping of the unified customer waybill data are performed, including: analyzing the service mode field and the transportation mode field in the customer waybill data, matching the value of the service mode field with the preset full truckload transportation service identifier, and matching the value of the transportation mode field with the preset direct flight transportation mode identifier and air-to-air transportation mode identifier; for waybill data matching the full truckload transportation identifier of the service mode or matching the direct flight or air-to-air identifier of the transportation mode, logical deletion or marked filtering state is performed; field standardization processing is performed on the unfiltered waybill data, including mapping data items with the same meaning but different field names in different customer systems to a unified standard field, converting unstructured data to a structured data format, and filling in default values or marking abnormal data for missing fields.

[0154] In some embodiments, the to-be-straightened waybill data is excluded from the distribution flow frequency less than or equal to 1, the corresponding attribution distribution information of each to-be-straightened waybill data is obtained, including: for the to-be-straightened waybill data filtered according to the sender company and sender point, extracting the distribution node flow log of the waybill, and counting the number of distribution center nodes actually passed by each waybill in the transportation process as the distribution flow frequency; waybill data with a distribution flow frequency greater than 1 is retained, and waybill data with a distribution flow frequency less than or equal to 1 is removed; for the retained waybill data, according to the last actual operation distribution center node information in the distribution node flow log, combined with the preset distribution center attribution rule, the attribution distribution center corresponding to the waybill is determined as the attribution distribution information, and the attribution rule includes attribution matching according to the geographical area, operation subject or network level of the distribution node.

[0155] In some embodiments, the corresponding non-straightened freight volume information and should-straightened freight volume information are obtained, including: filtering waybill data with a customer type field of sender company or payment company from the freight distribution configuration table, and extracting the freight weight field in the waybill; comparing the freight weight with a preset weight threshold, and retaining waybill data with a freight weight greater than or equal to the preset weight threshold; for the retained waybill data, the total weight of the freight of the waybills not marked as straightened is counted as the non-straightened freight volume information, and the total weight of the freight of all waybills meeting the customer type and weight conditions is counted as the should-straightened freight volume information, and the preset weight threshold is pre-configured according to the straightening cost and efficiency parameters of the transportation network.

[0156] In some embodiments, the generating the distribution configuration freight flow table according to the plurality of home distribution information comprises: grouping the to-be-straightened shipping order data with the same home distribution information, and aggregating and counting the transportation destination, transportation line and transportation time limit requirement of each group of shipping order data; generating a freight flow record starting from the home distribution center and ending at the transportation destination according to the aggregation result corresponding to the aggregation and counting, each freight flow record containing the number of shipping orders, the weight distribution of the freight and the configuration information of the commonly used transportation line type of the flow; and performing deduplication and standardization processing on the freight flow record to form a distribution configuration freight flow table containing the distribution center code, the destination code, the transportation line rule and the freight type adaptation condition, the distribution configuration freight flow table being used to guide the freight sorting and transportation line planning of the distribution center.

[0157] In some embodiments, the generating the analysis result information according to the to-be-straightened shipping order data, the non-straightened freight volume information and the should-straightened freight volume information comprises: establishing an analysis data model, and associating the sender company information, the sender point department information and the home distribution information in the to-be-straightened shipping order data with the non-straightened freight volume information and the should-straightened freight volume information; calculating the ratio of the non-straightened freight volume to the should-straightened freight volume as a straightening rate index, and analyzing the straightening rate difference of different sender companies, sender point departments or home distribution centers; combining the transportation cost data and the distribution center processing efficiency data to generate a freight volume straightening priority ranking in each dimension, and forming the analysis result information containing the straightening potential evaluation, the cost-benefit analysis and the operation suggestion, the analysis result information being stored in the form of a structured data table or a visual report.

[0158] In some embodiments, the completing the straightening of the customer shipping order data according to the analysis result information comprises: generating a straightening operation instruction for a high-priority sender company or sender point department based on the straightening priority ranking in the analysis result information, the operation instruction including adjusting the freight sorting rule of the distribution center to match the optimal line in the distribution configuration freight flow table and sending a configuration request of a straightened transportation line to a transportation scheduling system; marking the shipping order data that executes the straightening operation as straightened, and recording the execution time, the operation subject and the transportation line change information of the straightening operation; and periodically performing backtracking verification on the straightened shipping order data, comparing the actual transportation time limit and transportation cost with the expected index in the analysis result information, and forming a straightening effect evaluation report for optimizing the subsequent straightening strategy.

[0159] In some embodiments, the application further provides a computer readable storage medium storing a computer program, the computer program comprising program instructions, and the processor executes the program instructions to implement the steps of the customer shipping order straightening method provided in the above embodiments.

[0160] The computer readable storage medium can be an internal storage unit of the computer device, such as a hard disk or a memory of the computer device. The computer readable storage medium can also be an external storage device of the computer device, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, and the like.

[0161] The above merely provides the specific implementation of the present application, but the protection scope of the present application is not limited to this. Any person skilled in the art can easily think of various equivalent modifications or replacements within the technical range disclosed by the present application, and these modifications or replacements shall be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A method for straightening customer waybills, characterized in that: include: Perform universal pre-filtering on the acquired customer waybill data, excluding waybills with full truckload service, direct flights, and air-to-air transport modes, and unify the data format and field mapping of the customer waybill data. The customer waybill data is then filtered by shipping company and shipping point to obtain the waybill data to be straightened. Excluding waybill data with a distribution flow number less than or equal to 1 from the waybill data to be straightened, obtaining the attribution distribution information corresponding to each of the waybill data to be straightened; generating a cargo distribution configuration cargo flow direction table based on the plurality of attribution distribution information; Extract the waybill data to be straightened from the cargo flow table of the cargo distribution configuration, whose customer type is a shipping company or a paying company and whose weight is greater than or equal to a preset weight, and obtain the corresponding information on the quantity of goods not straightened and the quantity of goods to be straightened; Analysis result information is generated based on the waybill data to be straightened, the information on the quantity of cargo that has not been straightened, and the information on the quantity of cargo that should be straightened, and the straightening of the customer waybill data is completed based on the analysis result information.

2. The method according to claim 1, characterized in that The obtained customer waybill data is subjected to general pre-filtering, excluding waybills with full truckload transportation as the service mode, and excluding waybills with direct flight and air-to-air transportation modes, and unifying the data format and field mapping of customer waybill data, including: Parse the service mode field and transportation mode field in the customer's waybill data, match the value of the service mode field with the preset full truckload transport service identifier, and match the value of the transportation mode field with the preset direct flight transport mode identifier and air-to-air transport mode identifier; Logically delete or mark the waybill data whose service mode matches the full truckload transport identifier or whose transport mode matches the direct flight or air-to-air identifier; Perform field standardization on unfiltered waybill data, including mapping data items with the same meaning but different field names in different customer systems to unified standard fields, converting unstructured data into structured data format, and filling missing fields with default values ​​or marking abnormal data.

3. The method according to claim 1, characterized in that The step of excluding the waybill data having a distribution flow number less than or equal to 1 from the waybill data to be straightened, and obtaining the attribution distribution information corresponding to each of the waybill data to be straightened includes: For the waybill data to be straightened, which is filtered by the shipping company and shipping point dimensions, extract the waybill's distribution node flow logs and count the number of distribution center nodes each waybill actually passes through during the transportation process as the distribution flow count; The waybill data with the number of distribution passes greater than 1 will be retained, and the waybill data with the number of distribution passes less than or equal to 1 will be discarded; For the retained waybill data, the attribution distribution center corresponding to the waybill is determined as the attribution distribution information based on the last actual operation distribution center node information in the distribution node flow log and the preset distribution center attribution rules. The attribution distribution center attribution rules include attribution matching based on the geographical area, operating entity or network level of the distribution node.

4. The method according to claim 1, wherein The obtaining of the corresponding information on the quantity of goods that have not been straightened and the quantity of goods that should be straightened includes: Filter the waybill data whose customer type field is the shipping company or the paying company from the cargo flow table of the cargo distribution configuration, and extract the cargo weight field in the waybill; The cargo weight is compared with a preset weight threshold, and the waybill data with cargo weight greater than or equal to the preset weight threshold is retained; for the retained waybill data, the total cargo weight of the waybills that are not marked as straightened is counted as the unstraightened cargo volume information, and the total cargo weight of all waybills that meet the customer type and weight conditions is counted as the straightened cargo volume information. The preset weight threshold is pre-configured according to the straightening cost and efficiency parameters of the transportation network.

5. The method according to claim 1, wherein The generating of a cargo distribution configuration cargo flow table according to the plurality of attribution allocation information includes: Group the waybill data to be straightened with the same attribution and distribution information, and aggregate statistics on the transportation destination, transportation route, and transportation time requirements of each group of waybill data; Generate cargo flow records based on the corresponding aggregation results of the aggregated statistics, starting from the distribution center and ending at the transportation destination. Each flow record contains the number of waybills in that direction, cargo weight distribution, and configuration information of common transportation route types; The flow records are deduplicated and standardized to form a cargo flow table containing the distribution center code, destination code, transportation route rules, and cargo type adaptation conditions. The cargo flow table is used to guide the cargo sorting and transportation route planning of the distribution center.

6. The method according to claim 1, characterized in that The generating of analysis result information based on the to-be-straightened waybill data, the unstraightened cargo quantity information, and the to-be-straightened cargo quantity information includes: Establish an analytical data model to associate the shipping company information, shipping point information, and allocation information in the waybill data to be straightened with the information on the volume of goods that have not been straightened and the volume of goods that should be straightened; The ratio of unstraightened cargo to the cargo that should be straightened is calculated as the straightening rate indicator, and the differences in straightening rates among different shipping companies, shipping points or affiliated distribution centers are analyzed. Combined with transportation cost data and distribution center processing efficiency data, a priority ranking of cargo straightening under each dimension is generated, forming analysis result information including straightening potential assessment, cost-benefit analysis, and operational recommendations. The analysis result information is stored in the form of structured data tables or visual reports.

7. The method according to claim 6, characterized in that The straightening of the customer waybill data according to the analysis result information includes: Based on the straightening priority ranking in the analysis result information, a straightening operation instruction is generated for the high-priority shipping companies or shipping points. The operation instruction includes adjusting the cargo sorting rules of the distribution center to match the optimal route in the cargo distribution configuration cargo flow table and sending a configuration request for the straightening transportation route to the transportation scheduling system; The waybill data that has been straightened is marked as straightened, and the execution time of the straightening operation, the operating entity, and the transportation route change information are recorded; Regularly conduct retrospective verification on the straightened waybill data, compare the actual transportation time and transportation cost with the expected indicators in the analysis result information, and form a straightening effect evaluation report to optimize subsequent straightening strategies.

8. A customer waybill straightening device, characterized in that: include: The pre-filtering unit is used to perform general pre-filtering on the acquired customer waybill data, excluding waybills with full truckload service mode and non-stop and air-to-air transport modes, unifying the data format and field mapping of the customer waybill data; and obtaining the waybill data to be straightened based on the dimensions of the shipping company and the shipping point of the customer waybill data; A flow direction generating unit is used to exclude the waybill data with a distribution flow number less than or equal to 1 from the waybill data to be straightened, and obtain the attribution distribution information corresponding to each of the waybill data to be straightened; Generate a cargo flow table for cargo distribution configuration according to the plurality of attribution allocation information; An information acquisition unit is configured to extract, from the cargo flow table of the cargo distribution configuration, the waybill data to be straightened, whose customer type is a shipping company or a paying company and whose weight is greater than or equal to a preset weight, and obtain the corresponding information on the quantity of goods not straightened and the quantity of goods to be straightened; The straightening completion unit is used to generate analysis result information based on the waybill data to be straightened, the information on the quantity of cargo that has not been straightened, and the information on the quantity of cargo that should be straightened, and complete the straightening of the customer waybill data based on the analysis result information.

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