Circulation collaboration method and system based on supply chain full-link data management

Abstract

The invention discloses a circulation collaboration method and system based on supply chain full-link data management, and particularly relates to the field of supply chain management, and the method comprises the steps: collecting multi-source original data generated in each circulation link of a supply chain, and carrying out format unification, field mapping, de-duplication verification and time and space correction processing on the multi-source original data, generating a unified data table; based on the unified data table, an index structure organized according to the main body dimension, the commodity dimension, the time dimension and the space dimension is constructed, and a unified data base layer is formed. An event set is constructed after multi-source circulation data is treated in a unified mode, continuous modeling is conducted on the number and attribution changes of commodities in the splitting, merging and transferring processes through a measurable source share unit, a checkable circulation behavior chain is rebuilt, and collaborative decision making and closed-loop updating are achieved according to the checkable circulation behavior chain.

Description

Technical Field

[0001] This invention relates to the field of supply chain management technology, and more specifically, to a circulation collaboration method and system based on end-to-end supply chain data governance. Background Technology

[0002] In the supply chain process, existing technologies typically use batch numbers, document numbers, or transaction records as the basis for identifying the source of goods, and reconstruct the distribution chain by aggregating and sorting events at each stage, thereby achieving traceability and supervision. This approach is effective in scenarios where the distribution units remain relatively stable, but in real business, goods often undergo multiple splits, mergers, and reorganizations, and their form and ownership relationships continue to change. Taking agricultural products or fresh produce as an example, after a truckload of goods enters the wholesale market, it is often first broken down into several smaller portions, and then recombined by different merchants for distribution as needed. In this process, the same batch of original goods may be divided into multiple sales units, and multiple small batches from different sources may be mixed and packed into new packaging units, accompanied by multiple weighing, relabeling, or system re-entry operations. In these stages, existing systems typically only assign a new batch or identifier to each newly generated circulation unit, but lack the ability to continuously express its source relationship. This results in an implicit business relationship between the original batch and the new unit, while the data structure has been broken down. As splitting and merging operations accumulate, although a seemingly complete event record or business chain can still be formed in the system, this chain actually only reflects the most recent transaction or processing process and can no longer accurately point to the original source. When quality spot checks, problem recalls, or responsibility tracing occur, the system can often only trace back to the current merchant or the current batch, and it is difficult to further restore the true evolution path of the product in each upstream link; Therefore, the fundamental problem with existing technologies is that the current circulation management method, which is based on a single batch identifier and a linear event chain, cannot adapt to the evolutionary characteristics of circulation units under repeated splitting and merging. Its essential defect is the lack of a calculable expression mechanism that can depict the continuous changes in the source relationship of goods during multiple reconstructions, which causes the circulation link to lose the continuity of the real lineage and the ability to trace back reversibly in complex scenarios. Summary of the Invention

[0003] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a circulation collaboration method and system based on end-to-end supply chain data governance. By unifying and governing multi-source circulation data to construct an event set, and using measurable source share units to continuously model the changes in quantity and ownership of goods during the splitting, merging, and transfer processes, a verifiable circulation behavior chain is reconstructed, and collaborative decision-making and closed-loop updates are achieved accordingly, thereby solving the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a circulation collaboration method based on end-to-end supply chain data governance, comprising: S1. Collect multi-source raw data generated from various circulation links in the supply chain, perform format unification, field mapping, deduplication verification, and time and space correction on the multi-source raw data to generate a unified data table; and based on the unified data table, construct an index structure organized by subject dimension, product dimension, time dimension, and space dimension to form a unified data foundation layer. S2. Based on the data foundation layer and index structure, extract data records related to circulation actions and convert the data records into event records containing subject identifiers, product identifiers, time information, spatial information, and business parameters; perform aggregation processing on the event records according to the conditions of consistent subject, adjacent time, and similar space to form an event set; S3. Represent the source of goods as several source share units for measurement, and based on the event set, identify splitting operations, merging operations and transfer operations, perform quantity allocation and ownership change processing on the source share units, and generate and update the share status table for recording the current subject ownership and spatial ownership of each source share unit. S4. Based on the continuous update results of the source share units in the share status table, construct a share evolution chain that represents the transfer relationship of the source share units between different subjects and different spaces; and generate the corresponding circulation behavior chain according to the time sequence of each transfer relationship in the share evolution chain. S5. Based on the circulation behavior chain, generate circulation scheduling results, abnormal handling results, and source tracing results; write back the execution feedback data of circulation scheduling results, abnormal handling results, and source tracing results to the data foundation layer; and based on the execution feedback data, trigger the update processing of the event set, share status table, and share evolution chain.

[0005] In a preferred embodiment, S1 includes: S1-1. Collect multi-source raw data from various circulation links in the supply chain. The multi-source raw data includes at least equipment-reported data and business system-recorded data. The equipment-reported data includes weighing data, identification data, and positioning data. The business system-recorded data includes transaction data, inbound and outbound data, and entity registration data. The multi-source raw data is processed sequentially with format unification, field mapping, deduplication verification, and time and space correction. A unified data table is constructed in a row-column manner, wherein the unified fields include at least subject identifier, product identifier, time information, spatial information, and business parameters; the row-column manner includes: each row corresponds to one circulation record, and each column corresponds to one unified field; S1-2. Based on the unified data table, construct index structures using the subject identifier, product identifier, time information, and spatial information as keys, and establish a mapping relationship between each index structure and the corresponding row record in the unified data table; S1-3. Organize and store the unified data table and the index structure to form a unified data foundation layer that can perform retrieval according to subject dimension, product dimension, time dimension and space dimension.

[0006] In a preferred embodiment, S2 includes: S2-1. Based on a unified data foundation layer and index structure, perform index retrieval according to the subject identifier and product identifier, and limit the time information to be within a preset time window and the spatial information to be within a preset spatial neighborhood. Extract the corresponding row records from the unified data table to form a set of event candidate data records. S2-2. For each row record in the event candidate data record set, extract the subject identifier, product identifier, time information, spatial information and business parameters according to the unified field mapping rules, and reorganize the extraction results into structured event records to form an event record set; S2-3. Using the same subject identifier in the event record, time information difference not exceeding the preset time window threshold, and spatial information distance not exceeding the preset spatial neighborhood threshold as aggregation conditions, perform grouping aggregation processing on the event record set, and merge the event records that meet the aggregation conditions into the same event set.

[0007] In a preferred embodiment, S3 includes: S3-1. Based on the event record representing the first appearance of the product in the event set, read the quantity value corresponding to the business parameter in the event record, and use the quantity value as a measurable benchmark to represent the source of the corresponding product as several source share units, wherein each source share unit corresponds to a part of the quantity value, and register the subject identifier and spatial information corresponding to the event record for each source share unit to generate an initial share status table.

[0008] In a preferred embodiment, S3 further includes: S3-2. Based on the relationship between the number of event records for the same product identifier in adjacent time periods within the event set, the following determination is made: When the number of subsequent event records corresponding to the same subject is less than the number of preceding event records, it is identified as a split operation. When the number of subsequent event records corresponding to the same subject is greater than the number of preceding event records and originates from multiple subjects, it is identified as a merging operation. When the quantity value remains the same but the main identifier changes, it is identified as a transfer operation; And based on the splitting operation, merging operation, or transfer operation, calculate the change in the quantity value to be allocated to each source share unit respectively; S3-3. According to the change in quantity value, perform quantity update processing on the source share unit in the share status table, generate a new source share unit when a split operation occurs, accumulate the quantity values ​​corresponding to multiple source share units into a share record under the same subject when a merge operation occurs, and update the subject affiliation and spatial affiliation of the source share unit when a transfer operation occurs, thereby generating and updating a share status table that records the current subject affiliation, current spatial affiliation and corresponding quantity value of each source share unit.

[0009] In a preferred embodiment, S4 includes: S4-1. During the continuous update of the share status table, the records of the same source share unit before and after two adjacent updates are compared, and the difference in quantity, change in subject affiliation and change in spatial affiliation are calculated one by one. The change results obtained from each comparison are combined with the corresponding time information to generate a share status change sequence sorted by time information. S4-2. Based on the share state change sequence, each state change is represented as a transfer relationship record. The transfer relationship record includes at least the source share unit identifier, the subject affiliation before the change, the subject affiliation after the change, the spatial affiliation before the change, the spatial affiliation after the change, the quantity value change, and time information. Using the source share unit identifier as the association key, the transfer relationship records are organized into an adjacency structure according to the time information sequence to construct a share evolution chain that represents the transfer process of the source share unit between different subjects and different spaces.

[0010] In a preferred embodiment, S4 further includes: S4-3. In the share evolution chain, for the continuous transfer relationship of the same source share unit, the following verification process is performed sequentially: To verify whether the quantity value after the change of the previous record in the adjacent transfer relationship is consistent with the quantity value before the change of the next record, the quantity conservation verification is completed. To verify whether the subject ownership of the previous record after the change is consistent with the subject ownership of the next record before the change in the adjacent transfer relationship, the continuity of subject ownership is verified. To verify whether the spatial ownership of the previous record after the change is consistent with the spatial ownership of the next record before the change in the adjacent transfer relationship, the continuity of spatial ownership is verified. When any verification result is inconsistent, the corresponding transfer relationship will be eliminated, and the continuous transfer relationships that have passed all verifications will be solidified into valid evolution paths in chronological order. S4-4. Based on the effective evolution path, multiple transfer relationships with the same subject affiliation and continuous time information are processed by path compression and merged into behavior segments representing continuous flow processes within the same subject. The behavior segments are then connected sequentially according to the time information to generate a flow behavior chain.

[0011] In a preferred embodiment, S5 includes: S5-1. Based on the aforementioned circulation behavior chain, construct at least two different collaborative strategy behavior deduction paths for the same source share unit in the current state. The two different collaborative strategy behavior deduction paths include the maintenance strategy of maintaining the original circulation path and the adjustment strategy of changing the subject affiliation or spatial affiliation. Calculate the corresponding quantity stability, time efficiency value and spatial deviation for each deduction path to form a multi-strategy deduction result set. The quantity stability is calculated by comparing the ratio of the quantity value difference between the starting point and the ending point of the deduced path to the quantity value of the starting point. The time efficiency value is calculated by ratioing the sum of the time intervals of each transition relationship in the deduced path to the number of transitions contained in the path; The spatial deviation is calculated by accumulating or weighting the distances between the spatial positions corresponding to each transfer relationship in the deduction path and the preset target spatial position.

[0012] In a preferred embodiment, S5 further includes: S5-2, Perform the following determination on the set of multi-strategy deduction results: When the time efficiency value of the adjustment strategy is better than that of the maintenance strategy and the quantity stability is not lower than that of the maintenance strategy, a circulation scheduling result is generated. Otherwise, if the quantity stability of any inference path is lower than the preset stability threshold or the spatial deviation is higher than the preset deviation threshold, an abnormal handling result is generated. Otherwise, determine the source tracing result; S5-3. Perform the corresponding actual handling operation on the circulation scheduling result, abnormal handling result or source tracing result, and collect the actual quantity change, actual time consumption and actual spatial offset after execution; When the deviation between the actual quantity change, actual time consumption, and actual spatial offset and the predicted quantity change, time efficiency value, and spatial deviation in the corresponding inference path exceeds the preset deviation range, the preset stability threshold or preset deviation threshold is updated. Otherwise, the preset stable threshold or preset deviation threshold remains unchanged to generate execution feedback data; S5-4. Write the execution feedback data into a unified data infrastructure layer and trigger two-level updates simultaneously: Level 1 updates include updating the event set based on the feedback data from the execution; The secondary update includes updating the share status table in conjunction with the event set when the event set is updated, and updating the share evolution chain in conjunction with the share status table after the share evolution chain is updated.

[0013] A circulation collaboration system based on end-to-end supply chain data governance includes a construction module, a recording module, a share module, a generation module, and an update module; The construction module is used to collect multi-source raw data generated in various circulation links of the supply chain, perform format unification, field mapping, deduplication and verification, and time and space correction on the multi-source raw data to generate a unified data table; and based on the unified data table, construct an index structure organized by subject dimension, product dimension, time dimension and space dimension to form a unified data foundation layer. The recording module includes extracting data records related to circulation actions based on the data foundation layer and index structure, and converting the data records into event records containing subject identifiers, product identifiers, time information, spatial information and business parameters; and performing aggregation processing on the event records according to the conditions of consistent subject, adjacent time and spatial proximity to form event sets. The share module is used to represent the source of goods as several source share units for measurement, and based on the event set, it identifies splitting, merging and transfer operations, performs quantity allocation and ownership change processing on the source share units, and generates and updates the share status table to record the current subject ownership and spatial ownership of each source share unit. The generation module includes constructing a share evolution chain that represents the transfer relationship of the source share unit between different subjects and different spaces based on the continuous update results of the source share unit in the share status table; and generating the corresponding circulation behavior chain according to the time sequence of each transfer relationship in the share evolution chain. The update module includes generating circulation scheduling results, anomaly handling results, and source tracing results based on the circulation behavior chain; writing back the execution feedback data of the circulation scheduling results, anomaly handling results, and source tracing results to the data foundation layer; and triggering update processing of the event set, share status table, and share evolution chain based on the execution feedback data.

[0014] The technical effects and advantages of this invention are as follows: 1. This invention abstracts the source of goods into measurable source share units and continuously tracks the changes in their quantity and ownership during the splitting, merging and transfer processes, constructing a share evolution chain and a circulation behavior chain. This maintains the continuity of the source genealogy at the data structure level and solves the fundamental problem that existing batch identification and linear event chains cannot truly trace the source in complex reconstruction scenarios. 2. This invention unifies the processing of multi-source data such as rebalancing, identification, positioning, transactions, warehousing, and entity registration, and constructs a multi-dimensional index to form a unified data foundation layer. This enables heterogeneous data to be aligned and associated in the same structure, providing consistent and reliable data support for end-to-end circulation modeling. 3. This invention transforms and aggregates discrete raw data records into a structured event set, enabling the flow of the same subject within a short time and near space to be restored into computable event units, thus avoiding the link distortion problem caused by event fragmentation. 4. This invention extracts the transfer relationship from the continuous changes in share status and performs consistency checks on quantity, subject, and space, thus solidifying the effective evolution path and making the reconstructed circulation link physically reasonable and business consistent, thereby improving the credibility of the traceability results. 5. This invention introduces a multi-strategy deduction and counterfactual comparison mechanism on the basis of the circulation behavior chain, automatically generates scheduling, anomaly handling or tracing results, and forms a closed-loop update through execution feedback write-back, realizing the transformation of circulation collaboration from static recording to dynamic decision-making and adaptive optimization. Attached Figure Description

[0015] Figure 1 This is a flowchart of the method steps of the present invention.

[0016] Figure 2 This is a schematic diagram of the system modules of the present invention. Detailed Implementation

[0017] 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.

[0018] Refer to the instruction manual appendix Figure 1-2 An embodiment of the present invention provides a circulation collaboration method based on end-to-end supply chain data governance, comprising: S1. Unified Data Foundation Layer Construction Steps: Collect multi-source raw data generated from various circulation links in the supply chain, perform format unification, field mapping, deduplication verification, and time and space correction on the multi-source raw data to generate a unified data table; and based on the unified data table, construct an index structure organized by subject dimension, product dimension, time dimension, and space dimension to form a unified data foundation layer. S2. Event record generation and event set construction steps: Based on the data foundation layer and index structure, extract data records related to circulation actions, and convert the data records into event records containing subject identifiers, product identifiers, time information, spatial information, and business parameters; perform aggregation processing on the event records according to the conditions of consistent subjects, adjacent times, and similar spaces to form event sets; S3. Source share unit modeling and share status table update steps: The source of goods is represented as several source share units for measurement, and based on the event set, splitting operation, merging operation and transfer operation are identified. Quantity allocation and ownership change processing are performed on the source share units, and a share status table is generated and updated to record the current subject ownership and spatial ownership of each source share unit. S4. Steps for constructing the share evolution chain and circulation behavior chain: Based on the continuous update results of the source share units in the share status table, construct a share evolution chain that represents the transfer relationship of the source share units between different subjects and different spaces; and generate the corresponding circulation behavior chain according to the time sequence of each transfer relationship in the share evolution chain. S5. Steps for generating collaborative results and writing back to the unified data foundation layer: Based on the circulation behavior chain, generate circulation scheduling results, abnormal handling results, and source tracing results; write back the execution feedback data of circulation scheduling results, abnormal handling results, and source tracing results to the data foundation layer; and based on the execution feedback data, trigger the update processing of the event set, share status table, and share evolution chain to form a circulation collaboration closed loop.

[0019] S1 includes: S1-1, Multi-source raw data collection and unified data table generation steps: Collect multi-source raw data from various circulation links of the supply chain. The multi-source raw data includes at least equipment reported data and business system recorded data. The equipment reported data includes weighing data, identification data and positioning data. The business system recorded data includes transaction data, warehouse entry and exit data and entity registration data. Weighing data comprises: a set of data variables directly related to weight measurement generated by weighing equipment, including weighing time, weighing location coordinates, weighing entity identifier, product identifier, gross weight, tare weight, and net weight; identification data comprises a set of data variables for object identification generated by barcode scanning, RFID, or image recognition equipment, including identification time, identification location coordinates, entity identifier, product identifier, batch identifier, label code, and identification confidence level; positioning data comprises a set of spatial measurement data variables generated by positioning terminals or vehicle positioning devices, including positioning time, longitude coordinates, latitude coordinates, carrier identifier, and associated entity identifier; and transaction data comprises... The transaction or settlement system generates a set of data variables reflecting transaction behavior, including transaction time, buyer entity identifier, seller entity identifier, product identifier, transaction quantity value, transaction order value, transaction amount value, and transaction status code; inventory entry and exit data is a set of data variables generated by the inventory management system reflecting inventory changes, including inventory entry and exit time, inventory entry and exit type code, inventory node identifier, entity identifier, product identifier, inventory entry and exit quantity value, and inventory balance quantity value; entity registration data is a set of basic information data variables maintained by the entity management system, including entity identifier, entity type code, entity name, entity location coordinates, and entity validity status identifier; The multi-source raw data is sequentially processed with format unification, field mapping, deduplication verification, and time and space correction. A unified data table is constructed according to a row-column arrangement. The unified fields include at least the subject identifier, product identifier, time information, spatial information, and business parameters. The row-column arrangement includes: each row corresponds to one circulation record, and each column corresponds to one unified field. Format unification refers to converting the raw data from different systems and devices according to preset data type rules, so that the time field is unified to the standard timestamp format, the numerical field is unified to the numerical format, and the coordinate field is unified to the latitude and longitude or location coordinate format, generating data records with a consistent structure. Field mapping refers to mapping source fields in the original data one by one to target fields in a unified data table based on a pre-established field correspondence table, and supplementing missing fields by associating them with subject registration data or business rules to form data records with consistent field semantics. Deduplication verification refers to using time information, subject identifier, product identifier and business parameters as joint judgment conditions to perform similarity comparison or primary key comparison on data records, and identify and delete data records that are repeatedly reported or entered within a set time window. Time and space correction processing refers to performing a reasonableness check on the time information and spatial coordinates in the data record. When it is detected that the time information exceeds the allowable error range of the equipment or does not meet the site constraints, the corresponding time value and coordinate value are corrected or replaced through equipment clock offset correction and spatial constraint mapping rules. It should be noted that the subject identifier is a data field used to uniquely identify the subject participating in the circulation, corresponding to the subject identifier in the subject registration data, and used to uniformly associate different subjects involved in weighing data, identification data, transaction data, and inbound / outbound data with the same subject object; Product identification is a data field used to uniquely identify circulating goods. It includes product identification variables in corresponding weighing data, identification data, transaction data, and inbound / outbound data, and is used to establish a consistent reference for the same product across data from different sources. Time information is a data field used to describe the moment when circulation actions occur, including variables such as corresponding weighing time, identification time, positioning time, transaction time, and entry / exit time, which are used to sort and align multi-source data on a unified time axis; Spatial information is a data field used to describe the location where circulation actions occur. It includes variables such as the coordinates of the corresponding weighing location, the coordinates of the identification location, the coordinates of the positioning location, and the location of the inventory node. It is used to correlate and verify multi-source data in a unified spatial dimension. Business parameters are a set of numerical or status fields that are directly related to specific circulation actions. They correspond to variables such as weight value, identification confidence level, transaction quantity, transaction amount, inbound and outbound quantity, and inventory balance. They are used to retain the parameters required for various business calculations in a unified structure. Among them, circulation records refer to a single data record formed by unified processing of multi-source raw data, which describes a circulation action performed by a certain entity on a certain commodity at a certain time and in a certain space; S1-2, Index Structure Construction Steps: Based on the unified data table, construct index structures using the subject identifier, product identifier, time information, and spatial information as keys, and establish a mapping relationship between each index structure and the corresponding row record in the unified data table; the mapping relationship in S1-2 refers to associating each index key value with the storage location of the row record in the unified data table that has the same subject identifier, product identifier, time information, or spatial information when constructing the index structure, so that the corresponding row record can be directly located through any index key value; S1-3, Steps for forming a unified data foundation layer: Organize and store the unified data table and the index structure to form a unified data foundation layer that can perform retrieval according to the subject dimension, product dimension, time dimension and space dimension.

[0020] S2 includes: S2-1, Event Candidate Data Record Extraction Steps: Based on a unified data foundation layer and index structure, index retrieval is performed according to the subject identifier and product identifier, and the time information is limited to within a preset time window and the spatial information is limited to within a preset spatial neighborhood. The corresponding row records are extracted from the unified data table to form a set of event candidate data records. Wherein, the time information is within the preset time window means that, with the time information of a certain candidate row record as the center, other row records in the unified data foundation layer whose time information does not exceed the length of the preset time window are retrieved. Wherein, the preset time window is a fixed time interval value preset by the system according to the circulation link type or the equipment reporting frequency. The spatial information being located within a preset spatial neighborhood means calculating the spatial distance between the spatial information of a candidate row record and the spatial information of other row records, and limiting the spatial distance to not exceeding the radius of the preset spatial neighborhood, wherein the preset spatial neighborhood is a fixed distance threshold preset by the system based on the scope of the circulation area or the coverage of the equipment. In addition, extracting the corresponding row records from the unified data table refers to reading and selecting the row data in the unified data table that meet the conditions of the preset time window and preset spatial neighborhood based on the index retrieval results of the subject identifier, product identifier, time information and spatial information. S2-2, Event Record Conversion Steps: For each row record in the event candidate data record set, extract the subject identifier, product identifier, time information, spatial information, and business parameters according to the unified field mapping rules, and reorganize the extraction results into structured event records to form an event record set; wherein, extracting the subject identifier, product identifier, time information, spatial information, and business parameters according to the unified field mapping rules means reading the values ​​of the corresponding source fields from the row records of the unified data table based on the pre-established correspondence table between source fields and unified fields, and assigning them to the subject identifier, product identifier, time information, spatial information, and business parameters, wherein the unified field mapping rules include at least the source field name, the target unified field name, and necessary data type conversion rules; S2-3, Event Set Aggregation Construction Steps: Using the same subject identifier, time information difference not exceeding a preset time window threshold, and spatial information distance not exceeding a preset spatial neighborhood threshold as aggregation conditions, perform grouping aggregation processing on the event record set, and merge the event records that meet the aggregation conditions into the same event set; The aggregation condition is set in this way to merge multiple event records generated by the same subject on the same product within a short time and near space into a set of events corresponding to the same circulation action, thereby avoiding the erroneous merging of unrelated records from different subjects or different locations. In addition, the time information difference refers to the absolute value of the difference between the time information of two event records under the same time unit, and the spatial information distance refers to the geographical distance calculated by the spatial information of two event records under the same coordinate system; the preset time window threshold in application includes taking a fixed number of seconds or minutes according to the typical action cycle of the circulation link, and the preset spatial neighborhood threshold includes taking a fixed number of meters or the radius of the site partition according to the radius of the smallest identifiable working area of ​​the circulation site.

[0021] S3 includes: S3-1, Source Share Unit Generation Steps: Based on the event records representing the first appearance of a product in the event set, read the quantity values ​​corresponding to the business parameters in the event records, and use the quantity values ​​as a measurable benchmark to represent the source of the corresponding product as several source share units. Each source share unit corresponds to a portion of the quantity value, and registers the subject identifier and spatial information corresponding to the event record for each source share unit, generating an initial share status table. The measurable benchmark refers to a unified measurement basis used for quantifying and calculating source share units. Using the quantity value as a measurable benchmark means using the product quantity value corresponding to the business parameters in the event record as the measurement scale, decomposing and allocating this quantity value to each source share unit, so that each source share unit corresponds to a clear calculable quantity. Furthermore, each source share unit corresponding to a portion of the quantity value means splitting the total product quantity value in the event record and assigning a specific sub-quantity value to each source share unit, so that the sum of the sub-quantity values ​​of each source share unit equals the total product quantity value. Finally, the initial share status table is a data table that stores the corresponding quantity value, subject identifier, and spatial information for each source share unit as a record.

[0022] S3 also includes: S3-2, Operation Type Identification and Quantity Allocation Calculation Steps: Based on the relationship between the number of event records for the same product identifier in adjacent time periods within the event set, the following determination is performed: When the number of subsequent event records corresponding to the same subject is less than the number of preceding event records, it is identified as a splitting operation; that is, this condition indicates that under the same subject, the quantity of the same product decreases in the event records of adjacent time periods, indicating that the original quantity has been divided into multiple destinations, and thus the change in quantity is judged as a splitting operation. When the number of subsequent event records corresponding to the same subject is greater than the number of preceding event records and originates from multiple subjects, it is identified as a merging operation; that is, when the number of goods in the event records of adjacent times under the same subject increases, and the increase corresponds to the aggregation of quantities from multiple different subjects, it indicates that the quantities originally scattered across different subjects have been concentrated into the same subject, thus the change in quantity is determined as a merging operation. When the quantity remains the same but the subject identifier changes, it is identified as a transfer operation; that is, when the quantity of goods in the event records of adjacent time periods remains unchanged, but the corresponding subject identifier changes from one subject to another, it indicates that the entire quantity has been transferred from the original subject to the new subject, and thus the change in quantity is determined to be a transfer operation. Based on the splitting, merging, or transferring operations, the change in the quantity value allocated to each source share unit is calculated respectively. Specifically, in the splitting operation, the quantity value corresponding to the preceding event record is decomposed according to the proportion of quantity in subsequent event records, and the resulting sub-quantities are used as the quantity value change of each source share unit (reduction in quantity or creation of new share units). In the merging operation, the quantity values ​​of source share units from different entities are accumulated according to the merged quantities in subsequent event records, and the corresponding accumulation result is used as the quantity value change of the source share units under the target entity. In the transferring operation, the quantity value of the source share units corresponding to the preceding event record is deducted from the original entity and increased by an equal amount under the new entity, so that this quantity value is used as the quantity value change of the corresponding source share unit when the entity ownership changes. S3-3, Share Status Table Update Steps: According to the change in the quantity value, perform quantity update processing on the source share units in the share status table, generate new source share units when a split operation occurs, accumulate the quantity values ​​corresponding to multiple source share units into share records under the same subject when a merge operation occurs, and update the subject affiliation and spatial affiliation of the source share units when a transfer operation occurs, thereby generating and updating the share status table that records the current subject affiliation, current spatial affiliation and corresponding quantity value of each source share unit; In practical applications of S3-3, the share status table can be stored in the database as an updatable data table. Each row corresponds to a source share unit, containing the quantity value, subject affiliation, and spatial affiliation fields. After calculating the change in quantity value based on the event set, the system performs addition and subtraction updates on the quantity field of the corresponding row through database update operations. When a split operation occurs, a new row record is inserted to represent the newly generated source share unit. When a merge operation occurs, the quantity fields of multiple rows are accumulated and updated to a single row record under the same subject and spatial affiliation, and the merged row record is deleted. When a transfer operation occurs, only the subject affiliation and spatial affiliation fields in the corresponding row record are updated without changing the quantity field. Thus, after each event processing, a share status table reflecting the current status can be obtained directly.

[0023] S4 includes: S4-1, Steps for Extracting Share Status Changes: During the continuous updating of the share status table, the records of the same source share unit before and after two adjacent updates are compared. The difference in quantity, change in subject affiliation, and change in spatial affiliation are calculated one by one. The change results obtained from each comparison are combined with the corresponding time information to generate a share status change sequence sorted by time information. The difference in quantity refers to the difference in quantity values ​​of the same source share unit in two adjacent records of the share status table, which is used to characterize the increase or decrease in quantity of the source share unit during the time period. The change in subject affiliation refers to the change in subject affiliation of the same source share unit from one subject identifier to another in two adjacent records. The change in spatial affiliation refers to the change in spatial affiliation of the same source share unit from one spatial location to another in two adjacent records. These three together constitute the complete change result of the same source share unit in one update process. S4-2, Steps for constructing share evolution relationships: Based on the share state change sequence, each state change is represented as a transfer relationship record. The transfer relationship record includes at least the source share unit identifier, the subject affiliation before the change, the subject affiliation after the change, the spatial affiliation before the change, the spatial affiliation after the change, the change in quantity value, and time information. Using the source share unit identifier as the association key, the transfer relationship records are organized into an adjacency structure according to the time information order to construct a share evolution chain that represents the transfer process of the source share unit between different subjects and different spaces. Regarding S4-2, it should be noted that in the transfer relationship record, the source share unit identifier is an identifier value used to uniquely point to a record in the share status table, used to determine the specific source share unit corresponding to this change; the subject affiliation before and after the change stores the subject identifier value of the source share unit before and after the change, used to indicate from which subject the quantity is transferred to which subject; the spatial affiliation before and after the change stores the spatial location coordinates or spatial node identifier of the source share unit before and after the change, used to indicate from which location the quantity migrates to which location; the quantity value change stores the increase or decrease value of the quantity field of the source share unit in this change, used to quantify whether the quantity decreases, increases, or remains unchanged during the transfer process; the time information stores the timestamp value corresponding to this change, used to sort multiple transfer relationship records sequentially. All the above fields are jointly stored in the form of a structured record, enabling the system to directly calculate, sort, and reconstruct the transfer process of the source share unit based on the subject identifier, spatial identifier, quantity change, and timestamp.

[0024] S4 also includes: S4-3. Evolutionary Chain Consistency Verification and Path Fixation Steps: In the share evolutionary chain, for the continuous transfer relationship of the same source share unit, the following verification processes are performed sequentially: The quantity conservation check is performed to verify whether the quantity value after the change of the previous record in the adjacent transfer relationship is consistent with the quantity value before the change of the next record. The significance of the quantity conservation check is that by checking whether the quantity value of the same source share unit after the end of the previous transfer is equal to the quantity value before the start of the next transfer, it is ensured that the share does not increase or decrease without cause during the continuous transfer process, thereby ensuring the conservation consistency of quantity in the evolution chain. The consistency of subject attribution after the change in the previous record and the subject attribution before the change in the next record in adjacent transfer relationships is checked to complete the subject attribution continuity check. The significance of subject attribution continuity check is: by comparing whether the subject attribution after the change in the previous transfer relationship and the subject attribution before the change in the next transfer relationship of the same source share unit are consistent, it is ensured that the source share unit always connects from the correct subject to the next subject in the continuous evolution process, avoiding subject break or jump, and thus relatively ensuring the continuity and rationality of subject transfer path. The spatial attribution of the preceding record in the adjacent transfer relationship is checked to ensure consistency between the spatial attribution of the preceding record and the spatial attribution of the following record. The significance of the spatial attribution continuity check is to ensure that the spatial attribution of the same source share unit is connected in space during the continuous evolution process by comparing whether the spatial attribution of the same source share unit in the preceding transfer relationship is consistent with the spatial attribution of the following transfer relationship, thus avoiding spatial jumps without physical basis. When any verification result is inconsistent, the corresponding transfer relationship will be removed, and the continuous transfer relationships that have passed all verifications will be solidified into valid evolution paths in chronological order. Here, the adjacent transfer relationship refers to two consecutive transfer relationship records in the share evolution chain corresponding to the same source share unit, which are sorted by time information and represent the state change process of the source share unit at two consecutive time points. It should be noted that in practical applications, this includes sorting the continuous transition relationships that have passed all verifications in ascending order of their time information, and establishing a forward-backward pointing relationship between adjacent transition relationships in the system to form an indivisible ordered chain record, thereby solidifying the ordered transition sequence into an effective evolutionary path. S4-4, Steps for generating the circulation behavior chain: Based on the effective evolution path, multiple transfer relationships with the same subject affiliation and continuous time information are compressed and merged into behavior segments representing continuous circulation processes within the same subject. The behavior segments are then connected sequentially according to the time information to generate a circulation behavior chain describing the continuous flow process of source share units across subjects and spaces. In S4-4, it should be noted that path compression processing refers to traversing the transfer relationships of the same valid evolution path in the system according to the time information sequence. When multiple consecutive transfer relationships are found to have the same subject affiliation and the time information is connected end to end, these transfer relationships are merged into a single record in the data structure. The subject affiliation and the earliest time information are retained as the start time, and the latest time information is retained as the end time. The change in quantity value is accumulated, thereby compressing the original multiple transfer relationships into a single behavior segment. Subsequently, connecting each behavior segment in sequence according to the time information sequence means sorting all behavior segments according to their start time and establishing a forward and backward pointing relationship in the system. The end time of the previous behavior segment is sequentially connected with the start time of the next behavior segment to form a chain structure ordered by time, thereby generating a circulation behavior chain describing the continuous flow process of the source share unit across subjects and across spaces.

[0025] S5 includes: S5-1, Multi-strategy Parallel Inference Calculation Steps: Based on the circulation behavior chain, construct at least two different collaborative strategy behavior inference paths for the same source share unit in the current state. The two different collaborative strategy behavior inference paths include the maintenance strategy of maintaining the original circulation path and the adjustment strategy of changing the subject affiliation or spatial affiliation. Calculate the corresponding quantity stability, time efficiency value and spatial deviation for each inference path to form a multi-strategy inference result set. The strategy of maintaining the original circulation path means that the existing subject affiliation and spatial affiliation transfer order in the current circulation behavior chain is used without change in the simulation calculation. Only the time and quantity of the subsequent transfer process are continued to be calculated to evaluate the natural circulation results under non-intervention conditions. The other strategy of changing subject affiliation or spatial affiliation means that one or more transfer nodes in the current circulation behavior chain are virtually replaced with new subject affiliation or new spatial affiliation in the simulation calculation, and the subsequent transfer relationship is recalculated accordingly to evaluate the potential circulation results after the implementation of scheduling or rerouting. The quantity stability is calculated by comparing the ratio of the difference between the quantity values ​​at the start and end of the deduced path to the quantity value at the start. The quantity stability is used to reflect the degree to which the quantity remains unchanged under the path. The calculation process of quantity stability is based on the quantity value at the start of the deduced path. The absolute value of the difference between the quantity values ​​at the start and end is divided by the quantity value at the start, thereby obtaining a proportional index that reflects the magnitude of quantity change under the path. The time efficiency value is calculated by dividing the sum of the time intervals between each transfer relationship in the deduction path by the number of transfers in the path. The time efficiency value is used to reflect the average time cost required for a unit transfer. The calculation process of the time efficiency value is to sum the time intervals between adjacent transfer relationships in the deduction path and then divide by the number of transfers in the path to obtain the average time consumption index corresponding to each transfer. The spatial deviation is calculated by accumulating or weighting the distances between the spatial positions corresponding to each transfer relationship in the deduction path and the preset target spatial position. The spatial deviation is used to reflect the degree of deviation of the path in space. The calculation process of the spatial deviation is to calculate the distances between the spatial positions corresponding to each transfer relationship in the deduction path and the preset target spatial position one by one, and then accumulate or weight the distances to obtain a numerical index reflecting the overall degree of spatial deviation.

[0026] S5 also includes: S5-2, Counterfactual Comparison Determination Step: Perform the following determination on the set of multi-strategy deduction results: When the time efficiency value of the adjustment strategy is better than that of the maintenance strategy and the quantity stability is not lower than that of the maintenance strategy, a circulation scheduling result is generated. Otherwise, if the quantity stability of any inference path is lower than the preset stability threshold or the spatial deviation is higher than the preset deviation threshold, an abnormal handling result is generated. Otherwise, determine the source tracing result; For S5-2, it should be noted that the maintenance strategy refers to using the existing subject affiliation and spatial affiliation in the current circulation behavior chain without changing them during the simulation process, and only calculating the subsequent quantity and time evolution according to the existing transfer order to obtain the benchmark simulation path; the benchmark simulation path or maintenance strategy includes directly selecting the most recently confirmed effective evolution path in the current circulation behavior chain as the initial path, and calculating it by extending it forward according to the existing transfer order without changing its subject affiliation and spatial affiliation, so as to serve as the benchmark simulation result for comparison with the adjustment strategy; In practical applications, "better than" means that the time efficiency value of the adjustment strategy is less than that of the maintenance strategy, and "not less than" means that the quantity stability of the adjustment strategy is greater than or equal to that of the maintenance strategy. The determination of generating circulation scheduling results refers to extracting the subject ownership change and spatial ownership change schemes from the deduced path corresponding to the adjustment strategy when the above comparison conditions are met, forming a set of scheduling instructions to guide the actual circulation path adjustment. This set of scheduling instructions is the circulation scheduling result, which covers the target subject, target space, and corresponding quantity arrangement. The circulation scheduling result can be presented in the form of structured instruction records. The circulation scheduling result includes at least the target subject identifier, target space location, scheduling quantity value, and scheduling execution time, which is used to guide subject docking and path adjustment in the actual circulation process. In addition, the abnormal handling result refers to the set of handling instructions calculated based on the corresponding circulation behavior chain and share evolution chain when it is determined that there is a quantity abnormality, subject jump, or spatial discontinuity. It is presented in the form of structured instruction records, which include at least the abnormal subject, abnormal spatial location, abnormal quantity, and corresponding handling action identifier. The source tracing result refers to the source path information obtained by tracing back upstream along the circulation behavior chain and the corresponding share evolution path when no scheduling or abnormal conditions are triggered. It is presented in the form of a structured path record, which includes at least each upstream entity, spatial location, corresponding time information and cumulative quantity. S5-3, Adaptive Threshold Correction and Execution Feedback Generation Step: Execute the corresponding actual handling operation on the circulation scheduling result, abnormal handling result, or source tracing result, and collect the actual quantity change, actual time consumption, and actual spatial offset after execution; wherein the actual handling operation refers to scheduling, intercepting, or backtracking the actual circulation process according to the instructions corresponding to the circulation scheduling result, abnormal handling result, or source tracing result; the actual quantity change generated is calculated from the weighing or inventory record after execution; the actual time consumption is calculated from the time difference between the execution start time and end time; and the actual spatial offset is calculated from the distance between the corresponding spatial location information before and after execution. When the deviation between the actual quantity change, actual time consumption, and actual spatial offset and the predicted quantity change, time efficiency value, and spatial deviation in the corresponding extrapolation path exceeds a preset deviation range, the preset stability threshold or preset deviation threshold is updated. When the deviation exceeds the preset deviation range, the system adjusts the corresponding preset stability threshold or preset deviation threshold according to the deviation direction and magnitude to make the threshold converge to the actual execution result. The update is achieved by adding a correction amount proportional to the deviation to the original threshold to make subsequent extrapolation calculations closer to the real circulation state. Otherwise, the preset stable threshold or preset deviation threshold remains unchanged to generate execution feedback data; The predicted quantity change, time efficiency value, and spatial deviation in the corresponding deduction path refer to the predicted indicators calculated for the deduction path generated by the maintenance strategy or adjustment strategy in the aforementioned multi-strategy parallel deduction calculation steps. The predicted quantity change is obtained by calculating the difference between the quantity values ​​corresponding to the start and end points of the deduction path. The predicted time efficiency value is obtained by summing the time intervals of each adjacent transfer relationship in the deduction path and then dividing by the number of transfers. The predicted spatial deviation is obtained by accumulating or weighted averaging the distances between the spatial positions corresponding to each transfer relationship in the deduction path and the preset target spatial position. The execution of the deduction path refers to the virtual application of the maintenance strategy or adjustment strategy to replace or continue the subject affiliation and spatial affiliation based on the current circulation behavior chain and share evolution chain within the system. Without triggering actual circulation operations, the subsequent quantity, time, and spatial changes are calculated step by step according to the existing transfer rules to obtain the predicted indicators. Execution feedback data refers to a structured record consisting of actual quantity changes, actual time consumption, actual spatial offset, and corresponding execution status identifiers, used to characterize the actual execution results of circulation scheduling, anomaly handling, or source tracing. S5-4, Dual-loop write-back and cascading update steps: Write the execution feedback data to a unified data base layer and simultaneously trigger the following two-level updates: Level 1 updates include updating the event set based on the feedback data from the execution; The second-level update includes updating the share status table in conjunction with the event set when the event set is updated, and further updating the share evolution chain in conjunction with the update of the share status table; For S5-4, it should be noted that the specific process of the first-level update includes: the system first writes the execution feedback data back to the unified data foundation layer to form a new unified data table row record; then the system extracts the subject identifier, product identifier, time information, spatial information, and business parameters from the row record according to the unified field mapping rules to generate a new event record; then the system searches in the existing event set to see if there is an event set that meets the requirements of subject consistency, time contiguousness, and spatial proximity with the event record. If it exists, the system adds the event record to the event set; if it does not exist, the system creates a new event set with the event record and registers it in the event set list, thereby completing the incremental update of the event set. The second-level update first updates the share status table and then updates the share evolution chain. The specific process includes: the system takes the newly added event records after the first-level update as input, reads the quantity value in its subject identifier, spatial information and business parameters, and matches it with the existing source share unit corresponding to the same product identifier in the share status table. Then, based on the increase or decrease of the quantity value and whether the subject ownership and spatial ownership have changed, it is identified as a split operation, a merge operation or a transfer operation. Subsequently, the system performs a database update on the share status table based on the identification results: during splitting, the quantity of the original source share unit is deducted and a new source share unit record is inserted; during merging, the quantities of multiple source share unit records are accumulated and merged into a record under the same subject affiliation and spatial affiliation; during transfer, the quantity remains unchanged and only the subject affiliation and spatial affiliation are updated; after the share status table is updated, the system compares the subject affiliation, spatial affiliation, and quantity value of the same source share unit before and after the update, generates a transfer relationship record containing the subject affiliation before the change, the subject affiliation after the change, the spatial affiliation before the change, the spatial affiliation after the change, the change in quantity value, and time information, and appends this transfer relationship record to the share evolution chain in chronological order, thereby completing the incremental update of the share evolution chain.

[0027] A circulation collaboration system based on end-to-end supply chain data governance includes a construction module, a recording module, a share module, a generation module, and an update module; The construction module is used to collect multi-source raw data generated in various circulation links of the supply chain, perform format unification, field mapping, deduplication and verification, and time and space correction on the multi-source raw data to generate a unified data table; and based on the unified data table, construct an index structure organized by subject dimension, product dimension, time dimension and space dimension to form a unified data foundation layer. The recording module includes extracting data records related to circulation actions based on the data foundation layer and index structure, and converting the data records into event records containing subject identifiers, product identifiers, time information, spatial information and business parameters; and performing aggregation processing on the event records according to the conditions of consistent subject, adjacent time and spatial proximity to form event sets. The share module is used to represent the source of goods as several source share units for measurement, and based on the event set, it identifies splitting, merging and transfer operations, performs quantity allocation and ownership change processing on the source share units, and generates and updates the share status table to record the current subject ownership and spatial ownership of each source share unit. The generation module includes constructing a share evolution chain that represents the transfer relationship of the source share unit between different subjects and different spaces based on the continuous update results of the source share unit in the share status table; and generating the corresponding circulation behavior chain according to the time sequence of each transfer relationship in the share evolution chain. The update module includes generating circulation scheduling results, anomaly handling results, and source tracing results based on the circulation behavior chain; writing back the execution feedback data of circulation scheduling results, anomaly handling results, and source tracing results to the data foundation layer; and triggering the update processing of event sets, share status tables, and share evolution chains based on the execution feedback data, forming a circulation collaboration closed loop.

[0028] Working principle: This invention achieves end-to-end supply chain collaboration through a process of "unified data, event aggregation, share evolution, behavior reconstruction, collaborative decision-making, and feedback loop." This invention first gathers multi-source data from various circulation links, including weighing, identification, positioning, transactions, warehousing, and entity registration. After unified processing, it forms a unified data foundation layer that can be retrieved by entity, commodity, time, and space. On this basis, the scattered data records are reorganized into event records reflecting specific circulation actions and aggregated into event sets according to time and space relationships. Subsequently, the quantity of commodities is broken down into measurable source share units, and their quantity changes in different entities and spaces are continuously tracked to form a share status table. Transfer relationships are extracted from continuous changes to construct a share evolution chain and circulation behavior chain, thereby realistically restoring the circulation trajectory of commodities in the supply chain. Based on this, the system performs multi-strategy simulations and comparisons on the behavior chain, automatically generating scheduling, anomaly handling, or source tracing results. Finally, the execution results are written back to the unified data foundation layer, which updates the event set, share status table, and share evolution chain in a coordinated manner, enabling the model to continuously self-correct and forming an integrated closed-loop operation mechanism covering data governance, circulation process reconstruction, and collaborative decision-making.

[0029] The above description is merely 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 circulation collaboration method based on supply chain full-link data governance, characterized in that, Comprise: S1, collect multi-source original data generated in each circulation link of the supply chain, perform format unification, field mapping, de-duplication verification, and time and space correction processing on the multi-source original data, and generate a unified data table; and based on the unified data table, build an index structure organized by subject dimension, commodity dimension, time dimension, and space dimension, and form a unified data foundation layer; S2, based on the data foundation layer and the index structure, extract data records related to circulation actions, and convert the data records into event records containing subject identification, commodity identification, time information, space information, and business parameters; according to the conditions of consistent subject, adjacent time, and close space, perform aggregation processing on the event records to form an event set; S3, express the commodity source as a plurality of source share units for measurement, and based on the event set, identify splitting operations, merging operations, and transfer operations, perform quantity allocation and ownership change processing on the source share units, generate and update a share state table for recording the current subject ownership and spatial ownership of each source share unit; S4, based on the continuous update results of the source share units in the share state table, build a share evolution chain representing the transfer relationship between the source share units in different subjects and different spaces; and according to the time sequence of each transfer relationship in the share evolution chain, generate the corresponding circulation behavior chain; S5, based on the circulation behavior chain, generate circulation scheduling results, abnormal disposal results, and source tracing results; write the execution feedback data of the circulation scheduling results, abnormal disposal results, and source tracing results back to the data foundation layer; and based on the execution feedback data, trigger the update processing of the event set, the share state table, and the share evolution chain.

2. The circulation collaboration method based on supply chain full-link data governance according to claim 1, characterized in that: In S1, it comprises: S1-1, collect multi-source original data from each circulation link of the supply chain, the multi-source original data at least including device reporting data and business system record data, the device reporting data including weighing data, identification data and positioning data, the business system record data including transaction data, warehouse in-out data and subject registration data; perform format unification, field mapping, de-duplication verification, and time and space correction processing on the multi-source original data in sequence, and build a unified data table in row-column mode, wherein the unified field at least includes subject identification, commodity identification, time information, space information and business parameters; the row-column mode includes: each row corresponds to a circulation record, and each column corresponds to a unified field; S1-2, based on the unified data table, build index structures respectively taking the subject identification, commodity identification, time information and space information as key values, and establish the mapping relationship between each index structure and the corresponding row record in the unified data table; S1-3, organize and store the unified data table and the index structure to form a unified data foundation layer that can be searched by subject dimension, commodity dimension, time dimension and space dimension.

3. The circulation collaboration method based on supply chain full-link data governance according to claim 2, characterized in that: In S2, it comprises: S2-1, based on the unified data base layer and index structure, performing index retrieval according to the subject identifier and the commodity identifier, and limiting the time information to be within the preset time window and the space information to be within the preset space neighborhood, extracting the corresponding row records from the unified data table to form an event candidate data record set; S2-2, for each row record in the event candidate data record set, extracting the subject identifier, the commodity identifier, the time information, the space information and the business parameter according to the unified field mapping rule, and reorganizing the extraction result into a structured event record to form an event record set; S2-3, taking the same subject identifier in the event record, the time information difference not exceeding the preset time window threshold and the space information distance not exceeding the preset space neighborhood threshold as the aggregation condition, performing grouping aggregation processing on the event record set, and merging the event records meeting the aggregation condition into the same event set.

4. The circulation collaboration method based on supply chain full-link data governance according to claim 3, characterized in that: S3 includes: S3-1, based on the event record representing the first appearance of the commodity in the event set, reading the quantity value corresponding to the business parameter in the event record, and taking the quantity value as a measurable reference, representing the source of the corresponding commodity as a plurality of source share units, wherein each source share unit corresponds to a part of the quantity value, and each source share unit is registered with the subject identifier and the space information corresponding to the event record, and an initial share state table is generated.

5. The circulation collaboration method based on supply chain full-link data governance according to claim 4, characterized in that: S3 also includes: S3-2, based on the number of event records of the same commodity identifier in the event set within the adjacent time, the following determinations are performed: When the number of subsequent event records corresponding to the same subject is less than the number of previous event records, it is identified as a split operation; When the number of subsequent event records corresponding to the same subject is greater than the number of previous event records and derived from multiple subjects, it is identified as a merge operation; When the number of values remains the same but the subject identifier changes, it is identified as a transfer operation; And according to the split operation, the merge operation or the transfer operation, the quantity value change of each source share unit should be calculated respectively; S3-3, according to the quantity value change, performing quantity update processing on the source share unit in the share state table, and generating a new source share unit when the split operation occurs, adding the quantity values corresponding to multiple source share units as a share record under the same subject when the merge operation occurs, and updating the subject attribution and space attribution of the source share unit when the transfer operation occurs, thereby generating and updating the share state table recording the current subject attribution, the current space attribution and the corresponding quantity value of each source share unit.

6. The circulation collaboration method based on supply chain full-link data governance according to claim 5, characterized in that: S4 includes: S4-1、in the continuous updating process of the share state table, the records of the same source share unit before and after the adjacent two updates are compared, the quantity value difference, the subject attribution change amount and the space attribution change amount are calculated one by one, and the change results obtained each time are combined with the corresponding time information to generate a share state change sequence sorted by time information; S4-2, based on the share state change sequence, each state change is represented as a transfer relationship record, and the transfer relationship record at least includes source share unit identification, pre-change subject attribution, post-change subject attribution, pre-change space attribution, post-change space attribution, quantity value change amount and time information; and taking the source share unit identification as the association key, the transfer relationship records are organized in the order of time information to form an adjacent structure, and a share evolution chain representing the transfer process of the source share unit between different subjects and different spaces is constructed.

7. The circulation collaboration method based on supply chain full-link data governance according to claim 6, characterized in that: In S4, further comprising: S4-3, in the share evolution chain, for the continuous transfer relationship of the same source share unit, the following checking processes are sequentially performed: whether the post-change quantity value of the first record in the adjacent transfer relationship is consistent with the pre-change quantity value of the second record is checked to complete the quantity conservation check; whether the post-change subject attribution of the first record in the adjacent transfer relationship is consistent with the pre-change subject attribution of the second record is checked to complete the subject attribution continuity check; whether the post-change space attribution of the first record in the adjacent transfer relationship is consistent with the pre-change space attribution of the second record is checked to complete the space attribution continuity check; when any check result is inconsistent, the corresponding transfer relationship is removed, and the continuous transfer relationship that passes all checks is solidified as an effective evolution path in time sequence; S4-4, based on the effective evolution path, the path compression process is performed on multiple transfer relationships with the same subject attribution and continuous time information, which are merged into behavior segments representing the continuous flow process within the same subject, and each behavior segment is connected in time sequence to generate a circulation behavior chain.

8. The circulation collaboration method based on supply chain full-link data governance according to claim 7, characterized in that: In S5, comprising: S5-1, based on the circulation behavior chain, at least two types of different collaboration strategies are constructed for the same source share unit in the current state, and the two types of different collaboration strategies include the maintenance strategy of maintaining the original circulation path and the adjustment strategy of changing the subject attribution or the space attribution, and the quantity stability, the time efficiency value and the space deviation of each deduction path are calculated to form a multi-strategy deduction result set; the quantity stability is calculated by comparing the proportion of the quantity value difference between the start point and the end point of the deduction path to the quantity value of the start point; the time efficiency value is calculated by comparing the sum of the time intervals of each transfer relationship in the deduction path with the number of transfers included in the path; the time efficiency value is calculated by comparing the sum of the time intervals of each transfer relationship in the deduction path with the number of transfers included in the path; The spatial deviation is obtained by adding or weightedly averaging the distances between the spatial positions corresponding to each transfer relationship in the deduced path and the preset target spatial position.

9. The circulation collaboration method based on supply chain whole-link data governance according to claim 8, characterized in that: In S5, further comprising: S5-2, performing the following determination on the set of multi-strategy deduction results: When the time efficiency value of the adjustment strategy is better than that of the maintenance strategy and the quantity stability degree is not lower than that of the maintenance strategy, it is determined to generate a circulation scheduling result; Otherwise, when the quantity stability degree of any deduced path is lower than the preset stability threshold or the spatial deviation is higher than the preset deviation threshold, it is determined to generate an abnormal handling result; Otherwise, it is determined to generate a source tracing result; S5-3, performing corresponding actual handling operations on the circulation scheduling result, the abnormal handling result or the source tracing result, and collecting actual quantity changes, actual time consumptions and actual spatial deviations after execution; When the deviations between the actual quantity changes, actual time consumptions and actual spatial deviations and the predicted quantity changes, time efficiency values and spatial deviations in the corresponding deduced path exceed the preset deviation range, the preset stability threshold or the preset deviation threshold is updated; Otherwise, the preset stability threshold or the preset deviation threshold remains unchanged, forming execution feedback data; S5-4, writing the execution feedback data into the unified data base layer and synchronously triggering two-level updates: The first-level update includes updating the event set based on the execution feedback data; The second-level update includes updating the share state table when the event set is updated, and updating the share evolution chain after the share state table is updated.

10. A circulation collaboration system based on supply chain whole-link data governance, comprising a construction module, a recording module, a share module, a generation module and an update module, characterized in that: The construction module is used to collect multi-source original data generated in each circulation link of the supply chain, perform format unification, field mapping, de-duplication verification and time and space correction processing on the multi-source original data, and generate a unified data table; and based on the unified data table, an index structure organized according to subject dimension, commodity dimension, time dimension and space dimension is constructed, forming a unified data base layer; The recording module includes extracting data records related to circulation actions based on the data base layer and the index structure, and converting the data records into event records containing subject identifier, commodity identifier, time information, space information and business parameters; the event records are aggregated according to the conditions of consistent subject, adjacent time and close space, forming an event set; The share module is used to represent the commodity source as a plurality of source share units for measurement, and based on the event set, to identify splitting operations, merging operations and transfer operations, to perform quantity allocation and ownership change processing on the source share units, to generate and update a share state table for recording the current subject ownership and spatial ownership of each source share unit; The generation module includes constructing a share evolution chain representing the transfer relationship between the source share units in different subjects and different spaces based on the continuous update results of the source share units in the share state table; The update module is used to update the event set based on the execution feedback data, and to update the share state table when the event set is updated, and to update the share evolution chain after the share state table is updated. And according to the time sequence of each transfer relationship in the share evolution chain, the corresponding circulation behavior chain is generated; The updating module includes generating circulation scheduling results, abnormal handling results and source tracing results based on the circulation behavior chain; The execution feedback data of the circulation scheduling results, the abnormal handling results and the source tracing results are written back to the data base layer; And based on the execution feedback data, the update processing of the event set, the share state table and the share evolution chain is triggered.

Images