Search

CN-121581900-B - Intelligent traceability management method for electronic commerce supply chain commodity

CN121581900BCN 121581900 BCN121581900 BCN 121581900BCN-121581900-B

Abstract

The invention relates to the technical field of electronic commerce supply chains, and discloses an intelligent traceability management method for commodities in an electronic commerce supply chain. The method comprises the steps of accessing a multi-source asynchronous data stream generated in the whole period of supply chain stream, and generating a synchronous tracing basic data set after time stamp alignment and cleaning. Based on this, a space-time trajectory grid of commodity circulation is constructed, the cells of which contain multidimensional commodity state vectors. Carrying out state evolution deduction on the grid, generating a commodity circulation behavior sequence map formed by behavior nodes and edges, carrying out multi-mode deep matching on the map and a preset path template, outputting a circulation decision path and decision reliability, and dynamically optimizing packaging protection parameters according to the result to generate a protection instruction set. The method realizes continuous and refined perception and intelligent path decision-making of commodity circulation states, and improves the accuracy of traceability analysis and the initiative of risk coping.

Inventors

  • HUANG BIN

Assignees

  • 莆田学院

Dates

Publication Date
20260508
Application Date
20260126

Claims (9)

  1. 1. An intelligent traceability management method for electronic commerce supply chain commodities is characterized by comprising the following steps: Accessing a multi-source asynchronous data stream generated by a target commodity in the whole period of supply chain circulation, wherein the multi-source asynchronous data stream comprises a commodity position coordinate sequence, a carrier running state record, a packaging microenvironment parameter and multi-angle packaging visual image data; performing time stamp alignment and data cleaning on the multi-source asynchronous data stream to generate a synchronous tracing basic data set; Constructing a space-time track grid of commodity circulation based on the synchronous tracing basic data set, wherein each grid unit of the space-time track grid corresponds to a geographic area and a time window and contains commodity state vectors in the unit; Carrying out state evolution deduction on the space-time track grid to generate a commodity circulation behavior sequence map, wherein the commodity circulation behavior sequence map is composed of a series of behavior nodes and behavior edges, the behavior nodes represent the states of specific space-time grid units, and the behavior edges represent the paths and conditions of state transition; Carrying out multi-mode deep matching on the commodity circulation behavior sequence map and a preset commodity circulation path template, and generating a circulation decision path and corresponding decision credibility of a target commodity according to a matching result; dynamically optimizing the commodity package protection parameters according to the circulation decision path to generate a protection optimization instruction set containing package reinforcement parameters and environment isolation parameters; the step of performing multi-mode deep matching on the commodity circulation behavior sequence map and a preset commodity circulation path template, and generating circulation decision paths and corresponding decision credibility of target commodities according to matching results comprises the following steps: retrieving one or more commodity circulation path templates corresponding to the target commodity class from a knowledge base, wherein each commodity circulation path template is stored in the form of a standard behavior sequence map and comprises a series of standard behavior nodes and standard behavior edges in an ideal state; comparing the commodity circulation behavior sequence patterns with each commodity circulation path template in a pattern structure and node attributes; The map structure comparison comprises the steps of calculating the topological similarity between the editing distance of the behavior node sequence and the behavior edge connection; the node attribute comparison comprises the steps of calculating multi-dimensional difference metrics between commodity state vectors of corresponding behavior nodes; Fusing the map structure comparison result and the node attribute comparison result to generate an overall matching degree score of the commodity circulation behavior sequence map and each commodity circulation path template; Selecting a commodity circulation path template with the highest overall matching degree score as a reference template; Identifying a behavior node sequence with significant deviation from the reference template in the commodity circulation behavior sequence map, wherein the determination of the significant deviation is based on whether the multi-dimensional difference metric exceeds a preset deviation threshold; Marking the identified behavior node sequences with significant deviation as decision points in the commodity circulation behavior sequence map; Starting from the initial behavior node, penetrating through all decision points until the termination behavior node, and forming a circulation decision path formed by key decision points; And associating a decision credibility with each decision point in the circulation decision path, wherein the decision credibility is calculated based on the matching degree scores of all upstream behavior nodes reaching the decision point.
  2. 2. The intelligent traceability management method of electronic commerce supply chain commodity according to claim 1, wherein said constructing a space-time trajectory grid of commodity circulation based on said synchronized traceability base data set comprises: Dividing the physical path and the time axis of the whole supply chain into a plurality of space-time grid units according to the preset geographic spatial resolution and the time slice length; for each space-time grid cell, the following processing is performed: extracting all data points falling within the time and geographic range of the space-time grid unit from the synchronous tracing basic data set; calculating the distribution center and the dispersion of the commercial grade position coordinates in the space-time grid unit as a position state vector; Counting the frequency distribution characteristics of the carrier running state records in the space-time grid unit, and taking the frequency distribution characteristics as carrier state vectors; calculating the average value, peak value and fluctuation variance of the packing micro-environment parameters in the space-time grid unit as an environment state vector; Performing damage feature extraction and texture change analysis on the multi-angle package visual image data to generate a visual state vector of the space-time grid unit; Splicing the position state vector, the carrier state vector, the environment state vector and the vision state vector to form a commodity state vector representing the complete state of the space-time grid unit; And combining all the space-time grid units and the corresponding commodity state vectors to generate the space-time track grid.
  3. 3. The method for intelligent traceability management of electronic commerce supply chain commodities according to claim 2, wherein said performing state evolution deduction on said space-time trajectory grid to generate commodity circulation behavior sequence patterns comprises: establishing a transfer relation among units according to the front-to-back sequence of adjacent space-time grid units in the space-time track grid on a time axis; two space-time grid units adjacent to each other in front and back are taken as a transfer pair; According to the transfer pair, extracting the variation quantity from the commodity state vector of the previous time space grid unit to the commodity state vector of the next time space grid unit, and calculating a state transfer vector; Mapping the state transition vector into specific behavior type codes based on preset commodity circulation behavior coding rules, wherein the behavior type codes comprise transportation behavior codes, storage behavior codes and loading and unloading behavior codes; Defining each space-time grid cell as a behavior node, wherein the attribute of the behavior node comprises a commodity state vector of the cell; Defining the connection between two behavior nodes with time sequence as a behavior edge, wherein the attribute of the behavior edge comprises a behavior type code and a state transition vector for connecting the two nodes; all behavior nodes in the whole period of the supply chain are connected with the behavior edges to form a commodity circulation behavior sequence map representing the continuous evolution of commodity states.
  4. 4. The method for intelligent traceability management of e-commerce supply chain commodities according to claim 3, wherein said dynamically optimizing commodity packaging protection parameters according to said circulation decision path generates a protection optimization instruction set including packaging reinforcement parameters and environment isolation parameters, comprising: analyzing the circulation decision path, and extracting all decision points with decision reliability lower than an alarm value from the path; Acquiring commodity state vectors in the behavior nodes corresponding to decision points lower than the warning value, and separating a carrier state vector, an environment state vector and a visual state vector from the commodity state vectors; Aiming at the component indicating abnormal vibration or impact in the carrier state vector, calculating the density and the distribution position of the buffer material to be increased, and generating a first type of package reinforcement parameter; aiming at components indicating that temperature, humidity or other harmful factors exceed standards in the environmental state vector, calculating the thickness and material properties of the isolation layer to be enhanced, and generating environmental isolation parameters; Aiming at components indicating package abrasion or deformation in the visual state vector, calculating the strength of the abrasion-resistant coating or a structural reinforcement scheme to be added at the corresponding position, and generating second-type package reinforcement parameters; combining and conflict resolution are carried out on the first type package reinforcement parameters, the environment isolation parameters and the second type package reinforcement parameters generated aiming at the same decision point or a plurality of continuous relevant decision points, so that a protection optimization sub-instruction applicable to the decision point or the circulation path segments corresponding to the continuous relevant decision points is formed; summarizing all protection optimization sub-instructions corresponding to the decision points lower than the warning value, and sequencing according to the sequence of the decision points in the circulation decision path to generate the protection optimization instruction set.
  5. 5. The method for intelligent traceability management of e-commerce supply chain commodities according to claim 4, wherein said combining and conflict resolution of said first type of package reinforcement parameters, environment isolation parameters and second type of package reinforcement parameters generated for the same decision point or a plurality of consecutive relevant decision points to form a protection optimization sub-instruction applicable to the circulation path segment corresponding to the decision point or the plurality of consecutive relevant decision points comprises: listing all package reinforcement parameters and environment isolation parameters generated for the same decision point; Checking whether physical conflicts or performance offsets exist between different parameters, wherein the physical conflicts comprise space occupation conflicts, and the performance offsets comprise reinforcement of one material to reduce the isolation performance of the other material; If physical conflict exists, parameter adjustment is carried out according to a preset conflict resolution rule, and the conflict resolution rule preferentially guarantees a protection dimension with the greatest influence on commodity safety; If the performance offset exists, calculating the offset comprehensive protection efficiency, and iteratively adjusting related parameters until the comprehensive protection efficiency meets the protection requirement of the decision point; And combining the final parameters after conflict resolution and performance cancellation, and binding with corresponding decision point information and circulation path segment information to form the protection optimization sub-instruction.
  6. 6. The intelligent traceability management method of electronic commerce supply chain commodity according to claim 5, further comprising: after the traceability management of a batch of target commodities is completed, collecting a real-time circulation feedback data set of the batch of commodities; Converting the real-time circulation feedback data set into a feedback behavior sequence map; Calculating the overall matching degree score of the feedback behavior sequence map and the reference template as an actual matching score; comparing the actual matching score with a previously predicted matching score to obtain decision reliability deviation; According to the decision reliability deviation, adjusting the standard behavior node attribute, the standard behavior edge connection and the deviation threshold in the commodity circulation path template to realize the self-adaptive optimization of the commodity circulation path template and the deviation judgment rule; And updating the optimized commodity circulation path template and the deviation judging rule to a knowledge base.
  7. 7. The method for intelligent traceability management of e-commerce supply chain commodities according to claim 6, wherein said adjusting the standard behavior node attribute, standard behavior edge connection and said deviation threshold in said commodity circulation path template according to said decision confidence deviation includes: if the decision reliability deviation shows that the actual circulation is better than the expected circulation, analyzing the state vector characteristics of the feedback behavior sequence map which are better than the standard behavior node attributes; Performing forward fine adjustment on attribute vectors of corresponding standard behavior nodes in the commodity circulation path template by using state vector features superior to the standard; simultaneously, releasing a deviation threshold value of the dimension related to the state vector features which are superior to the standard; if the decision reliability deviation indicates that the actual circulation is inferior to the expected circulation, analyzing key behavior nodes and state vectors thereof which lead to the reduced matching degree in the feedback behavior sequence map; standard attribute constraint conditions corresponding to the key behavior nodes in the commodity circulation path template are enhanced; Meanwhile, a deviation threshold value of the dimension related to the key behavior node attribute is tightened; And synchronously adjusting the state transition conditions represented by the standard behavior edges between the key behavior nodes in the commodity circulation path template to enable the state transition conditions to be suitable for the adjusted standard behavior node attributes.
  8. 8. The method for intelligent traceability management of e-commerce supply chain commodities according to claim 1, wherein said method further comprises: Establishing a commodity tracing management knowledge base, wherein the knowledge base is used for storing commodity circulation path templates, behavior type coding rules, state vector comparison rules, deviation thresholds and historical protection optimization instruction sets of different types of commodities; the knowledge base receives new behavior pattern data generated in the commodity circulation behavior sequence pattern matching process; the knowledge base receives effect feedback data from the protection optimization instruction set after execution; incremental updating is carried out on commodity circulation path templates of corresponding classes by utilizing the new behavior pattern data; Verifying and calibrating a behavior type coding rule, a state vector comparison rule and a deviation threshold value by utilizing the effect feedback data; The knowledge base provides updated commodity circulation path templates, behavior type coding rules, state vector comparison rules and deviation thresholds for the next traceability management of the same or similar commodities.
  9. 9. The method for intelligent traceability management of e-commerce supply chain commodities according to claim 1, wherein said method comprises: Accessing a multi-source asynchronous data stream of a target commodity in a supply chain, and processing to obtain a synchronous tracing basic data set; Constructing a space-time track grid for describing the space-time state evolution of the commodity, and deducting and generating a commodity circulation behavior sequence map; carrying out deep matching on the commodity circulation behavior sequence map and a commodity circulation path template in a knowledge base to generate a circulation decision path and decision credibility; dynamically generating a commodity package protection optimization instruction set based on the circulation decision path; And carrying out self-adaptive optimization updating on the commodity circulation path templates and the matching rules by using actual circulation feedback data, and storing experience into a knowledge base.

Description

Intelligent traceability management method for electronic commerce supply chain commodity Technical Field The invention relates to the technical field of electronic commerce supply chains, in particular to an intelligent tracing management method for commodities in an electronic commerce supply chain. Background Commodity traceability management of current e-commerce supply chains generally relies on collection and recording of isolated logistic event data at key nodes. These data typically contain only time, place and simple operating states, forming a record of the flow consisting of discrete points. The lack of effective correlation and continuous registration between the actual state of the article in transit and the operational state of the carrier. The transparency degree of the supply chain is limited to the node level by the recording mode, and the complete state evolution process of the commodity in the space-time continuum cannot be described, so that abnormal positioning, risk early warning and responsibility definition lack of fine data support. The existing path verification and decision support technology is mostly based on preset fixed path rules or simple space-time sequence comparison. Such methods have difficulty handling complex, non-standard flow scenarios in reality. When the circulation process is not completely matched with the preset template, the system can only give general judgment of 'abnormality' or 'mismatch', cannot intelligently infer the path which the commodity is most likely to actually follow, and is difficult to quantitatively evaluate the credibility of the inferred result. The stiffness of path analysis makes the system unable to provide a deep, intelligent basis with confidence assessment for dynamic pack guard optimization and supply chain decisions. Disclosure of Invention The invention aims to provide an intelligent traceability management method for electronic commerce supply chain commodities, which aims to solve the problems in the background technology. In order to achieve the above purpose, the present invention provides an intelligent traceability management method for electronic commerce supply chain commodities, the method comprising: Accessing a multi-source asynchronous data stream generated by a target commodity in the whole period of supply chain circulation, wherein the multi-source asynchronous data stream comprises a commodity position coordinate sequence, a carrier running state record, a packaging microenvironment parameter and multi-angle packaging visual image data; performing time stamp alignment and data cleaning on the multi-source asynchronous data stream to generate a synchronous tracing basic data set; Constructing a space-time track grid of commodity circulation based on the synchronous tracing basic data set, wherein each grid unit of the space-time track grid corresponds to a geographic area and a time window and contains commodity state vectors in the unit; Carrying out state evolution deduction on the space-time track grid to generate a commodity circulation behavior sequence map, wherein the commodity circulation behavior sequence map is composed of a series of behavior nodes and behavior edges, the behavior nodes represent the states of specific space-time grid units, and the behavior edges represent the paths and conditions of state transition; Carrying out multi-mode deep matching on the commodity circulation behavior sequence map and a preset commodity circulation path template, and generating a circulation decision path and corresponding decision credibility of a target commodity according to a matching result; And dynamically optimizing the commodity package protection parameters according to the circulation decision path to generate a protection optimization instruction set containing package reinforcement parameters and environment isolation parameters. Preferably, the constructing a space-time trajectory grid of commodity circulation based on the synchronized tracing basic data set includes: Dividing the physical path and the time axis of the whole supply chain into a plurality of space-time grid units according to the preset geographic spatial resolution and the time slice length; for each space-time grid cell, the following processing is performed: extracting all data points falling within the time and geographic range of the space-time grid unit from the synchronous tracing basic data set; calculating the distribution center and the dispersion of the commercial grade position coordinates in the space-time grid unit as a position state vector; Counting the frequency distribution characteristics of the carrier running state records in the space-time grid unit, and taking the frequency distribution characteristics as carrier state vectors; calculating the average value, peak value and fluctuation variance of the packing micro-environment parameters in the space-time grid unit as an environment state vector; Performing damage