CN-121981465-A - Agricultural industry chain management method and system based on multi-module cooperation

CN121981465ACN 121981465 ACN121981465 ACN 121981465ACN-121981465-A

Abstract

The embodiment of the invention relates to the technical field of agricultural management and discloses an agricultural industry chain management method based on multi-module cooperation, which comprises the steps of generating an electronic order associated with an order placing operation; the method comprises the steps of automatically intercepting a specific live video clip associated with an order placing operation time, uniquely binding the specific live video clip with an electronic order, storing a first digital fingerprint of the specific live video clip, core information of the electronic order, and a second digital fingerprint of a logistics tracking information and a quality inspection report generated in a subsequent supply chain link into a blockchain network as a group of associated data records, dynamically determining a delivery node and a logistics path of the electronic order after the user finishes the order placing, and executing optimized scheduling of supply chain resources based on aggregation information of a plurality of electronic orders collected in a preset time window. According to the scheme provided by the embodiment of the invention, the live video and the full-link traceability information corresponding to the order are verified through the blockchain, so that the consumption trust sense is enhanced.

Inventors

TAN HAIJUN
CHENG YUNZE
CHENG YUNFENG

Assignees

乐禾食品集团股份有限公司

Dates

Publication Date: 20260505
Application Date: 20260121

Claims (10)

1. The agricultural industry chain management method based on multi-module cooperation is characterized by comprising the following steps of: Generating an electronic order associated with the ordering operation in response to a user ordering operation triggered in an agricultural product live broadcast streaming media interface, automatically intercepting a specific live broadcast video clip associated with the ordering operation time, and uniquely binding the specific live broadcast video clip with the electronic order; Storing the first digital fingerprint of the specific live video clip, the core information of the electronic order, and the second digital fingerprint of the logistic tracking information and the quality inspection report generated in the subsequent supply chain link as a set of associated data records into a blockchain network; after the user finishes placing an order, dynamically determining a delivery node and a logistics path of the electronic order; based on the aggregated information of the plurality of electronic orders collected within the preset time window, an optimized scheduling of supply chain resources is performed.
2. The multi-module collaboration-based agricultural industry chain management method of claim 1, wherein the dynamically determining shipping nodes and logistics paths of the electronic order comprises: The method comprises the steps of accessing a real-time inventory database, comprehensively analyzing the geographic position of a user, the inventory distribution of selected commodities and a logistics cost matrix by using a path optimization algorithm with the goal of lowest total cost of performance or shortest promised time, and selecting an optimal delivery node and a corresponding logistics service provider from a plurality of candidate delivery nodes; the performing an optimized schedule of supply chain resources includes: Calling an operation optimizing engine, and taking the aggregation information, the state of the resource pool of the pretreatment center and the state of the physical distribution resource pool as inputs; Grouping orders with similar geographic attributes by applying a clustering algorithm; A cost-minimized or age-optimized resource allocation model is solved for each order group to determine the optimal pretreatment center and logistics path to service the order group and to generate scheduling instructions.
3. The multi-module collaboration-based agricultural industry chain management method of claim 1, wherein the automatically intercepting a particular live video clip associated with the order operation time comprises: In response to receiving the user ordering operation associated with the target commodity at an ordering time point, starting a backtracking analysis process; Acquiring a multi-mode live broadcast data stream in a first preset time window before an ordering time point, wherein the multi-mode live broadcast data stream comprises a video frame sequence, a corresponding audio stream and an interaction event log; Dynamically determining a starting point in time of the respective content segment by analyzing the multi-modal live data stream, wherein the determining of the starting point in time is based on detecting a starting event in which the target commodity is in focus in the live data stream, the starting event including a time at which the target commodity is visually identified as a salient object in a video frame, a time at which a specific descriptive phrase related to the target commodity is identified in an audio stream, or a time at which an interactivity event indicating that the target commodity starts to be promoted is recorded; Live data within a time period from a start time point to a next time point is defined as a specific live video clip.
4. The multi-module collaboration-based agricultural industry chain management method as claimed in claim 3, wherein dynamically determining the starting point in time of the respective content segment comprises: performing real-time object detection on the video frame sequence to identify and track the time of occurrence, size and location of the target commodity in the picture; When the proportion of the picture area occupied by the target commodity in the continuous video frames exceeds a first threshold value and the duration exceeds a second threshold value, judging that a first type of initial event occurs; Performing real-time speech recognition on the audio stream to convert to a text stream; determining that a second type of initiating event occurs when one or more predefined sets of key phrases are identified in the text stream that are strongly associated with the target commodity; Analyzing the interaction event log, and judging that a third type of initial event occurs when the operation of setting the top, highlighting or issuing the associated coupon of the target commodity in the commodity list is identified; And determining the earliest occurring time point in the first type of initial event, the second type of initial event and the third type of initial event as an initial time point.
5. The multi-module collaboration-based agricultural industry chain management method of claim 4, wherein the management method further comprises: Transcribing the audio stream in the specific live video segment to generate a transcribed text; extracting sentences containing attribute claims, quality commitments or source statements about the target commodity from the transcribed text; performing key frame extraction and analysis on video frames in the specific live video segment to generate text description describing visual presentation of the target commodity; Fusing the extracted sentences with the text description to generate a section of natural language content abstract related to the specific live video clip, and storing the hash value of the content abstract in association with the electronic order; The management method further comprises the following steps: in the live broadcast streaming media interface, presenting a thumbnail of a specific live broadcast video clip or the content abstract to a user triggering the ordering operation; receiving acknowledgement or objection feedback from the respective user; If the objection feedback is received, a manual auditing flow is started, and whether the starting and ending time of the specific live video segment is corrected or the backtracking analysis process is re-executed is determined according to an auditing result; binding the confirmation feedback or final audit confirmation result of the user as a verification evidence with the electronic order; the first digital fingerprint is generated by: packaging video data, audio data and associated metadata corresponding to the specific live video clip into a data object; calculating a cryptographic hash value of the data object as a first digital fingerprint; and storing the data object in a distributed file storage system, and recording the index of the storage address and the encrypted hash value in the distributed account book.
6. The multi-module collaboration-based agricultural industry chain management method of claim 1, wherein the management method further comprises: Analyzing video streams from a field unmanned aerial vehicle or a fixed camera to perform early yield and quality predictions, wherein the analyzing comprises identifying crop plants in video frames and counting the number of plants or fruits per unit area, and detecting pest and disease features on leaves or fruits; estimating the total yield of the future harvest time and the expected proportion of each quality grade by using a time sequence prediction model based on the counted number, the detected pest and disease characteristics and the accessed time sequence meteorological data; before said performing the optimized scheduling of supply chain resources, further comprising: Analyzing live broadcast plan text of the anchor to identify potential high-demand commodities and target sales areas; The historical sales data and the live broadcast plan analysis result are combined, and the demand quantity of the sub-regional and sub-commodity in a specific time period in the future is predicted; Based on the predicted demand, the current inventory, and the deployment cost, a pre-deployment instruction to the pre-warehouse is generated to deploy the inventory in advance to a warehouse proximate to the predicted high demand area.
7. The multi-module collaboration-based agricultural industry chain management method of claim 6, wherein the management method further comprises: establishing attribute portraits for a plurality of live account numbers, and analyzing regional sales trends based on real-time user interaction data and regional information of each live account number; According to the agricultural product yield prediction result and the attribute portrait of the live account, different live content materials are distributed for different live accounts; During live broadcasting, based on the user interaction data analyzed in real time, targeted sales strategy prompts are pushed to a host; The supply chain collaborative execution module is used for responding to an order generated by live broadcast sales, and integrating block chain tracing and intelligent matching of storage logistics resources so as to finish the delivery of agricultural products from a production end to a consumption end; After the user interaction data based on the real-time analysis, further comprising: monitoring interactive behaviors of users from different geographical areas in each living broadcast room, wherein the interactive behaviors comprise clicking, stay time length and barrage comments of specific commodities; combining external data, including weather data and historical consumption data of a target area, and constructing a short-term area demand prediction model; And identifying an area presenting high potential demands for the specific agricultural product class based on the output of the short-term area demand prediction model, and generating a sales trend report.
8. The multi-module collaboration-based agricultural industry chain management method of claim 7, further comprising, after the analyzing the video stream from the field drone or the fixed camera to perform early yield and quality predictions: generating and dynamically updating a multi-dimensional supply prediction curve for the corresponding production unit based on the continuously received agricultural condition data and agricultural events, the supply prediction curve comprising a probability distribution of expected yield quantity and quality attributes of the production unit at a future point in time; generating and dynamically updating a probabilistic demand prediction curve of the sales area or channel for the agricultural products at a future point in time based on the market data and the user interaction data; Matching the latest supply prediction curve with the demand prediction curve to generate a corresponding collaborative decision instruction set, wherein the collaborative decision instruction set comprises agricultural intervention advice aiming at the production unit, pricing and marketing strategy advice aiming at a sales terminal and preset dialing advice aiming at a logistics terminal; And sending the collaborative decision instruction set to a corresponding user terminal.
9. The multi-module collaboration-based agricultural industry chain management method of claim 8, wherein the matching the latest supply prediction curve with the demand prediction curve comprises: Comparing a current or future supply prediction curve with a reference demand prediction curve, classifying the current state into one of a plurality of predefined difference scenes according to the difference modes of total quantity, time sequence and quality structure, wherein the predefined difference scenes comprise a first scene, a second scene, a third scene, a fourth scene and a fifth scene, the first scene is used for representing that the total quantity of supply exceeds the reference demand total quantity, the peak time and the quality structure are unchanged, the second scene is used for representing that the total quantity of supply is lower than the reference demand total quantity, the peak time and the quality structure are unchanged, the third scene is used for representing that the peak time of supply is significantly advanced or delayed compared with the peak value of the reference demand, the fourth scene is used for representing that the high-grade product occupation ratio in the total quantity of supply is significantly improved or reduced compared with the reference expectation, and the fifth scene is used for representing that the supply curve is severely fluctuated due to disastrous events; Based on the classified difference scenes, a corresponding targeted live broadcast marketing strategy instruction set is retrieved from a strategy knowledge base and combined to generate, wherein the instruction set comprises a channel distribution strategy, a content theme strategy and a pricing promotion strategy.
10. An agricultural industry chain management system based on multi-module cooperation, which is characterized by comprising: The generation module is used for responding to the user ordering operation triggered in the agricultural product live broadcast streaming media interface, generating an electronic order associated with the ordering operation, automatically intercepting a specific live broadcast video segment associated with the ordering operation time, and uniquely binding the specific live broadcast video segment with the electronic order; The storage module is used for storing the first digital fingerprint of the specific live video clip, the core information of the electronic order, and the second digital fingerprint of the logistics tracking information and the quality inspection report generated in the subsequent supply chain link into a blockchain network as a group of associated data records; The logistics determining module is used for dynamically determining a delivery node and a logistics path of the electronic order after the user finishes placing the order; and the scheduling module is used for executing the optimized scheduling of the supply chain resources based on the aggregation information of the plurality of electronic orders collected in the preset time window.

Description

Agricultural industry chain management method and system based on multi-module cooperation Technical Field The invention relates to the technical field of agricultural industry management, in particular to an agricultural industry chain management method and system based on multi-module cooperation. Background At present, with the rapid development of the internet of things, big data and electronic commerce technologies, traditional agriculture is undergoing a digital transformation from production to sales. However, the digital management of the current agricultural industry chain still faces a serious challenge, and obvious information islands and coordination barriers exist among links, so that the problems of low system efficiency, serious resource waste, uncontrollable market risk and the like are caused. Disclosure of Invention Aiming at the defects, the embodiment of the invention discloses an agricultural industry chain management method based on multi-module cooperation, which can improve the whole industry cooperation efficiency and promote the sales of agricultural products. The first aspect of the embodiment of the invention discloses an agricultural industry chain management method based on multi-module cooperation, which comprises the following steps: Generating an electronic order associated with the ordering operation in response to a user ordering operation triggered in an agricultural product live broadcast streaming media interface, automatically intercepting a specific live broadcast video clip associated with the ordering operation time, and uniquely binding the specific live broadcast video clip with the electronic order; Storing the first digital fingerprint of the specific live video clip, the core information of the electronic order, and the second digital fingerprint of the logistic tracking information and the quality inspection report generated in the subsequent supply chain link as a set of associated data records into a blockchain network; after the user finishes placing an order, dynamically determining a delivery node and a logistics path of the electronic order; based on the aggregated information of the plurality of electronic orders collected within the preset time window, an optimized scheduling of supply chain resources is performed. As an optional implementation manner, in the first aspect of the embodiment of the present invention, the dynamically determining a shipping node and a logistics path of the electronic order includes: The method comprises the steps of accessing a real-time inventory database, comprehensively analyzing the geographic position of a user, the inventory distribution of selected commodities and a logistics cost matrix by using a path optimization algorithm with the goal of lowest total cost of performance or shortest promised time, and selecting an optimal delivery node and a corresponding logistics service provider from a plurality of candidate delivery nodes; the performing an optimized schedule of supply chain resources includes: Calling an operation optimizing engine, and taking the aggregation information, the state of the resource pool of the pretreatment center and the state of the physical distribution resource pool as inputs; Grouping orders with similar geographic attributes by applying a clustering algorithm; A cost-minimized or age-optimized resource allocation model is solved for each order group to determine the optimal pretreatment center and logistics path to service the order group and to generate scheduling instructions. As an optional implementation manner, in the first aspect of the embodiment of the present invention, the automatically intercepting a specific live video clip associated with the ordering operation time includes: In response to receiving the user ordering operation associated with the target commodity at an ordering time point, starting a backtracking analysis process; Acquiring a multi-mode live broadcast data stream in a first preset time window before an ordering time point, wherein the multi-mode live broadcast data stream comprises a video frame sequence, a corresponding audio stream and an interaction event log; Dynamically determining a starting point in time of the respective content segment by analyzing the multi-modal live data stream, wherein the determining of the starting point in time is based on detecting a starting event in which the target commodity is in focus in the live data stream, the starting event including a time at which the target commodity is visually identified as a salient object in a video frame, a time at which a specific descriptive phrase related to the target commodity is identified in an audio stream, or a time at which an interactivity event indicating that the target commodity starts to be promoted is recorded; Live data within a time period from a start time point to a next time point is defined as a specific live video clip. As an optional implementation manner, in the fi