A method for enterprise-product relationship analysis based on material coding

Abstract

This invention discloses a method for analyzing enterprise-goods relationships based on material coding, applicable to the analysis of material flow between enterprises within a comprehensive bonded zone. It comprises three stages: data parsing, graph relationship construction, and graph relationship querying. Addressing the traditional reliance on relational databases for data querying in conventional relationship analysis methods, this invention supports data querying for various formats, including text and database types, and efficiently connects to multi-source heterogeneous file data from different upstream and downstream enterprises and systems. This invention stores all data in a graph database using point-edge relationships, solving the problem of relationship mining under massive datasets. Currently, due to changes in material coding relationships, customs officers do not support analyzing the material flow between upstream and downstream import / export enterprises; this invention provides a solution to support the analysis of previous material flow relationships between enterprises.

Description

Technical Field

[0001] This invention relates to the field of analysis of the flow of material codes among enterprises within a comprehensive bonded zone, and particularly to a method for analyzing enterprise-goods relationships based on material codes. Background Technology

[0002] Materials are a general term for all tangible objects that an enterprise purchases, manufactures, and sells, such as raw materials, blanks, parts, components, and products. Material codes are unique identifiers used within an enterprise to identify materials and their names, specifications, categories, and other information. Enterprises use material codes to replace textual records in managing the issuance, acceptance, requisition, follow-up, inventory, and storage of materials. According to customs regulations, the customs management system in mainland comprehensive bonded zones primarily establishes electronic ledgers based on the internal commodity material codes of import and export enterprises. It controls and analyzes relevant reports at the three stages of registration, customs declaration, and verification, thereby managing the enterprise's customs declaration activities, strengthening control over customs affairs, and improving the efficiency and management level of enterprise storage.

[0003] However, the import and export business within the comprehensive bonded zone involves numerous enterprises, with a wide variety of material codes exchanged between them. The principles governing material coding differ between enterprises, and complex material code conversion relationships exist between upstream and downstream enterprises in the supply chain. The same material may have multiple different codes across different upstream and downstream companies. Furthermore, the increasing diversification of enterprise products leads to an exponential growth in the number of materials for individual enterprises. Currently, the customs management system only supports functions such as reporting, recording, reviewing, and archiving enterprise business data. Customs officers lack effective analytical tools to perform relationship analysis and mining on massive amounts of data. Analyzing the flow relationships between import and export enterprises has always been a challenging problem for customs officers, requiring a technology application to analyze the relationships between production enterprises, logistics enterprises, and downstream enterprises corresponding to material codes. Summary of the Invention

[0004] The technical problem to be solved by this invention is to provide a general solution for visual analysis of the flow of goods at the material number level between upstream and downstream production and logistics enterprises in customs comprehensive bonded zones, and to provide a method for analyzing enterprise-goods relationships based on material codes.

[0005] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0006] A method for analyzing enterprise-goods relationships based on material coding includes three stages: data parsing, graph relationship construction, and graph relationship query. These three stages are progressively related.

[0007] (I) The data parsing stage is the data preparation stage, which includes data preparation, data exploration, data cleaning, data processing and data aggregation. It is used to uniformly access, clean and process all heterogeneous data uploaded and declared by all enterprises, and form enterprise, goods and material format data in a unified format for storage, so as to prepare the data in the required format for subsequent graph relationship construction.

[0008] (ii) In the graph relationship construction stage, the relationships between enterprises, goods and materials are constructed using enterprises, goods and materials as nodes. This includes the relationship between points between enterprises, the relationship between edges between enterprises and goods, and the relationship between edges between goods and materials. This is used to represent the graph relationship between enterprises, goods and materials with points and edges, and the graph relationship is stored in the graph database.

[0009] (III) Graph Relationship Query Step: Construct a graph relationship query service to set graph query relationship parameters and define the graph query return result format. Customs officers can use the graph relationship query service to query the flow of goods between upstream and downstream production and logistics enterprises in the comprehensive bonded zone with one click and display it visually.

[0010] As a further preferred embodiment of the enterprise-goods relationship analysis method based on material coding of the present invention, the data parsing step specifically includes the following steps;

[0011] Step 1.1: Unify the access to the original verification list headers, bonded list data, work order file data, and relational database structured data uploaded by enterprises, and store them as file data. Use offline MapReduce data cleaning algorithms to clean the data, and then transform and benchmark the data fields according to unified standards to generate standardized bonded list structured data.

[0012] Step 1.2: Screening key dimensions to determine the uniqueness of enterprise data. Since there are differences in data format, enterprise coding, and enterprise name during the cleaning process, a technical solution is designed to merge and cluster enterprise node data from multiple dimensions such as unified credit code and enterprise registration information. The results are merged using the unified credit code to ensure the uniqueness of the enterprise in the data.

[0013] Step 1.3: Based on the bonded list data and work order data generated in Step 1.1, merge the data. According to the conclusions of Step 1.2, using the ETPS_SCCD Enterprise Unified Credit Code as the unique key, perform standard cleaning and benchmarking to extract enterprise node data:

[0014] Step 1.4: Based on the bonded list structured data generated in Step 1.1, normalize the data with the data in the work order, clean and benchmark according to the standard, and extract the goods / materials node data.

[0015] As a further preferred embodiment of the enterprise-goods relationship analysis method based on material coding of the present invention, the graph relationship construction step is as follows:

[0016] Based on the structured bonded list data, enterprise node data, and goods / materials node data generated in the data parsing process, a graph relationship between enterprises, goods, and materials is constructed, including: the relationship between points between enterprises, the relationship between edges between enterprises and goods, and the relationship between edges between goods and materials.

[0017] The specific steps are as follows:

[0018] Step 2.1: Investigate the actual import and export business needs within the comprehensive bonded zone, define the three entities of enterprises, goods, and materials, and the relationships between these entities, including: the point relationships between enterprises, the edge relationships between enterprises and goods, and the edge relationships between goods and materials. Draw a point-edge attribute graph as shown, and use this attribute graph as a graph model to create a data structure.

[0019] Step 2.2: Based on the attribute graph generated in Step 2.1, complete the construction of the relationships between nodes, primary keys, attributes, etc. in the graph.

[0020] As a further preferred embodiment of the enterprise-goods relationship analysis method based on material coding of the present invention, the graph relationship query step is specifically as follows;

[0021] Using the graph data generated in the data parsing and graph relationship construction stages as the data to be queried, a graph relationship query process is designed to provide customs supervision users with a one-click query service to query the flow of goods and materials between upstream and downstream production and logistics enterprises in the comprehensive bonded zone during a certain period of time.

[0022] Specifically, it includes the following steps:

[0023] Step 3.1: Construct a standard process for a single enterprise's single goods transfer, as follows:

[0024] In all paths that conform to the above cargo flow process, calculate whether the imported or exported goods in the path carry the query keywords, thereby filtering out the target goods / materials that meet the conditions and are to be queried, adding query time as a filtering condition to filter the list, and finally obtaining the bonded list set result;

[0025] Step 3.2: Investigate the actual import and export cargo flow relationships of enterprises within the comprehensive bonded zone, and design flowcharts to calculate import and export transactions respectively;

[0026] Step 3.3: Denote the query results of inbound transactions generated in Step 3.2 as A, the query results of outbound transactions generated in Step 3.2 as B, and the query results of a single transaction within the zone generated in Step 3.1 as C, as follows:

[0027] R1 = A + C, where R1 represents all data from the moment the goods enter the zone until they are traded within the zone.

[0028] R2 = B + C, where R2 represents all data related to the circulation and transactions within the zone, tracing back from where the goods left the zone.

[0029] As a further preferred embodiment of the enterprise-goods relationship analysis method based on material coding of the present invention, in step 3.2, for import transactions: all import lists that meet the following conditions are selected: the starting enterprise is a list type that circulates within the region, the traded goods contain the query keywords, the traded material number contains the specified material number, and the list declaration time is within the specified time range. The goods associated with the selected import lists are used as the import transaction query results.

[0030] Export transactions: Filter all export lists whose starting enterprises meet the following conditions: the associated list is not a list circulating within the zone, the associated list number is empty, the transaction goods in the above lists contain the query keywords, the transaction material number contains the specified material number, and the list declaration time is within the specified time range. Use the associated goods in the filtered export lists as the export transaction query results.

[0031] As a further preferred embodiment of the enterprise-goods relationship analysis method based on material coding of the present invention, the detailed calculation rules in step 3.3 are as follows:

[0032] If no starting or ending companies are specified, then you only need to iterate from the companies entering the zone to the end, regardless of whether the ending company is a company leaving the zone.

[0033] There are starting companies and no ending companies. The process starts from the starting company and continues until the end, regardless of whether the ending company is a company that leaves the zone.

[0034] If there are terminating companies but no starting companies, determine if there is a line from the companies entering the zone to the terminating companies. If not, end the process; otherwise, traverse the line.

[0035] If there are starting and ending companies, determine if there is a line from the company entering the zone to the company ending the zone. If not, end the process; otherwise, traverse the line.

[0036] If there is a designated terminating enterprise, then take the intersection R1∩R2; otherwise, take R1+(R1∩B).

[0037] Compared with the prior art, the present invention, employing the above technical solution, has the following technical effects:

[0038] 1. This invention provides a material coding-based enterprise-goods relationship analysis method, which is a general solution for analyzing the flow of goods and materials between upstream and downstream production and logistics enterprises in customs comprehensive bonded zones. It mainly consists of three stages: data parsing, graph relationship construction, and graph relationship query. It provides a corresponding solution for traditional relationship analysis models that rely on relational databases for data query and analysis. It supports data querying of various formats such as text and database types, and supports efficient connection of multi-source heterogeneous file data from different upstream and downstream enterprises and systems.

[0039] 2. Traditional relationship construction methods rely on databases and support a limited amount of data. This invention stores all data in a graph database as point and edge relationships, solving the problem of relationship mining under massive data.

[0040] 3. Currently, due to changes in material coding relationships and other reasons, customs officers do not support the analysis of upstream and downstream import and export processes.

[0041] In the process of material flow between enterprises, this invention provides a solution to support the analysis of material flow relationships between enterprises. Attached Figure Description

[0042] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0043] Figure 1 This is a business process diagram of the data parsing stage of this invention;

[0044] Figure 2 This is a schematic diagram of the standardized bonded list structured data of this invention;

[0045] Figure 3 This is a diagram illustrating the principle of enterprise mergers in this invention.

[0046] Figure 4 This invention provides the standard for extracting enterprise node data, performing cleaning and benchmarking processing.

[0047] Figure 5 This invention provides a standard for extracting and cleaning / calibrating cargo / material node data.

[0048] Figure 6 This invention is for drawing point-edge attribute maps;

[0049] Figure 7 This is a flowchart illustrating the standard process for constructing a single-enterprise, single-transaction goods flow in accordance with the present invention.

[0050] Figure 8 This is a flowchart illustrating the process of obtaining the bonded list set results according to the present invention;

[0051] Figure 9 This is the query result of the transaction entering the zone generated by this invention;

[0052] Figure 10 This is the query result of outbound transactions generated by this invention. Detailed Implementation

[0053] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings:

[0054] The present invention will be further described below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only for explaining the present invention and are not intended to limit the present invention. Those skilled in the art should understand that various substitutions and modifications made to the present invention based on basic technical knowledge and means in the art without departing from the above-described technical concept should be included within the protection scope of the present invention.

[0055] This invention provides a general solution for analyzing the flow of goods and materials among upstream and downstream production and logistics enterprises in customs comprehensive bonded zones. The specific implementation of this invention mainly consists of three stages: data parsing, graph relationship construction, and graph relationship querying.

[0056] Step 1: Data Analysis

[0057] The data parsing stage mainly involves cleaning and processing the raw, multi-source, heterogeneous data to produce structured data. The business process is as follows: Figure 1 As shown, the specific implementation steps are as follows:

[0058] Step 1: Unify the access to the original verification list headers, bonded list data, work order file data, and relational database structured data uploaded by enterprises, and store them as file data. Use offline MapReduce data cleaning algorithms to clean the data, and then sort the data according to... Figure 2 Data fields are transformed and benchmarked using unified standards to ultimately generate standardized bonded list structured data.

[0059] The final data is stored as structured bonded manifest data in JSON format as follows:

[0060]

[0061] Step 2: Screening Key Dimensions to Determine the Uniqueness of Enterprise Data. Due to inconsistencies in data format, changes in enterprise coding, and differences in enterprise names during the data cleaning process, a technical solution was designed to merge and cluster enterprise node data from multiple dimensions, including the unified enterprise credit code and enterprise registration information. The results are then merged using the unified credit code to ensure the uniqueness of each enterprise within the data. The principle of enterprise merging is as follows. Figure 3 as follows:

[0062] Step 3: Merge the bonded list data and work order data generated in Step 1. Based on the conclusions of Step 2, use the ETPS_SCCD Enterprise Unified Credit Code as the unique key, and then... Figure 4 The standard is cleaned and benchmarked to extract enterprise node data:

[0063] The resulting JSON-type structured enterprise data after benchmarking:

[0064]

[0065] Step 4: Normalize the structured bonded manifest data generated in Step 1 and the data in the work order, according to... Figure 5 Standardized cleaning and benchmarking processes are performed to extract cargo / material node data:

[0066] The processed JSON data is as follows:

[0067]

[0068] Step Two: Graph Relationship Construction

[0069] Based on the structured bonded list data, enterprise node data, and goods / materials node data generated in the first step above, a graph relationship between enterprises, goods, and materials is constructed, including: the relationship between points between enterprises, the relationship between edges between enterprises and goods, and the relationship between edges between goods and materials.

[0070] Specifically, it consists of the following two implementation steps:

[0071] Step 1: Investigate the actual import and export business needs within the comprehensive bonded zone, define the three entities—enterprises, goods, and materials—and the relationships between them, including: the point relationships between enterprises, the edge relationships between enterprises and goods, and the edge relationships between goods and materials. Draw a point-edge attribute graph as follows: Figure 6 As shown, the data structure is ultimately created using this attribute graph as the graph model.

[0072] Step 2: Based on the attribute graph generated in Step 1, complete the construction of the relationships between nodes, primary keys, attributes, etc. in the graph:

[0073] Perform graph data structure creation and data writing:

[0074] Step 3: Graph Relationship Inquiry

[0075] Using the graph data generated in steps one and two as the data to be queried, a graph relationship query process is designed to provide customs supervision users with a one-click query service to query the flow of goods and materials between upstream and downstream production and logistics enterprises in the comprehensive bonded zone during a certain period.

[0076] Specifically, the implementation process consists of the following three steps:

[0077] Step 1: Establish a standard process for a single enterprise's single goods transfer, as follows: Figure 7 As shown:

[0078] In all paths that conform to the above cargo flow process, it is calculated whether the imported or exported goods in the path carry the query keywords, thereby filtering out the target goods / materials that meet the conditions and are to be queried. The query time is added as a filtering condition to filter the list, and finally the bonded list set result is obtained.

[0079] The detailed flowchart is as follows: Figure 8 As shown.

[0080] Step 2: Investigate the actual import and export cargo flow relationships of enterprises within the comprehensive bonded zone, and design flowcharts to calculate import and export transactions respectively.

[0081] 1. Import Transactions: Filter all import lists that meet the following conditions: the originating enterprise is a list type circulating within the zone, the traded goods contain the query keywords, the traded part number contains the specified part number, and the list declaration time is within the specified time range. Use the goods associated with the filtered import lists as the import transaction query results.

[0082] 2. Export Transactions: Select all export lists from the starting enterprise that meet the following conditions: the associated list is not circulated within the zone, the associated list number is empty, the transaction goods in the above lists contain the query keywords, the transaction material number contains the specified material number, and the list declaration time is within the specified time range. The associated goods in the selected export lists will be used as the export transaction query results.

[0083] As shown below Figure 9 As shown, the query results for transactions entering the zone are generated; see below. Figure 10 As shown, the query results for outbound transactions are generated.

[0084] Step 3: Record the query results of inbound transactions generated in Step 2 as A, the query results of outbound transactions generated in Step 2 as B, and the query results of single transactions within the zone generated in Step 1 as C.

[0085] R1 = A + C, where R1 represents all data from the moment the goods enter the zone until they are traded within the zone.

[0086] R2 = B + C, where R2 represents all data related to the circulation and transactions within the zone, tracing back from where the goods left the zone.

[0087] The detailed calculation rules are as follows:

[0088] If no starting or ending company is specified, then you only need to iterate from the companies entering the zone to the end, regardless of whether the ending company is a company leaving the zone.

[0089] There are starting companies but no ending companies. The process starts from the starting company and continues until the end, regardless of whether the ending company is a company that leaves the zone.

[0090] If there are terminating companies but no starting companies, determine if there is a line from the companies entering the zone to the terminating companies. If not, end the process; otherwise, traverse the line.

[0091] If there are starting and ending companies, determine if there is a line from the company entering the zone to the company ending. If not, end the process; otherwise, traverse the line.

[0092] In summary, if there are designated terminating companies, the intersection R1∩R2 is taken; otherwise, R1+(R1∩B) is taken (that is, the transaction data of companies in B that have already appeared in R1 are taken, so as to prevent isolated out-of-zone data).

[0093] It will be understood by those skilled in the art that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. It should also be understood that terms such as those defined in general dictionaries should be understood to have the same meaning as in the context of the prior art, and should not be interpreted in an idealized or overly formal sense unless defined as herein.

[0094] The above embodiments are merely illustrative of the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made to the technical solution based on the technical concept proposed in this invention shall fall within the scope of protection of this invention. The embodiments of the present invention have been described in detail above, but the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention.

Claims

1. A method for analyzing enterprise-product relationships based on material coding, characterized in that: It includes three stages: data parsing, graph relationship construction, and graph relationship querying, and these three stages are progressively related. (I) The data parsing stage is the data preparation stage, which includes data preparation, data exploration, data cleaning, data processing and data aggregation. It is used to uniformly access, clean and process all heterogeneous data uploaded and declared by all enterprises, and form enterprise, goods and material format data in a unified format for storage, so as to prepare the data in the required format for subsequent graph relationship construction. (ii) Graph relationship construction stage: The relationship between enterprises, goods and materials is constructed with enterprises, goods and materials as nodes. This includes the relationship between points between enterprises, the relationship between edges between enterprises and goods, and the relationship between edges between goods and materials. The graph relationship is used to represent the graph relationship between enterprises, goods and materials with points and edges, and the graph relationship is stored in the graph database. (III) Graph Relationship Query Step: Construct a graph relationship query service to set graph query relationship parameters and define the graph query return result format. Customs officers can use the graph relationship query service to query the flow of goods between upstream and downstream production and logistics enterprises in the comprehensive bonded zone with one click and display it visually. The graph relationship query process is detailed as follows; Using the graph data generated in the data parsing and graph relationship construction stages as the data to be queried, a graph relationship query process is designed to provide customs supervision users with a one-click query service to query the flow of goods and materials between upstream and downstream production and logistics enterprises in the comprehensive bonded zone during a certain period of time. Specifically, it includes the following steps: Step 3.1: Construct a standard process for a single enterprise's single goods transfer, as follows: In all paths that conform to the cargo flow process, calculate whether the imported or exported goods in the path carry the query keywords, thereby filtering out the target goods / materials that meet the conditions and are to be queried, adding query time as a filtering condition to filter the list, and finally obtaining the bonded list set result; Step 3.2: Investigate the actual import and export cargo flow relationships of enterprises within the comprehensive bonded zone, and design flowcharts to calculate import and export transactions respectively; Step 3.3: Denote the query results of inbound transactions generated in Step 3.2 as A, the query results of outbound transactions generated in Step 3.2 as B, and the query results of a single transaction within the zone generated in Step 3.1 as C, as follows: R1 = A + C, where R1 represents all data from the moment the goods enter the zone until they are traded within the zone. R2 = B + C, where R2 represents all data related to the circulation and transactions within the zone, tracing back from where the goods left the zone.

2. The method for enterprise-goods relationship analysis based on material coding according to claim 1, characterized in that: The data parsing process specifically includes the following steps; Step 1.1: Unify the access to the original verification list headers, bonded list data, work order file data, and relational database structured data uploaded by enterprises, and store them as file data. Use offline MapReduce data cleaning algorithms to clean the data, and then transform and benchmark the data fields according to unified standards to generate standardized bonded list structured data. Step 1.2: Screening key dimensions to determine the uniqueness of enterprise data. Since there are differences in data format, enterprise code changes and enterprise name differences during the cleaning process of enterprise node data, a technical solution is designed to merge and cluster enterprise node data from multiple dimensions such as unified credit code and enterprise registration information. The results are merged using the unified credit code to ensure the uniqueness of enterprises in the data. Step 1.3: Based on the bonded list data and work order data generated in Step 1.1, merge the data. According to the conclusions of Step 1.2, using the ETPS_SCCD Enterprise Unified Credit Code as the unique key, perform standard cleaning and benchmarking to extract enterprise node data: Step 1.4: Based on the bonded list structured data generated in Step 1.1, normalize the data with the data in the work order, clean and benchmark according to the standard, and extract the goods / materials node data.

3. The enterprise-goods relationship analysis method based on material coding according to claim 1, characterized in that: The specific steps for constructing the graph relationship are as follows: Based on the structured bonded list data, enterprise node data, and goods / materials node data generated in the data parsing process, a graph relationship between enterprises, goods, and materials is constructed, including: the relationship between points between enterprises, the relationship between edges between enterprises and goods, and the relationship between edges between goods and materials. The specific steps are as follows: Step 2.1: Investigate the actual import and export business needs within the comprehensive bonded zone, define the three entities of enterprises, goods, and materials, and the relationships between these entities, including: the point relationships between enterprises, the edge relationships between enterprises and goods, and the edge relationships between goods and materials. Draw a point-edge attribute graph and use this attribute graph as a graph model to create a data structure. Step 2.2: Based on the attribute graph generated in Step 2.1, complete the construction of nodes, primary keys, and attribute relationships in the graph relationship.

4. The enterprise-goods relationship analysis method based on material coding according to claim 1, characterized in that: In step 3.2, for import transactions: filter all import lists that are of the type of list circulating within the zone, contain the query keywords in the traded goods, contain the specified material number in the traded material number, and are declared within the specified time range. Use the goods associated with the selected import lists as the import transaction query results. exit Transaction: Filter all export lists that meet the following conditions: the originating enterprise is not a list circulating within the zone, the list number is empty, the transaction goods in the list contain the query keywords, the transaction part number contains the specified part number, and the list declaration time is within the specified time range. Use the associated goods in the filtered export lists as the export transaction query results.

5. The enterprise-goods relationship analysis method based on material coding according to claim 1, characterized in that: In step 3.3, the detailed calculation rules are as follows: If no starting or ending companies are specified, then you only need to iterate from the companies entering the zone to the end, regardless of whether the ending company is a company leaving the zone. There are starting companies and no ending companies. The process starts from the starting company and continues until the end, regardless of whether the ending company is a company that leaves the zone. If there are terminating companies but no starting companies, determine if there is a line from the companies entering the zone to the terminating companies. If not, end the process; otherwise, traverse the line. If there are starting and ending companies, determine if there is a line from the company entering the zone to the company ending the zone. If not, end the process; otherwise, traverse the line. If there is a designated terminating enterprise, then take the intersection R1∩R2; otherwise, take R1+(R1∩B).

Images