A method and system for the management and valuation of data assets across the entire logistics chain.

Abstract

This invention discloses a method and system for the management and value assessment of logistics data assets across the entire supply chain, relating to the field of artificial intelligence-based logistics information management technology. The method includes constructing a knowledge graph for the entire logistics supply chain to achieve structured association and lineage tracing of multi-source heterogeneous data; defining ownership, governing quality and compliance, and dynamically calibrating data assets; using a targeted, optimized incremental GraphSAGE model for graph reasoning to output dynamic asset value scores, hierarchical classifications, and ownership value decomposition; implementing differentiated full lifecycle management and permission configuration based on a hierarchical control rule engine; and iteratively optimizing the knowledge graph and value model through closed-loop feedback. This invention solves the technical problems of difficult ownership confirmation, opaque value assessment, and lack of end-to-end control for logistics data assets, achieving a closed-loop transformation of logistics data assets from resources to assets, and improving the operational efficiency and value realization capabilities of data assets.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence logistics information management technology, and in particular to a method and system for the management and value assessment of data assets across the entire logistics chain. Background Technology

[0002] With the deepening of digital transformation in the logistics industry, the scale of multi-source heterogeneous data generated across the entire supply chain is growing exponentially, and data has become a core production factor. However, existing logistics data asset management faces the following technical challenges: The data ownership is unclear: logistics data involves multiple entities such as cargo owners, carriers, and warehouse operators. The data flow is complex, and the lack of a unified ownership definition and entity association mechanism makes it difficult to confirm the ownership of data assets and results in unfair value distribution.

[0003] Data quality and compliance are difficult to guarantee: Logistics data comes from diverse sources and has different formats, resulting in problems such as missing data, inconsistencies, and poor timeliness. Furthermore, there is a lack of automated quality inspection and compliance verification methods, making it difficult to meet the quality requirements for data assetization.

[0004] Value assessment is opaque and not dynamic: Traditional data asset valuation often uses static and subjective methods, which do not fully consider the full-link correlation characteristics and dynamic value changes of logistics data, making it difficult to achieve accurate valuation and ownership separation.

[0005] Lack of full lifecycle management: The lack of differentiated management strategies for data assets of different value levels makes it impossible to achieve full lifecycle management from data collection, processing, and use to archiving and decommissioning, resulting in the inability to maximize the value of data assets.

[0006] To address these issues, there is an urgent need for a method and system for the management and value assessment of data assets across the entire logistics chain. Summary of the Invention

[0007] To address the aforementioned issues, this application proposes a method and system for the management and valuation of logistics data assets across the entire supply chain. This system enables the confirmation of ownership, governance, dynamic valuation, hierarchical control, and closed-loop iteration of logistics data assets. It resolves the problems of difficulty in defining ownership, inaccurate valuation, and lack of full lifecycle control in existing technologies, thereby improving the operational efficiency and value realization capabilities of logistics data assets.

[0008] On the one hand, this application proposes a method for the management and value assessment of data assets across the entire logistics chain, including the following steps: S1. Obtain multi-source heterogeneous raw data from the entire logistics chain, perform knowledge extraction and fusion processing on the multi-source heterogeneous raw data from the entire logistics chain, and obtain the logistics data asset map base library. S2. Based on the four-in-one lineage map, perform full-link graph traversal processing on the logistics data asset map base database to obtain the asset full-link lineage traceability relationship and ownership boundary definition results. S3. Based on the asset lineage tracing relationship and ownership definition results, the graph's native quality inspection and compliance verification algorithm is used to carry out full life cycle governance processing to obtain highly reliable data assets that meet both quality and compliance standards, and simultaneously complete the dynamic calibration of the asset value coefficient. S4. Based on highly reliable data assets and full-link graph data, the GraphSAGE model is used to perform graph reasoning calculations to obtain real-time dynamic asset value scoring, hierarchical classification results and ownership value splitting strategies. S5. Based on the dynamic asset value scoring, hierarchical classification results and ownership splitting strategy, the hierarchical control rule engine of the graph is used to perform differentiated management strategy matching and processing, so as to obtain the full life cycle control strategy and permission configuration strategy that are adapted to the asset value level. S6. Obtain high-value data assets and business scenario requirements after hierarchical management and control, and perform scenario-based asset matching processing based on graph semantic retrieval and association recommendation algorithms to obtain asset application strategies that meet business requirements. S7. Obtain incremental business data and value realization results generated during asset business applications, and perform closed-loop iterative processing based on the incremental update and value feedback mechanism of the asset map to obtain the updated asset map base library and optimized value assessment model parameters.

[0009] Preferably, the specific content of S1 includes: Acquire multi-source heterogeneous raw data from the entire logistics chain, perform data classification, verification, and format normalization processing to obtain the raw data pool D. clean ; Construct an asset-specific ontology system, wherein the asset-specific ontology system defines the top-level core entity logistics data asset E. core The four related domains are business processes, ownership entities, lifecycle, and compliance control, and the value assessment attribute A is pre-embedded. value The valuation attributes include the initial value V0 and the business weight W. b Timing decay coefficient k; Based on the asset-specific ontology system, knowledge extraction and processing of entities and relations are performed, and knowledge fusion and semantic unification are carried out to obtain triples; Based on the asset type and the relevant attributes of the business process to which it belongs, the asset nodes and relationships embedded with the initial value are obtained, resulting in a fusion triplet with the initial value. By writing the fusion triples with initial value into a graph database for materialized storage, a standardized, semantically unified, and natively embedded value logistics data asset graph base library is obtained.

[0010] Preferably, the specific content of S2 includes: Edge relationship expansion processing is performed on the underlying database G0 of the data asset graph, adding technical lineage edges, business link edges, ownership edges, and value transfer edges to obtain a four-in-one lineage graph G. blood ; Four-in-one bloodline map G blood Perform graph traversal parameter configuration processing, set traversal depth and node filtering conditions, and obtain graph traversal execution steps adapted to logistics scenarios; Based on the graph traversal execution steps, a full-link graph traversal process is performed on each asset entity to trace the four major flow links of technology, business, ownership, and value, thereby obtaining the asset's full-link lineage tracing relationship T. race ; T for tracing the lineage of assets across the entire value chain race The ownership and ownership relationships of each asset are analyzed and processed to clarify the ownership subject, participating and contributing subject, and basic dimensions of contribution of each asset, thus obtaining the ownership boundary definition result O. T for tracing the lineage of assets across the entire value chain race The results of the ownership boundary delineation (O) are integrated and synchronously updated to the four-in-one bloodline map (G). blood This yields an updated graph G1 containing blood relations and ownership definitions, and outputs a blood tracing table and an ownership boundary definition table.

[0011] Preferably, the specific content in S3 includes: T for tracing the lineage of assets across the entire value chain race Result of ownership boundary delineation O ={( E core , O i , d i The updated graph G1 is then used for data integration and core information extraction to obtain the full asset dataset D. base ; Based on the full asset dataset D base The assets undergo end-to-end verification, calculating integrity, accuracy, and consistency metrics. Low-quality asset nodes are then identified, resulting in the asset quality verification results and a list of low-quality assets (D). lowQ ; Based on the asset quality verification results and the list of low-quality assets D lowQ Low-quality assets are corrected and all quality indicators are re-verified to obtain a set of assets that meet quality standards. E Q ; Based on the collection of quality-compliant assets EQ By associating data with the compliance control domain in the graph G1, the assets undergo comprehensive verification processing for sensitive information and circulation compliance. This process calculates compliance indicators for sensitive information and circulation, marks non-compliant asset nodes, and yields the asset compliance verification results and a list of non-compliant assets, D. lowC ; Obtain asset compliance verification results and a list of non-compliant assets (D) lowC Non-compliant assets were rectified to meet compliance requirements, and all compliance indicators were re-verified to obtain an asset set that met both quality and compliance standards. E QC ; Based on the quality-value and compliance-value mapping rules, the quality value coefficient Q and compliance value coefficient C are calculated. The calculation results are labeled as asset value attributes, and the asset value attributes are dynamically calibrated to obtain highly reliable data assets that meet both quality and compliance standards and have calibrated value coefficients. E trust ; Will E trust Q, C are simultaneously updated to G1, completing the iterative processing of the graph data, resulting in the updated graph G2 = G1∪{Q,C,E}. trust It also outputs highly reliable data assets and value coefficients.

[0012] Preferably, the specific content in S4 includes: For highly reliable data assets E trust The basic dataset D is obtained by performing feature extraction on Q, C, and G2. model The underlying dataset D model This includes asset topology characteristics, business characteristics, quality compliance characteristics, and time-series characteristics; For the basic dataset D model The input feature vector X is obtained by standardization. t ,X b ,X qc ,X time The GraphSAGE model is used to perform graph inference calculations on the input feature vectors to obtain real-time dynamic asset value scores, thereby obtaining a set of real-time dynamic asset value scores V. all ; There are preset asset thresholds. Based on the real-time dynamic value score of the assets, the assets are classified and graded to obtain the asset classification result Grade={Grade1, Grade2, Grade3}. Based on the asset classification results, the real-time dynamic asset value score set, and the ownership boundary delineation results, the Shapley function is used to calculate the value contribution ratio of each ownership entity. Strategies for Decomposing Ownership Value ; The asset real-time dynamic value score set V all Asset classification results (Grade) and ownership value splitting strategy The graph is synchronously updated to graph G2 to complete the graph iteration process, resulting in the updated graph G3 = G2∪{V, Grade, V i It also outputs an asset dynamic value scoring table, a classification table, and an ownership value breakdown table.

[0013] Preferably, the specific content in S5 includes: Real-time dynamic value scoring set V of assets all Asset classification results (Grade) and ownership value splitting strategy The information extraction and processing of the graph G3 yields the value characteristics, ownership entities, and contribution ratios of assets at each level, resulting in the hierarchical management and control basic dataset D. control ; Based on the graph-based hierarchical control rule engine, combined with the needs of logistics data asset management, differentiated control rules are configured for assets of different levels, and a hierarchical control rule system Rule={Rule1, Rule2, Rule3} is constructed to obtain control rule strategies adapted to each asset level. Based on hierarchical control rules and strategies and hierarchical control basic dataset D control The storage strategy is matched with different asset levels to obtain the asset-tiered storage strategy Store. rule ; Store based on hierarchical storage strategy rule Ownership value splitting strategy Based on the asset classification results (Grade), a three-tier permission system (subject-asset-permission) is constructed, and access, usage, and revenue permissions for each subject are configured to obtain the permission configuration strategy P. erm ={(O i E core P ij ) }, P ij Access level; Based on permission configuration policy P erm Real-time dynamic asset value scoring set V all By combining the time-series decay factor T, archiving and decommissioning thresholds are set for assets of different levels, triggering full lifecycle management actions, thus obtaining the asset full lifecycle management system Life. control ; Asset Tiered Storage Strategy Store rulePermission configuration strategy (Perm) and full lifecycle management system (Life) control The system is integrated to obtain a full lifecycle management and control system and permission configuration strategy. At the same time, the management and control rules and permission information are updated to graph G3, completing the graph iteration and obtaining the updated graph G4 = G3∪{Rule, Perm, V}. new}, V new Based on the value score obtained after the feedback on the control effect, output a control system strategy and permission configuration table.

[0014] The preferred expression for real-time dynamic asset value scoring is: ; Where V is the real-time dynamic asset value score, V0 is the initial value, Q is the quality value coefficient, C is the compliance value coefficient, and B is the business value-added coefficient (B=1+0.1). Call frequency T is the time decay factor. t is the asset duration, and k is the decay coefficient (k=0.05 for time-series assets, k=0.01 for other assets). The expression for the value contribution ratio of each ownership entity is as follows: ; Where N is the set of all ownership entities, and S is a subset that does not include entity i. Let S be the asset value corresponding to subset S. The percentage of value contribution of subject i; ; The value due to subject i .

[0015] On the other hand, this application proposes a logistics end-to-end data asset management and value assessment system, including: Data acquisition unit: Acquires multi-source heterogeneous raw data from the entire logistics chain, performs knowledge extraction and fusion processing based on the innovatively constructed asset-specific ontology system, and obtains a logistics data asset map base library; The graph construction unit acquires a standardized logistics data asset graph base library, performs full-link graph traversal processing based on the innovatively constructed four-in-one lineage graph, obtains the asset full-link lineage traceability relationship and ownership boundary definition results, performs full life cycle governance processing based on the graph's native quality inspection and compliance verification algorithms, obtains highly reliable data assets that meet both quality and compliance standards, and simultaneously completes the dynamic calibration of the asset value coefficient. Value scoring configuration unit: acquires high-reliability data assets and full-link graph data after quality compliance calibration, performs graph inference calculation based on the targeted optimization incremental GraphSAGE model to obtain real-time dynamic asset value scoring, hierarchical classification results and ownership value splitting strategy, and performs differentiated management strategy matching processing based on the graph-based hierarchical control rule engine to obtain full life cycle control strategy and permission configuration strategy adapted to asset value level. Strategy generation unit: Acquires high-value data assets and business scenario requirements after hierarchical management and control, performs scenario-based asset matching processing based on graph semantic retrieval and association recommendation algorithms, and obtains asset application strategies that adapt to business requirements; Model optimization unit: Acquires incremental business data and value realization results generated during asset business applications, performs closed-loop iterative processing based on the incremental update of the asset map and the value feedback mechanism, and obtains the updated asset map base library and the optimized value assessment model parameters.

[0016] An electronic device includes a memory and a processor, wherein the memory stores a computer program, and the processor invokes the computer program in the memory to implement a method for the management and valuation of data assets across the entire logistics chain.

[0017] A storage medium storing computer-executable instructions, which, when loaded and executed by a processor, realize a method for data asset management and value assessment across the entire logistics chain.

[0018] In summary, the logistics end-to-end data asset management and value assessment method and system of the present invention has the following advantages compared with traditional technologies: 1. By constructing a full-link network of logistics data through knowledge graphs, the ownership boundaries of multiple entities are clarified, the problem of ambiguous ownership of logistics data is solved, and the foundation for data assetization is laid. 2. Employ automated quality inspection and compliance verification algorithms to achieve full lifecycle governance of data assets, ensuring that data assets meet the quality and compliance requirements for assetization and reducing compliance risks; 3. Based on the incremental GraphSAGE model and Shapley value calculation with targeted optimization, it fully considers the full-link correlation characteristics and dynamic value changes of logistics data, and realizes real-time assessment of asset value and fair allocation of ownership. 4. Configure differentiated management and control strategies based on asset value levels to achieve full lifecycle management of data assets from collection to decommissioning, maximize the value of data assets, and improve operational efficiency.

[0019] 5. Continuously optimize the knowledge graph, value model, and control rules through management feedback to ensure the adaptability and effectiveness of the strategy and to meet the dynamic development needs of the logistics industry.

[0020] The technical method of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0021] Figure 1 This is a flowchart illustrating the steps of a method for managing and assessing the value of end-to-end logistics data assets according to the present invention. Figure 2 This is a unit diagram of a logistics end-to-end data asset management and value assessment system according to the present invention. Detailed Implementation

[0022] The technical method of the present invention will be further described below with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of this application.

[0023] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the scope of this application and its application or use.

[0024] Techniques, systems, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, they should be considered part of the instruction manual.

[0025] In all the examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0026] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0027] Example 1 This embodiment proposes a method for the asset-based management and value assessment of data across the entire logistics chain, such as... Figure 1 As shown, it includes the following steps: S1. Obtain multi-source heterogeneous raw data from the entire logistics chain, perform knowledge extraction and fusion processing on the multi-source heterogeneous raw data from the entire logistics chain, and obtain the logistics data asset map base library.

[0028] Addressing the pain points of traditional logistics knowledge graphs, which focus on business entities and treat data as secondary attributes, thus failing to support asset management, we reconstruct the ontology modeling logic to achieve integrated design of asset management and value assessment from the source, solving the core problems of fragmented logistics data, inconsistent semantics, and missing asset attributes.

[0029] Furthermore, the specific content of S1 includes: Acquire multi-source heterogeneous raw data from the entire logistics chain, perform data classification, verification, and format normalization processing to obtain the raw data pool D. clean ; Build an asset-specific ontology system The asset-specific ontology system defines the top-level core entity logistics data asset E. core , c1 is the primary data asset, c2 is the business data asset, c3 is the time-series data asset, c4 is the unstructured data asset, and c5 is the compliant data asset; R is the set of relationships between entities (such as "belongs to", "belongs to", "transfers to", "contributes to", etc.); A is the set of attributes (including value attributes). , Basic attributes (asset ID, data type, etc.).

[0030] Business processes Ownership ,life cycle Compliance Management Four major related domains, with pre-embedded value assessment attribute A. value The valuation attributes include the initial value V0 and the business weight W. b Timing decay coefficient k; Understandably, the reconstructed ontology architecture innovatively constructs a logistics-specific ontology system with one core, four domains, and native value attributes. Logistics data assets are designated as the top-level, sole core entity, rather than business entities like orders and vehicles in traditional strategies. This fundamentally changes the underlying modeling logic of the graph. The core entity comprises five subcategories: master data assets, business data assets, time-series data assets, unstructured data assets, and compliance data assets. It natively embeds value assessment attributes such as initial value, quality value coefficient, compliance value coefficient, and business contribution weight. This completes the underlying binding of asset management and value assessment from the ontology construction stage. It also constructs four related entity domains: business processes, ownership entities, lifecycle, and compliance control. This binds each asset with its business origin, ownership, lifecycle nodes, and compliance requirements, achieving full-dimensional standardization of asset semantics.

[0031] Based on the asset-specific ontology system, knowledge extraction and processing of entities and relations are performed, and knowledge fusion and semantic unification are carried out to obtain triples; Based on the asset type and the relevant attributes of the business process to which it belongs, the asset nodes and relationships embedded with the initial value are obtained, resulting in a fusion triplet with the initial value. By writing the fusion triples with initial value into a graph database for materialized storage, a standardized, semantically unified, and natively embedded value logistics data asset graph base library is obtained.

[0032] S2. Based on the four-in-one lineage map, the logistics data asset map base database is processed by full-link graph traversal to obtain the asset full-link lineage traceability relationship and ownership boundary definition results.

[0033] Traditional data lineage mapping only covers the technical level of table-field associations, which cannot solve the problems of business flow tracing and ownership definition across logistics entities and links. This paper innovatively expands the coverage dimensions of the lineage map to achieve simultaneous tracing of asset links, value sources, and ownership entities.

[0034] Furthermore, the specific content of S2 includes: Edge relationship expansion processing is performed on the underlying database G0 of the data asset graph, adding technical lineage edges, business link edges, ownership edges, and value transfer edges to obtain a four-in-one lineage graph G. blood ; Four-in-one bloodline map G blood Perform graph traversal parameter configuration processing, set traversal depth and node filtering conditions, and obtain graph traversal execution steps adapted to logistics scenarios; Based on the graph traversal execution steps, a full-link graph traversal process is performed on each asset entity to trace the four major flow links of technology, business, ownership, and value, thereby obtaining the asset's full-link lineage tracing relationship T. race ; T for tracing the lineage of assets across the entire value chain race The ownership and ownership relationships of each asset are analyzed and processed to clarify the ownership subject, participating and contributing subject, and basic dimensions of contribution of each asset, thus obtaining the ownership boundary definition result O. T for tracing the lineage of assets across the entire value chain race The results of the ownership boundary delineation (O) are integrated and synchronously updated to the four-in-one bloodline map (G). blood This yields an updated graph G1 containing blood relations and ownership definitions, and outputs a blood tracing table and an ownership boundary definition table.

[0035] S3. Based on the asset lineage tracing relationship and ownership definition results, the graph's native quality inspection and compliance verification algorithm is used to carry out full life cycle governance processing to obtain highly reliable data assets that meet both quality and compliance standards, and simultaneously complete the dynamic calibration of the asset value coefficient. Furthermore, the specific content of S3 includes: T for tracing the lineage of assets across the entire value chain race Result of ownership boundary delineation O ={( E core, O i , d i The updated graph G1 is then used for data integration and core information extraction to obtain the full asset dataset D. base ; Based on the full asset dataset D base The assets undergo end-to-end verification, calculating integrity, accuracy, and consistency metrics. Low-quality asset nodes are then identified, resulting in the asset quality verification results and a list of low-quality assets (D). lowQ ; Integrity index Q c : ; Accuracy index Q a : ; Consistency index Qs: ; Quality value coefficient Q: ; in, To meet quality standards.

[0036] Based on the asset quality verification results and the list of low-quality assets D lowQ Low-quality assets are corrected and all quality indicators are re-verified to obtain a set of assets that meet quality standards. E Q ; Based on the collection of quality-compliant assets E Q By associating data with the compliance control domain in the graph G1, the assets undergo comprehensive verification processing for sensitive information and circulation compliance. This process calculates compliance indicators for sensitive information and circulation, marks non-compliant asset nodes, and yields the asset compliance verification results and a list of non-compliant assets, D. lowC ; Sensitive information compliance indicators: ; Circulation compliance indicators: ; Compliance Value Coefficient: ; in, To achieve compliance and standards.

[0037] Obtain asset compliance verification results and a list of non-compliant assets (D) lowC Non-compliant assets were rectified to meet compliance requirements, and all compliance indicators were re-verified to obtain an asset set that met both quality and compliance standards. E QC ; Based on the quality-value and compliance-value mapping rules, the quality value coefficient Q and compliance value coefficient C are calculated. The calculation results are labeled as asset value attributes, and the asset value attributes are dynamically calibrated to obtain highly reliable data assets that meet both quality and compliance standards and have calibrated value coefficients. E trust ; Will E trust Q, C are simultaneously updated to G1, completing the iterative processing of the graph data, resulting in the updated graph G2 = G1∪{Q,C,E}. trust It also outputs highly reliable data assets and value coefficients.

[0038] S4. Based on highly reliable data assets and full-link graph data, the GraphSAGE model is used to perform graph reasoning calculations to obtain real-time dynamic asset value scoring, hierarchical classification results and ownership value splitting strategies. Furthermore, the specific content of S4 includes: For highly reliable data assets E trust The basic dataset D is obtained by performing feature extraction on Q, C, and G2. model The underlying dataset D model This includes asset topology characteristics, business characteristics, quality compliance characteristics, and time-series characteristics; For the basic dataset D model The input feature vector X is obtained by standardization. t ,X b ,X qc ,X time The GraphSAGE model is used to perform graph inference calculations on the input feature vectors to obtain real-time dynamic asset value scores, thereby obtaining a set of real-time dynamic asset value scores V. all ; There are preset asset thresholds. Based on the real-time dynamic value score of the assets, the assets are classified and graded to obtain the asset classification result Grade={Grade1, Grade2, Grade3}. Based on the asset classification results, the real-time dynamic asset value score set, and the ownership boundary delineation results, the Shapley function is used to calculate the value contribution ratio of each ownership entity. Strategies for Decomposing Ownership Value ; The asset real-time dynamic value score set V all Asset classification results (Grade) and ownership value splitting strategy The graph is synchronously updated to graph G2 to complete the graph iteration process, resulting in the updated graph G3 = G2∪{V, Grade, V i It also outputs an asset dynamic value scoring table, a classification table, and an ownership value breakdown table.

[0039] Furthermore, the expression for the real-time dynamic value score of an asset is as follows: ; Where V is the real-time dynamic asset value score, V0 is the initial value, Q is the quality value coefficient, C is the compliance value coefficient, and B is the business value-added coefficient (B=1+0.1). Call frequency T is the time decay factor. t is the asset duration, and k is the decay coefficient (k=0.05 for time-series assets, k=0.01 for other assets). The expression for the value contribution ratio of each ownership entity is as follows: ; Where N is the set of all ownership entities, and S is a subset that does not include entity i. Let S be the asset value corresponding to subset S. The percentage of value contribution of subject i; ; The value due to subject i .

[0040] S5. Based on the dynamic asset value scoring, hierarchical classification results and ownership splitting strategy, the hierarchical control rule engine of the graph is used to perform differentiated management strategy matching and processing, so as to obtain the full life cycle control strategy and permission configuration strategy that are adapted to the asset value level. Furthermore, the specific content of S5 includes: Real-time dynamic value scoring set V of assets all Asset classification results (Grade) and ownership value splitting strategy The information extraction and processing of the graph G3 yields the value characteristics, ownership entities, and contribution ratios of assets at each level, resulting in the hierarchical management and control basic dataset D. control ; Based on the graph-based hierarchical control rule engine, combined with the needs of logistics data asset management, differentiated control rules are configured for assets of different levels, and a hierarchical control rule system Rule={Rule1, Rule2, Rule3} is constructed to obtain control rule strategies adapted to each asset level. Based on hierarchical control rules and strategies and hierarchical control basic dataset D control The storage strategy is matched with different asset levels to obtain the asset-tiered storage strategy Store. rule ; Store based on hierarchical storage strategyrule Ownership value splitting strategy Based on the asset classification results (Grade), a three-tier permission system (subject-asset-permission) is constructed, and access, usage, and revenue permissions for each subject are configured to obtain the permission configuration strategy P. erm ={(O i E core P ij ) }, P ij Access level; Based on permission configuration policy P erm Real-time dynamic asset value scoring set V all By combining the time-series decay factor T, archiving and decommissioning thresholds are set for assets of different levels, triggering full lifecycle management actions, thus obtaining the asset full lifecycle management system Life. control ; Asset Tiered Storage Strategy Store rule Permission configuration strategy (Perm) and full lifecycle management system (Life) control The system is integrated to obtain a full lifecycle management and control system and permission configuration strategy. At the same time, the management and control rules and permission information are updated to graph G3, completing the graph iteration and obtaining the updated graph G4 = G3∪{Rule, Perm, V}. new}, V new Based on the value score obtained after the feedback on the control effect, output a control system strategy and permission configuration table.

[0041] It is understandable that a value level-control strategy-mapping rule system is constructed, which automatically matches differentiated storage strategies, access permissions, circulation rules and operation strategies based on the dynamic value scoring and classification results of assets (core / important / general assets), to replace the traditional fixed-level control model. Based on the ownership splitting strategy, a three-level permission graph of subject-asset-permission is constructed, which automatically matches the access, use and revenue permissions of each subject with the corresponding value share, so as to realize the refined permission control of cross-subject assets. Based on the asset value decay cycle, lifecycle node management technology automatically triggers the archiving, offline, and update processes of assets to ensure the value density of the asset pool; The results of the control strategy are written back to the asset attributes in the graph in real time, which in turn affects the dynamic value score of the asset, forming a two-way linkage between control and value.

[0042] S6. Obtain high-value data assets and business scenario requirements after hierarchical management and control, and perform scenario-based asset matching processing based on graph semantic retrieval and association recommendation algorithms to obtain asset application strategies that meet business requirements. Understandably, building a business scenario-asset value association model involves pre-constructing the association between scenarios and asset types and value weights based on the value requirements of core logistics business scenarios (route optimization, inventory forecasting, network planning, supply chain finance, etc.). Using semantic retrieval and subgraph matching algorithms, combined with dynamic asset value scoring, it accurately matches high-value, highly adaptable data assets to business scenarios, replacing traditional keyword retrieval and full data retrieval. Based on association recommendation algorithms, it automatically mines high-value derivative assets strongly associated with scenario requirements, improving asset reuse rate and value release efficiency. The frequency and effectiveness of asset business applications are written back to the asset attributes of the graph in real time, synchronously updating the asset's business contribution weight and dynamic value score.

[0043] S7. Obtain incremental business data and value realization results generated during asset business applications, and perform closed-loop iterative processing based on the incremental update and value feedback mechanism of the asset map to obtain the updated asset map base library and optimized value assessment model parameters.

[0044] Example 2 This application proposes a system for the management and value assessment of data assets across the entire logistics chain, such as... Figure 2 As shown, it includes: Data acquisition unit: Acquires multi-source heterogeneous raw data from the entire logistics chain, performs knowledge extraction and fusion processing based on the innovatively constructed asset-specific ontology system, and obtains a logistics data asset map base library; The graph construction unit acquires a standardized logistics data asset graph base library, performs full-link graph traversal processing based on the innovatively constructed four-in-one lineage graph, obtains the asset full-link lineage traceability relationship and ownership boundary definition results, performs full life cycle governance processing based on the graph's native quality inspection and compliance verification algorithms, obtains highly reliable data assets that meet both quality and compliance standards, and simultaneously completes the dynamic calibration of the asset value coefficient. Value scoring configuration unit: acquires high-reliability data assets and full-link graph data after quality compliance calibration, performs graph inference calculation based on the targeted optimization incremental GraphSAGE model to obtain real-time dynamic asset value scoring, hierarchical classification results and ownership value splitting strategy, and performs differentiated management strategy matching processing based on the graph-based hierarchical control rule engine to obtain full life cycle control strategy and permission configuration strategy adapted to asset value level. Strategy generation unit: Acquires high-value data assets and business scenario requirements after hierarchical management and control, performs scenario-based asset matching processing based on graph semantic retrieval and association recommendation algorithms, and obtains asset application strategies that adapt to business requirements; Model optimization unit: Acquires incremental business data and value realization results generated during asset business applications, performs closed-loop iterative processing based on the incremental update of the asset map and the value feedback mechanism, and obtains the updated asset map base library and the optimized value assessment model parameters.

[0045] An electronic device includes a memory and a processor, wherein the memory stores a computer program, and the processor invokes the computer program in the memory to implement a method for the management and valuation of data assets across the entire logistics chain.

[0046] A storage medium storing computer-executable instructions, which, when loaded and executed by a processor, realize a method for data asset management and value assessment across the entire logistics chain.

[0047] Finally, it should be noted that the above embodiments are only used to illustrate the technical methods of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical methods of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical methods to deviate from the spirit and scope of the technical methods of the present invention.

Claims

1. A method for the management and value assessment of data assets across the entire logistics chain, characterized in that, Includes the following steps: S1. Obtain multi-source heterogeneous raw data from the entire logistics chain, perform knowledge extraction and fusion processing on the multi-source heterogeneous raw data from the entire logistics chain, and obtain the logistics data asset map base library. S2. Construct a four-in-one lineage map, and perform full-link graph traversal processing on the logistics data asset map base database based on the four-in-one lineage map to obtain the asset full-link lineage traceability relationship and ownership boundary definition results. S3. Based on the asset lineage tracing relationship and ownership definition results, the graph's native quality inspection and compliance verification algorithm is used to carry out full life cycle governance processing to obtain highly reliable data assets that meet both quality and compliance standards, and simultaneously complete the dynamic calibration of the asset value coefficient. S4. Based on highly reliable data assets and full-link graph data, the GraphSAGE model is used to perform graph reasoning calculations to obtain real-time dynamic asset value scoring, hierarchical classification results and ownership value splitting strategies. S5. Based on the dynamic asset value scoring, hierarchical classification results and ownership splitting strategy, the hierarchical control rule engine of the graph is used to perform differentiated management strategy matching and processing, so as to obtain the full life cycle control strategy and permission configuration strategy that are adapted to the asset value level. S6. Obtain high-value data assets and business scenario requirements after hierarchical management and control, and perform scenario-based asset matching processing based on graph semantic retrieval and association recommendation algorithms to obtain asset application strategies that meet business requirements. S7. Obtain incremental business data and value realization results generated during asset business applications, and perform closed-loop iterative processing based on the incremental update and value feedback mechanism of the asset map to obtain the updated asset map base library and optimized value assessment model parameters.

2. The method for managing and evaluating the value of end-to-end logistics data assets according to claim 1, characterized in that, The specific content of S1 includes: Acquire multi-source heterogeneous raw data from the entire logistics chain, perform data classification, verification, and format normalization processing to obtain the raw data pool D. clean ; Construct an asset-specific ontology system, wherein the asset-specific ontology system defines the top-level core entity logistics data asset E. core The four related domains are business processes, ownership entities, lifecycle, and compliance control, and the value assessment attribute A is pre-embedded. value The valuation attributes include the initial value V0 and the business weight W. b Timing decay coefficient k; Based on the asset-specific ontology system, knowledge extraction and processing of entities and relations are performed, and knowledge fusion and semantic unification are carried out to obtain triples; Based on the asset type and the relevant attributes of the business process to which it belongs, the asset nodes and relationships embedded with the initial value are obtained, resulting in a fusion triplet with the initial value. By writing the fusion triples with initial value into a graph database for materialized storage, a standardized, semantically unified, and natively embedded value logistics data asset graph base library is obtained.

3. The method for managing and evaluating the value of end-to-end logistics data assets according to claim 2, characterized in that, The specific content of S2 includes: Edge relationship expansion processing is performed on the underlying database G0 of the data asset graph, adding technical lineage edges, business link edges, ownership edges, and value transfer edges to obtain a four-in-one lineage graph G. blood ; Four-in-one bloodline map G blood Perform graph traversal parameter configuration processing, set traversal depth and node filtering conditions, and obtain graph traversal execution steps adapted to logistics scenarios; Based on the graph traversal execution steps, a full-link graph traversal process is performed on each asset entity to trace the four major flow links of technology, business, ownership, and value, thereby obtaining the asset's full-link lineage tracing relationship T. race ; T for tracing the lineage of assets across the entire value chain race The ownership and ownership relationships of each asset are analyzed and processed to clarify the ownership subject, participating and contributing subject, and basic dimensions of contribution of each asset, thus obtaining the ownership boundary definition result O. T for tracing the lineage of assets across the entire value chain race The results of the ownership boundary delineation (O) are integrated and synchronously updated to the four-in-one bloodline map (G). blood This yields an updated graph G1 containing blood relations and ownership definitions, and outputs a blood tracing table and an ownership boundary definition table.

4. The method for managing and evaluating the value of end-to-end logistics data assets according to claim 2, characterized in that, The specific contents of S3 include: T for tracing the lineage of assets across the entire value chain race Result of ownership boundary delineation O ={( E core , O i , d i The updated graph G1 is then used for data integration and core information extraction to obtain the full asset dataset D. base ; Based on the full asset dataset D base The assets undergo end-to-end verification, calculating integrity, accuracy, and consistency metrics. Low-quality asset nodes are then identified, resulting in the asset quality verification results and a list of low-quality assets (D). lowQ ; Based on the asset quality verification results and the list of low-quality assets D lowQ Low-quality assets are corrected and all quality indicators are re-verified to obtain a set of assets that meet quality standards. E Q ; Based on the collection of quality-compliant assets E Q By associating data with the compliance control domain in the graph G1, the assets undergo comprehensive verification processing for sensitive information and circulation compliance. This process calculates compliance indicators for sensitive information and circulation, marks non-compliant asset nodes, and yields the asset compliance verification results and a list of non-compliant assets, D. lowC ; Obtain asset compliance verification results and a list of non-compliant assets (D) lowC Non-compliant assets were rectified to meet compliance requirements, and all compliance indicators were re-verified to obtain an asset set that met both quality and compliance standards. E QC ; Based on the quality-value and compliance-value mapping rules, the quality value coefficient Q and compliance value coefficient C are calculated. The calculation results are marked into the asset value attribute, and the asset value attribute is dynamically calibrated to obtain a highly reliable data asset that meets both quality and compliance standards and has its value coefficient calibrated. E trust ; Will E trust Q, C are simultaneously updated to G1, completing the iterative processing of the graph data, resulting in the updated graph G2 = G1∪{Q,C,E}. trust It also outputs highly reliable data assets and value coefficients.

5. The method for managing and evaluating the value of end-to-end logistics data assets according to claim 4, characterized in that, The specific content in S4 includes: For highly reliable data assets E trust The basic dataset D is obtained by performing feature extraction on Q, C, and G2. model The underlying dataset D model This includes asset topology characteristics, business characteristics, quality compliance characteristics, and time-series characteristics; For the basic dataset D model The input feature vector X is obtained by standardization. t ,X b ,X qc ,X time The GraphSAGE model is used to perform graph inference calculations on the input feature vectors to obtain real-time dynamic asset value scores, thereby obtaining a set of real-time dynamic asset value scores V. all ; There are preset asset thresholds. Based on the real-time dynamic value score of the assets, the assets are classified and graded to obtain the asset classification result Grade={Grade1, Grade2, Grade3}. Based on the asset classification results, the real-time dynamic asset value score set, and the ownership boundary delineation results, the Shapley function is used to calculate the value contribution ratio of each ownership entity. Strategies for Decomposing Ownership Value ; The asset real-time dynamic value score set V all Asset classification results (Grade) and ownership value splitting strategy The graph is synchronously updated to graph G2 to complete the graph iteration process, resulting in the updated graph G3 = G2∪{V, Grade, V i It also outputs an asset dynamic value scoring table, a classification table, and an ownership value breakdown table.

6. The method for managing and assessing the value of end-to-end logistics data assets according to claim 5, characterized in that, The specific content of S5 includes: Real-time dynamic value scoring set V of assets all Asset classification results (Grade) and ownership value splitting strategy The information extraction and processing of the graph G3 yields the value characteristics, ownership entities, and contribution ratios of assets at each level, resulting in the hierarchical management and control basic dataset D. control ; Based on the graph-based hierarchical control rule engine, combined with the needs of logistics data asset management, differentiated control rules are configured for assets of different levels, and a hierarchical control rule system Rule={Rule1, Rule2, Rule3} is constructed to obtain control rule strategies adapted to each asset level. Based on hierarchical control rules and strategies and hierarchical control basic dataset D control The storage strategy is matched with different asset levels to obtain the asset-tiered storage strategy Store. rule ; Store based on hierarchical storage strategy rule Ownership value splitting strategy Based on the asset classification results (Grade), a three-tier permission system (subject-asset-permission) is constructed, and access, usage, and revenue permissions for each subject are configured to obtain the permission configuration strategy P. erm ={(O i E core P ij ) }, P ij Access level; Based on permission configuration policy P erm Real-time dynamic asset value scoring set V all By combining the time-series decay factor T, archiving and decommissioning thresholds are set for assets of different levels, triggering full lifecycle management actions, thus obtaining the asset full lifecycle management system Life. control ; Asset Tiered Storage Strategy Store rule Permission configuration strategy (Perm) and full lifecycle management system (Life) control The system is integrated to obtain a full lifecycle management and control system and permission configuration strategy. At the same time, the management and control rules and permission information are updated to graph G3, completing the graph iteration and obtaining the updated graph G4 = G3∪{Rule, Perm, V}. new }, V new Based on the value score obtained after the feedback on the control effect, output a control system strategy and permission configuration table.

7. The method for managing and evaluating the value of end-to-end logistics data assets according to claim 5, characterized in that, The expression for the real-time dynamic value score of an asset is: ； Where V is the real-time dynamic asset value score, V0 is the initial value, Q is the quality value coefficient, C is the compliance value coefficient, B is the business value-added coefficient, and T is the time-series decay factor. t is the asset's duration, and k is the decay coefficient; The expression for the value contribution ratio of each ownership entity is as follows: ； Where N is the set of all ownership entities, and S is a subset that does not include entity i. Let S be the asset value corresponding to subset S. The percentage of value contribution of subject i; ； The value due to subject i .

8. A system for managing and evaluating the value of data assets across the entire logistics chain, characterized in that, include: Data acquisition unit: Acquires multi-source heterogeneous raw data from the entire logistics chain, performs knowledge extraction and fusion processing based on the innovatively constructed asset-specific ontology system, and obtains a logistics data asset map base library; The graph construction unit acquires a standardized logistics data asset graph base library, performs full-link graph traversal processing based on the innovatively constructed four-in-one lineage graph, obtains the asset full-link lineage traceability relationship and ownership boundary definition results, performs full life cycle governance processing based on the graph's native quality inspection and compliance verification algorithms, obtains highly reliable data assets that meet both quality and compliance standards, and simultaneously completes the dynamic calibration of the asset value coefficient. Value scoring configuration unit: acquires high-reliability data assets and full-link graph data after quality compliance calibration, performs graph inference calculation based on the targeted optimization incremental GraphSAGE model to obtain real-time dynamic asset value scoring, hierarchical classification results and ownership value splitting strategy, and performs differentiated management strategy matching processing based on the graph-based hierarchical control rule engine to obtain full life cycle control strategy and permission configuration strategy adapted to asset value level. Strategy generation unit: Acquires high-value data assets and business scenario requirements after hierarchical management and control, performs scenario-based asset matching processing based on graph semantic retrieval and association recommendation algorithms, and obtains asset application strategies that adapt to business requirements; Model optimization unit: Acquires incremental business data and value realization results generated during asset business applications, performs closed-loop iterative processing based on the incremental update of the asset map and the value feedback mechanism, and obtains the updated asset map base library and the optimized value assessment model parameters.

9. An electronic device, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program, and the processor, when calling the computer program in the memory, implements the content of the logistics end-to-end data asset management and value assessment method as described in any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium stores computer-executable instructions, which, when loaded and executed by a processor, implement the content of the logistics end-to-end data asset management and value assessment method as described in any one of claims 1 to 7.

Images