A supply chain data quality automatic checking and closed-loop management method and system

Abstract

The application discloses a kind of supply chain data quality automatic check and closed loop management method and system, it is related to supply chain digital technology field, including the following steps: the supply chain authority source list obtained is input into the preset check rule set, according to rule check result extraction problem field set;Problem field set is injected into DCAN responsibility mapping mechanism, and the ability matrix is output through the pre-constructed multi-modal responsibility intelligent agent, and the responsibility proportion weight of responsibility subject is calculated based on ability matrix;The responsibility proportion weight is input into responsibility allocation decision model, and the allocation responsibility of corresponding responsibility subject is output;According to allocation responsibility, dispatch problem rectification task to corresponding responsibility subject;If it is detected that rectification fails, then extract rectification failure field set, and according to the preset responsibility transmission path, rectification failure field set is transmitted to secondary responsibility subject, and the problem is solved until execution cycle is performed.This application is used to solve the problem that data responsibility attribution is not clear, rectification task is dispatched, and there is lack of tracking after rectification process is broken.

Description

Technical Field

[0001] This invention relates to the field of supply chain digitalization technology, and more specifically, to a method and system for automatic verification and closed-loop management of supply chain data quality. Background Technology

[0002] As enterprises undergo increasingly profound digital transformation, supply chain management systems are becoming more complex, encompassing multiple business systems. Modern supply chain management relies on the collaborative work of multi-source heterogeneous information systems, generating massive amounts of structured and semi-structured data. This data permeates procurement, manufacturing, warehousing, transportation, and after-sales processes, forming the foundation for inventory management, order fulfillment, cost control, and risk monitoring. To ensure the correctness of business decisions and automated processes, it is essential to verify and govern the quality attributes of supply chain data, including its integrity, accuracy, consistency, and timeliness. Current supply chain data quality management methods include: improving data quality during the data access phase through transformation, mapping, and standardization; recording metadata, lineage relationships, and responsible parties, providing viewing and auditing functions; utilizing historical distribution, time-series models, or supervised / unsupervised algorithms to detect abnormal records or events; and manually analyzing anomalies, identifying responsibilities, and driving remediation, often relying on cross-departmental coordination.

[0003] Currently, the problems in supply chain data quality management include: reliance on manual spot checks or static verification methods; data from various business systems being scattered and lacking unified standards and authoritative data sources; unclear data responsibility; and insufficient adaptive and intelligent capabilities, resulting in low supply chain data quality and hindering the digital transformation and operational efficiency improvement of the supply chain.

[0004] To address the above problems, this invention proposes a solution. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a method and system for automatic verification and closed-loop governance of supply chain data quality. By constructing an authoritative source list, adopting the DCAN responsibility mapping mechanism and responsibility allocation decision model, and combining the responsibility transmission path, the method solves the problems of unclear data responsibility attribution in various business systems, lack of tracking after the assignment of rectification tasks, and broken rectification processes.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] An automatic verification and closed-loop governance method for supply chain data quality includes the following steps: inputting an acquired authoritative source list of the supply chain into a preset verification rule set, and extracting a problem field set based on the rule verification results; injecting the problem field set into the DCAN responsibility mapping mechanism, outputting a capability matrix through a pre-constructed multimodal responsibility agent, and calculating the responsibility proportion weight of the responsible entity based on the capability matrix; inputting the responsibility proportion weight into the responsibility allocation decision model, and outputting the allocated responsibility of the corresponding responsible entity; dispatching problem rectification tasks to the corresponding responsible entity according to the allocated responsibility; if rectification failure is detected, extracting the rectification failure field set, and transmitting the rectification failure field set to the secondary responsible entity according to the preset responsibility transmission path, executing the loop until the problem is resolved.

[0008] In a preferred embodiment, the step of inputting the acquired list of authoritative supply chain sources into a preset set of verification rules and extracting a set of problematic fields based on the rule verification results specifically involves: fusing the acquired multi-source supply chain system data, calculating the field similarity of the fused system data, and generating a standard dictionary based on the similarity clustering results; calculating the data source authority score based on the standard dictionary and filtering to obtain the list of authoritative sources; inputting the list of authoritative sources into the preset set of rule verification for verification, generating a data quality score, and when the data quality score is lower than a preset authority threshold, locating problematic fields and constructing a set of problematic fields.

[0009] In a preferred embodiment, the step of injecting the problem field set into the DCAN responsibility mapping mechanism, outputting a capability matrix through a pre-built multimodal responsibility agent, and calculating the responsibility proportion weight of the responsible subject based on the capability matrix specifically involves: inputting the problem field set into the pre-built multimodal responsibility agent; outputting the capability matrix of the responsible subject based on the multimodal responsibility agent, wherein the capability matrix contains the processing capability parameters of each subject for the field type; calculating the subject effectiveness value of the responsible subject in processing the problem field according to a preset subject effectiveness formula and combining the parameters in the capability matrix, and calculating the marginal contribution value based on the effectiveness value; calculating the Shapley value based on the marginal contribution value, and normalizing the Shapley value of each subject by field to obtain the responsibility proportion weight.

[0010] In a preferred embodiment, the construction of the multimodal responsible agent specifically involves: extracting features from cross-system metadata to obtain a domain vector and a subject vector; constructing an initial capability matrix based on the matching degree between the domain vector and the subject vector; and constructing an agent template based on the domain vector and the capability matrix, combined with the identity kernel, reputation value, and historical task rectification pass rate.

[0011] In a preferred embodiment, the step of inputting the responsibility proportion weight into the responsibility allocation decision model and outputting the allocated responsibility of the corresponding responsible entity specifically involves: inputting the responsibility proportion weight, the entity's technical capability strength, the penalty amount, and the repair workload into the responsibility allocation decision model, wherein the entity's technical capability strength is calculated based on the capability matrix; based on the entity's technical capability strength and the responsibility proportion weight, the repair responsibility is obtained by calculating the normalized allocation of the repair workload; and based on the responsibility proportion weight, the penalty responsibility is obtained by weighted allocation of the penalty amount.

[0012] In a preferred embodiment, if rectification failure is detected, the rectification failure field set is extracted and transmitted to the secondary responsible entity according to a preset responsibility transmission path. This process is repeated until the problem is resolved. Specifically, the following steps are taken: the rectification pass rate is calculated based on the task assignment results and rectification feedback data; when the rectification pass rate is lower than a preset threshold, the rectification failure field set is extracted from the rectification feedback data; the cross-domain responsibility transmission weight of the rectification failure fields is calculated to obtain the responsibility transmission path; based on the responsibility transmission path, the rectification failure fields are re-entered into the DCAN responsibility mapping mechanism to be assigned to a new responsible entity, until the re-rectified rectification failure fields pass the verification rule set review and the data quality score is not lower than the authority threshold.

[0013] In a preferred embodiment, the multimodal responsible agent further includes an agent evolution mechanism, specifically: based on the rectification pass rate, the reputation increment of the responsible entity is calculated through a reputation increment calculation model, and the reputation value is dynamically updated in combination with the decay index; a prediction loss function is calculated based on the rectification pass rate and the predicted rectification pass rate, and the capability matrix parameters are adjusted through gradient descent, wherein the predicted rectification pass rate is calculated from the reputation value; and the capability matrix of similar domains is transferred to the newly added domain according to the constructed cross-domain capability transfer formula.

[0014] A system for automatic verification and closed-loop governance of supply chain data quality includes: a problem extraction module, used to input the acquired authoritative source list of the supply chain into a preset verification rule set, and extract a problem field set based on the rule verification results;

[0015] The responsibility mapping module is used to inject the problem field set into the DCAN responsibility mapping mechanism, output the capability matrix through the pre-built multimodal responsibility agent, and calculate the responsibility proportion weight of the responsible subject based on the capability matrix;

[0016] The responsibility allocation module is used to input the responsibility proportion weights into the responsibility allocation decision model and output the allocated responsibilities of the corresponding responsible entities.

[0017] The task assignment module is used to assign problem rectification tasks to the corresponding responsible parties based on the assigned responsibilities.

[0018] The closed-loop governance module is used to extract the set of rectification failure fields if rectification failure is detected, and to pass the set of rectification failure fields to the secondary responsible entities according to the preset responsibility transmission path, and execute the loop until the problem is resolved.

[0019] The technical effects and advantages of the automatic verification and closed-loop management method and system for supply chain data quality of this invention are as follows:

[0020] This invention constructs reliable and authoritative source data by fusing multi-source data. Employing a DCAN responsibility mapping mechanism and responsibility allocation decision model, it assigns data responsibility to relevant responsible parties. Combined with responsibility transmission paths, it ensures that in the event of rectification failure, responsibility can be effectively transferred to relevant areas for redistribution, avoiding blind spots in data governance. It effectively solves the problems of unclear data governance responsibility and delayed rectification of data quality issues in existing technologies. It can effectively improve the digital transformation and operational efficiency of the supply chain. Through automated quality scoring and problem reporting mechanisms, coupled with dynamic reputation adjustment and intelligent task allocation, it significantly improves the efficiency and accuracy of data governance. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of a method for automatic verification and closed-loop management of supply chain data quality provided in an embodiment of the present invention.

[0022] Figure 2 This is a schematic diagram of the system structure of an automatic verification and closed-loop management method for supply chain data quality provided in an embodiment of the present invention.

[0023] Figure 3 A logic diagram of an automatic verification and closed-loop management method for supply chain data quality provided in an embodiment of the present invention. Detailed Implementation

[0024] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] Example 1, Figure 1 This invention presents an automatic verification and closed-loop management method for supply chain data quality, comprising the following steps:

[0026] S1. Input the obtained authoritative source list of the supply chain into the preset verification rule set, and extract the problem field set based on the rule verification results;

[0027] S2, inject the problem field set into the DCAN responsibility mapping mechanism, output the capability matrix through the pre-built multimodal responsibility agent, and calculate the responsibility proportion weight of the responsible subject based on the capability matrix;

[0028] S3, input the responsibility proportion weight into the responsibility allocation decision model, and output the allocation responsibility of the corresponding responsible subject;

[0029] S4. Based on the assigned responsibilities, issue problem rectification tasks to the corresponding responsible entities;

[0030] S5. If rectification failure is detected, the rectification failure field set is extracted and transmitted to the secondary responsible entity according to the preset responsibility transmission path. The loop continues until the problem is resolved.

[0031] This embodiment integrates multi-source data to construct reliable and authoritative source data. Employing the DCAN responsibility mapping mechanism and responsibility allocation decision model, data responsibility is assigned to relevant responsible parties. Combined with responsibility transmission paths, it ensures that in the event of rectification failure, responsibility can be effectively transferred to relevant areas for redistribution, avoiding blind spots in data governance. This effectively solves the problems of unclear data governance responsibility and delayed rectification of data quality issues in existing technologies. It can effectively improve the digital transformation and operational efficiency of the supply chain. Through automated quality scoring and problem reporting mechanisms, coupled with dynamic reputation adjustment and intelligent task allocation, it significantly improves the efficiency and accuracy of data governance.

[0032] S1: Input the obtained authoritative source list of the supply chain into the preset verification rule set, and extract the problem field set based on the rule verification results.

[0033] In this embodiment, the step of inputting the obtained authoritative supply chain source list into a preset verification rule set and extracting the problem field set based on the rule verification results is specifically as follows:

[0034] The system integrates multi-source supply chain system data, calculates the field similarity of the integrated system data, and generates a standard dictionary based on the similarity clustering results.

[0035] The authority score of the data source is calculated based on the standard dictionary, and a list of authoritative sources is obtained by filtering.

[0036] The authoritative source list is input into the preset rule verification set for verification, and a data quality score is generated. When the data quality score is lower than the preset authoritative threshold, the problem fields are located and a problem field set is constructed.

[0037] In this embodiment, data from multiple supply chain systems is integrated, the field similarity of the integrated data is calculated, and highly similar fields are merged to generate a standard dictionary. Specifically:

[0038] By connecting multiple supply chain business systems through an API gateway, data is extracted from each system, field information from different data sources is merged, and the similarity of fields from different data sources is calculated. Greater than the preset similarity threshold When clustering fields into a unified standard field, called a standard dictionary. .

[0039] The specific formula for data fusion is as follows:

[0040]

[0041] In the formula, For the integrated multi-source supply chain system data; For field alignment operators; To indicate the first A collection of fields from a data source; For the first Data content from each data source.

[0042] The specific formula for field similarity is as follows:

[0043]

[0044] In the formula, For fields and Similarity; , For fields and The corresponding value range; , For fields and Data types, For field type matching degree.

[0045] In this embodiment, the authority score of the data source is calculated based on the standard dictionary, and data sources that exceed the threshold of high-scoring data sources are selected to form an authoritative source list.

[0046] The formula for scoring the authority of a data source is as follows:

[0047]

[0048] In the formula, As a data source, Assess the authority of the data source. Scoring for data integrity Scoring for field consistency Scoring based on data timeliness, , , These are the weighting coefficients, and .

[0049] When the data source is authoritatively rated At that time, retain the data source in the list of authoritative sources, among which The threshold for high-scoring data sources.

[0050] In this embodiment, the authoritative source list is input into a preset rule verification set for verification, and a data quality score is generated, specifically as follows:

[0051] The specific formula for rule validation is as follows:

[0052]

[0053] In the formula, Authoritative source list data Regarding the rules The verification result.

[0054] The specific formula for data quality scoring is as follows:

[0055]

[0056] In the formula, Score the data quality. This represents the total number of data entries in the authoritative source list. The total number of rules in the preset rule validation set. This is the verification result of the authoritative source list data.

[0057] In this embodiment, when the data quality score is lower than a preset authority threshold, the problematic fields are located and a problem field set is constructed, specifically as follows:

[0058] The preset authority threshold is ,like The system triggered an alarm, indicating that attention and action are needed;

[0059] Locate the violation fields in the rule validation and generate a set of violation fields.

[0060] It should be noted that a problem field refers to a field that is determined to have data anomalies, errors, or non-compliance with standard specifications.

[0061] S2 injects the problem field set into the DCAN responsibility mapping mechanism, outputs the capability matrix through the pre-built multimodal responsibility agent, and calculates the responsibility proportion weight of the responsible subject based on the capability matrix.

[0062] In this embodiment, the step of injecting the problem field set into the DCAN responsibility mapping mechanism, outputting a capability matrix through a pre-built multimodal responsibility agent, and calculating the responsibility proportion weight of the responsible subject based on the capability matrix is ​​as follows:

[0063] Input the problem field set into the pre-built multimodal responsible agent;

[0064] The multimodal responsible intelligent agent outputs a capability matrix of the responsible subjects, which includes the processing capability parameters of each subject for field types;

[0065] Based on the preset subject effectiveness formula, the subject effectiveness value of the responsible subject's problem-handling field is calculated by combining the parameters in the capability matrix, and the marginal contribution value is calculated based on the effectiveness value.

[0066] The Shapley value is calculated based on the marginal contribution value, and the Shapley values ​​of each entity are normalized by field to obtain the responsibility proportion weight.

[0067] In this embodiment, the construction of the multimodal responsible intelligent agent is specifically as follows:

[0068] Feature extraction is performed on cross-system metadata to obtain domain vectors and subject vectors;

[0069] Construct an initial capability matrix based on the matching degree between the domain vector and the subject vector;

[0070] Based on the domain vector and capability matrix, an intelligent agent template is constructed by combining the identity core, reputation value, and historical task rectification pass rate.

[0071] In this embodiment, metadata is extracted from the HR system, permission system, and operation logs. Feature filtering is performed on the metadata to remove redundant or low-relevance features. Metadata features are extracted by statistically analyzing the "operation type" in the logs, the "skill tags" in the HR system, and the "role name" in the permission system, resulting in domain vectors and subject vectors. A capability matrix is ​​then constructed based on the matching degree between the domain vectors and the subject vectors, as detailed below:

[0072] For each field The domain vector is obtained by counting the number of occurrences of each label in the domain.

[0073] The domain vector is specifically:

[0074] ,

[0075] in This represents the number of times each tag appears in this field.

[0076] For each subject The main vector is obtained by counting the number of times the associated tags appear.

[0077] The main vector is specifically: ,

[0078] in This represents the number of times the tag is related to the subject.

[0079] It should be noted that the domain vector reflects the importance of an operation in that domain by counting the number of associations between the domain and the operation label representing that domain;

[0080] The subject vector determines the frequency of association between the subject and the operation by statistically analyzing the number of times the responsible subject is associated with the operation tag, thus reflecting the frequency of the responsible subject's use of the operation.

[0081] The specific formula for constructing the capability matrix is ​​as follows: ,

[0082] in, This is a capability matrix. This represents the matching degree between the domain vector and the subject vector.

[0083] In this embodiment, an intelligent agent template is constructed using an identity core, reputation value, domain vector, capability matrix, and historical task rectification pass rate. The identity core is obtained through a CA digital certificate authentication system.

[0084] The specific agent template is as follows:

[0085]

[0086] in, As an agent template, For identity verification, For reputation value, For the domain vector, This is a capability matrix. The pass rate for rectification of historical tasks.

[0087] In this embodiment, the subject effectiveness value of the responsible subject handling problem field is calculated based on the preset subject effectiveness formula and the parameters in the capability matrix, and the marginal contribution value is calculated based on the effectiveness value.

[0088] Shapley values ​​are calculated based on marginal contribution values, and the Shapley values ​​of each entity are normalized by field to obtain the responsibility proportion weights, specifically:

[0089] Set of question fields Input a multimodal responsible intelligent agent, the set of responsible entities is: ;

[0090] For each question field Identify the task type associated with each question field. For each subject, extract its capability vector. ,and , This is a capability matrix;

[0091] The subject effectiveness value of the responsible subject's problem-handling field is calculated based on the preset subject effectiveness formula;

[0092] Calculate the responsible party subset of responsible entities In the question field Marginal contribution;

[0093] Based on the marginal contribution value, for each responsible entity and problem fields Calculate the Shapley value;

[0094] Based on the Shapley value, generate the responsibility proportion weight. .

[0095] The specific formula for the main body's effectiveness is as follows:

[0096]

[0097] In the formula, Subset of responsible entities For the question field The main performance value, A subset of the responsible entities, For the question field The importance weight of the business As the responsible party, It is the L2 norm. This is the capability vector.

[0098] It should be noted that the subject effectiveness value refers to the overall performance of the responsible subject in handling the problem field.

[0099] The specific formula for marginal contribution is as follows:

[0100]

[0101] In the formula, This is the marginal contribution value.

[0102] The specific formula for calculating the Shapley value is as follows:

[0103]

[0104] In the formula, Shapley value, The total number of responsible entities, A subset of the responsible entities, To remove the responsible party The complete list of responsible parties.

[0105] The specific formula for calculating the responsibility ratio weighting is as follows:

[0106] ,

[0107] in

[0108] In the formula, as the main body For the question field Responsibility proportion weighting Shapley value, This is the sum of the Shapley values ​​for all subjects.

[0109] S3 inputs the responsibility proportion weights into the responsibility allocation decision model and outputs the allocation responsibility of the corresponding responsible subject.

[0110] In this embodiment, the step of inputting the responsibility proportion weights into the responsibility allocation decision model and outputting the allocated responsibility of the corresponding responsible subject specifically involves:

[0111] The responsibility allocation decision model is input with the responsibility ratio weight, the main technical capability strength, the penalty amount and the repair workload. The main technical capability strength is calculated based on the capability matrix.

[0112] Based on the strength of the main technical capabilities and the weight of the responsibility ratio, the repair responsibility is obtained by calculating the normalized allocation of the repair workload;

[0113] Based on the weighted proportion of responsibility, the penalty amount is weighted and allocated to determine the penalty responsibility.

[0114] It should be noted that the strength of the main technical capabilities is calculated based on the capability matrix, the actual repair workload, and the dynamic complexity factor; the penalty amount is obtained through the contract management system and the financial settlement system; and the repair workload is obtained through the data quality management system and the work order management system.

[0115] In this embodiment, a responsibility ratio weight matrix is ​​constructed based on the calculated responsibility ratio weights. Obtain the task repair workload vector Vector of penalty amount and the matrix of technical capabilities of responsible entities The responsibilities for restoration and punishment are quantified.

[0116] It should be noted that only responsible entities with the capability to perform repair tasks can allocate repair responsibilities. The capability to perform repair tasks is judged based on the strength of the responsible entity's technical ability to repair the problem field. When the strength of the technical ability is higher than the technical feasibility threshold, repair responsibility is allocated.

[0117] The formula for normalizing repair responsibility is as follows:

[0118]

[0119] in,

[0120] In the formula, the constraints are satisfied. , To quantify the responsibility for repair, As a standard repair workload, As a weighted proportion of responsibility, As the responsible party Fix the problematic fields Technical capabilities As a technical feasibility threshold, Responsible Entity Fix the problematic fields The capability vector, This is a capability matrix. This is a technical strength adjustment factor.

[0121] The weighted formula for calculating penalties is as follows:

[0122]

[0123] In the formula, To quantify punitive liability and meet constraints , The standard penalty amount, This refers to the weighting of responsibility proportions.

[0124] S4. Based on the assigned responsibilities, assign problem rectification tasks to the corresponding responsible entities.

[0125] In this embodiment, the step of assigning problem rectification tasks to the corresponding responsible parties based on the allocated responsibilities specifically includes:

[0126] The intelligent dispatch formula generates subject-level task instructions, and the task instructions are used to dispatch problem rectification tasks to the responsible subjects.

[0127] The intelligent distribution formula is as follows:

[0128]

[0129] In the formula, As the main task, As the responsible party, This represents the number of fields in the question.

[0130] S5. If rectification failure is detected, the rectification failure field set is extracted and transmitted to the secondary responsible entity according to the preset responsibility transmission path. The loop continues until the problem is resolved.

[0131] In this embodiment, if rectification failure is detected, the rectification failure field set is extracted and transmitted to the secondary responsible entity according to the preset responsibility transmission path, and the process is repeated until the problem is resolved. Specifically:

[0132] Calculate the rectification pass rate based on the task assignment results and rectification feedback data;

[0133] When the rectification pass rate is lower than the preset threshold, extract the rectification failure field set from the rectification feedback data;

[0134] Calculate the cross-domain responsibility transmission weight for the rectification failure field to obtain the responsibility transmission path;

[0135] Based on the responsibility transmission path, the rectification failure field is re-entered into the DCAN responsibility mapping mechanism to assign a new responsible subject until the rectification failure field after re-rectification is reviewed by the verification rule set and the data quality score is not lower than the authoritative threshold.

[0136] It should be noted that the purpose of calculating the cross-domain responsibility transfer weight is to re-enter the rectification failure field into other domains similar to the original domain in the DCAN responsibility mapping mechanism to re-map responsibility when the problem rectification fails.

[0137] In this embodiment, the formula for the rectification pass rate is as follows:

[0138]

[0139] In the formula, The current rectification compliance rate of the responsible party. To assign to the subject The problem field subset This is the result of field verification based on the set of validation rules.

[0140] In this embodiment, the preset rectification pass rate threshold is: When the rectification pass rate At that time, extract the set of fields for rectification failure from the rectification feedback data. This triggers a cross-domain responsibility transfer process, transferring responsibility for the problematic fields that failed to be rectified. The transferred problematic fields are then re-entered into the DCAN responsibility mapping mechanism and responsibility allocation decision model to assign new responsible entities until the rectified data is reviewed by the verification rule set and the quality score is not lower than the authority threshold.

[0141] The specific formula for cross-domain responsibility transmission is as follows:

[0142]

[0143]

[0144] In the formula, For rectification failure field Transmission to the field The weight, For fields Original domain vector, This is the domain vector of the target domain. represents the candidate neighborhood vector during the iteration.

[0145] The specific formula for the conduction path is as follows:

[0146]

[0147] In the formula, For rectification failure field The optimal transmission path, For rectification failure field Transmission to the field The weight.

[0148] In this embodiment, the multimodal responsible agent further includes an agent evolution mechanism, specifically:

[0149] Based on the rectification compliance rate, the credit increment of the responsible entity is calculated through the credit increment calculation model, and the credit value is dynamically updated in combination with the decay index.

[0150] The prediction loss function is calculated based on the rectification pass rate and the predicted rectification pass rate. The capability matrix parameters are adjusted by the gradient descent method. The predicted rectification pass rate is calculated from the reputation value.

[0151] Based on the constructed cross-domain capability transfer formula, the capability matrix of similar domains is transferred to the newly added domain.

[0152] In this embodiment, the credit increment calculation model is as follows:

[0153]

[0154] In the formula, To increase reputation; For the number of fields in the question, The current rectification compliance rate of the responsible parties is the actual problem. The percentage of issues rectified by the responsible party in the previous instance; For capability marker matrix elements, As the responsible party Repair fields Technical capabilities This is the threshold for technical feasibility. As a weight of responsibility, For adaptive reputation learning rate and , , This represents the current credit score of the responsible entity.

[0155] Furthermore, the specific formula for calculating the reputation score update is as follows:

[0156]

[0157] Attenuation coefficient dynamically adjusted:

[0158] In the formula, For the updated reputation value, , To increase reputation, The reputation value before the main update. The time decay coefficient, For time step.

[0159] In this embodiment, the specific formula for calculating the prediction loss function is as follows:

[0160]

[0161] The logit function:

[0162] The formula for calculating the predicted rectification pass rate is:

[0163] In the formula, For the prediction loss function, For the number of fields in the question, The predicted rectification compliance rate for this field. The actual rectification pass rate of the field. For fields The predicted rectification pass rate This represents the current credit score of the responsible entity.

[0164] Furthermore, the specific formula for the gradient of the computational capability matrix is ​​as follows:

[0165]

[0166] in, , This is achieved through automatic differentiation.

[0167] In the formula, Let be the gradient matrix of the loss function. For the elements of the capability matrix, The coordinates of the matrix elements For partial differential operators, As the responsible party Repair fields Technical capabilities This is the technical feasibility threshold.

[0168] Furthermore, the formula for updating the capability matrix is ​​as follows:

[0169]

[0170] In the formula, For the updated capability matrix, This is the current capability matrix. Let be the learning rate, and the constraint be... , From the perspective of capabilities, For the number of fields in the question, The gradient matrix of the loss function .

[0171] In this embodiment, the updated capability matrix Feedback is sent to the multimodal responsible agent in the responsibility mapping mechanism; the updated agent reputation value is then... The multimodal responsible agent is fed back to the responsibility mapping mechanism for prediction in the next cycle.

[0172] In this embodiment, the capability matrix of similar domains is migrated to the newly added domain according to the constructed cross-domain capability transfer formula.

[0173] The specific formula for cross-domain capability transfer is as follows:

[0174]

[0175] The formula for the transfer factor is:

[0176] In the formula, To update the capability score of the target domain after migration, In the source domain capability matrix, the first... The ability in the first The capability score of each source domain For migration strength scalar, For the first The ability to the first The adaptability factor of each source domain, For source domain With the target domain similarity, As the transfer factor, For the first The confidence level of each source domain. The time difference between the last update of the source domain's capabilities and the current time. This is the time-series decay coefficient. For source domain The capability vector, For the target domain The characteristic subspace, To project vectors into a subspace The linear projection operator, It is a constant divided by zero.

[0177] It should be noted that the cross-domain capability transfer formula obtains the original capability score of the target domain by calculating the similarity between the source domain capability matrix and the target domain capability matrix. Based on the original capability score, the transfer increment from the source domain capability to the target domain capability is calculated by using the similarity between the source domain capability matrix and the target domain capability matrix, the source domain confidence, the source domain adaptability factor, the transferability scalar, and the transfer factor. This increment is then added to the original capability score, thereby achieving effective cross-domain capability transfer.

[0178] Example 2, Figure 2 This invention provides a system for automatic verification and closed-loop management of supply chain data quality, comprising:

[0179] The problem extraction module is used to input the obtained authoritative source list of the supply chain into a preset set of verification rules, and extract a set of problem fields based on the rule verification results;

[0180] The responsibility mapping module is used to inject the problem field set into the DCAN responsibility mapping mechanism, output the capability matrix through the pre-built multimodal responsibility agent, and calculate the responsibility proportion weight of the responsible subject based on the capability matrix;

[0181] The responsibility allocation module is used to input the responsibility proportion weights into the responsibility allocation decision model and output the allocated responsibilities of the corresponding responsible entities.

[0182] The task assignment module is used to assign problem rectification tasks to the corresponding responsible parties based on the assigned responsibilities.

[0183] The closed-loop governance module is used to extract the set of rectification failure fields if rectification failure is detected, and to pass the set of rectification failure fields to the secondary responsible entities according to the preset responsibility transmission path, and execute the loop until the problem is resolved.

[0184] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0185] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0186] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0187] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0188] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0189] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for supply chain data quality automatic checking and closed-loop governance, characterized in that, The method comprises the following steps: inputting the obtained supply chain authoritative source list into a preset verification rule set, extracting a problem field set according to a rule verification result; injecting the problem field set into a DCAN responsibility mapping mechanism, outputting a capability matrix through a pre-built multi-modal responsibility intelligent agent, calculating a responsibility proportion weight of a responsibility subject based on the capability matrix, comprising: inputting the problem field set into the pre-built multi-modal responsibility intelligent agent, outputting a capability matrix containing processing capability parameters of each subject on the field type; calculating the subject performance value based on the pre-set subject performance formula combined with the parameters in the capability matrix, and calculating the marginal contribution value based on the performance value; calculating the Shapley value based on the marginal contribution value, and normalizing the Shapley value of each subject according to the field to obtain the responsibility proportion weight; subject performance formula: In the formula Subset of responsible entities For the question field The main performance value, Assigning weights based on business importance. As the responsible party, This is a capability vector; marginal contribution formula: In the formula Marginal contribution value; Shapley value calculation formula: wherein is the Shapley value, is the total number of responsible subjects, is the full set of responsible subjects after removing the responsible subject . inputting the responsibility proportion weight into a responsibility allocation decision model to output the allocation responsibility of the corresponding responsibility subject, comprising: inputting the responsibility proportion weight, the subject technical capability strength calculated according to the capability matrix, the penalty amount and the repair workload into the responsibility allocation decision model; based on the subject technical capability strength and the responsibility proportion weight, the repair responsibility is obtained by calculating the normalized allocation of the repair workload; and based on the responsibility proportion weight, the penalty responsibility is obtained by weighted allocation of the penalty amount; according to the allocation responsibility, assigning the problem rectification task to the corresponding responsibility subject; if the rectification fails, extracting the rectification failure field set, and conducting the rectification failure field set to the secondary responsibility subject according to the preset responsibility transmission path, and executing the cycle until the problem is solved.

2. The supply chain data quality auto-validation and closed-loop governance method of claim 1, wherein, The method comprises the following steps: fuse the obtained multi-source supply chain system data, calculate the field similarity of the fused system data, and generate a standard dictionary based on the clustering result of the similarity; calculate the data source authority score according to the standard dictionary, and select the authoritative source list; input the authoritative source list into the preset rule verification set for verification, generate a data quality score, and when the data quality score is lower than the preset authority threshold, locate the problem field to build the problem field set.

3. The supply chain data quality auto-validation and closed-loop governance method of claim 2, wherein, The construction of the multi-modal responsibility intelligent agent comprises: extracting features from cross-system metadata to obtain domain vectors and subject vectors; constructing an initial capability matrix according to the matching degree of the domain vectors and the subject vectors; constructing an intelligent agent template based on the domain vectors and the capability matrix, combined with the identity core, the reputation value and the historical task rectification qualification rate.

4. The supply chain data quality auto-validation and closed-loop governance method of claim 3, wherein, The method comprises the following steps: generate a subject-level task instruction through an intelligent dispatch formula, and dispatch the problem rectification task to the responsibility subject through the task instruction; wherein the intelligent dispatch formula is: In the formula, is the main task, is the responsible subject, is the number of problem fields.

5. The supply chain data quality auto-validation and closed-loop governance method of claim 4, wherein, If the rectification fails, the rectification failure field set is extracted, the rectification failure field set is conducted to the secondary responsibility subject according to the preset responsibility transmission path, and the cycle is executed until the problem is solved, comprising: calculate the rectification qualification rate according to the task dispatch result and the rectification feedback data; when the rectification qualification rate is lower than the preset threshold, extract the rectification failure field set from the rectification feedback data; The cross-domain responsibility conduction weight of the rectification failure field is calculated to obtain a responsibility conduction path; Based on the responsibility conduction path, the rectification failure field is re-input into the DCAN responsibility mapping mechanism to allocate a new responsibility subject until the re-rectified rectification failure field passes the review of the check rule set and the data quality score is not lower than the authority threshold.

6. The supply chain data quality auto-validation and closed-loop governance method of claim 5, wherein, The multi-modal responsibility intelligent agent further comprises an intelligent agent evolution mechanism, specifically: Based on the rectification eligibility rate, the responsibility subject credit increment is calculated through a credit increment calculation model, and the credit value is dynamically updated in combination with a decay index; According to the rectification eligibility rate and the predicted rectification eligibility rate, a predicted loss function is calculated, and the capability matrix parameters are adjusted through the gradient descent method, wherein the predicted rectification eligibility rate is calculated from the credit value; According to the constructed cross-domain capability migration formula, the capability matrix of the similar field is migrated to the new field.

7. A system using the supply chain data quality automatic check and closed-loop governance method according to any one of claims 1-6, comprising: a problem extraction module configured to input the obtained supply chain authority source list into a preset check rule set, and extract a problem field set according to the rule check result; a responsibility mapping module configured to inject the problem field set into a DCAN responsibility mapping mechanism, and output a capability matrix through a pre-constructed multi-modal responsibility intelligent agent, and calculate a responsibility proportion weight of a responsibility subject based on the capability matrix; a responsibility allocation module configured to input the responsibility proportion weight into a responsibility allocation decision model, and output an allocated responsibility of the corresponding responsibility subject; a task dispatching module configured to dispatch a problem rectification task to the corresponding responsibility subject according to the allocated responsibility; a closed-loop governance module configured to extract a rectification failure field set if the rectification fails, and conduct the rectification failure field set to a secondary responsibility subject according to a preset responsibility conduction path, and perform a loop until the problem is solved.

Images