Operation and maintenance processing method and system applied to supply chain financial platform

Abstract

The invention provides an operation and maintenance processing method and system applied to a supply chain financial platform, and relates to the technical field of supply chain finance, in particular to an operation and maintenance processing method applied to the supply chain financial platform. The system comprises a transaction subject business dependency recording unit, a transaction subject circulation dependency recording unit and a right voucher association dependency recording unit. A transaction link dependency graph is generated based on the set, nodes in the graph correspond to transaction subjects, transaction objects or right and interest vouchers, and connection edges represent dependency relationship types and association frequencies; and then a dependency attenuation coefficient is calculated according to a connection edge dependency association frequency change trend, and an abnormal dependency node is positioned. And aiming at the operation and maintenance strategy adapted to the abnormally dependent node, generating an operation and maintenance execution instruction comprising a node processing flow, a dependency relationship adjustment scheme and a strategy execution time sequence, and sending the operation and maintenance execution instruction to the operation and maintenance terminal, thereby realizing efficient operation and maintenance processing of the supply chain financial platform.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of supply chain finance, in particular to a method and system for operation and maintenance of a supply chain finance platform. BACKGROUND

[0002] In the development process of the supply chain finance platform, its operation and maintenance faces many complex and critical challenges. The operation and maintenance of the traditional supply chain finance platform often focuses on the monitoring and management of a single transaction link or a specific business function. However, supply chain finance business has high complexity and correlation, with numerous transaction subjects, diverse transaction objects, frequent transfer of rights and interests certificates, and close and complex dependencies between links.

[0003] Existing operation and maintenance technologies mostly fail to fully consider these complex dependencies, usually only treating each transaction node and business operation in isolation, lacking in-depth analysis and comprehensive understanding of the overall dependency of the transaction link. This leads to difficulties in quickly and accurately locating the root cause of problems when problems occur in the transaction link, failing to timely detect potential abnormal dependencies, and thus failing to timely take effective operation and maintenance strategies for intervention and processing. This not only affects the stable operation of the supply chain finance platform, but also may cause economic losses to the parties involved in the transaction, and restricts the healthy development of supply chain finance business. SUMMARY

[0004] In view of the above-mentioned problems, in combination with the first aspect of the present application, the embodiments of the present application provide a method for operation and maintenance of a supply chain finance platform, comprising: obtaining a transaction link dependency information set of the supply chain finance platform, the transaction link dependency information set comprising a business dependency record unit between transaction subjects, a transfer dependency record unit of transaction objects, and an association dependency record unit of rights and interests certificates; generating a transaction link dependency graph based on the transaction link dependency information set, wherein the nodes in the transaction link dependency graph correspond to transaction subjects, transaction objects or rights and interests certificates, and the connection edges between the nodes represent dependency relationship types and dependency association frequencies; calculating dependency decay coefficients corresponding to each connection edge according to the dependency association frequency change trend of the connection edge in the transaction link dependency graph, the dependency decay coefficients being used to reflect the weakening degree of the dependency relationship; locating abnormal dependency nodes in the transaction link based on the dependency decay coefficients, the abnormal dependency nodes being nodes whose dependency relationship weakening degree corresponding to the dependency decay coefficients exceeds the normal range; An operation and maintenance strategy is adapted to the abnormal dependency node, operation and maintenance execution instructions containing a node processing flow, a dependency relationship adjustment scheme and a strategy execution time sequence are generated, and the operation and maintenance execution instructions are sent to an operation and maintenance terminal of the supply chain financial platform.

[0005] In another aspect, the embodiment of the present application also provides an operation and maintenance processing system applied to a supply chain financial platform, comprising a processor, a machine readable storage medium, the machine readable storage medium is connected with the processor, the machine readable storage medium is used for storing programs, instructions or codes, and the processor is used for running the programs, instructions or codes in the machine readable storage medium to realize the above method.

[0006] Based on the above aspects, the embodiment of the present application obtains a transaction link dependency information set containing business dependency between transaction subjects, transaction target flow transfer dependency and right certificate association dependency, generates a transaction link dependency graph based on the transaction link dependency information set, presents the dependency relationship type and association frequency between the transaction subjects, the transaction targets and the right certificates in an intuitive graphical manner, calculates a dependency attenuation coefficient according to the dependency association frequency change trend of the connection edge, can reflect the weakening degree of the dependency relationship, locates the abnormal dependency node based on the dependency attenuation coefficient, can quickly and accurately find out the node with potential risks in the transaction link, and avoids problem expansion. An operation and maintenance strategy is adapted to the abnormal dependency node, and operation and maintenance execution instructions containing a node processing flow, a dependency relationship adjustment scheme and a strategy execution time sequence are generated, which realizes the standardization and normalization of operation and maintenance processing, improves operation and maintenance efficiency and accuracy. Finally, the operation and maintenance execution instructions are sent to an operation and maintenance terminal, which ensures that the operation and maintenance strategy can be executed in time and effectively, guarantees the stable operation of the supply chain financial platform, and improves the reliability and security of the entire supply chain financial business. BRIEF DESCRIPTION OF DRAWINGS

[0007] Figure 1 is an execution flow schematic diagram of the operation and maintenance processing method applied to the supply chain financial platform provided by the embodiment of the present application.

[0008] Figure 2 is a hardware architecture schematic diagram of the operation and maintenance processing system applied to the supply chain financial platform provided by the embodiment of the present application. DETAILED DESCRIPTION

[0009] The present application will be described in detail below with reference to the accompanying drawings, Figure 1 is a flow schematic diagram of the operation and maintenance processing method applied to the supply chain financial platform provided by an embodiment of the present application, and the operation and maintenance processing method applied to the supply chain financial platform will be described in detail below.

[0010] Step S110: Obtain the transaction link dependency information set of the supply chain finance platform. The transaction link dependency information set includes business dependency record units between transaction entities, transfer dependency record units of transaction targets, and association dependency record units of equity certificates.

[0011] This embodiment uses a supply chain finance scenario involving suppliers, manufacturers, distributors, and financial institutions as an example to illustrate the process of obtaining the transaction chain dependency information set. In this scenario, suppliers provide raw materials to manufacturers, manufacturers process the raw materials into finished products and sell them to distributors, and financial institutions provide financing and other financial services to each entity. The related transactions involve the generation and transfer of equity certificates. To obtain a complete transaction chain dependency information set, relevant records need to be extracted from multiple functional modules of the supply chain finance platform. First, the platform's business interaction module is accessed. This module stores various interaction data generated by each transaction entity during business collaboration, including but not limited to information on order initiation, confirmation, and fulfillment. Next, the target asset transfer module is accessed. This module records the entire transfer trajectory of the transaction target from the initial holder to the subsequent holder, such as the transfer of raw materials from suppliers to manufacturers and the transfer of finished products from manufacturers to distributors. Then, the certificate management module is accessed. This module saves records of the generation, association, and transfer of equity certificates, such as accounts receivable certificates generated based on orders and warehouse receipts generated based on the transfer of ownership of goods. By extracting information from these three modules, we can initially obtain the basic data that constitutes the transaction chain dependency information set. Subsequently, we will organize and associate these data to form complete business dependency record units, flow dependency record units, and association dependency record units.

[0012] Step S111: Access the business interaction module of the supply chain finance platform and extract the interaction records generated by the transaction entities during the business collaboration process. The interaction records include the identifier of the collaboration initiating entity, the identifier of the collaboration responding entity, and the collaboration business type.

[0013] In this scenario, the business interaction module stores records of raw material purchase order interactions between suppliers and manufacturers, finished product sales order interactions between manufacturers and distributors, and financing application and approval interactions between various entities and financial institutions. When extracting these interaction records, it's necessary to clearly identify the initiating entity, responding entity, and business type of each record. For example, in a supplier-initiated raw material purchase order interaction record, the initiating entity is the supplier's unique code, the responding entity is the manufacturer's unique code, and the business type is labeled as "raw material purchase." Similarly, in a manufacturer's financing application interaction record received by a financial institution, the initiating entity is the manufacturer's unique code, the responding entity is the financial institution's unique code, and the business type is labeled as "financing application." Using this method, all relevant interaction records can be extracted from the business interaction module.

[0014] Step S112: Select record segments with continuous collaborative relationships from the interaction records, and organize them into business dependency record units between transaction entities according to the time sequence of the collaboration. Each business dependency record unit corresponds to a group of transaction entities with a fixed collaboration mode.

[0015] From the extracted interaction records, record segments with continuous collaborative relationships are selected. For example, if a supplier and manufacturer have engaged in multiple raw material procurement transactions over a period of time, each transaction, from order initiation and confirmation to fulfillment, forms a complete interaction record. These consecutively occurring, similar interaction records constitute a record segment. These record segments are then organized chronologically to form business dependency record units between the trading entities. Each business dependency record unit includes the identifiers of the trading entities, the type of collaborative transaction, and a summary of each collaborative transaction in chronological order. For example, the business dependency record unit between a supplier and a manufacturer includes the supplier identifier, manufacturer identifier, the collaborative transaction type as raw material procurement, and summary information such as the order number, initiation time, and completion time of each purchase order, arranged chronologically. Through this organization, each business dependency record unit clearly reflects a fixed collaborative pattern and historical collaboration between a group of trading entities.

[0016] Step S113: Access the target transfer module of the supply chain finance platform and extract the transfer trajectory record of the target from the initial holder to the subsequent holder. The transfer trajectory record includes the target identifier, the transfer initiator identifier, the transfer recipient identifier, and the transfer completion time.

[0017] The asset transfer module stores transfer information for various transaction assets, including records of raw materials and finished products in this scenario. When extracting transfer trajectory records, it's necessary to obtain the asset identifier, initiating entity identifier, receiving entity identifier, and completion time for each record. For example, in a record of raw materials transferring from a supplier to a manufacturer, the asset identifier is the unique code for that batch of raw materials, the initiating entity identifier is the supplier's unique code, the receiving entity identifier is the manufacturer's unique code, and the completion time is the time the raw materials were delivered and accepted. Similarly, in a record of finished products transferring from a manufacturer to a distributor, the asset identifier is the unique code for that batch of finished products, the initiating entity identifier is the manufacturer's unique code, the receiving entity identifier is the distributor's unique code, and the completion time is the time the finished products were delivered and accepted. By extracting this information, a complete understanding of the transfer path and timelines of the transaction assets among the various entities can be obtained.

[0018] Step S114: Based on the flow trajectory records, sort out the transfer order of the transaction target between different entities, integrate the continuous flow records containing the same transaction target into the flow dependency record unit of the transaction target, and each flow dependency record unit corresponds to the complete flow path of a transaction target.

[0019] Based on the extracted circulation records, the transfer sequence of each transaction target between different entities is traced. For the same transaction target, there may be multiple transfers. For example, raw materials flow from the supplier to the manufacturer, are processed into finished products, and then the finished products flow from the manufacturer to the distributor. This constitutes a continuous circulation process of the raw material-related target. Continuous circulation records containing the same transaction target are integrated together to form a circulation dependency record unit for the transaction target. Each circulation dependency record unit includes the target identifier, the initiating entity identifier of each circulation in chronological order, the receiving entity identifier, and the circulation completion time. For example, in the circulation dependency record unit for a batch of raw materials, the first record is of the supplier transferring the raw materials to the manufacturer, followed by the record of the manufacturer transferring the processed finished products to the distributor. These records are arranged in chronological order, completely presenting the circulation path of the transaction target.

[0020] Step S115: Access the voucher management module of the supply chain finance platform and extract the association records between the equity voucher and the transaction entity and the transaction target. The association records include the voucher identifier, the associated entity identifier and the associated target identifier.

[0021] The voucher management module stores information on various equity vouchers, including accounts receivable vouchers and warehouse receipts in this scenario. When extracting related records, it's necessary to obtain the voucher identifier, related entity identifier, and related object identifier for each record. For example, for an accounts receivable voucher generated based on a raw material purchase order between a supplier and a manufacturer, the voucher identifier in its related record is the unique code of that accounts receivable voucher; the related entity identifiers are the unique codes of the supplier (payee) and the manufacturer (payer); and the related object identifier is the unique code of that batch of raw materials. Similarly, for a warehouse receipt generated based on finished products stored in a manufacturer's warehouse, the voucher identifier in its related record is the unique code of that warehouse receipt; the related entity identifiers are the unique codes of the manufacturer (owner) and the warehouse manager; and the related object identifier is the unique code of that batch of finished products. By extracting these related records, the correspondence between equity vouchers and the transaction entities and transaction objects is clarified.

[0022] Step S116: Associate the associated record with the corresponding business dependency record unit and flow dependency record unit to form an associated dependency record unit of the rights certificate. Each associated dependency record unit corresponds to a rights certificate and a set of matching relationships between business dependencies and flow dependencies.

[0023] The extracted related records are associated with their corresponding business dependency record units and circulation dependency record units. For example, the business dependency record unit corresponding to the related record of an accounts receivable voucher is the raw material procurement business dependency record unit between the supplier and the manufacturer, and the corresponding circulation dependency record unit is the circulation dependency record unit for the flow of that batch of raw materials from the supplier to the manufacturer. These relationships are integrated to form the related dependency record units for equity documents. Each related dependency record unit includes a voucher identifier, a corresponding business dependency record unit identifier, a corresponding circulation dependency record unit identifier, and a description of the correspondence between the related entity identifier and the related asset identifier. Through these associations, the equity document is closely linked to the business collaboration and asset circulation process that generated it, forming a complete related dependency record.

[0024] Step S117: Connect the business dependency record unit, the flow dependency record unit and the association dependency record unit in series according to the upstream and downstream collaboration order of the transaction link to form a transaction link dependency information set containing multiple association dependencies.

[0025] Following the upstream and downstream collaboration sequence of the transaction chain, business dependency recording units, circulation dependency recording units, and related dependency recording units are linked together. In this scenario, the transaction chain starts with the supplier, followed by the manufacturer, then the distributor, with financial institutions providing financial services at each stage. First, the business dependency recording unit between the supplier and the manufacturer is taken as the starting point. Next, the raw material circulation dependency recording unit corresponding to this business is linked, followed by the related dependency recording unit of equity certificates generated based on this business and circulation. Then, the business dependency recording unit between the manufacturer and the distributor, as well as the corresponding finished product circulation dependency recording unit and related equity certificate related dependency recording unit, are linked. For financing and other businesses involving financial institutions, their business dependency recording units and corresponding equity certificate related dependency recording units are inserted into the corresponding transaction stages. Through this linking method, a transaction chain dependency information set containing multiple related dependencies is formed, comprehensively reflecting the transaction chain dependency situation in the supply chain finance platform.

[0026] Step S1171: Identify the collaboration initiator and collaboration responder in each business dependency record unit, determine that the collaboration initiator is the upstream entity and the collaboration responder is the downstream entity, and establish the upstream and downstream relationship of business dependencies.

[0027] For each business dependency record unit, identify the initiating and responding entities. In the raw material procurement business dependency record unit between a supplier and a manufacturer, the initiating entity is the supplier, and the responding entity is the manufacturer. Therefore, the supplier is identified as the upstream entity, and the manufacturer as the downstream entity, establishing an upstream-downstream business dependency relationship between them. In the finished product sales business dependency record unit between a manufacturer and a distributor, the initiating entity is the manufacturer, and the responding entity is the distributor. The manufacturer is identified as the upstream entity, and the distributor as the downstream entity, establishing a corresponding upstream-downstream business dependency relationship. Through this method, the upstream and downstream positions of the entities in each business dependency record unit are clearly defined.

[0028] Step S1172: Identify the initiating entity and receiving entity in each flow dependency record unit, determine that the initiating entity is the upstream entity and the receiving entity is the downstream entity, and establish the upstream and downstream relationship of flow dependency.

[0029] In each flow dependency record unit, the initiating entity and the receiving entity are identified. For a flow dependency record unit where raw materials flow from a supplier to a manufacturer, the initiating entity is the supplier, and the receiving entity is the manufacturer. The supplier is identified as the upstream entity, and the manufacturer as the downstream entity, establishing the upstream and downstream relationship of the flow dependency. For a flow dependency record unit where finished products flow from a manufacturer to a distributor, the initiating entity is the manufacturer, and the receiving entity is the distributor. The manufacturer is identified as the upstream entity, and the distributor as the downstream entity, establishing the corresponding upstream and downstream relationship of the flow dependency. Through the above identification and determination, the upstream and downstream relationships of the entities involved in the flow of the transaction target are clarified.

[0030] Step S1173: Identify the association order between the associated subject and the equity certificate in each associated dependency record unit, determine whether business dependency or circulation dependency is generated first, and then form equity certificate associated dependency, and establish the upstream and downstream relationship between associated dependency and business dependency, circulation dependency.

[0031] Analyze the association order between the related entity and the equity document in each related dependency record unit. The creation of equity documents is typically based on existing business dependencies or liquidity dependencies. For example, accounts receivable documents are created after a business dependency occurs (a purchase order is established) and the underlying asset is transferred (raw material delivery). Therefore, determine that business dependencies or liquidity dependencies occur first, followed by equity document related dependencies, establishing a downstream relationship between related dependencies and business dependencies / liquidity dependencies. That is, business dependencies and liquidity dependencies are upstream, and related dependencies are downstream, forming an upstream-downstream relationship with their corresponding business dependencies and liquidity dependencies.

[0032] Step S1174: Starting from the upstream entity of the transaction link, associate the corresponding business dependency record unit, flow dependency record unit and associated dependency record unit in sequence according to the upstream and downstream relationship to form a dependency information chain for a single transaction link.

[0033] Starting with the upstream supplier in this scenario, the various record units are linked according to their upstream and downstream relationships. First, the business dependency record unit between the supplier and the manufacturer is linked. Next, the raw material flow dependency record unit corresponding to this business is linked. Then, the equity certificate dependency record unit generated based on this business and flow is linked. Afterward, starting with the manufacturer (as a downstream entity relative to the supplier) as a new upstream entity, the business dependency record unit between the manufacturer and the distributor is linked. Then, the finished product flow dependency record unit is linked, followed by the related equity certificate dependency record unit. Through this sequential linking, a dependency information chain is formed, starting from the supplier, passing through the manufacturer, and ending at the distributor.

[0034] Step S1175: Check if there are multiple parallel transaction link dependency information chains. If so, analyze the relationship between each information chain, merge the information chains with shared nodes, and form a dependency information network containing multiple branch links.

[0035] In supply chain finance platforms, multiple parallel transaction chains may exist, each with its own dependency information chain. For example, a supplier may simultaneously provide raw materials to multiple manufacturers, forming multiple parallel business dependency record units originating from that supplier, along with corresponding flow and association dependency record units. After examining these parallel information chains and analyzing their relationships, it's discovered that these chains share the supplier node. Merging these information chains with shared nodes—for example, merging the supplier's chain with manufacturer A, and the supplier's chain with manufacturer B—creates a dependency information network with the supplier as the shared node, containing multiple branch chains (pointing to manufacturer A and manufacturer B respectively). This merging provides a more comprehensive reflection of the overall dependency situation of the transaction chain.

[0036] Step S1176: Sequentially number the record units in the dependency information network to ensure that the position of each record unit is consistent with the actual collaboration order of the transaction link, and finally form a transaction link dependency information set containing multiple related dependencies.

[0037] All record units in the merged dependency information network are sequentially numbered. The numbering order follows the actual collaboration order of the transaction link; for example, the business dependency record unit that occurs first is numbered first, followed by the corresponding flow dependency record unit, and the associated dependency record unit is numbered after the flow dependency. For record units in parallel branch links, they are numbered sequentially after the shared node according to the branch order. For example, after the supplier, the business dependency record units between the supplier and manufacturer A are numbered first, followed by the corresponding flow and associated dependency record units; then the business dependency record units between the supplier and manufacturer B are numbered, along with their corresponding flow and associated dependency record units. Through this sequential numbering, the position of each record unit is ensured to be consistent with the actual collaboration order of the transaction link, ultimately forming a complete set of transaction link dependency information.

[0038] Step S120: Generate a transaction link dependency graph based on the transaction link dependency information set. In the transaction link dependency graph, nodes correspond to transaction entities, transaction targets, or equity certificates, and the connecting edges between nodes represent the dependency relationship type and dependency association frequency.

[0039] Based on the existing set of transaction dependency information, a transaction dependency graph is generated. First, the identifiers of all transaction entities, transaction targets, and equity certificates are extracted from the information set; these identifiers will serve as nodes in the graph. Then, based on the dependencies in each record unit, the connection edges between nodes are determined. The type of connection edge is determined according to the type of dependency relationship, categorized as business dependency, flow dependency, and association dependency. Simultaneously, the frequency of each dependency relationship in the historical records is counted and marked as the dependency association frequency on the corresponding connection edge. For example, if the business dependency between a supplier and a manufacturer has occurred multiple times in the information set, the dependency association frequency of the connection edge will be marked with that number of occurrences. Through this method, the transaction dependency information set is transformed into a visual transaction dependency graph, intuitively displaying the dependencies and association frequencies between each node.

[0040] Step S121: Extract the identifiers of all transaction entities, transaction targets, and equity certificates from the transaction link dependency information set, and create transaction entity nodes, transaction target nodes, and equity certificate nodes respectively, with each node assigned a unique identifier code.

[0041] From the transaction dependency information set, the identifiers of all transaction entities are extracted one by one, such as unique codes for suppliers, manufacturers, distributors, and financial institutions. A corresponding transaction entity node is created for each entity, and a unique identifier code is assigned to each node. This identifier code is related to but independent of the original identifier of the transaction entity, and is used to uniquely identify the node in the dependency graph. Similarly, the identifiers of all transaction objects are extracted, such as unique codes for raw materials and finished products, and transaction object nodes are created with unique identifier codes. The identifiers of all equity certificates are extracted, such as unique codes for accounts receivable certificates and warehouse receipts, and equity certificate nodes are created with unique identifier codes. Through the above methods, it is ensured that each node in the dependency graph has a unique identifier, facilitating subsequent connections and management.

[0042] Step S122: For the business dependency record unit, identify the transaction subject node corresponding to the collaboration initiator identifier and the collaboration responder identifier, draw a connection edge between the two transaction subject nodes, and label the type of the connection edge as business dependency. At the same time, count the number of times collaboration occurs in the business dependency record unit as the dependency association frequency and label it on the connection edge.

[0043] For each business dependency record unit, based on the collaboration initiator identifier and collaboration responder identifier, the corresponding node is found among the created transaction entity nodes. For example, in the supplier and manufacturer business dependency record unit, the transaction entity nodes corresponding to the supplier and the manufacturer are found, and a connection edge is drawn between these two nodes. The dependency relationship type is labeled as "business dependency" on the connection edge. Simultaneously, the total number of collaborations occurring in this business dependency record unit is counted, and this number is marked as the dependency association frequency on the side of the connection edge or at a specific location. For example, if there are several raw material procurement collaborations in this business dependency record unit, the dependency association frequency is marked as this number, allowing viewers to intuitively understand the frequency of the business dependency.

[0044] Step S123: For the flow dependency record unit, identify the transaction subject node corresponding to the flow initiator identifier, the flow receiver identifier, and the transaction target node corresponding to the transaction target identifier. Draw connection edges between the flow initiator node and the transaction target node, and between the transaction target node and the flow receiver node. The type of the connection edge is labeled as flow dependency. Count the number of times the same flow path occurs in the flow dependency record unit as the dependency association frequency and label it on the connection edge.

[0045] For each flow dependency record unit, first locate the corresponding transaction entity node based on the flow initiator's identifier, and then locate the corresponding transaction target node based on the transaction target's identifier. Draw a connection edge between the flow initiator node and the transaction target node. Next, locate the corresponding transaction entity node based on the flow receiver's identifier, and draw another connection edge between the transaction target node and the flow receiver node. Both of these connection edges are labeled as flow dependency. Count the number of occurrences of the same flow path (i.e., from the same flow initiator through the same transaction target to the same flow receiver) in this flow dependency record unit, and label this number as the dependency association frequency on both connection edges, or, depending on the actual situation, on the primary connection edge, to reflect the frequency of occurrence of this flow path.

[0046] Step S124: For the related dependency record unit, identify the transaction entity node corresponding to the related entity identifier, the transaction target node corresponding to the related target identifier, and the equity certificate node corresponding to the equity certificate identifier. Draw connection edges between the equity certificate node and the transaction entity node, and between the equity certificate node and the transaction target node. The type of the connection edge is labeled as related dependency. Count the number of times the same relationship occurs in the related dependency record unit as the dependency relationship frequency and label it on the connection edge.

[0047] In this embodiment, for a dependency record unit, such as the dependency record unit of an accounts receivable document, the associated entity identifier includes the unique codes of the supplier and manufacturer, the associated object identifier is the unique code of the batch of raw materials, and the equity certificate identifier is the unique code of the accounts receivable document. First, based on the unique codes of the supplier and manufacturer, the corresponding supplier node and manufacturer node are found in the created transaction entity nodes; based on the unique code of the raw materials, the corresponding transaction object node is found; and based on the unique code of the accounts receivable document, the corresponding equity certificate node is found. Then, a connection edge is drawn between the accounts receivable document node and the supplier node, another connection edge is drawn between the accounts receivable document node and the manufacturer node, and a connection edge is drawn between the accounts receivable document node and the raw material object node. All these connection edges are labeled as dependency. The number of times the same accounts receivable document is associated with the supplier, manufacturer, and raw material object in this dependency record unit is counted, and this number is used as the dependency frequency and labeled on the corresponding connection edges to reflect the frequency of the association.

[0048] Step S125: Adjust the layout of all nodes and connecting edges so that nodes of the same type are clustered together and nodes of different types are arranged in order of dependency relationship to form the transaction link dependency graph.

[0049] After completing the initial drawing of all nodes and connecting edges, their layout needs to be adjusted. First, the transaction entity nodes, transaction target nodes, and equity certificate nodes are categorized and clustered, ensuring that nodes of the same type are concentrated in specific areas of the graph, avoiding visual clutter caused by a mixed distribution of different types of nodes. Then, according to the order of dependencies, different types of nodes are arranged; for example, the transaction entity nodes corresponding to business dependencies are placed first, followed by the transaction target nodes corresponding to flow dependencies, and the equity certificate nodes corresponding to association dependencies are placed near the corresponding transaction entity and transaction target nodes. Through these layout adjustments, the structure of the entire transaction chain dependency graph is clear, and the flow of dependencies is explicit, facilitating subsequent analysis and viewing of dependencies.

[0050] Step S1251: Assign different visual identifiers to the transaction entity node, the transaction target node, and the equity certificate node, so that nodes of the same type have consistent visual characteristics, while nodes of different types have different visual characteristics.

[0051] To clearly distinguish different types of nodes in the transaction dependency graph, distinct visual identifiers are assigned to transaction entity nodes, transaction target nodes, and equity certificate nodes. For example, transaction entity nodes are set as circles with a blue fill color; transaction target nodes are set as squares with a green fill color; and equity certificate nodes are set as triangles with a red fill color. Furthermore, different border thicknesses are assigned to each type of node: thicker borders for transaction entity nodes, medium-thicker borders for transaction target nodes, and thinner borders for equity certificate nodes. These different visual features enable viewers to quickly identify node types, improving the efficiency of understanding the dependency graph.

[0052] Step S1252: Divide the graph canvas into three areas, which are used to place the transaction subject node, the transaction target node, and the equity certificate node, respectively. The three areas are arranged in the order of business dependency, circulation dependency, and association dependency.

[0053] On the graph canvas, three contiguous areas are divided from left to right. The first area on the left is dedicated to all transaction entity nodes, corresponding to the main area where business dependencies occur; the second area in the middle is used to place all transaction target nodes, corresponding to the area where circulation dependencies occur; and the third area on the right is used to place all equity certificate nodes, corresponding to the area where association dependencies occur. A certain distance is maintained between the three areas to clearly distinguish the distribution range of different types of nodes, and also to reserve sufficient space for drawing connecting edges between nodes, avoiding overly dense connecting edges at the boundaries of the areas.

[0054] Step S1253: Arrange the transaction entity nodes in the corresponding area according to the upstream and downstream cooperation order, with the upstream entity node located on the left side of the area and the downstream entity node located on the right side of the area.

[0055] Within the area to the left of the transaction entity's node, nodes are arranged according to the upstream and downstream collaboration order in the transaction chain. For example, in this scenario, the supplier, as the starting upstream entity, has its node placed on the far left of this area; the manufacturer, as the supplier's downstream entity, has its node placed to the right of the supplier's node; the distributor, as the manufacturer's downstream entity, has its node placed to the right of the manufacturer's node; and the financial institution, as a related entity providing services to each entity, has its node placed near the corresponding entity node based on its primary service recipients, or in the middle if it serves multiple entities. This arrangement visually demonstrates the upstream and downstream relationships between transaction entities, making the flow of business dependencies immediately clear.

[0056] Step S1254: Assign the target node to the right side of the main node in the order of transfer, so that the connection edge between the target node and the corresponding upstream and downstream main node is horizontal.

[0057] In the middle area where the transaction target nodes are located, they are arranged according to the flow order of the transaction targets. For raw material target nodes flowing from suppliers to manufacturers, they are placed in the corresponding positions to the right of the supplier node and the manufacturer node, so that the connection edges between raw material target nodes and supplier nodes, and between raw material target nodes and manufacturer nodes, are as horizontal as possible. For finished product target nodes flowing from manufacturers to distributors, they are placed in the corresponding positions to the right of the manufacturer node and the distributor node, also so that the relevant connection edges are horizontal. This arrangement makes the flow dependency path clearer and reduces the curvature and intersection of connection edges.

[0058] Step S1255: Place the equity certificate node in the area between the corresponding transaction subject node and the transaction target node, so that the length of the connecting edge between the equity certificate node and the associated node is similar.

[0059] Place the equity document node in the transition zone between the right-hand area and the middle and left-hand areas. For accounts receivable document nodes based on suppliers, manufacturers, and raw materials, place them in the area between the supplier node, manufacturer node, and raw material node, ensuring that the lengths of the connecting edges between the accounts receivable document node and the supplier node, manufacturer node, and raw material node are roughly similar to avoid any connecting edge being too long or too short and affecting the visual effect. Place other equity document nodes according to the same principle to clearly present the connection relationship between equity documents and related nodes.

[0060] Step S1256: Check whether there is any intersection or overlap of the connecting edges after the layout. If so, fine-tune the point positions until the intersection and overlap of the connecting edges are eliminated, forming a transaction link dependency graph.

[0061] After completing the initial layout, a comprehensive check of the connecting edges throughout the entire graph is conducted. If any intersections or overlaps are found between different nodes—for example, if a connecting edge dependent on a certain business intersects with a connecting edge dependent on a certain flow, or if two related dependent connecting edges overlap—the positions of the affected nodes are fine-tuned. During fine-tuning, the approximate area of ​​the nodes remains unchanged; only the horizontal or vertical positions of the nodes are slightly moved to gradually eliminate intersections and overlaps. After multiple fine-tunings, until all connecting edges are clearly displayed without significant intersections or overlaps, a transaction chain dependency graph with a clear structure and well-defined relationships is finally formed.

[0062] Step S130: Based on the frequency change trend of dependency associations of the connecting edges in the transaction link dependency graph, calculate the dependency decay coefficient corresponding to each connecting edge. The dependency decay coefficient is used to reflect the degree of weakening of the dependency relationship.

[0063] Based on the generated transaction dependency graph, the dependency decay coefficient of each connection edge is calculated. First, dependency frequency data for each connection edge is collected over a certain period, and the changing trends of this data are analyzed to determine whether the dependency relationship is weakening, strengthening, or stable. For connection edges showing a weakening trend, the dependency decay coefficient level is determined based on the duration and magnitude of the weakening; for connection edges showing a strengthening or stable trend, a basic dependency decay coefficient level is set. Through the above calculations, a corresponding dependency decay coefficient is assigned to each connection edge to quantify the degree of weakening of the dependency relationship.

[0064] Step S131: Extract the dependency association frequency data of each connection edge in the transaction link dependency graph within a preset time period, and organize it into a dependency association frequency sequence in chronological order.

[0065] The preset time period is set to the past six months, divided into six time intervals. For each connection edge in the transaction chain dependency graph, the monthly dependency frequency data is extracted from the historical records of the supply chain finance platform. For example, for the business dependency connection edge between suppliers and manufacturers, the number of raw material procurement collaborations per month in the past six months is extracted; for the flow dependency connection edge of raw materials from suppliers to manufacturers, the number of raw material flows per month is extracted; for the association dependency connection edge between accounts receivable vouchers and related nodes, the number of associations occurring per month is extracted. The extracted data is then organized into a dependency frequency sequence in chronological order (from the first month to the sixth month), with each sequence containing six data points, corresponding to the dependency frequency of each month.

[0066] Step S132: Analyze the changing direction of the dependency association frequency sequence. If the dependency association frequency in the later time period is lower than that in the previous time period, the dependency association frequency of the connection edge is determined to be a weakening trend. If the dependency association frequency in the later time period is higher than that in the previous time period, it is determined to be a strengthening trend. If the dependency association frequency remains stable in different time periods, it is determined to be a stable trend.

[0067] Analyze the dependency association frequency sequence of each connection edge segment by segment, and compare the dependency association frequencies in two adjacent time periods. For example, for the frequency sequence of a certain connection edge, the frequency in the first month is A1, and in the second month is A2. If A2 < A1, the change direction from the first month to the second month is downward; if A2 > A1, the change direction is upward; if A2 is basically equal to A1 (the difference is within a preset small range), the change direction is stable. By analyzing the change directions of all adjacent time periods in the entire sequence, if the change directions of most adjacent time periods are downward, it is determined that the change trend of the dependency association frequency of this connection edge is a weakening trend; if the change directions of most adjacent time periods are upward, it is determined as a strengthening trend; if the change directions of most adjacent time periods are stable, it is determined as a stable trend.

[0068] Step S133: For the connection edges with a weakening trend, further analyze the duration and amplitude of the weakening. The amplitude of the weakening is the degree of difference between the initial dependency association frequency and the current dependency association frequency.

[0069] For the connection edges determined to have a weakening trend, determine the time point when the weakening starts and the duration of the weakening. For example, if the frequency of a certain connection edge starts to decline from the third month and is still declining until the sixth month, the duration of the weakening is four months (from the third month to the sixth month). When calculating the amplitude of the weakening, use the initial dependency association frequency (the frequency in the first month) in the sequence as the benchmark, subtract the current frequency (the frequency in the sixth month) from the initial frequency, and then divide by the initial frequency. The resulting ratio is the amplitude of the weakening. For example, the initial frequency is B1, the current frequency is B6, and the amplitude of the weakening is (B1 - B6) / B1. The larger this ratio, the greater the amplitude of the weakening. Through the above analysis, clarify the duration and amplitude of the weakening for each connection edge with a weakening trend.

[0070] Step S134: Determine the level of the dependency attenuation coefficient according to the duration and amplitude of the weakening. The longer the duration of the weakening and the greater the amplitude of the weakening, the higher the level corresponding to the dependency attenuation coefficient.

[0071] Step S1341: Divide the duration of the weakening into multiple consecutive duration intervals, and each duration interval corresponds to a basic level. The longer the duration interval, the higher the basic level.

[0072] The duration of the weakening effect is divided into three consecutive duration intervals: the first interval is 1 to 2 months, corresponding to a base level of 1; the second interval is 3 to 4 months, corresponding to a base level of 2; and the third interval is 5 months or more, corresponding to a base level of 3. For example, if the weakening duration of a connecting edge is 3 months, it falls into the second interval, and its base level is 2; if the weakening duration of another connecting edge is 6 months, it falls into the third interval, and its base level is 3. Through this division, different weakening durations correspond to different base levels, and the longer the duration, the higher the base level.

[0073] Step S1342: Divide the weakening range into multiple consecutive range intervals. Each range interval corresponds to an adjustment level. The larger the range interval, the higher the adjustment level.

[0074] The weakening effect is divided into three consecutive ranges: the first range is 0% to 20%, corresponding to an adjustment level of 1; the second range is 21% to 50%, corresponding to an adjustment level of 2; and the third range is 51% and above, corresponding to an adjustment level of 3. For example, if the weakening effect of one connecting edge is 30%, it falls into the second range, corresponding to an adjustment level of 2; if the weakening effect of another connecting edge is 60%, it falls into the third range, corresponding to an adjustment level of 3. Thus, the larger the weakening effect, the higher the corresponding adjustment level.

[0075] Step S1343: Overlay the base level and the adjustment level to obtain the initial dependency decay coefficient level of each weakening trend connection edge.

[0076] For each weakening trend connecting edge, its corresponding base level and adjustment level are added together to obtain the initial dependency decay coefficient level. For example, if a connecting edge has a base level of 2 (weakening duration of 3 months) and an adjustment level of 2 (weakening magnitude of 30%), then the initial dependency decay coefficient level is 2 + 2 = 4; if another connecting edge has a base level of 3 (weakening duration of 6 months) and an adjustment level of 3 (weakening magnitude of 60%), then the initial dependency decay coefficient level is 3 + 3 = 6. This superposition method comprehensively considers the impact of both the duration and magnitude of weakening on the dependency decay coefficient level.

[0077] Step S1344: Refer to the dependency type corresponding to the connection edge. If the dependency type is a business dependency, increase the level by one based on the initial level; if it is a flow dependency, keep the initial level; if it is an association dependency, decrease the level by one based on the initial level.

[0078] The initial dependency decay coefficient level is adjusted based on the dependency type of the connection edge. For connection edges with a business dependency type, due to their core role in the transaction chain, the initial level is increased by one level; for example, if the initial level is 4, the adjusted level is 5. For connection edges with a flow dependency type, their importance is relatively stable, so the initial level remains unchanged; for example, if the initial level is 4, the adjusted level remains 4. For connection edges with an association dependency type, their importance is greatly affected by business and flow dependencies, and their own weakening effect is relatively small; therefore, the initial level is decreased by one level; for example, if the initial level is 4, the adjusted level is 3.

[0079] Step S1345: Check whether the adjusted dependency attenuation coefficient level conforms to the level range of similar historical weakening cases. If it exceeds the level range, make fine adjustments to ensure that the adjusted dependency attenuation coefficient level is within a reasonable range.

[0080] Retrieve the dependency decay coefficient levels of connection edges from historical data related to similar weakening scenarios (same dependency type, similar duration and magnitude of weakening) on ​​the supply chain finance platform. For example, if the level range for connection edges with similar business dependencies under similar weakening scenarios is 4 to 6, and the current adjusted level of a connection edge is 7, exceeding this range, then fine-tune it to 6; if the adjusted level is 3, below this range, then fine-tune it to 4. Through the above checks and fine-tuning, ensure that the adjusted dependency decay coefficient levels are consistent with historical data and remain within a reasonable range.

[0081] Step S1346: Associate and store the final determined dependency attenuation coefficient level with the connection edge identifier, dependency relationship type, weakening duration and weakening magnitude to form a complete attenuation coefficient list.

[0082] In addition to the edge identifier and the final dependency decay level, the decay coefficient list also includes information such as dependency type, weakening duration, and weakening magnitude. For example, a record for a certain edge might include the edge identifier "Business Dependency - Supplier - Manufacturer", dependency decay level 5, dependency type "Business Dependency", weakening duration 3 months, and weakening magnitude 30%. All this information for all edges is stored together to form a complete decay coefficient list, facilitating subsequent querying and analysis of the decay status of each edge.

[0083] Step S135: For the connecting edges whose changing trend is a strengthening trend or a stable trend, set a basic dependency decay coefficient. The level corresponding to the basic dependency decay coefficient is lower than the dependency decay coefficient level of the connecting edges with a weakening trend.

[0084] For connections exhibiting a strengthening trend, regardless of the degree of strengthening, the basic dependency decay coefficient is uniformly set to level one. Similarly, for connections exhibiting a stable trend, the basic dependency decay coefficient is also set to level one. This is because a strengthening trend indicates an increasing dependency, while a stable trend indicates that the dependency has not weakened. Their dependency decay is much lower than that of connections exhibiting a weakening trend, thus they are assigned a lower level to clearly distinguish them from connections exhibiting a weakening trend.

[0085] Step S136: Associate the dependency attenuation coefficient level corresponding to each connection edge with the connection edge identifier to form an attenuation coefficient list containing all connection edges and their corresponding dependency attenuation coefficients.

[0086] Each connection edge in the transaction dependency graph is assigned a unique identifier, which includes information such as node type and dependency relationship type. The dependency decay coefficient level of each connection edge calculated earlier is associated with its corresponding connection edge identifier. For example, a connection edge with the identifier "Business Dependency - Supplier - Manufacturer" has a dependency decay coefficient level of three; a connection edge with the identifier "Circulation Dependency - Raw Materials - Supplier - Manufacturer" has a dependency decay coefficient level of five, and so on. This association information is compiled into a list, where each record contains the connection edge identifier and its corresponding dependency decay coefficient level, forming a complete decay coefficient list.

[0087] Step S140: Locate abnormal dependency nodes in the transaction link based on the dependency decay coefficient. The abnormal dependency nodes are nodes whose dependency relationship weakening degree exceeds the normal range corresponding to the dependency decay coefficient.

[0088] Using the previously obtained list of attenuation coefficients, we first determine the normal range of the attenuation coefficients, based on the attenuation coefficient levels under historical normal operating conditions. Then, we filter out connection edges whose levels exceed the normal range from the list and mark them as anomalous connection edges. Next, we trace the nodes corresponding to these anomalous connection edges, analyze the number and type of anomalous connection edges associated with each node, and determine whether the node is a core anomalous dependency node or a regular anomalous dependency node. Finally, we count the number and distribution of these nodes, sort them by the scope of their impact, and form a list of anomalous dependency nodes, thus completing the location of the anomalous dependency nodes.

[0089] Step S141: Determine the normal range of the dependency attenuation coefficient, which is determined based on the dependency attenuation coefficient level of the connection edge in the transaction link under historical normal operation.

[0090] In this embodiment, the dependency decay coefficient level data of all connection edges during the past year when the supply chain finance platform was operating normally is retrieved. Statistical analysis is performed on this historical data to calculate the distribution of dependency decay coefficient levels for different dependency relationship types (business dependency, circulation dependency, and association dependency). For example, the distribution of business dependency connection edge levels in the historical data shows that most levels are concentrated in the range of 1 to 3; most circulation dependency connection edge levels are concentrated in the range of 1 to 4; and most association dependency connection edge levels are concentrated in the range of 1 to 2. The range in which 95% of the connection edge levels in the historical data fall under each dependency relationship type is determined as the normal range of dependency decay coefficients for that type of connection edge. For example, the normal range for business dependency connection edges is set to levels 1 to 3, the normal range for circulation dependency connection edges is set to levels 1 to 4, and the normal range for association dependency connection edges is set to levels 1 to 2. This method ensures that the determination of the normal range is data-supported and conforms to the actual situation under historical normal operation.

[0091] Step S142: Filter out the connection edges whose dependence on the attenuation coefficient level exceeds the normal range from the attenuation coefficient list and mark them as abnormal connection edges.

[0092] Each connection edge record in the attenuation coefficient list is examined one by one, and its corresponding normal range is found according to its dependency type. For example, if a connection edge's dependency type is business dependency, its dependency attenuation coefficient level is 5, while the normal range for business dependencies is 1 to 3, level 5 exceeds this range, so the connection edge is marked as an abnormal connection edge. For flow dependency connection edges, if its level is 6, exceeding the normal range of 1 to 4, it is also marked as an abnormal connection edge. For association dependency connection edges, if its level is 4, exceeding the normal range of 1 to 2, it is also marked as an abnormal connection edge. All connection edges exceeding their corresponding normal ranges are filtered out and uniformly marked as abnormal connection edges, forming an abnormal connection edge list.

[0093] Step S143: Track the two nodes corresponding to the abnormal connection edge, and record the positions of the two nodes in the transaction link dependency graph and their connection relationships with other nodes.

[0094] For each abnormal connection edge in the abnormal connection edge list, determine the two corresponding nodes based on its connection edge identifier. For example, if the abnormal connection edge identifier is "Business Dependency - Supplier - Manufacturer", then the two corresponding nodes are the supplier node and the manufacturer node. Locate the specific positions of these two nodes in the transaction link dependency graph and record their coordinate information or regional position in the graph. At the same time, query the connection relationships between these two nodes and other nodes. For example, the supplier node also has business dependency connections with other manufacturer nodes, and circulation dependency connections with raw material target nodes, etc.; the manufacturer node also has business dependency connections with distributor nodes, and circulation dependency connections with finished product target nodes, etc., thereby recording the above location information and connection relationships in detail.

[0095] Step S144: Analyze the dependency decay coefficient between the node corresponding to the abnormal connection edge and the connection edges of other nodes. If multiple connection edges associated with the node are all abnormal connection edges, then the node is determined to be a core abnormal dependency node; if the node is only associated with one abnormal connection edge, then the node is determined to be a normal abnormal dependency node.

[0096] Step S1441: For the first node corresponding to each abnormal connection edge, traverse all connection edges connected to that node in the transaction link dependency graph.

[0097] For each abnormal connection edge in the list, the first of its two corresponding nodes is used as the analysis object. For example, in the abnormal connection edge "Business Dependency - Supplier - Manufacturer", the first node is the supplier node. All connections in the transaction dependency graph that are linked to the supplier node are traversed, including business dependency connections between the supplier and other manufacturers, flow dependency connections between the supplier and raw material suppliers, and association dependency connections between the supplier and accounts receivable documents. This ensures that all connections are traversed without missing any dependencies related to that node.

[0098] Step S1442: Check if each traversed connection edge exists in the list of abnormal connection edges, and count the number of abnormal connection edges associated with the node.

[0099] For each connection edge encountered that links to a node, check if its identifier exists in the list of abnormal connections. For example, when iterating through the connections of a supplier node, check if the identifier of the connection edge with business dependency of manufacturer A is in the abnormal list, if the identifier of the connection edge with business dependency of manufacturer B is in the abnormal list, and so on. For each connection edge found in the abnormal list, increment the abnormal connection edge count for that node by 1. For example, if a supplier node has 5 connections, and 3 of them are in the abnormal list, then the count of abnormal connections is 3. This method accurately counts the number of abnormal connections associated with each node.

[0100] Step S1443: If the number of abnormal connection edges counted exceeds the preset ratio of the total number of connection edges of the node, the dependency relationship type of the abnormal connection edges is further analyzed. If it includes two or more types of business dependency, flow dependency and association dependency, the node is determined to be a core abnormal dependency node.

[0101] The preset ratio is set to 40%. For each abnormal connection edge counted, its proportion of the total number of connections to that node is calculated. For example, if a supplier node has 10 total connections and 5 abnormal connections, the ratio is 50%, exceeding the preset 40%. In this case, the dependency types of these 5 abnormal connections are further analyzed. If they include both business dependencies and workflow dependencies, then the condition of including two or more types is met, and the supplier node is determined to be a core abnormal dependency node. If the abnormal connection edge is of only one type, even if the ratio exceeds the preset value, it needs to be judged comprehensively in conjunction with other conditions. However, if multiple types are met in this step, it is directly determined to be a core anomaly.

[0102] Step S1444: If the number of abnormal connection edges counted is only one, and the dependency type of the abnormal connection edge is a single type, then check the historical abnormal records of the node. If the node has not had any abnormal connection edge associations in the historical records, then the node is determined to be a normal abnormal dependency node.

[0103] If a node has only one abnormal connection edge counted, and the dependency type of that abnormal connection edge is a single type, such as only an association dependency, then the node's historical anomaly records are retrieved. If there has never been a case of an abnormal connection edge associated with that node in the historical records, then this anomaly is considered an occasional, single anomaly, and it is classified as a normal abnormal dependency node. For example, if a distributor node only has abnormal connection edges with the manufacturer's business dependencies, and there are no anomaly records for that node in the historical records, then it is classified as a normal abnormal dependency node.

[0104] Step S1445: If the number of abnormal connection edges counted is one, but the node's historical records show multiple abnormal associations of different connection edges, then based on the dependency decay coefficient level of the current abnormal connection edge, if the level is higher than the level of the historical abnormal connection edge, then the node is determined to be a core abnormal dependency node; if the level is the same as the level of the historical abnormal connection edge, then it is determined to be an ordinary abnormal dependency node.

[0105] For nodes with only one abnormal connection edge but multiple abnormal connection edges recorded in the historical data (e.g., a manufacturer node had three abnormal connection edge records in the past six months, and now a new abnormal connection edge has appeared), the dependency decay coefficient level of the current abnormal connection edge is compared with the levels of historical abnormal connection edges. If the current level is 5 and the highest historical level is 3, and the current level is higher than the historical level, it is determined to be a core abnormal dependency node. If the current level is 3, consistent with the highest historical level, it is determined to be a normal abnormal dependency node. The abnormal nature of the node is determined by comprehensively considering historical data using the above method.

[0106] Step S1446: After determining the nodes corresponding to all abnormal connection edges, classify them into core abnormal dependency nodes and ordinary abnormal dependency nodes to form sub-lists of the two types of abnormal nodes.

[0107] After determining the nodes corresponding to each abnormal connection edge, the nodes identified as core abnormal dependency nodes are grouped together to form a core abnormal node sublist. This list includes information such as node identifier, the number and type of associated abnormal connection edges, and the determination criteria. Similarly, ordinary abnormal dependency nodes are categorized to form an ordinary abnormal node sublist. For example, the core abnormal node sublist includes supplier nodes and manufacturer nodes, while the ordinary abnormal node sublist includes distributor nodes, nodes for a specific raw material, etc. These two sublists together constitute the initial classification of abnormal dependency nodes.

[0108] Step S145: Count the number of core abnormal dependent nodes and ordinary abnormal dependent nodes and their distribution in the transaction chain. Sort the abnormal dependent nodes according to the size of the abnormal impact range to form an abnormal dependent node list. The abnormal impact range is determined based on the number of connection edges associated with the node and the type of dependency relationship.

[0109] The system counts the number of core and ordinary abnormal dependent nodes. For example, there are 2 core abnormal dependent nodes and 5 ordinary abnormal dependent nodes. Based on the recorded node location information, their distribution in the transaction chain is determined. For example, core abnormal dependent nodes are located at the beginning and key intermediate links of the transaction chain, while ordinary abnormal dependent nodes are distributed in various branch links. The abnormal impact range of each abnormal dependent node is assessed. Nodes with more associated connection edges have a larger impact range; nodes associated with business dependency connection edges have a larger impact range than nodes associated with association dependency connection edges. For example, a core abnormal dependent node supplier is associated with 10 connection edges, 5 of which are abnormal connection edges, and most of them are business dependency types, so its impact range is large; an ordinary abnormal dependent node is associated with only 3 connection edges, 1 of which is an abnormal connection edge of association dependency type, so its impact range is small. The abnormal dependent nodes are sorted from largest to smallest in impact range to form an abnormal dependent node list. The list includes information such as node identifier, abnormal type (core or ordinary), and impact range assessment.

[0110] Step S150: For the abnormal dependent node, adapt the corresponding operation and maintenance strategy, generate an operation and maintenance execution instruction that includes the node processing flow, dependency adjustment scheme and strategy execution sequence, and send the operation and maintenance execution instruction to the operation and maintenance terminal of the supply chain finance platform.

[0111] Based on the list of core and ordinary abnormal dependency nodes, different operational strategies are adapted. For core abnormal dependency nodes, a more comprehensive and in-depth handling process and adjustment plan are adopted; for ordinary abnormal dependency nodes, relatively simple handling measures are used. Specific handling procedures are formulated for each node, clarifying the adjustment methods for dependencies, and planning the order and timing of strategy execution. The above content is integrated into operational execution instructions, ensuring the instructions are complete and clear, and then sent to the operational terminal of the supply chain finance platform to guide operational personnel in their operations.

[0112] Step S151: For core anomaly dependent nodes, retrieve the preset core anomaly handling strategy library, and filter the corresponding node processing flow according to the anomaly connection edge type and dependency decay coefficient level associated with the core anomaly dependent node. The node processing flow includes node information update, node association relationship investigation and node collaboration permission adjustment.

[0113] The core exception handling strategy library stores processing flow templates for different types and levels of exception connections. For example, for core exception dependency nodes with high-level exceptions related to business dependencies and workflow dependencies, the corresponding comprehensive processing flow template is retrieved from the strategy library. This node processing flow first includes a node information update step, requiring the verification and updating of the node's basic information, qualification status, etc.; next is a node relationship investigation step, comprehensively checking the connection edge status and historical collaboration records of this node with all other nodes; finally, a node collaboration permission adjustment step, temporarily restricting or adjusting some of the node's collaboration permissions based on the investigation results, such as restricting the permission to initiate new business orders. By selecting suitable processing flows, it is ensured that core exception dependency nodes are fully handled.

[0114] Step S152: For ordinary abnormal dependency nodes, retrieve the preset ordinary abnormal handling strategy library, and determine the corresponding dependency adjustment scheme according to the dependency relationship type of the abnormal connection edge. The dependency adjustment scheme includes connection edge dependency frequency optimization, dependency association object replacement and dependency relationship re-establishment.

[0115] The general exception handling strategy library contains adjustment scheme templates for single-type exception connection edges. For example, if the exception connection edge of a general exception dependency node is of the association dependency type, the association dependency adjustment scheme template is retrieved from the strategy library. This association dependency adjustment scheme first includes optimization of the connection edge dependency frequency, analyzing the reasons for the frequency anomaly, such as whether it is due to data recording errors; if so, the frequency data is corrected. Secondly, it includes replacement of the dependency association object, checking whether there are other suitable rights certificates that can replace the currently associated certificate. Finally, it includes the reconstruction of the dependency relationship; if there are problems with the original relationship, it guides the re-initiation of the association process. Through these adjustment measures, the problem of general exception dependency nodes is resolved.

[0116] Step S153: Determine the priority of strategy execution based on the position of the abnormal dependent node in the transaction chain. The strategy execution priority of the core abnormal dependent node located at a critical position in the transaction chain is higher than that of abnormal dependent nodes in other positions.

[0117] Assess the importance of abnormal dependent nodes within the transaction chain. Nodes located at the beginning of the transaction chain, key inflection points, or connecting multiple branch chains are designated as critical nodes. For example, a core abnormal dependent node supplier, located at the beginning of the transaction chain, is the source of collaboration for multiple downstream entities, and its policy execution priority is set to the highest. Ordinary abnormal dependent nodes, such as distributors, located at the end of branch chains, have a lower policy execution priority. Core abnormal dependent nodes generally have a higher priority than ordinary abnormal dependent nodes. If an ordinary abnormal dependent node is located in a critical position, its priority can be appropriately increased, but it will still be lower than that of a core abnormal dependent node in a critical position. This prioritization ensures that anomalies at important nodes are handled first.

[0118] Step S154: Based on the policy execution priority, plan the execution time sequence of the operation and maintenance policies corresponding to each abnormal dependent node to form a policy execution sequence. The policy execution sequence includes the execution start time, execution duration and execution connection requirements of each operation and maintenance policy.

[0119] The execution time of operation and maintenance strategies is planned according to their execution priority from high to low. For example, the operation and maintenance strategy for the core anomaly dependent node supplier with the highest priority is scheduled in the first execution period, starting within 2 hours of receiving the instruction, with an estimated execution time of 8 hours; the strategy for the core anomaly dependent node manufacturer with the next highest priority is scheduled to begin after the supplier's strategy is completed, starting within 1 hour of the supplier's strategy completion, with an estimated execution time of 6 hours; the strategies for ordinary anomaly dependent nodes are scheduled in subsequent periods according to their priority. The strategy execution sequence should clearly specify the start time, estimated execution time, and connection requirements between the completion of the previous strategy and the start of the next strategy, such as whether data synchronization and result confirmation are required.

[0120] Step S155: Integrate the node processing flow, dependency adjustment plan and strategy execution sequence according to the preset format, add instruction identifier, generation time and execution subject information to form operation and maintenance execution instructions.

[0121] Following the preset format of the supply chain finance platform's operation and maintenance instructions, the node processing flow, dependency adjustment plan, and strategy execution sequence are integrated. A unique instruction identifier is added at the beginning of each instruction, such as "Operation and Maintenance Instruction-20250821-001"; the instruction generation time is added, accurate to the minute; and the executing entity is clearly identified as a specific subgroup within the platform's operation and maintenance team, such as the core operation and maintenance group responsible for executing strategies for core abnormal dependency nodes, and the general operation and maintenance group responsible for executing strategies for ordinary abnormal dependency nodes. This ensures that the integrated instruction structure is clear, the content is complete, all information is accurate, and it conforms to the platform's format requirements for operation and maintenance instructions.

[0122] Step S156: Send the operation and maintenance execution instruction to the operation and maintenance terminal of the supply chain finance platform.

[0123] The generated operation and maintenance (O&M) execution instructions are sent to the O&M terminal via the internal instruction transmission system of the supply chain finance platform. Encryption technology is used during transmission to ensure that the instruction content is not leaked or tampered with. Upon receiving the instruction, the O&M terminal automatically decrypts and verifies its format to confirm its integrity and legality. The terminal interface displays the main content of the instruction, including a list of abnormal dependency nodes, strategy execution priority, and timing arrangements, reminding O&M personnel to review and execute it promptly. Simultaneously, the system records information such as the instruction's sending time and receiving status for future traceability.

[0124] Figure 2 This illustration shows the hardware structure of an operation and maintenance processing system 100 for implementing the above-described operation and maintenance processing method for a supply chain finance platform, as provided in an embodiment of the present invention. Figure 2 As shown, the operation and maintenance processing system 100 applied to the supply chain finance platform may include a processor 110, a machine-readable storage medium 120, a bus 130, and a communication unit 140.

[0125] In one possible design, the operation and maintenance processing system 100 applied to the supply chain finance platform can be a single server or a server group. The server group can be centralized or distributed (e.g., the operation and maintenance processing system 100 applied to the supply chain finance platform can be a distributed system). In some embodiments, the operation and maintenance processing system 100 applied to the supply chain finance platform can be local or remote. For example, the operation and maintenance processing system 100 applied to the supply chain finance platform can access information and / or data stored in machine-readable storage medium 120 via a network. As another example, the operation and maintenance processing system 100 applied to the supply chain finance platform can directly connect to machine-readable storage medium 120 to access the stored information and / or data.

[0126] Machine-readable storage medium 120 may store data and / or instructions. In some embodiments, machine-readable storage medium 120 may store data acquired from an external terminal. In some embodiments, machine-readable storage medium 120 may store data and / or instructions used by the operation and maintenance processing system 100 of the supply chain finance platform to perform or use in order to accomplish the exemplary methods described in this invention.

[0127] In the specific implementation process, one or more processors 110 execute computer-executable instructions stored in machine-readable storage medium 120, so that processor 110 can execute the operation and maintenance processing method applied to the supply chain finance platform as described in the above method embodiment. The processor 110, machine-readable storage medium 120 and communication unit 140 are connected through bus 130. The processor 110 can be used to control the sending and receiving actions of communication unit 140.

[0128] The specific implementation process of processor 110 can be found in the various method embodiments executed by the operation and maintenance processing system 100 applied to the supply chain finance platform. The implementation principle and technical effect are similar, and will not be repeated here.

[0129] Furthermore, this embodiment of the invention also provides a readable storage medium containing computer-executable instructions. When the processor executes the computer-executable instructions, the above-described operation and maintenance method applied to the supply chain finance platform is implemented.

[0130] It should be noted that, in order to simplify the description of this invention and thus aid in the understanding of one or more embodiments, the foregoing description of the embodiments of this invention sometimes combines multiple features into a single embodiment, drawing, or description thereof. Similarly, it should be noted that, in order to simplify the description of this invention and thus aid in the understanding of one or more embodiments, the foregoing description of the embodiments of this invention sometimes combines multiple features into a single embodiment, drawing, or description thereof.

Claims

1. A method for operation and maintenance of a supply chain finance platform, characterized in that, The method includes: Obtain the transaction link dependency information set of the supply chain finance platform, wherein the transaction link dependency information set includes business dependency record units between transaction entities, transfer dependency record units of transaction targets, and association dependency record units of equity certificates; A transaction link dependency graph is generated based on the transaction link dependency information set. Nodes in the transaction link dependency graph correspond to transaction entities, transaction targets, or equity certificates, and the connecting edges between nodes represent the dependency relationship type and dependency association frequency. Based on the frequency change trend of dependency associations of the connecting edges in the transaction link dependency graph, the dependency decay coefficient corresponding to each connecting edge is calculated. The dependency decay coefficient is used to reflect the degree of weakening of the dependency relationship. Based on the dependency decay coefficient, abnormal dependency nodes in the transaction link are located. The abnormal dependency nodes are nodes whose dependency relationship weakening degree exceeds the normal range corresponding to the dependency decay coefficient. For the corresponding operation and maintenance strategy adapted to the abnormal dependent node, an operation and maintenance execution instruction is generated, which includes the node processing flow, dependency adjustment scheme and strategy execution sequence, and the operation and maintenance execution instruction is sent to the operation and maintenance terminal of the supply chain finance platform.

2. The operation and maintenance method for a supply chain finance platform according to claim 1, characterized in that, The set of transaction chain dependency information obtained from the supply chain finance platform includes: Access the business interaction module of the supply chain finance platform to extract the interaction records generated by the transaction entities during the business collaboration process. The interaction records include the collaboration initiator identifier, the collaboration responder identifier, and the collaboration business type. Record segments with continuous collaborative relationships are selected from the interaction records and organized into business dependency record units between transaction entities according to the time sequence of the collaboration. Each business dependency record unit corresponds to a group of transaction entities with a fixed collaboration mode. Access the target transfer module of the supply chain finance platform to extract the transfer trajectory record of the transaction target from the initial holder to the subsequent holder. The transfer trajectory record includes the target identifier, the transfer initiator identifier, the transfer recipient identifier, and the transfer completion time. Based on the circulation trajectory records, the order of transfer of the transaction target between different entities is sorted out, and continuous circulation records containing the same transaction target are integrated into circulation dependency record units of the transaction target. Each circulation dependency record unit corresponds to a complete circulation path of a transaction target. Access the voucher management module of the supply chain finance platform to extract the association records between equity vouchers and transaction entities and transaction targets. The association records include voucher identifiers, associated entity identifiers, and associated target identifiers. The associated records are associated with the corresponding business dependency record units and circulation dependency record units to form an associated dependency record unit for the rights certificate. Each associated dependency record unit corresponds to a rights certificate and a set of matching relationships between business dependencies and circulation dependencies. According to the upstream and downstream collaboration sequence of the transaction link, the business dependency record unit, the flow dependency record unit and the association dependency record unit are connected in series to form a transaction link dependency information set containing multiple association dependencies.

3. The operation and maintenance method for a supply chain finance platform according to claim 2, characterized in that, The business dependency record unit, flow dependency record unit, and association dependency record unit are connected in series according to the upstream and downstream collaboration sequence of the transaction link to form a transaction link dependency information set containing multiple association dependencies, including: Identify the collaboration initiator and collaboration responder in each business dependency record unit, determine that the collaboration initiator is the upstream entity and the collaboration responder is the downstream entity, and establish the upstream and downstream relationship of business dependencies. Identify the initiating entity and receiving entity in each flow dependency record unit, determine that the initiating entity is the upstream entity and the receiving entity is the downstream entity, and establish the upstream and downstream relationship of flow dependency; Identify the association order between the associated subject and the equity certificate in each associated dependency record unit, determine whether business dependency or circulation dependency is generated first, and then form equity certificate associated dependency, and establish the upstream and downstream relationship between associated dependency and business dependency and circulation dependency; Starting from the upstream entity at the beginning of the transaction link, the corresponding business dependency record unit, flow dependency record unit and associated dependency record unit are sequentially associated according to the upstream and downstream relationship to form a dependency information chain for a single transaction link. Check if there are multiple parallel transaction links that depend on information chains. If so, analyze the relationship between each information chain and merge information chains with shared nodes to form a dependency information network containing multiple branch links. The record units in the dependency information network are sequentially numbered to ensure that the position of each record unit is consistent with the actual collaboration order of the transaction link, thus forming a transaction link dependency information set containing multiple related dependencies.

4. The operation and maintenance method for a supply chain finance platform according to claim 1, characterized in that, The step of generating a transaction link dependency graph based on the transaction link dependency information set includes: Extract the identifiers of all transaction entities, transaction targets, and equity certificates from the transaction chain dependency information set, and create transaction entity nodes, transaction target nodes, and equity certificate nodes respectively, with each node assigned a unique identifier code; For each business dependency record unit, identify the transaction entity node corresponding to the collaboration initiator identifier and the collaboration responder identifier, draw a connection edge between the two transaction entity nodes, label the type of the connection edge as business dependency, and count the number of times collaboration occurs in this business dependency record unit as the dependency association frequency and label it on the connection edge. For each flow dependency record unit, identify the transaction entity node corresponding to the flow initiator identifier, the flow receiver identifier, and the transaction target node corresponding to the transaction target identifier. Draw connection edges between the flow initiator node and the transaction target node, and between the transaction target node and the flow receiver node. The type of the connection edge is labeled as flow dependency. Count the number of times the same flow path occurs in the flow dependency record unit as the dependency association frequency and label it on the connection edge. For the related dependency record unit, identify the transaction entity node corresponding to the related entity identifier, the transaction target node corresponding to the related target identifier, and the equity certificate node corresponding to the equity certificate identifier. Draw connection edges between the equity certificate node and the transaction entity node, and between the equity certificate node and the transaction target node. The type of the connection edge is labeled as related dependency. Count the number of times the same relationship occurs in the related dependency record unit as the dependency relationship frequency and label it on the connection edge. The layout of all nodes and connecting edges is adjusted so that nodes of the same type are clustered together and nodes of different types are arranged in order of dependency relationship to form the transaction link dependency graph.

5. The operation and maintenance method for a supply chain finance platform according to claim 4, characterized in that, The process of adjusting the layout of all nodes and connecting edges to cluster nodes of the same type and arrange nodes of different types according to their dependencies, forming the transaction link dependency graph, includes: Different visual identifiers are assigned to the transaction entity nodes, transaction target nodes, and equity certificate nodes, so that nodes of the same type have consistent visual characteristics, while nodes of different types have different visual characteristics. The graph canvas is divided into three areas, which are used to place transaction entity nodes, transaction target nodes, and equity certificate nodes, respectively. The three areas are arranged in the order of business dependency, circulation dependency, and association dependency. The transaction entity nodes are arranged in the corresponding area according to the order of upstream and downstream cooperation, with upstream entity nodes located on the left side of the area and downstream entity nodes located on the right side of the area. The target nodes are mapped to the right side of the main transaction nodes according to the flow order, so that the connection edges between the target nodes and the corresponding upstream and downstream main transaction nodes are in the horizontal direction. Place the equity certificate node in the area between the corresponding transaction entity node and the transaction target node, so that the length of the connecting edge between the equity certificate node and the associated node is similar; Check if there are any overlapping or intersecting edges after the layout is completed. If so, fine-tune the point positions until the overlapping or intersecting edges are eliminated, thus forming a transaction link dependency graph.

6. The operation and maintenance method for a supply chain finance platform according to claim 1, characterized in that, The step of calculating the dependency decay coefficient corresponding to each connection edge based on the frequency change trend of dependency associations of the connection edges in the transaction link dependency graph includes: Extract the dependency association frequency data of each connection edge in the transaction link dependency graph within a preset time period, and organize it into a dependency association frequency sequence in chronological order; Analyze the changing direction of the dependency association frequency sequence. If the dependency association frequency in the later time period is lower than that in the previous time period, the dependency association frequency of the connection edge is determined to be a weakening trend. If the dependency association frequency in the later time period is higher than that in the previous time period, it is determined to be a strengthening trend. If the dependency association frequency remains stable in different time periods, it is determined to be a stable trend. For the connection edges whose changing trend is weakening, further analyze the duration and magnitude of the weakening, whereby the magnitude of the weakening is the degree of difference between the initial dependency frequency and the current dependency frequency. The level of the dependency attenuation coefficient is determined based on the duration and magnitude of the weakening. The longer the duration and the greater the magnitude of the weakening, the higher the level of the dependency attenuation coefficient. For connecting edges whose changing trend is a strengthening trend or a stable trend, a basic dependency decay coefficient is set, and the level corresponding to the basic dependency decay coefficient is lower than the dependency decay coefficient level of connecting edges with a weakening trend. Associate the dependency attenuation coefficient level of each connection edge with the connection edge identifier to form an attenuation coefficient list containing all connection edges and their corresponding dependency attenuation coefficients.

7. The operation and maintenance method for a supply chain finance platform according to claim 6, characterized in that, The method of determining the level of dependence on attenuation coefficient based on the duration and magnitude of attenuation includes: The duration of the weakening effect is divided into multiple consecutive duration intervals, each duration interval corresponding to a base level. The longer the duration interval, the higher the base level. The weakening range is divided into multiple consecutive range intervals, each range interval corresponds to an adjustment level, and the larger the range interval, the higher the adjustment level; The base level and the adjustment level are superimposed to obtain the initial dependency decay coefficient level of each weakening trend connection edge; Based on the dependency type corresponding to the connection edge, if the dependency type is a business dependency, the level is increased by one level from the initial level; if it is a flow dependency, the initial level is maintained; if it is an association dependency, the level is decreased by one level from the initial level. Check whether the adjusted dependency attenuation coefficient level conforms to the level range of similar historical weakening cases. If it exceeds the level range, make fine adjustments to ensure that the adjusted dependency attenuation coefficient level is within a reasonable range. The finalized dependency attenuation coefficient level is associated with the edge identifier, dependency type, attenuation duration, and attenuation magnitude and stored to form a complete attenuation coefficient list.

8. The operation and maintenance method for a supply chain finance platform according to claim 1, characterized in that, The step of locating abnormal dependency nodes in the transaction chain based on the dependency decay coefficient includes: Determine a general range for the dependency attenuation coefficient, which is based on the dependency attenuation coefficient level of the connecting edges in the transaction link under historical normal operating conditions; Filter out connection edges whose attenuation coefficient level exceeds the normal range from the attenuation coefficient list and mark them as abnormal connection edges; Track the two nodes corresponding to the abnormal connection edge, and record the positions of the two nodes in the transaction link dependency graph and their connection relationships with other nodes; Analyze the dependency decay coefficient between the node corresponding to the abnormal connection edge and the connection edges of other nodes. If multiple connection edges associated with the node are all abnormal connection edges, then the node is determined to be a core abnormal dependency node; if the node is associated with only one abnormal connection edge, then the node is determined to be a normal abnormal dependency node. The number of core abnormal dependent nodes and ordinary abnormal dependent nodes and their distribution in the transaction chain are counted. The abnormal dependent nodes are sorted according to the size of the abnormal impact range to form an abnormal dependent node list. The abnormal impact range is determined based on the number of connection edges associated with the node and the type of dependency relationship. Specifically, the analysis of the dependency decay coefficient between the node corresponding to the abnormal connection edge and other nodes determines whether the node is a core abnormal dependency node if multiple connections associated with the node are abnormal; otherwise, the node is a normal abnormal dependency node. For the first node corresponding to each abnormal connection edge, traverse all connection edges in the transaction dependency graph that are connected to that node. Check if each traversed connection edge exists in the list of abnormal connection edges, and count the number of abnormal connection edges associated with that node; If the number of abnormal connection edges counted exceeds the preset ratio of the total number of connection edges of the node, the dependency relationship type of the abnormal connection edges will be further analyzed. If it includes two or more types of business dependency, flow dependency and association dependency, the node will be determined as a core abnormal dependency node. If the number of abnormal connection edges counted is only one, and the dependency type of the abnormal connection edge is a single type, then check the historical abnormal records of the node. If the node has not had any abnormal connection edge associations in the historical records, then the node is determined to be a normal abnormal dependency node. If the number of abnormal connection edges counted is one, but the node's historical records show multiple abnormal associations with different connection edges, then the node is determined to be a core abnormal dependency node if the dependency decay coefficient level of the current abnormal connection edge is higher than that of the historical abnormal connection edges; otherwise, it is determined to be a normal abnormal dependency node. After determining the nodes corresponding to all abnormal connection edges, the nodes are classified into core abnormal dependency nodes and ordinary abnormal dependency nodes, forming two sub-lists of abnormal nodes.

9. The operation and maintenance method for a supply chain finance platform according to claim 1, characterized in that, The operation and maintenance strategy adapted to the abnormal dependent nodes generates operation and maintenance execution instructions that include node processing flow, dependency adjustment scheme, and strategy execution sequence, including: For core anomaly dependent nodes, a preset core anomaly handling strategy library is retrieved. Based on the anomaly connection edge type and dependency decay coefficient level associated with the core anomaly dependent node, the corresponding node processing flow is selected. The node processing flow includes node information update, node association relationship investigation and node collaboration permission adjustment. For ordinary abnormal dependent nodes, a preset ordinary abnormal handling strategy library is invoked, and a corresponding dependency adjustment scheme is determined according to the dependency relationship type of the abnormal connection edge. The dependency adjustment scheme includes connection edge dependency frequency optimization, dependency association object replacement, and dependency relationship re-establishment. Based on the position of the abnormal dependent nodes in the transaction chain, the priority of strategy execution is determined. The strategy execution priority of the core abnormal dependent nodes located in the key position of the transaction chain is higher than that of abnormal dependent nodes in other positions. Based on the policy execution priority, the execution time sequence of the operation and maintenance policies corresponding to each abnormal dependent node is planned to form a policy execution sequence. The policy execution sequence includes the execution start time, execution duration and execution connection requirements of each operation and maintenance policy. The node processing flow, dependency adjustment plan, and strategy execution sequence are integrated according to a preset format, and instruction identifiers, generation time, and execution subject information are added to form operation and maintenance execution instructions.

10. An operation and maintenance system for a supply chain finance platform, characterized in that, The system includes a processor and a memory, the memory being connected to the processor. The memory is used to store programs, instructions, or code, and the processor is used to run the programs, instructions, or code in the memory to implement the operation and maintenance processing method for a supply chain finance platform as described in any one of claims 1-9.

Images