Agricultural product supply chain management method and system based on order data

Abstract

The invention provides an agricultural product supply chain management method and system based on order data. The method comprises the following steps: acquiring agricultural product order data, agricultural product harvesting data and agricultural product storage data; respectively processing the data in batches; constructing a comprehensive supply chain network based on all the processed data; the comprehensive supply chain network is a heterogeneous graph model added with a timestamp and a time window, heterogeneous nodes in the comprehensive supply chain network represent entities in a comprehensive supply chain, including order nodes, greenhouse nodes and warehouse nodes, and heterogeneous edges represent relationships among the entities; generating a plurality of undetermined supply chain management strategies based on the comprehensive supply chain network; and evaluating the plurality of undetermined supply chain management strategies, and selecting the undetermined supply chain management strategy with the highest evaluation value as a target supply chain management strategy. According to the technical scheme provided by the invention, the efficiency and flexibility of supply chain management are improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of agricultural product supply chain management, and in particular, to an agricultural product supply chain management method and system based on order data. Background Art

[0002] With the rapid development of the global economy and the diversification of consumer demand, agricultural product supply chain management has become increasingly complex. Especially in the agricultural product supply chain, due to the seasonality, perishability and volatility of agricultural products and market demand, agricultural product supply chain management faces greater challenges.

[0003] Traditional agricultural product supply chain management methods often rely on manual experience and simple information systems, which make it difficult to respond quickly to market changes and customer needs, thus affecting the timeliness and accuracy of decision-making. Summary of the invention

[0004] The embodiments of the present application provide a method and system for agricultural product supply chain management based on order data, so as to solve the problems of low efficiency and poor flexibility of supply chain management existing in the prior art.

[0005] In a first aspect, an embodiment of the present application provides an agricultural product supply chain management method based on order data, comprising: Obtain agricultural product order data in the order system, agricultural product harvest data collected by the greenhouse monitoring system, and agricultural product storage data recorded by the warehouse monitoring system; respectively batch processing the agricultural product order data, agricultural product harvesting data and agricultural product storage data; A comprehensive supply chain network is constructed based on processed agricultural product order data, processed agricultural product harvest data, and processed agricultural product storage data; the comprehensive supply chain network is a heterogeneous graph model with timestamps and time windows added, and the heterogeneous nodes in the comprehensive supply chain network represent various entities in the comprehensive supply chain, including order nodes, greenhouse nodes, and warehouse nodes, and the heterogeneous edges represent the relationships between the entities; generating a plurality of pending supply chain management strategies based on the integrated supply chain network; Evaluate multiple pending supply chain management strategies, and select the pending supply chain management strategy with the highest evaluation value as the target supply chain management strategy.

[0006] Optionally, the building of a comprehensive supply chain network based on the processed agricultural product order data, the processed agricultural product harvest data, and the processed agricultural product storage data includes: Define order nodes based on processed agricultural product order data, define greenhouse nodes based on processed agricultural product harvesting data, and define warehouse nodes based on processed agricultural product storage data; Based on the processed agricultural product order data and the processed agricultural product harvesting data, the heterogeneous edges between the order nodes and the greenhouse nodes are defined; based on the processed agricultural product order data and the processed agricultural product storage data, the heterogeneous edges between the order nodes and the warehouse nodes are defined; based on the processed agricultural product harvesting data and the processed agricultural product storage data, the heterogeneous edges between the greenhouse nodes and the warehouse nodes are defined; Add corresponding timestamps to each heterogeneous node and heterogeneous edge to obtain heterogeneous nodes and heterogeneous edges with timestamps; Define a time window, and divide all heterogeneous nodes and heterogeneous edges with timestamps according to the time window, to obtain a set of heterogeneous nodes and a set of heterogeneous edges divided by the time window; Based on the heterogeneous node sets and heterogeneous edge sets divided by time windows, the sliding window method and time series analysis technology are used to obtain the time dependency relationship of heterogeneous nodes and the time dependency relationship of heterogeneous edges; A comprehensive supply chain network is constructed based on heterogeneous nodes, heterogeneous edges, and the time dependencies of the heterogeneous nodes and the time dependencies of the heterogeneous edges.

[0007] Optionally, the method of obtaining the time dependency of the heterogeneous nodes and the time dependency of the heterogeneous edges based on the heterogeneous node set and the heterogeneous edge set divided by the time window by using a sliding window method and a time series analysis technology includes: Based on the heterogeneous node set and heterogeneous edge set divided by time window, the size and step size of the sliding window are defined; Based on the size and step size of the sliding window, sliding window processing is performed on the timestamps of the heterogeneous nodes and the heterogeneous edges, and based on the timestamps of the heterogeneous nodes and the heterogeneous edges in the sliding window, statistical features in the sliding window are calculated; Perform time series analysis based on the statistical characteristics within the sliding window to obtain the time series analysis results; Construct a time-dependency target matrix based on the time series analysis results; The time dependencies of heterogeneous nodes and the time dependencies of heterogeneous edges are determined based on the time dependency target matrix.

[0008] Optionally, constructing a time dependency target matrix based on the time series analysis results includes: Select the target dependency measurement method based on the time series analysis results; The target dependency measurement method is used to calculate the dependency strength values ​​between heterogeneous nodes and heterogeneous edges; Based on the dependency strength values ​​between heterogeneous nodes and heterogeneous edges, an initial matrix of temporal dependencies is constructed; Based on the initial matrix of time dependency, the uncertainty of time dependency is evaluated using the Bayesian method to obtain the uncertainty quantification result, and the time dependency target matrix is ​​constructed. Each element in the time dependency target matrix represents a dependency, including the dependency strength value and its uncertainty quantification result.

[0009] Optionally, the determining the time dependencies of heterogeneous nodes and the time dependencies of heterogeneous edges based on the time dependency target matrix includes: Based on the time dependency target matrix, a graph clustering algorithm is used to cluster heterogeneous nodes and edges respectively to obtain heterogeneous node clustering results and heterogeneous edge clustering results. Using a community detection method to perform community detection on the heterogeneous node clustering results and the heterogeneous edge clustering results, respectively, to obtain the community structure of the heterogeneous nodes and the community structure of the heterogeneous edges; Based on the community structure of the heterogeneous nodes and the community structure of the heterogeneous edges, a graph convolutional network is used to determine the temporal dependencies of the heterogeneous nodes and the temporal dependencies of the heterogeneous edges.

[0010] Optionally, constructing a comprehensive supply chain network based on heterogeneous nodes, heterogeneous edges, and the time dependencies of the heterogeneous nodes and the time dependencies of the heterogeneous edges includes: Obtain customer relationship data and customer rating data; Building a customer relationship network based on the customer relationship data, and determining customer grade attributes based on the customer grade data; the nodes in the customer relationship network are customer nodes, and the edges of the customer relationship network represent the relationships between customers; Building an initial network based on heterogeneous nodes, heterogeneous edges, and the time dependencies of the heterogeneous nodes and the time dependencies of the heterogeneous edges, and integrating the customer relationship network and the customer grade attribute into the initial network to obtain an integrated network; Prioritize customers based on the integrated network and determine order processing and resource scheduling plans based on the customer priority ranking results; Based on the order processing and resource scheduling solution, a comprehensive supply chain network is constructed.

[0011] Optionally, the evaluating of multiple pending supply chain management strategies comprises the following steps: Determine the evaluation indicators and evaluation methods for the proposed supply chain management strategy; Based on the evaluation index, each pending supply chain management strategy is evaluated using the evaluation method.

[0012] In a second aspect, the embodiment of the present application provides an agricultural product supply chain management system based on order data, including: The acquisition module is used to obtain the agricultural product order data in the order system, the agricultural product harvesting data collected by the greenhouse monitoring system, and the agricultural product storage data recorded by the warehouse monitoring system; A processing module, used for batch processing the agricultural product order data, agricultural product harvesting data and agricultural product storage data respectively; A construction module is used to construct a comprehensive supply chain network based on processed agricultural product order data, processed agricultural product harvest data, and processed agricultural product storage data; the comprehensive supply chain network is a heterogeneous graph model with timestamps and time windows added, and the heterogeneous nodes in the comprehensive supply chain network represent various entities in the comprehensive supply chain, including order nodes, greenhouse nodes, and warehouse nodes, and the heterogeneous edges represent the relationships between the entities; A generation module, used for generating a plurality of pending supply chain management strategies based on the comprehensive supply chain network; The evaluation and selection module is used to evaluate multiple pending supply chain management strategies and select the pending supply chain management strategy with the highest evaluation value as the target supply chain management strategy.

[0013] In a third aspect, an embodiment of the present application provides a computing device, comprising a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement an agricultural product supply chain management method based on order data as described in any one of the first aspects.

[0014] In a fourth aspect, an embodiment of the present application provides a computer storage medium storing a computer program, which, when executed by a computer, implements an agricultural product supply chain management method based on order data as described in any one of the first aspects.

[0015] In an embodiment of the present application, a method for managing an agricultural product supply chain based on order data is provided, the method comprising: obtaining agricultural product order data in an order system, agricultural product harvesting data collected by a greenhouse monitoring system, and agricultural product storage data recorded by a warehouse monitoring system; batch processing the agricultural product order data, agricultural product harvesting data, and agricultural product storage data respectively; constructing a comprehensive supply chain network based on the processed agricultural product order data, processed agricultural product harvesting data, and processed agricultural product storage data; the comprehensive supply chain network is a heterogeneous graph model with timestamps and time windows, and the heterogeneous nodes in the comprehensive supply chain network represent various entities in the comprehensive supply chain, including order nodes, greenhouse nodes, and warehouse nodes, and the heterogeneous edges represent the relationship between entities; generating multiple pending supply chain management strategies based on the comprehensive supply chain network; evaluating multiple pending supply chain management strategies, and selecting the pending supply chain management strategy with the highest evaluation value as the target supply chain management strategy. In the embodiment of the present application, the centralized management and batch processing of multiple data improve the accuracy of the data and provide a reliable basis for subsequent analysis and decision-making. The heterogeneous graph model with timestamps and time windows can dynamically reflect the time dependency in the supply chain, improve the flexibility and adaptability of supply chain management, and ensure rapid response to market changes and customer needs. Generate multiple pending supply chain management strategies, provide multiple options, increase the flexibility of supply chain management, and select the most appropriate pending supply chain management strategy according to market changes and customer needs, thereby improving the adaptability of supply chain management. Furthermore, in the process of building a comprehensive supply chain network, the embodiment of the present application uses a sliding window method and time series analysis technology to monitor each link in the supply chain in real time, dynamically capture the time dependency of each node and edge in the supply chain, ensure that the model can reflect the latest data changes, and dynamically adjust according to the latest data, thereby improving the flexibility and adaptability of agricultural product supply chain management.

[0016] These and other aspects of the present application will become more apparent from the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 A flowchart of a method for managing an agricultural product supply chain based on order data provided in an embodiment of the present application; Figure 2A schematic diagram of the structure of an agricultural product supply chain management system based on order data provided in an embodiment of the present application; Figure 3 A schematic diagram of the structure of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0019] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0020] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The sequence numbers of the operations, such as 11, 12, etc., are only used to distinguish between different operations, and the sequence numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to different types.

[0021] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0022] Figure 1 A flowchart of a method for managing an agricultural product supply chain based on order data is provided in an embodiment of the present application, such as Figure 1 As shown, the method includes: S11. Acquire the agricultural product order data in the order system, the agricultural product harvesting data collected by the greenhouse monitoring system, and the agricultural product storage data recorded by the warehouse monitoring system.

[0023] It should be understood that agricultural product order data may refer to order information about agricultural products obtained from the order system, including: order number, customer name, scheduled agricultural product type, scheduled agricultural product quantity, scheduled delivery time, scheduled delivery location, etc. Agricultural product harvesting data may refer to harvesting information about agricultural products collected from the greenhouse monitoring system, including: greenhouse number, overall greenhouse planting area, planting area of ​​each agricultural product, harvesting time, harvesting personnel, harvesting tools, harvesting vehicle location, harvesting quantity, harvesting quality and commodity rate, etc. Agricultural product storage data may refer to storage information about agricultural products recorded from the warehouse monitoring system, including storage location, storage time, outbound time, inventory, inventory turnover rate, inventory accuracy, storage conditions, preservation measures, storage cost, loss rate, order matching, inventory warning, etc.

[0024] S12. Batch process the agricultural product order data, agricultural product harvest data and agricultural product storage data respectively. Batch processing can refer to the centralized processing of a large amount of data of the same type, including data cleaning, format conversion, deduplication and other operations to ensure the integrity and accuracy of the data. Among them, data cleaning can refer to removing duplicate data, correcting erroneous data, and filling missing data. Format conversion refers to converting data into a unified format. Deduplication can ensure the uniqueness of each piece of data.

[0025] S13. Construct a comprehensive supply chain network based on processed agricultural product order data, processed agricultural product harvest data, and processed agricultural product storage data; the comprehensive supply chain network is a heterogeneous graph model that adds timestamps and time windows. The heterogeneous nodes in the comprehensive supply chain network represent various entities in the comprehensive supply chain, including order nodes, greenhouse nodes, and warehouse nodes. The heterogeneous edges represent the relationships between entities.

[0026] It should be understood that the integrated supply chain network can refer to a heterogeneous graph model containing multiple entities (such as order nodes, greenhouse nodes, warehouse nodes) and their relationships (such as the relationship between orders and greenhouses, the relationship between orders and warehouses, and the relationship between greenhouses and warehouses). The model adds timestamps and time windows, which can dynamically reflect the time dependencies in the supply chain. Among them, the timestamp can refer to the specific time when the corresponding events of each heterogeneous node and heterogeneous edge occur, which is used to identify the time attribute of the data. The time window can refer to a fixed time period, which is used to divide and analyze the time dependencies of data.

[0027] This embodiment can adjust heterogeneous nodes and heterogeneous edges according to business needs. For example, if there is a business demand for controlling the transportation route of the harvesting vehicle (including path planning, time optimization, transportation cost optimization, inventory cost optimization, transportation capacity optimization, personnel scheduling optimization, transportation safety, emergency handling, etc.), the heterogeneous node can also include a harvesting and transportation node. The heterogeneous edge can also include: the relationship between harvesting and transportation and orders, the relationship between harvesting and transportation and greenhouses, and the relationship between harvesting and transportation and warehouses. In another example, if there is a business demand for controlling the transportation route of the order transport vehicle, the heterogeneous node can also include an order transportation node. The heterogeneous edge can also include: the relationship between order transportation and orders, the relationship between order transportation and greenhouses, and the relationship between order transportation and warehouses. Therefore, the integrated supply chain network can realize dynamic adjustment of nodes and edges, and improve adaptability in different scenarios. In another example, the heterogeneous node can also include an agricultural product node, and the attributes of the agricultural product node include the growth cycle of the agricultural product, the average yield of the agricultural product in the historical time period, etc.

[0028] S14. Generate multiple pending supply chain management strategies based on the comprehensive supply chain network.

[0029] It should be understood that the pending supply chain management strategy may refer to multiple possible supply chain management strategies generated based on the comprehensive supply chain network. The supply chain management strategy may include strategies for at least one of the links of agricultural product planting, order receiving, harvesting, storage, and transportation. For example: The supply chain management strategy includes the following three specific strategies: Strategy 1: Increase the planting area of ​​a certain agricultural product and reduce the planting area of ​​another agricultural product. Strategy 2: Adjust the harvesting time of the greenhouse to ensure that the order is delivered on time. Strategy 3: Optimize the inventory management of the warehouse and reduce inventory costs. Strategy 4: Adjust the transportation route of the order transport vehicle.

[0030] S15. Evaluate multiple pending supply chain management strategies, and select the pending supply chain management strategy with the highest evaluation value as the target supply chain management strategy.

[0031] It should be understood that the evaluation value may refer to the score obtained after evaluating each pending supply chain management strategy, which is used to select the optimal supply chain management strategy. The target supply chain management strategy may refer to the pending supply chain management strategy with the highest evaluation value.

[0032] In a specific embodiment, this embodiment can dynamically adjust the planting plan of crops corresponding to agricultural products in the greenhouse through agricultural product order data, and use it as the content of the supply chain management strategy. Optionally, the greenhouse monitoring system includes a sensor network arranged in the connecting corridor between the greenhouses. This embodiment can use the sensor network to monitor the location information of the agricultural product harvesting vehicle, and predict the time period from picking to packing of agricultural products based on the location information. Optionally, this embodiment can identify and record the type of harvested agricultural products and their weight in real time, and provide predictive delivery capabilities for orders. Optionally, this embodiment can record the proportion of the harvested weight and the finished fruit (or commodity fruit) in the greenhouse in real time, and then provide strategies for harvesting time, transportation routes, etc. for the harvesting link according to the commodity rate. Optionally, agricultural product order data, agricultural product harvesting data, and agricultural product storage data provide big data support for the improvement of the commodity rate and storage efficiency of agricultural products.

[0033] Therefore, the centralized management and batch processing of multiple data in the embodiments of the present application improve the integrity and accuracy of the data, and provide a reliable basis for subsequent analysis and decision-making. The heterogeneous graph model with timestamps and time windows can dynamically reflect the time dependencies in the supply chain, improve the flexibility and adaptability of supply chain management, and ensure rapid response to market changes and customer needs. Generate multiple pending supply chain management strategies, provide multiple options, increase the flexibility of supply chain management, ensure that the target supply chain management strategy is optimal through evaluation and selection, and improve the efficiency and adaptability of supply chain management.

[0034] In some optional embodiments, S13, building a comprehensive supply chain network based on the processed agricultural product order data, the processed agricultural product harvest data, and the processed agricultural product storage data includes: Step 131: define an order node based on the processed agricultural product order data, define a greenhouse node based on the processed agricultural product harvesting data, and define a warehouse node based on the processed agricultural product storage data.

[0035] Among them, in the process of defining the order node, each order can be regarded as an order node, and the order node attributes include at least one of the following: order number, customer name, scheduled agricultural product type, scheduled agricultural product quantity, scheduled delivery time, scheduled delivery location, etc.; in the process of defining the greenhouse node, each greenhouse can be regarded as a greenhouse node, and the greenhouse node attributes include at least one of the following: greenhouse number, overall greenhouse planting area, planting area of ​​each agricultural product, harvesting time, harvesting personnel, harvesting tools, harvesting vehicle location, harvesting quantity, harvesting quality and commodity rate; warehouse node attributes include at least one of the following: warehouse number, storage location, warehousing time, outbound time, inventory, inventory turnover rate, inventory accuracy, storage conditions, preservation measures, storage cost, loss rate, order matching, inventory warning, etc.

[0036] Step 132: define heterogeneous edges between order nodes and greenhouse nodes based on the processed agricultural product order data and the processed agricultural product harvesting data, define heterogeneous edges between order nodes and warehouse nodes based on the processed agricultural product order data and the processed agricultural product storage data, and define heterogeneous edges between greenhouse nodes and warehouse nodes based on the processed agricultural product harvesting data and the processed agricultural product storage data.

[0037] It should be understood that the heterogeneous edge between the order node and the greenhouse node represents the association relationship between the order and the greenhouse, including the order number and the greenhouse number. The heterogeneous edge between the order node and the warehouse node represents the association relationship between the order and the warehouse, including the order number and the warehouse number. The heterogeneous edge between the greenhouse node and the warehouse node represents the association relationship between the greenhouse and the warehouse, including the greenhouse number and the warehouse number.

[0038] Step 133: Add corresponding timestamps to each heterogeneous node and heterogeneous edge to obtain heterogeneous nodes and heterogeneous edges with timestamps.

[0039] Exemplarily, an order node includes order node attributes and a timestamp of order generation. A greenhouse node includes greenhouse node attributes and a timestamp indicating harvest time. A warehouse node includes warehouse node attributes and a timestamp indicating storage time and storage time.

[0040] In another exemplary embodiment, the heterogeneous edge between the order node and the greenhouse node includes a timestamp indicating the order generation time and the harvest time. The heterogeneous edge between the order node and the warehouse node includes a timestamp indicating the order generation time, the storage time, and the outbound time. The heterogeneous edge between the greenhouse node and the warehouse node includes a timestamp indicating the harvest time, the storage time, and the outbound time.

[0041] Step 134: define a time window, and divide all heterogeneous nodes and heterogeneous edges with timestamps according to the time window, to obtain a heterogeneous node set and a heterogeneous edge set divided by the time window.

[0042] For example, the time window is set to one week, that is, the length of each time window is 7 days. All heterogeneous nodes and heterogeneous edges with timestamps are divided according to the time window. For example, order nodes and related edges are divided into multiple 7-day time windows.

[0043] Step 135: Based on the heterogeneous node set and heterogeneous edge set divided by time windows, a sliding window method and time series analysis technology are used to obtain the time dependency relationship of the heterogeneous nodes and the time dependency relationship of the heterogeneous edges.

[0044] The sliding window method may refer to capturing the dynamic changes of data by setting a time window of a fixed length (e.g., 7 days) and gradually moving the window. The time series analysis technology may refer to predicting future development trends by analyzing historical data. In this embodiment, the time series analysis technology is used to analyze the relationship between the order generation time and the harvest time, and predict the future harvest time; it is also used to analyze the relationship between the harvest time and the warehousing time, and predict the future warehousing time; it is also used to analyze the relationship between the order generation time and the warehousing time, and predict the future warehousing time.

[0045] Step 136: construct a comprehensive supply chain network based on heterogeneous nodes, heterogeneous edges, time dependencies of heterogeneous nodes, and time dependencies of heterogeneous edges.

[0046] This embodiment ensures that the time attributes of each heterogeneous node and heterogeneous edge are clear by adding timestamps, and can dynamically capture the time dependencies in the supply chain. Through the sliding window method, the data changes in each time window are gradually analyzed to ensure that the model can reflect the latest data changes. The use of time series analysis technology can improve the prediction accuracy of key indicators such as market demand, inventory levels, and transportation time. In addition, through time window division, real-time monitoring of each link in the supply chain can be achieved, and dynamic adjustments can be made according to the latest data to ensure the smooth operation of the supply chain. By analyzing time dependencies, powerful decision support is provided for supply chain management. The sliding window method enables the system to flexibly respond to market changes and customer needs quickly, improves the flexibility of supply chain management, and improves the efficiency and effectiveness of supply chain management.

[0047] In the above embodiment, step 135, based on the heterogeneous node set and the heterogeneous edge set divided by the time window, the sliding window method and the time series analysis technology are used to obtain the time dependency relationship of the heterogeneous nodes and the time dependency relationship of the heterogeneous edges, including: Step a1: Based on the heterogeneous node set and heterogeneous edge set divided by the time window, define the size and step size of the sliding window. The size of the sliding window can be understood as the length of the sliding window, such as 7 days. The step size of the sliding window can refer to the distance the sliding window moves each time, such as 1 day.

[0048] Step a2: Based on the size and step length of the sliding window, the timestamps of the heterogeneous nodes and edges are subjected to sliding window processing, and based on the timestamps of the heterogeneous nodes and edges in the sliding window, the statistical features in the sliding window are calculated. The sliding window processing may refer to gradually moving the sliding window by one step length each time, and extracting the timestamps of the heterogeneous nodes and edges in the sliding window. The statistical features include the mean of the timestamp, the variance of the timestamp, the maximum value, the minimum value, the median value of the timestamp, etc. The variance of the timestamp indicates the degree of fluctuation of the timestamp.

[0049] Step a3: Perform time series analysis based on the statistical features in the sliding window to obtain time series analysis results. In step a3, a time series analysis technology is used, which can be an autoregressive integrated moving average model, a long short-term memory network, etc.

[0050] For example, in order to predict the harvesting time, the autoregressive integrated moving average model in this embodiment can adopt the following formula: ; in, Indicates time harvest time; is a constant term, represents the autoregressive order, which represents the order of the autoregressive part in the autoregressive integrated moving average model; is the autoregressive coefficient, indicating time The value and past The linear relationship between the values ​​at each time point; Indicates time harvest time; For time The error term of is the moving average order, which indicates the order of the moving average part in the autoregressive integrated moving average model; is the moving average coefficient, indicating time The error term and the past The linear relationship between the error terms at each time point; Indicates time The error term of Indicates time The amount of agricultural products harvested; For time Storage conditions (e.g. storage temperature, etc.); Indicates time The time when agricultural products enter the warehouse; It refers to the coefficient of harvest quantity, which indicates the influence of harvest quantity on harvest time; It refers to the coefficient of storage conditions, which indicates the degree of influence of storage conditions on harvest time; It can refer to the coefficient of storage time, which indicates the influence of storage time on harvest time.

[0051] It should be noted that the prediction of order generation time and warehousing time can also be obtained using the above-mentioned autoregressive integrated moving average model. The formula used is similar to the above-mentioned formula and will not be repeated in this embodiment.

[0052] Step a4: construct a time dependency target matrix based on the time series analysis results.

[0053] It should be understood that constructing a time dependency target matrix can convert the time series analysis results into a matrix form, and each element in the time dependency target matrix represents the degree of time dependency between two heterogeneous nodes or heterogeneous edges.

[0054] Step a5: Determine the time dependencies of heterogeneous nodes and the time dependencies of heterogeneous edges based on the time dependency target matrix.

[0055] Through the above steps, the embodiments of the present application can dynamically capture the time dependencies in the supply chain, improve prediction accuracy, realize real-time monitoring and adjustment, enhance decision support, improve the flexibility of agricultural product supply chain management, and improve the efficiency and effectiveness of supply chain management, which can help agricultural product production companies better cope with market changes and respond quickly to customer needs.

[0056] In the above embodiment, step a4, constructing a time dependency target matrix based on the time series analysis results, includes: Step a41: Select a target dependency measurement method based on the time series analysis result. The target dependency measurement method may be a method for quantifying the strength of the dependency between heterogeneous nodes and heterogeneous edges, including correlation coefficient, mutual information, causal relationship analysis, etc.

[0057] Exemplarily, the correlation coefficient is selected as the target dependency measurement method. The correlation coefficient can quantify the strength of the linear relationship between two variables, ranging from -1 to 1, where 1 represents a perfect positive correlation, -1 represents a perfect negative correlation, and 0 represents no correlation.

[0058] Step a42: Calculate the dependency strength value between heterogeneous nodes and heterogeneous edges using the target dependency measurement method. The dependency strength value is a value calculated using the target dependency measurement method, indicating the strength of the dependency between two heterogeneous nodes or heterogeneous edges.

[0059] Exemplarily, the correlation coefficient is used to calculate the dependency strength value between the order node and the greenhouse node, the order node and the warehouse node, and the greenhouse node and the warehouse node.

[0060] Step a43: construct a time dependency initial matrix based on the dependency strength values ​​between heterogeneous nodes and heterogeneous edges, wherein the time dependency initial matrix is ​​a matrix used to represent the dependency strength values ​​between heterogeneous nodes and heterogeneous edges.

[0061] Step a44: Based on the initial time dependency matrix, use the Bayesian method to evaluate the uncertainty of the time dependency, obtain the uncertainty quantification result, and construct the time dependency target matrix. Each element in the time dependency target matrix represents a dependency, including the dependency strength value and its uncertainty quantification result.

[0062] It should be understood that the uncertainty quantification result can be understood as the uncertainty estimate of the dependency strength value obtained by the Bayesian method, which is usually expressed in the form of probability distribution. For example, the confidence interval of each correlation coefficient can be calculated. The time dependency target matrix is ​​a matrix used to represent the time dependency between heterogeneous nodes and heterogeneous edges, and each element contains the dependency strength value and its uncertainty quantification result. Among them, the Bayesian method can evaluate the uncertainty of the time dependency and provide the confidence interval or probability distribution of each dependency strength value, which helps managers understand the reliability of the dependency. By quantifying uncertainty, managers can better control and manage risks and avoid decision-making errors caused by the uncertainty of the dependency. The Bayesian method can fuse multi-source data, combine prior knowledge and new data for update, and provide a more comprehensive and accurate dependency evaluation. The Bayesian method allows the model to be dynamically updated according to new data to ensure that the dependency evaluation always reflects the latest data changes. The Bayesian method provides a clear probabilistic explanation, making the dependency evaluation results easy to understand and interpret.

[0063] Therefore, by introducing the Bayesian method, the construction of the time-dependent target matrix can more accurately quantify uncertainty, improve the scientificity and reliability of decision-making, achieve dynamic updates and adaptability, handle small sample problems, and provide explanation and transparency. These benefits jointly improve the efficiency and effectiveness of supply chain management and help agricultural product manufacturers better respond to market changes and customer needs quickly.

[0064] In the above embodiment, step a5, determining the time dependencies of heterogeneous nodes and the time dependencies of heterogeneous edges based on the time dependency target matrix, includes: Step a51: Based on the time dependency target matrix, a graph clustering algorithm is used to cluster heterogeneous nodes and heterogeneous edges respectively to obtain heterogeneous node clustering results and heterogeneous edge clustering results.

[0065] It should be understood that a graph clustering algorithm is an algorithm for dividing heterogeneous nodes or heterogeneous edges into several clusters. Graph clustering algorithms include spectral clustering, K-means clustering, and the like. This embodiment uses a graph clustering algorithm to divide heterogeneous nodes and heterogeneous edges into several clusters, respectively. The heterogeneous nodes or heterogeneous edges within each cluster have similar time dependencies. Specifically, heterogeneous node clustering refers to clustering order nodes, greenhouse nodes, and warehouse nodes, respectively, to obtain the cluster to which each heterogeneous node belongs. Heterogeneous edge clustering refers to clustering the heterogeneous edges between order nodes and greenhouse nodes, the heterogeneous edges between order nodes and warehouse nodes, and the heterogeneous edges between greenhouse nodes and warehouse nodes, respectively, to obtain the cluster to which each edge belongs.

[0066] Step a52: Use a community detection method to perform community detection on the heterogeneous node clustering results and the heterogeneous edge clustering results respectively to obtain the community structure of the heterogeneous nodes and the community structure of the heterogeneous edges.

[0067] It should be understood that the community detection method is a method for identifying community structure. A community refers to a group of heterogeneous nodes in a supply chain network. The connections between these heterogeneous nodes are relatively dense, while the connections between the nodes in this community and other communities are relatively sparse. Community detection methods include modularity maximization, label propagation algorithms, etc. The community structure of heterogeneous nodes refers to the set of nodes in each community obtained by performing community detection on the clustering results of heterogeneous nodes. The community structure of heterogeneous edges refers to the set of edges in each community obtained by performing community detection on the clustering results of heterogeneous edges.

[0068] Step a53: Based on the community structure of the heterogeneous nodes and the community structure of the heterogeneous edges, a graph convolutional network is used to determine the time dependencies of the heterogeneous nodes and the time dependencies of the heterogeneous edges.

[0069] It should be understood that the graph convolutional network is a deep learning method for processing graph structured data, which extracts the features of heterogeneous nodes and heterogeneous edges by performing convolution operations on the community structures of heterogeneous nodes and the community structures of heterogeneous edges.

[0070] In this embodiment, the graph clustering algorithm divides the nodes and edges into several clusters, and the nodes or edges in each cluster have similar time dependencies, which provides a basis for subsequent analysis. Through the community detection method, the community structure of heterogeneous nodes and edges with similar characteristics can be identified, thereby providing and refining the community structure. The graph convolutional network can extract high-level features of heterogeneous nodes and edges, which can more accurately reflect the time dependencies between heterogeneous nodes and edges, thereby improving the accuracy of the time dependencies. Through the application of graph clustering algorithms, community detection methods and graph convolutional networks, time dependencies can be analyzed from multiple scales and levels, providing a comprehensive perspective, thereby improving the efficiency and effectiveness of supply chain management.

[0071] In the above embodiment, step S136, constructing a comprehensive supply chain network based on heterogeneous nodes, heterogeneous edges, and the time dependency of heterogeneous nodes and the time dependency of heterogeneous edges, includes: Step b1: Obtain customer relationship data and customer grade data.

[0072] Among them, customer relationship data may refer to information about relationships between customers, such as frequency of cooperation between customers, transaction records, recommendation relationships, etc. Customer rating data may refer to rating information about customers, such as credit ratings, purchase frequency, purchase amounts, etc. of customers.

[0073] Step b2: construct a customer relationship network based on the customer relationship data, and determine the customer grade attributes based on the customer grade data; the nodes in the customer relationship network are customer nodes, and the edges of the customer relationship network represent the relationships between customers.

[0074] The customer relationship network may refer to a graph model, in which nodes represent customers and edges represent relationships between customers. The customer level attribute refers to the level information of each customer node, which is used to distinguish the priorities of different customers.

[0075] Step b3: construct an initial network based on heterogeneous nodes, heterogeneous edges, and the time dependencies of heterogeneous nodes and heterogeneous edges, and integrate the customer relationship network and customer grade attributes into the initial network to obtain an integrated network.

[0076] Step b4: sorting customer priorities based on the integrated network, and determining order processing and resource scheduling plans based on the customer priority sorting results.

[0077] Among them, customer priority sorting refers to the priority sorting of customers according to the customer relationship network and customer level attributes. Order processing and resource scheduling plan refers to determining the order of order processing and resource scheduling plan based on the customer priority sorting results.

[0078] Step b5: Based on the order processing and resource scheduling solution, a comprehensive supply chain network is constructed. The comprehensive supply chain network refers to the comprehensive network that is finally constructed, including all nodes, edges and their time dependencies, and is used to optimize supply chain management.

[0079] In this embodiment, through customer relationship data, the cooperative relationship between customers can be understood, which is helpful to optimize customer management and service. Through customer level data, the priorities of different customers can be distinguished to ensure that orders from important customers are given priority. Based on the customer priority sorting results, ensure that orders from high-priority customers are given priority to improve customer satisfaction. Optimize resource scheduling plans to ensure the effective use of resources and improve service quality and response speed. By building a comprehensive supply chain network, comprehensively consider order nodes, greenhouse nodes, warehouse nodes and their time dependencies to optimize supply chain management. By building a comprehensive supply chain network, visual analysis can be performed to help managers intuitively understand the various links and relationships in the supply chain. Through the comprehensive supply chain network, the order processing process is optimized, efficiency is improved, and unnecessary delays and waste are reduced.

[0080] In some optional embodiments, S15, evaluating multiple pending supply chain management strategies, comprises the following steps: Step 151: Determine the evaluation indicators and evaluation methods for the supply chain management strategy to be determined.

[0081] It should be understood that the evaluation indicators may include cost, customer satisfaction, delivery time, inventory level, environmental impact, etc., where cost may refer to total cost, including procurement cost, transportation cost, inventory cost, etc. Customer satisfaction may refer to customer satisfaction score, which can be measured by customer surveys, order completion time, return rate, etc. Delivery time may refer to the average time from order placement to delivery. Inventory level may refer to inventory turnover rate, inventory capital, etc. Environmental impact may refer to carbon emissions, energy consumption, etc.

[0082] In this embodiment, the evaluation method includes a multi-objective optimization evaluation method, a weighted scoring method, a hierarchical analysis method, etc. Among them, the multi-objective optimization evaluation method uses a multi-objective optimization algorithm (such as a non-dominated sorting genetic algorithm, a multi-objective evolutionary algorithm based on decomposition, etc.) to evaluate the comprehensive performance of multiple evaluation indicators, which can improve the comprehensiveness and efficiency of decision-making. The weighted scoring method assigns weights to each evaluation indicator and calculates the comprehensive score, which can quickly calculate the comprehensive score and simplify the evaluation process. The hierarchical analysis method determines the weight of each scoring indicator through a hierarchical structure model and expert scoring, and performs a comprehensive evaluation.

[0083] Step 152: Based on the evaluation indicators, an evaluation method is used to evaluate each pending supply chain management strategy.

[0084] The embodiments of the present application improve the diversity of evaluation indicators. A variety of evaluation methods can be used to select the most appropriate evaluation method according to actual conditions, thereby improving the flexibility of the evaluation method and ensuring the accuracy and reliability of the evaluation results.

[0085] Figure 2 A schematic diagram of the structure of an agricultural product supply chain management system based on order data provided in an embodiment of the present application is shown in FIG. Figure 2 As shown, the system includes: The acquisition module 21 is used to acquire the agricultural product order data in the order system, the agricultural product harvesting data collected by the greenhouse monitoring system, and the agricultural product storage data recorded by the warehouse monitoring system.

[0086] The processing module 22 is used to batch process the agricultural product order data, agricultural product harvesting data and agricultural product storage data respectively.

[0087] Construction module 23 is used to construct a comprehensive supply chain network based on processed agricultural product order data, processed agricultural product harvesting data, and processed agricultural product storage data; the comprehensive supply chain network is a heterogeneous graph model with timestamps and time windows added. The heterogeneous nodes in the comprehensive supply chain network represent various entities in the comprehensive supply chain, including order nodes, greenhouse nodes, and warehouse nodes, and the heterogeneous edges represent the relationship between entities.

[0088] The generation module 24 is used to generate a plurality of pending supply chain management strategies based on the comprehensive supply chain network.

[0089] The evaluation and selection module 25 is used to evaluate multiple pending supply chain management strategies and select the pending supply chain management strategy with the highest evaluation value as the target supply chain management strategy.

[0090] Figure 2 The agricultural product supply chain management system based on order data can be executed Figure 1 The implementation principle and technical effect of the agricultural product supply chain management method based on order data described in the illustrated embodiment will not be described in detail. The specific manner in which each module and unit performs operations in the agricultural product supply chain management system based on order data in the above embodiment has been described in detail in the embodiment of the method, and will not be elaborated here.

[0091] In one possible design, Figure 2 The agricultural product supply chain management system based on order data of the embodiment shown can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32 .

[0092] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0093] The processing component 32 is used to: obtain agricultural product order data in the order system, agricultural product harvesting data collected by the greenhouse monitoring system, and agricultural product storage data recorded by the warehouse monitoring system; batch process the agricultural product order data, agricultural product harvesting data, and agricultural product storage data respectively; build a comprehensive supply chain network based on the processed agricultural product order data, processed agricultural product harvesting data, and processed agricultural product storage data; the comprehensive supply chain network is a heterogeneous graph model with timestamps and time windows added, and the heterogeneous nodes in the comprehensive supply chain network represent various entities in the comprehensive supply chain, including order nodes, greenhouse nodes, and warehouse nodes, and the heterogeneous edges represent the relationship between entities; generate multiple pending supply chain management strategies based on the comprehensive supply chain network; evaluate multiple pending supply chain management strategies, and select the pending supply chain management strategy with the highest evaluation value as the target supply chain management strategy.

[0094] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented by one or more application-specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors or other electronic components to perform the above method.

[0095] The storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as random access memory (RAM), static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0096] Of course, the computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0097] The input / output interface provides an interface between the processing component and the peripheral interface module, which may be an output device, an input device, etc.

[0098] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0099] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0100] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 The agricultural product supply chain management method based on order data of the illustrated embodiment.

[0101] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0102] The system embodiment described above is merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art may understand and implement it without creative work.

[0103] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0104] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for managing agricultural product supply chain based on order data, characterized in that: include: Obtain agricultural product order data in the order system, agricultural product harvest data collected by the greenhouse monitoring system, and agricultural product storage data recorded by the warehouse monitoring system; The agricultural product order data, agricultural product harvesting data and agricultural product storage data are batch processed respectively; A comprehensive supply chain network is constructed based on processed agricultural product order data, processed agricultural product harvest data, and processed agricultural product storage data; the comprehensive supply chain network is a heterogeneous graph model with timestamps and time windows added, and the heterogeneous nodes in the comprehensive supply chain network represent various entities in the comprehensive supply chain, including order nodes, greenhouse nodes, and warehouse nodes, and the heterogeneous edges represent the relationships between the entities; generating a plurality of pending supply chain management strategies based on the integrated supply chain network; Evaluate multiple pending supply chain management strategies, and select the pending supply chain management strategy with the highest evaluation value as the target supply chain management strategy.

2. The method according to claim 1, characterized in that The comprehensive supply chain network is constructed based on the processed agricultural product order data, the processed agricultural product harvest data, and the processed agricultural product storage data, including: Define order nodes based on processed agricultural product order data, define greenhouse nodes based on processed agricultural product harvesting data, and define warehouse nodes based on processed agricultural product storage data; Based on the processed agricultural product order data and the processed agricultural product harvesting data, the heterogeneous edges between the order nodes and the greenhouse nodes are defined; based on the processed agricultural product order data and the processed agricultural product storage data, the heterogeneous edges between the order nodes and the warehouse nodes are defined; based on the processed agricultural product harvesting data and the processed agricultural product storage data, the heterogeneous edges between the greenhouse nodes and the warehouse nodes are defined; Add corresponding timestamps to each heterogeneous node and heterogeneous edge to obtain heterogeneous nodes and heterogeneous edges with timestamps; Define a time window, and divide all heterogeneous nodes and heterogeneous edges with timestamps according to the time window, to obtain a set of heterogeneous nodes and a set of heterogeneous edges divided by the time window; Based on the heterogeneous node sets and heterogeneous edge sets divided by time windows, the sliding window method and time series analysis technology are used to obtain the time dependency relationship of heterogeneous nodes and the time dependency relationship of heterogeneous edges; A comprehensive supply chain network is constructed based on heterogeneous nodes, heterogeneous edges, and the time dependencies of the heterogeneous nodes and the time dependencies of the heterogeneous edges.

3. The method according to claim 2, characterized in that The method of obtaining the time dependency relationship of the heterogeneous nodes and the time dependency relationship of the heterogeneous edges based on the heterogeneous node set and the heterogeneous edge set divided by the time window by using the sliding window method and the time series analysis technology includes: Based on the heterogeneous node set and heterogeneous edge set divided by time window, the size and step size of the sliding window are defined; Based on the size and step size of the sliding window, sliding window processing is performed on the timestamps of the heterogeneous nodes and the heterogeneous edges, and based on the timestamps of the heterogeneous nodes and the heterogeneous edges in the sliding window, statistical features in the sliding window are calculated; Perform time series analysis based on the statistical characteristics within the sliding window to obtain the time series analysis results; Construct a time-dependency target matrix based on the time series analysis results; The time dependencies of heterogeneous nodes and the time dependencies of heterogeneous edges are determined based on the time dependency target matrix.

4. The method according to claim 3, characterized in that The step of constructing a time dependency target matrix based on the time series analysis results includes: Select the target dependency measurement method based on the time series analysis results; The target dependency measurement method is used to calculate the dependency strength values ​​between heterogeneous nodes and heterogeneous edges; Based on the dependency strength values ​​between heterogeneous nodes and heterogeneous edges, an initial matrix of temporal dependencies is constructed; Based on the initial matrix of time dependency, the Bayesian method is used to evaluate the uncertainty of time dependency, obtain the uncertainty quantification result, and construct the time dependency target matrix. Each element in the time dependency target matrix represents a dependency, including the dependency strength value and its uncertainty quantification result.

5. The method according to claim 3, characterized in that: The determining of the time dependencies of heterogeneous nodes and the time dependencies of heterogeneous edges based on the time dependency target matrix includes: Based on the time dependency target matrix, a graph clustering algorithm is used to cluster heterogeneous nodes and edges respectively to obtain heterogeneous node clustering results and heterogeneous edge clustering results. Using a community detection method to perform community detection on the heterogeneous node clustering results and the heterogeneous edge clustering results, respectively, to obtain the community structure of the heterogeneous nodes and the community structure of the heterogeneous edges; Based on the community structure of the heterogeneous nodes and the community structure of the heterogeneous edges, a graph convolutional network is used to determine the temporal dependencies of the heterogeneous nodes and the temporal dependencies of the heterogeneous edges.

6. The method according to claim 2, characterized in that The method of constructing a comprehensive supply chain network based on heterogeneous nodes, heterogeneous edges, and the time dependency of the heterogeneous nodes and the time dependency of the heterogeneous edges includes: Obtain customer relationship data and customer rating data; Building a customer relationship network based on the customer relationship data, and determining customer grade attributes based on the customer grade data; the nodes in the customer relationship network are customer nodes, and the edges of the customer relationship network represent the relationships between customers; Building an initial network based on heterogeneous nodes, heterogeneous edges, and the time dependencies of the heterogeneous nodes and the time dependencies of the heterogeneous edges, and integrating the customer relationship network and the customer grade attribute into the initial network to obtain an integrated network; Prioritize customers based on the integrated network and determine order processing and resource scheduling plans based on the customer priority ranking results; Based on the order processing and resource scheduling solution, a comprehensive supply chain network is constructed.

7. The method according to claim 6, characterized in that The evaluation of multiple pending supply chain management strategies includes the following steps: Determine the evaluation indicators and evaluation methods for the proposed supply chain management strategy; Based on the evaluation index, each pending supply chain management strategy is evaluated using the evaluation method.

8. An agricultural product supply chain management system based on order data, characterized in that: include: The acquisition module is used to obtain the agricultural product order data in the order system, the agricultural product harvesting data collected by the greenhouse monitoring system, and the agricultural product storage data recorded by the warehouse monitoring system; A processing module, used for batch processing the agricultural product order data, agricultural product harvesting data and agricultural product storage data respectively; A construction module is used to construct a comprehensive supply chain network based on processed agricultural product order data, processed agricultural product harvest data, and processed agricultural product storage data; the comprehensive supply chain network is a heterogeneous graph model with timestamps and time windows added, and the heterogeneous nodes in the comprehensive supply chain network represent various entities in the comprehensive supply chain, including order nodes, greenhouse nodes, and warehouse nodes, and the heterogeneous edges represent the relationships between the entities; A generation module, used for generating a plurality of pending supply chain management strategies based on the comprehensive supply chain network; The evaluation and selection module is used to evaluate multiple pending supply chain management strategies and select the pending supply chain management strategy with the highest evaluation value as the target supply chain management strategy.

9. A computing device, characterized in that It comprises a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement an agricultural product supply chain management method based on order data as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a computer, an agricultural product supply chain management method based on order data as described in any one of claims 1 to 7 is implemented.

Images