Cloud digital green supply chain management method and system

Abstract

The invention discloses a cloud digital green supply chain management method and system, and relates to the technical field of supply chain management. A recommendation algorithm recommends a corresponding cloud service platform for a supply chain; the method comprises the following steps: preprocessing supply chain data, constructing an environmental protection degree by the preprocessed supply chain data, obtaining a regression coefficient of each dependent variable of a supply chain by using multiple linear regression analysis, constructing an optimization coefficient in combination with the position of each dependent variable in a feature column, and optimizing the supply chain according to an optimization feature selected by the optimization coefficient. And testing the optimized supply chain by using the trained supply chain digital twin model, constructing an improvement degree according to test data, and if the improvement degree is lower than an improvement threshold, outputting a corresponding maintenance scheme for the supply chain by the supply chain maintenance knowledge graph, and maintaining the optimized supply chain. Through multiple optimization, the supply chain is improved and optimized, the carbon emission during the operation of the supply chain is reduced, and the green management of the supply chain is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of supply chain management, and in particular to a cloud-based digital green supply chain management method and system. Background Art

[0002] Cloud-based digital green supply chain management is a green supply chain management model that integrates new-generation information technologies such as cloud computing, big data, the Internet of Things, and artificial intelligence. It aims to optimize the operation of the supply chain through digital means, reduce environmental impact, and improve the efficiency and sustainability of the supply chain.

[0003] The main features of cloud-based digital green supply chain management include: Data-driven: By integrating cloud computing technology, real-time collection, storage, analysis and sharing of supply chain data can be achieved, making supply chain management more data-based and intelligent. This helps companies to grasp the operation of the supply chain in real time, optimize resource allocation, and improve the response speed and accuracy of the supply chain. Environmentally friendly: In all aspects of supply chain management, attention is paid to environmental protection and resource conservation, reducing the generation of waste and emissions, and improving resource utilization efficiency. For example, by optimizing the location and layout of warehousing and distribution centers, reducing transportation distance and time, reducing transportation costs and carbon emissions; introducing environmentally friendly transportation tools to reduce dependence on fossil energy; selecting environmentally friendly raw materials and products to reduce the impact on the environment, etc.

[0004] In the Chinese invention patent with application publication number CN116307456A, a supply chain design system and supply chain design method are disclosed, including a computer terminal and a processor, the computer terminal is used to store executable instructions, the processor is used to read the executable instructions stored in the computer terminal, and also includes a user information module, an indicator database, a supply chain information storage module, a system management module, an indicator management module and a solution analysis module. The present invention solves the situation that enterprises rely on "passive acceptance" of material information in material procurement, and makes reminders based on project progress information such as milestone plans, combined with material demand plan reserves, and referring to material procurement and supply cycles; fully consider environmental issues, and include dozens of indicators of the entire green supply chain process into the design and optimization scope.

[0005] Combined with the above application and the contents of the prior art:

[0006] After completing the design and allocation of the supply chain, the supply chain enters a normal operation or operating state. If the operating state of the supply chain, such as the response time of the supply chain and the order fulfillment rate, is abnormal, the supply chain managers will usually optimize or transform the supply chain in order to ensure the normal operation of the supply chain and speed up the logistics response. However, if the optimization is only aimed at economic factors such as the rapid response of the supply chain and the rapid fulfillment of orders, other problems may arise. For example, when the optimized supply chain is in operation, energy consumption and waste generation increase significantly. In the existing green supply chain management methods, in addition to the above-mentioned problems, it may be difficult to find the focus that needs to be optimized when optimizing the supply chain, resulting in a relatively narrow optimization coverage of the supply chain and difficulty in achieving the desired effect.

[0007] To this end, the present invention provides a cloud-based digital green supply chain management method and system. Summary of the invention

[0008] In view of the deficiencies of the prior art, the present invention provides a cloud-based digital green supply chain management method and system, which uses multiple linear regression analysis to obtain the regression coefficients of each dependent variable of the supply chain, and constructs an optimization coefficient based on the position of each dependent variable in the feature column. The supply chain is optimized according to the optimization features selected by the optimization coefficient, and the optimized supply chain is tested using the trained supply chain digital twin model. The improvement degree is constructed from the test data. If the improvement degree is lower than the improvement threshold, the supply chain maintenance knowledge graph outputs a corresponding maintenance plan for the supply chain, and the optimized supply chain is maintained. Through multiple optimizations, the supply chain is improved and optimized, and carbon emissions during supply chain operation are reduced, thereby solving the technical problems raised in the background technology.

[0009] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0010] A cloud-based digital green supply chain management method includes: monitoring and obtaining operation-related data during supply chain operation within a monitoring period, and constructing a supply chain operation data set after aggregation, constructing an operation degree Yo from the operation data set, and issuing a cloud service management instruction to the outside if the obtained operation degree Yo exceeds expectations;

[0011] Combining supply chain data and cloud service platform data, the recommendation algorithm recommends the corresponding cloud service platform for the supply chain. If the recommended cloud service platform does not match the supply chain, a new cloud service platform is selected for the supply chain based on the matching degree Pp and reliability To of the cloud service platform.

[0012] After matching the corresponding preprocessing strategy for the supply chain data, the supply chain data is preprocessed, and the environmental protection degree Bo is constructed from the preprocessed supply chain data. If the environmental protection degree Bo does not exceed the expectation, an optimization instruction is issued to the cloud service platform;

[0013] Use multiple linear regression analysis to obtain the regression coefficients of each dependent variable of the supply chain, and build the optimization coefficient Ho based on the position of each dependent variable in the feature column. Optimize the supply chain based on the optimization features selected by the optimization coefficient Ho, and issue prediction instructions to the outside.

[0014] The trained supply chain digital twin model is used to test the optimized supply chain. The environmental protection degree Bo at each prediction node is constructed based on the test data, and then the corresponding improvement degree Gs is generated. If the improvement degree Gs is lower than the improvement threshold, the supply chain is optimized again. If the optimized improvement degree Gs is still lower than the improvement threshold, the supply chain maintenance knowledge graph outputs the corresponding maintenance plan for the supply chain to maintain the optimized supply chain. The construction method of the improvement degree Gs is as follows:

[0015]

[0016] Among them, ABo i and BBo i are the environmental protection degree of the i-th node before and after optimization, ABo avg With BBo avg is the corresponding mean, i is the node number, i = 1, 2, ..., w, w is the number of nodes, Zp i is the median value of the environmental protection degree of the ith node, Zp avg is the corresponding mean.

[0017] Furthermore, when the supply chain is still in operation, relevant data of supply chain operation are monitored and obtained in continuous sub-periods, and a supply chain operation data set is constructed after aggregation; the operation degree Yo is constructed from the data in the supply chain operation data set, wherein the response time Xt and the order fulfillment rate Vt are linearly normalized, and the corresponding data values ​​are mapped to the interval [0,1], according to the following method:

[0018]

[0019] Where i = 1, 2, ..., p, p is the number of sub-periods, Xv i is the response time of the ith sub-cycle, Cj i is the order fulfillment rate in the i-th sub-period, with weight coefficients: 0≤F1≤1, 0≤F2≤1, and F1+F2=1.

[0020] Furthermore, after receiving the cloud service management instruction, various data of the supply chain are collected, and a supply chain data set is constructed after aggregation; cloud service platform data is collected, and a cloud service platform data set is constructed after aggregation; based on the correspondence between the supply chain data and the cloud service platform data, the content-based recommendation algorithm recommends the corresponding cloud service platform for the supply chain; the Pearson correlation coefficient between the supply chain and the cloud service platform is constructed from the collected data, and it is used as the matching degree Pp. If the matching degree Pp is lower than the preset matching threshold, a reselection instruction is issued to the outside.

[0021] Furthermore, after receiving the reselection instruction, several cloud service platforms are evaluated to obtain the reliability To of each cloud service platform in the following manner: linear normalization is performed on the fault response time Yt and the data backup speed Bt, and the corresponding data values ​​are mapped to the interval [0,1], and then the following method is used:

[0022]

[0023] Weight coefficient: 0≤ρ≤1, 0≤ζ≤1; sort several cloud service platforms according to reliability To, and take the cloud service platform ranked in the top three in reliability To and with the highest matching degree Pp as the target platform, and issue data processing instructions to the outside.

[0024] Furthermore, after receiving the data processing instruction, the data in the supply chain data set are identified to obtain the corresponding data features; according to the correspondence between the data features and the preprocessing strategies, the preprocessing strategy library matches the corresponding preprocessing strategies, and the data in the set are processed according to the matched preprocessing strategies;

[0025] The environmental protection index Bo is constructed from the preprocessed supply chain data in the following way: the carbon emissions Tp and waste generation Fp in each sub-period are linearly normalized, and the corresponding data values ​​are mapped to the interval [0, 1], and then the following method is used:

[0026]

[0027] Where i = 1, 2, ..., m, m is the number of sub-periods, weight coefficient: 0 ≤ S1 ≤ 1, 0 ≤ S2 ≤ 1, and S2 + S1 = 1; Tp i is the carbon emissions in the ith sub-period, Tp avg is the mean carbon emission, is the qualified reference value of carbon emissions; Fp i The amount of waste generated in the ith sub-period, Fp avg is the mean amount of waste generated, It is the qualified reference value for waste generation.

[0028] Furthermore, after receiving the optimization instruction, multiple linear regression analysis is performed with various supply chain parameters as independent variables and environmental protection degree Bo as dependent variable to obtain the corresponding linear regression equation; the regression coefficient corresponding to the independent variable in the linear regression equation is used as the influence coefficient Xn; the independent variables are sorted by the influence coefficient Xn to obtain the corresponding feature column, and after determining the position of each dependent variable in the feature column, the closeness center Un of each dependent variable is obtained;

[0029] In addition, the corresponding optimization model is obtained by training the optimization algorithm, with improving the environmental protection degree Bo of the supply chain as the optimization goal, and the supply chain is optimized according to the screened optimization features, at least including: optimizing the supply chain transportation routes and modes, using energy-saving processes and equipment in the production process, and outputting the optimized supply chain.

[0030] Furthermore, after obtaining the closeness center Un of the dependent variable, combined with the corresponding influence coefficient Xn, the optimization coefficient Ho is constructed under dimensionless conditions. If the obtained optimization coefficient Ho exceeds the optimization threshold, the corresponding dependent variable is determined as the optimization feature. The optimization coefficient Ho is constructed as follows:

[0031]

[0032] Among them, i is the closeness centrality of the ith dependent variable, Un avg is the mean of the closeness center, Xn i is the influence coefficient of the ith dependent variable, Xn avg is the mean of the influence coefficient, i=1, 2,…,q, q is the number of dependent variables, and the weight coefficient is: 0≤β≤1, 0≤α≤1, and α+β=1.

[0033] Furthermore, after the cloud service platform receives the prediction instruction, the collected supply chain and related data are used as sample data to train and construct a supply chain digital twin model; the trained supply chain digital twin model predicts the optimized supply chain, and obtains corresponding test data at each prediction node, and reconstructs the environmental protection degree Bo on each prediction node with the supply chain data obtained by the test; each environmental protection degree Bo is marked with a test node, and each node before and after optimization is sorted according to the timeline, and one by one, and the improvement degree Gs is constructed from the environmental protection degree Bo on the nodes before and after optimization. If the improvement degree Gs is lower than the preset improvement threshold, the supply chain is re-optimized.

[0034] Furthermore, if after several optimizations, the improvement degree of the optimized supply chain is still lower than the improvement threshold, a maintenance instruction is issued to the outside; after receiving the maintenance instruction, supply chain maintenance is used as the target word, and a supply chain maintenance knowledge graph is pre-constructed; after setting corresponding maintenance standards for each indicator parameter, various relevant data of the supply chain are identified to obtain corresponding maintenance features; based on the correspondence between the maintenance features and the supply chain, the supply chain maintenance knowledge graph outputs the corresponding maintenance plan for the supply chain, and the maintenance plan is executed to maintain the supply chain.

[0035] A cloud-based digital green supply chain management system, comprising:

[0036] The operation monitoring unit monitors and obtains the operation-related data of the supply chain during the monitoring period, and builds the supply chain operation data set after aggregation. The operation degree Yo is built from the operation data set. If the obtained operation degree Yo exceeds expectations, a cloud service management instruction is issued to the outside.

[0037] The matching unit combines the supply chain data and the cloud service platform data, and recommends the corresponding cloud service platform for the supply chain through the recommendation algorithm. If the recommended cloud service platform does not match the supply chain, a new cloud service platform is selected for the supply chain based on the matching degree Pp and reliability To of the cloud service platform;

[0038] The data processing unit matches the corresponding preprocessing strategy for the supply chain data, preprocesses the supply chain data, and constructs the environmental protection degree Bo from the preprocessed supply chain data. If the environmental protection degree Bo does not exceed the expectation, an optimization instruction is issued to the cloud service platform;

[0039] The data analysis unit uses multiple linear regression analysis to obtain the regression coefficients of each dependent variable in the supply chain, and constructs the optimization coefficient Ho in combination with the position of each dependent variable in the feature column. The supply chain is optimized according to the optimization features selected by the optimization coefficient Ho, and a prediction instruction is issued to the outside.

[0040] Optimize the maintenance unit, and use the trained supply chain digital twin model to test the optimized supply chain. Use the test data to construct the environmental protection degree Bo at each prediction node, and then generate the corresponding improvement degree Gs. If the improvement degree Gs is lower than the improvement threshold, the supply chain is optimized again. If the optimized improvement degree Gs is still lower than the improvement threshold, the supply chain maintenance knowledge graph outputs the corresponding maintenance plan for the supply chain to maintain the optimized supply chain.

[0041] The present invention provides a cloud-based digital green supply chain management method and system, which has the following beneficial effects:

[0042] 1. The trained recommendation algorithm or similar algorithm can match the corresponding cloud service platform for the supply chain, which can improve efficiency in subsequent automated processing and optimization and promote the efficient operation of the supply chain.

[0043] 2. Build reliability To based on the data parameters of each cloud service platform, use reliability To to evaluate the reliability of each cloud service platform, and combine the reliability To and matching degree Pp of the cloud service platform to screen each cloud service platform, so as to select the most suitable cloud service platform for the supply chain, which can improve management efficiency and effectiveness when managing the supply chain.

[0044] 3. Construct an environmental degree Bo from the collected data, and evaluate and judge whether the supply chain is green and environmentally friendly based on the environmental degree Bo. If there are more carbon emissions or waste emissions during supply chain operation, the supply chain can be managed and optimized to save energy, reduce emissions and protect the environment. After selecting several independent variables, use linear regression analysis to determine the degree of influence of each independent variable on the environmental degree. According to the degree of influence, several independent variables can be selected and evaluated, and the key factors and non-key factors can be determined.

[0045] 4. After obtaining the position and influence coefficient Xn of the dependent variable, the corresponding optimization coefficient Ho is constructed. The key factors are selected from several independent variables based on the optimization coefficient Ho. The key factors are used as optimization features. Based on the optimization features, the supply chain can be optimized in a targeted manner. The trained optimization model is used to optimize the supply chain, and the transportation methods, equipment and production processes attached to the supply chain are optimized to obtain the optimized supply chain, thereby improving the environmental protection of the supply chain.

[0046] 5. The constructed supply chain digital twin model predicts the optimized supply chain, and the corresponding environmental protection degree Bo is reconstructed based on the predicted data. Based on the changes in the environmental protection degree Bo before and after, it can be judged whether the optimized supply chain has reached the available expectations, thereby judging the optimization effect.

[0047] 6. Build the improvement degree based on the obtained environmental protection degree, and judge whether the optimized supply chain is available according to the improvement degree. If it is available, the management of the green supply chain is completed. If it is not available, it can be optimized again. Through multiple optimizations, the supply chain can be improved and optimized to ensure the optimization effect.

[0048] 7. After screening out the maintenance features, the supply chain maintenance knowledge graph outputs the corresponding maintenance plan for the supply chain, executes the maintenance plan to maintain the supply chain, optimizes and improves the supply chain at a deeper level, reduces carbon emissions during supply chain operations, and realizes green management of the supply chain. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 This is a flow chart of the green supply chain management method of the present invention;

[0050] Figure 2 It is a schematic diagram of the structure of the green supply chain management system of the present invention. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] See also Figure 1 The present invention provides a cloud-based digital green supply chain management method, including the following contents:

[0053] Step 1: Monitor and obtain operation-related data during supply chain operation during the monitoring period, and build a supply chain operation data set after aggregation. The operation degree Yo is built from the operation data set. If the obtained operation degree Yo exceeds expectations, a cloud service management instruction is issued to the outside.

[0054] The step 1 includes the following contents:

[0055] Step 101: When the current supply chain needs to be managed, a monitoring cycle including several sub-cycles is set. When the supply chain is still in operation, the response time, order fulfillment rate and other relevant data of the supply chain operation are monitored and obtained in continuous sub-cycles. The monitoring data in each sub-cycle is aggregated to construct a supply chain operation data set;

[0056] Step 102: construct the operation index Yo from the data in the supply chain operation data set, wherein the response time Xt and the order fulfillment rate Vt are linearly normalized, and the corresponding data values ​​are mapped to the interval [0, 1], according to the following method:

[0057]

[0058] Where i = 1, 2, ..., p, p is the number of sub-periods, Xv i is the response time of the ith sub-cycle, Cj i is the order fulfillment rate in the ith sub-period, weight coefficient: 0≤F1≤1, 0≤F2≤1, and F1+F2=1; the weight coefficient is obtained by referring to the hierarchical analysis method;

[0059] Based on historical data and supply chain management expectations, the operation threshold is set in advance; if the operation degree Yo exceeds expectations, that is, exceeds the operation threshold, it means that the supply chain is in a poor state of operation and needs to be adjusted in a timely manner. At this time, a cloud service management instruction is issued to the outside. When the operation effect cannot meet the current demand, the supply chain is managed so that the supply chain can meet the set management goals and problems can be dealt with in a timely manner;

[0060] When using, combine the contents in steps 101 and 102:

[0061] When the supply chain is in operation, the corresponding operation degree Yo is constructed from the corresponding operation data. The response time and order fulfillment rate of the supply chain are used as the starting point to judge whether the operation status of the supply chain is healthy. If it is relatively healthy and the operation efficiency is relatively high, its operation status can be maintained. If the operation status is poor, the supply chain needs to be optimized and its various parameters adjusted to reduce the operation cost of the supply chain, reduce the irrational use of resources, and improve customer satisfaction.

[0062] After completing the supply chain design and allocation, the supply chain enters a normal operation or operating state. If the operation state of the supply chain, such as the response time of the supply chain and the order fulfillment rate, is abnormal, the supply chain managers will usually optimize or transform the supply chain in order to ensure the normal operation of the supply chain and speed up the logistics response. However, if the optimization is only aimed at economic factors such as the rapid response of the supply chain and the rapid fulfillment of orders, other problems may arise. For example, when the optimized supply chain is in operation, energy consumption and waste generation increase significantly. In the existing green supply chain management methods, in addition to the above-mentioned problems, it may be difficult to find the focus that needs to be optimized when optimizing the supply chain, resulting in a relatively narrow optimization coverage of the supply chain and difficulty in achieving the desired effect.

[0063] Step 2: Combining the supply chain data and the cloud service platform data, the recommendation algorithm recommends the corresponding cloud service platform for the supply chain. If the recommended cloud service platform does not match the supply chain, a new cloud service platform is selected for the supply chain based on the matching degree Pp and reliability To of the cloud service platform.

[0064] The step 2 includes the following contents:

[0065] Step 201: After receiving the cloud service management instruction, various data of the supply chain are collected, including data on energy consumption, waste generation, transportation distance, raw material procurement, production, logistics, packaging, product use and waste treatment, and the data are summarized to construct a supply chain data set; cloud service platform data are collected, including: economic cost, data security, scalability and service quality, cloud server, storage and network resources, and the data are summarized to construct a cloud service platform data set;

[0066] Step 202: Based on the correspondence between the supply chain data and the cloud service platform data, a content-based recommendation algorithm is used to recommend a corresponding cloud service platform for the supply chain; and after the recommendation is completed, the Pearson correlation coefficient between the supply chain and the cloud service platform is constructed from the collected data, and is used as the matching degree Pp, in the following manner:

[0067]

[0068] Among them, x i and i are the observed values ​​of two variables, x avg and avg is the corresponding mean, n is the number of observations;

[0069] A matching threshold is pre-set based on historical data and management expectations for cloud service platforms. If the matching degree Pp is lower than the preset matching threshold, it means that the currently matched cloud service platforms are not suitable and need to be replaced. At this time, a reselection instruction is issued to the outside.

[0070] When in use, in order to be able to monitor and automatically adjust the supply chain in real time, it is necessary to introduce a cloud platform for the current supply chain to realize real-time data collection, monitoring and optimization and other operations. Based on the collected data, the trained recommendation algorithm or similar algorithm matches the corresponding cloud service platform for the supply chain. In the subsequent automated processing and optimization, it can improve efficiency and promote the efficient operation of the supply chain.

[0071] It should be noted that recommendation algorithms are a series of algorithms used to predict content or items that users may be interested in. The following are some common recommendation algorithms: Content-Based Recommendation Algorithm: This algorithm recommends items similar to users' past preferences by analyzing the content features (such as text, metadata, etc.) of items that users liked in the past.

[0072] Collaborative Filtering Recommendation Algorithm: It is divided into user-based collaborative filtering and item-based collaborative filtering.

[0073] User-based collaborative filtering: find other users with similar interests to the user, and then recommend items that these users like to the user. Item-based collaborative filtering: find other items similar to items that the user liked in the past based on the user's historical behavior, and then recommend them to the user.

[0074] Hybrid Recommendation: Combines content-based recommendations with collaborative filtering recommendations and other recommendation techniques to provide more accurate recommendations. For example, you can first use content-based recommendations to narrow down the range of candidate items, and then use collaborative filtering to further filter them.

[0075] Step 203: After receiving the reselection instruction, based on the data of the cloud service platform, several cloud service platforms are evaluated to obtain the reliability To of each cloud service platform in the following manner: linear normalization is performed on the fault response time Yt and data backup speed Bt of the cloud service platform obtained by query or test, and the corresponding data values ​​are mapped to the interval [0, 1], and then the following method is used:

[0076]

[0077] Weight coefficient: 0≤ρ≤1, 0≤ζ≤1; the weight coefficient is obtained by referring to the hierarchical analysis method;

[0078] Sort several cloud service platforms by reliability To, and select the cloud service platforms with the top three reliability To rankings and the highest matching degree Pp as the target platforms, and issue data processing instructions to the outside world;

[0079] When using, combine the contents in steps 201 to 203:

[0080] On the basis of matching several cloud service platforms, the reliability To is constructed based on the data parameters of each cloud service platform, and the reliability of each cloud service platform is evaluated by the reliability To. In combination with the reliability To and the matching degree Pp of the cloud service platform, the cloud service platform can be screened to select the most suitable cloud service platform for the supply chain, which can improve the management efficiency and effectiveness when managing the supply chain.

[0081] Step 3: After matching the corresponding preprocessing strategy for the supply chain data, preprocess the supply chain data, and construct the environmental protection degree Bo from the preprocessed supply chain data. If the environmental protection degree Bo does not exceed the expectation, an optimization instruction is issued to the cloud service platform;

[0082] The step three includes the following contents:

[0083] Step 301: After receiving the data processing instruction, the distribution status and data value of each data in the supply chain data set are identified to obtain the corresponding data features; after online deep search or offline collection, a data preprocessing strategy library is pre-built; according to the correspondence between the data features and the preprocessing strategy, the preprocessing strategy library matches the corresponding preprocessing strategy, and the data in the set is processed according to the matched preprocessing strategy;

[0084] When in use, after completing data collection, a targeted data processing strategy is matched for the collected data based on the status of the data, and the data processing effect is guaranteed by targeted data processing.

[0085] Step 302: construct an environmental index Bo from the pre-processed supply chain data, and evaluate the supply chain with the environmental index Bo. The construction method of the environmental index Bo is as follows: perform linear normalization on the carbon emissions Tp and waste generation Fp in each sub-period of the supply chain operation, and map the corresponding data values ​​to the interval [0, 1], and then follow the following method:

[0086]

[0087] Wherein, i=1, 2, ..., m, m is the number of sub-periods, weight coefficient: 0≤S1≤1, 0≤S2≤1, and S2+S1=1, weight coefficient is obtained by referring to the hierarchical analysis method; Tp i is the carbon emissions in the ith sub-period, Tp avg is the mean carbon emission, is the qualified reference value of carbon emissions; Fp i The amount of waste generated in the ith sub-period, Fp avg is the mean amount of waste generated, The qualified reference value for waste generation; based on the low-energy consumption management expectations and historical data of the supply chain, the environmental protection threshold is pre-set; if the environmental protection degree does not exceed the expectation, it means that the carbon footprint or resource waste generated by the supply chain during operation is also large. At this time, an optimization instruction is issued to the cloud service platform;

[0088] When used, combine the contents in steps 301 and 302:

[0089] After obtaining the relevant data of the supply chain, an environmental degree Bo is constructed based on the collected data. Based on the environmental degree Bo, the greenness and environmental protection of the supply chain are evaluated and judged. If there are more carbon emissions or waste emissions during the operation of the supply chain, the supply chain can be managed and optimized to achieve energy conservation, emission reduction and environmental protection.

[0090] Step 4: Use multiple linear regression analysis to obtain the regression coefficients of each dependent variable in the supply chain, and construct the optimization coefficient Ho based on the position of each dependent variable in the feature column. Optimize the supply chain based on the optimization features selected by the optimization coefficient Ho, and issue prediction instructions to the outside.

[0091] The step 4 includes the following contents:

[0092] Step 401: After receiving the optimization instruction, the cloud service platform uses the various supply chain parameters as independent variables and the environmental protection degree Bo as the dependent variable to perform multiple linear regression analysis and obtain the corresponding linear regression equation; the regression coefficient corresponding to the independent variable in the linear regression equation is used as the influence coefficient Xn;

[0093] Sort the independent variables by the influence coefficient Xn, obtain the corresponding feature column, determine the position of each dependent variable in the feature column, and calculate the closeness center Un of each dependent variable as follows:

[0094]

[0095] Among them, q is the number of dependent variables, U ij is the shortest distance between dependent variable i and dependent variable j; each dependent variable is marked with the closeness centrality Un;

[0096] When using it, when it is necessary to optimize the supply chain, after selecting several independent variables, the influence of each independent variable on environmental protection can be determined through linear regression analysis. According to the influence degree, several independent variables can be selected and evaluated, so as to determine the key factors and non-key factors.

[0097] Step 402: After obtaining the closeness center Un of the dependent variable, combined with the corresponding influence coefficient Xn, the optimization coefficient Ho is constructed in the following manner under the dimensionless condition:

[0098]

[0099] Among them, i is the closeness centrality of the ith dependent variable, Un avg is the mean of the closeness center, Xn i is the influence coefficient of the ith dependent variable, Xn avg is the mean of the influence coefficient, i = 1, 2, …, q, q is the number of dependent variables, weight coefficient: 0 ≤ β ≤ 1, 0 ≤ α ≤ 1, and α + β = 1;

[0100] Combined with historical data and management expectations for dependent variables, the optimization threshold is set in advance; if the obtained optimization coefficient Ho exceeds the optimization threshold, considering that the corresponding dependent variable has a greater impact on the change of environmental protection, the corresponding dependent variable is determined as the optimization feature;

[0101] When using it, after obtaining the position of the dependent variable and the influence coefficient Xn, the corresponding optimization coefficient Ho is constructed. According to the optimization coefficient Ho, the key factors are selected from several independent variables. The key factors are used as optimization features, and the supply chain can be optimized in a targeted manner according to the optimization features.

[0102] Step 403: Obtain a corresponding optimization model through optimization algorithm training, take improving the environmental protection degree Bo of the supply chain as the optimization target, and optimize the supply chain according to the selected optimization features, at least including: optimizing the supply chain transportation route and mode, reducing energy consumption and emissions during transportation; adopting energy-saving processes and equipment in the production process to reduce waste and emissions; outputting the optimized supply chain and issuing prediction instructions to the outside;

[0103] When used, combine the contents in steps 401 to 403:

[0104] After screening out several optimization features and optimizing the supply chain with the trained optimization model, the transportation methods, equipment and production processes attached to the supply chain are optimized to obtain the optimized supply chain, thereby achieving improvement and enhancement of the environmental protection of the supply chain.

[0105] Step 5: The trained supply chain digital twin model is used to test the optimized supply chain. The environmental protection degree Bo at each prediction node is constructed based on the test data, and then the corresponding improvement degree Gs is generated. If the improvement degree Gs is lower than the improvement threshold, the supply chain is optimized again. If the optimized improvement degree Gs is still lower than the improvement threshold, the supply chain maintenance knowledge graph outputs the corresponding maintenance plan for the supply chain, and the optimized supply chain is maintained.

[0106] The step five includes the following contents:

[0107] Step 501: After receiving the prediction instruction, the cloud service platform uses the collected supply chain and related data as sample data, and uses the convolutional neural network to build an initial model. After the initial model is trained and tested with the sample data, a supply chain digital twin model is built. The trained supply chain digital twin model predicts the optimized supply chain, and obtains corresponding test data at each prediction node, including carbon emissions, energy consumption, waste generation, etc. The supply chain data obtained by the test is used to reconstruct the environmental protection degree Bo at each prediction node.

[0108] When in use, after obtaining the optimized supply chain, the constructed supply chain digital twin model predicts the optimized supply chain, and the corresponding environmental protection degree Bo is reconstructed based on the predicted data. Based on the changes before and after the environmental protection degree Bo, it can be judged whether the optimized supply chain has reached the available expectations, so as to judge the optimization effect;

[0109] Step 502: Mark each environmental degree Bo with a test node, sort each node before and after optimization according to the time axis, and correspond one to one. Construct the improvement degree Gs from the environmental degrees Bo on the nodes before and after optimization, in the following way:

[0110]

[0111] Among them, ABo i and BBo i are the environmental protection degree of the i-th node before and after optimization, ABo avg With BBo avg is the corresponding mean, i is the node number, i = 1, 2, ..., w, w is the number of nodes, Zp i is the median value of the environmental protection degree of the ith node, Zp avg is the corresponding mean;

[0112] If the improvement degree Gs is lower than the preset improvement threshold, it means that the application effect of the currently optimized supply chain may still be difficult to meet the current management expectations and needs further optimization or improvement. At this time, a re-optimization instruction is issued to the outside.

[0113] After receiving the re-optimization instruction, the output test data is used as feedback data to re-optimize the supply chain;

[0114] When in use, an improvement degree is constructed based on several environmental protection degrees. The usability of the optimized supply chain is determined based on the improvement degree. If it is available, the management of the green supply chain is completed. If it is not available, it can be optimized again. Through multiple optimizations, the supply chain can be improved and optimized to ensure the optimization effect.

[0115] Step 503: If the improvement degree of the optimized supply chain is still lower than the improvement threshold after several optimizations, it means that the optimized supply chain may still be unable to meet the application requirements, and the optimized supply chain may have some defects or failures and needs maintenance. At this time, a maintenance instruction is issued to the outside; after receiving the maintenance instruction, supply chain maintenance is used as the target word, and after deep retrieval and entity relationship building, a supply chain maintenance knowledge graph is pre-built;

[0116] Step 504: After setting the corresponding maintenance standards for each indicator parameter, identify the relevant data of the supply chain and obtain the corresponding maintenance features; use the trained matching model, based on the correspondence between the maintenance features and the supply chain, output the corresponding maintenance plan for the supply chain from the supply chain maintenance knowledge graph, execute the maintenance plan to maintain the supply chain, and realize the management of the current supply chain;

[0117] When using, combine the contents in steps 501 to 504:

[0118] After optimizing the supply chain, if the degree of improvement obtained cannot meet expectations, it means that the supply chain has become difficult to optimize and requires greater maintenance and replacement. After screening out the maintenance features, the supply chain maintenance knowledge graph outputs the corresponding maintenance plan for the supply chain, and the maintenance plan is executed to maintain the supply chain, thereby optimizing and improving the supply chain at a deeper level, reducing carbon emissions during supply chain operations, and achieving green management of the supply chain.

[0119] See also Figure 2 The present invention provides a cloud-based digital green supply chain management system, comprising:

[0120] The operation monitoring unit monitors and obtains the operation-related data of the supply chain during the monitoring period, and builds the supply chain operation data set after aggregation. The operation degree Yo is built from the operation data set. If the obtained operation degree Yo exceeds expectations, a cloud service management instruction is issued to the outside.

[0121] The matching unit combines the supply chain data and the cloud service platform data, and recommends the corresponding cloud service platform for the supply chain through the recommendation algorithm. If the recommended cloud service platform does not match the supply chain, a new cloud service platform is selected for the supply chain based on the matching degree Pp and reliability To of the cloud service platform;

[0122] The data processing unit matches the corresponding preprocessing strategy for the supply chain data, preprocesses the supply chain data, and constructs the environmental protection degree Bo from the preprocessed supply chain data. If the environmental protection degree Bo does not exceed the expectation, an optimization instruction is issued to the cloud service platform;

[0123] The data analysis unit uses multiple linear regression analysis to obtain the regression coefficients of each dependent variable in the supply chain, and constructs the optimization coefficient Ho in combination with the position of each dependent variable in the feature column. The supply chain is optimized according to the optimization features selected by the optimization coefficient Ho, and a prediction instruction is issued to the outside.

[0124] Optimize the maintenance unit, and use the trained supply chain digital twin model to test the optimized supply chain. Use the test data to construct the environmental protection degree Bo at each prediction node, and then generate the corresponding improvement degree Gs. If the improvement degree Gs is lower than the improvement threshold, the supply chain is optimized again. If the optimized improvement degree Gs is still lower than the improvement threshold, the supply chain maintenance knowledge graph outputs the corresponding maintenance plan for the supply chain to maintain the optimized supply chain.

[0125] It should be noted that the Analytic Hierarchy Process (AHP) is a decision-making method that decomposes decision-related elements into levels such as goals, criteria, and plans, and conducts qualitative and quantitative analysis on this basis. It is particularly suitable for dealing with target systems with hierarchical and staggered evaluation indicators, and when the target value is difficult to describe quantitatively, the AHP is an effective decision-making tool.

[0126] The core of the hierarchical analysis method is to decompose the decision problem into multiple levels to form a hierarchical structure, which usually includes the target level, the criterion level, the sub-criterion level and the solution level. By solving the eigenvector of the judgment matrix, the priority weight of each element of each level to an element of the previous level is obtained, and the final weight of each alternative solution to the total goal is recursively merged by the weighted sum method, so as to find the optimal solution.

[0127] The construction method of the supply chain maintenance knowledge graph can refer to the following:

[0128] Demand analysis and definition: First, the goal and scope of the supply chain maintenance knowledge graph need to be clarified. This includes determining the supply chain links that need to be covered, key entities and relationships, and required data sources.

[0129] Data collection: Based on demand analysis, collect data related to supply chain maintenance. This may include supplier information, product information, logistics information, maintenance records, etc. Data sources may include internal databases, external public data, data provided by partners, etc.

[0130] Information extraction: Extracting entities, attributes, and relationships between entities from the collected data. This usually requires the use of natural language processing (NLP) and machine learning techniques to automatically or semi-automatically extract structured information from unstructured data such as text and images.

[0131] Knowledge modeling: Based on the application attributes, knowledge characteristics, and actual needs of the industry, business abstraction and business modeling are carried out. This includes defining entities, relationships, attributes, etc., and determining the associations between them. When modeling, it is necessary to consider the rationality of concept division, attribute definition methods, complex knowledge labels such as time sequence, and the difficulty of subsequent knowledge expansion.

[0132] Schema layer construction: On top of the data layer, the schema layer of the knowledge graph is constructed. The schema layer is the core of the knowledge graph and stores refined knowledge. An ontology library is usually used to manage the schema layer of the knowledge graph. The ontology library's support for axioms, rules, and constraints is used to standardize the connections between entities, relationships, and entity types and attributes.

[0133] Knowledge fusion: Integrate knowledge from different sources to solve problems such as entity alignment and relationship conflicts. This may require the use of technical means such as entity linking and relationship matching.

[0134] Knowledge processing: Further processing and handling of the integrated knowledge to improve the quality and usability of knowledge. This may include steps such as data cleaning, deduplication, and relational reasoning.

[0135] Knowledge storage and query: Store the processed knowledge in the knowledge base and provide efficient query and retrieval functions. This may require the use of specialized technical means such as graph databases to support the storage and query of large-scale knowledge graphs.

[0136] Security protection: Develop effective security protection plans for the security risks that knowledge graphs may face. This may include measures such as data backup, access control, and encrypted transmission.

[0137] Iterative update: As the supply chain environment and business needs change, the knowledge graph needs to be continuously updated. This can be achieved by regularly collecting new data, updating models, optimizing algorithms, etc.

[0138] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0139] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0140] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0141] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only some logical function divisions. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0142] 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 distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0143] In addition, each functional unit in each embodiment of the present application may be integrated into a processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application.

[0144] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A cloud-based digital green supply chain management method, characterized by: include, During the monitoring period, the operation-related data of the supply chain operation is obtained, and after aggregation, a supply chain operation data set is constructed. The operation degree Yo is constructed from the operation data set. If the obtained operation degree Yo exceeds expectations, a cloud service management instruction is issued to the outside. Combining supply chain data and cloud service platform data, the recommendation algorithm recommends the corresponding cloud service platform for the supply chain. If the recommended cloud service platform does not match the supply chain, a new cloud service platform is selected for the supply chain based on the matching degree Pp and reliability To of the cloud service platform. After matching the corresponding preprocessing strategy for the supply chain data, the supply chain data is preprocessed, and the environmental protection degree Bo is constructed from the preprocessed supply chain data. If the environmental protection degree Bo does not exceed the expectation, an optimization instruction is issued to the cloud service platform; Use multiple linear regression analysis to obtain the regression coefficients of each dependent variable of the supply chain, and build the optimization coefficient Ho based on the position of each dependent variable in the feature column. Optimize the supply chain based on the optimization features selected by the optimization coefficient Ho, and issue prediction instructions to the outside. The trained supply chain digital twin model is used to test the optimized supply chain. The environmental protection degree Bo at each prediction node is constructed based on the test data, and then the corresponding improvement degree Gs is generated. If the improvement degree Gs is lower than the improvement threshold, the supply chain is optimized again. If the optimized improvement degree Gs is still lower than the improvement threshold, the supply chain maintenance knowledge graph outputs the corresponding maintenance plan for the supply chain to maintain the optimized supply chain. The construction method of the improvement degree Gs is as follows: Among them, ABo i and BBo i are the environmental protection degree of the i-th node before and after optimization, ABo avg With BBo avg is the corresponding mean, i is the node number, i = 1, 2, ..., w, w is the number of nodes, Zp i is the median value of the environmental protection degree of the ith node, Zp avg is the corresponding mean.

2. A cloud-based digital green supply chain management method according to claim 1, characterized in that: When the supply chain is still in operation, relevant data of supply chain operation is obtained in continuous sub-periods, and the supply chain operation data set is constructed after aggregation; the operation degree Yo is constructed from the data in the supply chain operation data set, wherein the response time Xt and the order fulfillment rate Vt are linearly normalized, and the corresponding data values ​​are mapped to the interval [0, 1], according to the following method: Where i = 1, 2, ..., p, p is the number of sub-periods, Xv i is the response time of the ith sub-cycle, Cj i is the order fulfillment rate in the i-th sub-period, with weight coefficients: 0≤F1≤1, 0≤F2≤1, and F1+F2=1.

3. The cloud-based digital green supply chain management method according to claim 1, characterized in that: After receiving the cloud service management instruction, it collects various supply chain data and aggregates them to build a supply chain data set; Collect cloud service platform data and build a cloud service platform data set after aggregation; According to the correspondence between the supply chain data and the cloud service platform data, the content-based recommendation algorithm recommends the corresponding cloud service platform for the supply chain; the Pearson correlation coefficient between the supply chain and the cloud service platform is constructed based on the collected data and used as the matching degree Pp. If the matching degree Pp is lower than the preset matching threshold, a reselection instruction is issued to the outside.

4. The cloud-based digital green supply chain management method according to claim 3 is characterized by: After receiving the reselection instruction, several cloud service platforms are evaluated to obtain the reliability To of each cloud service platform in the following way: linear normalization is performed on the fault response time Yt and the data backup speed Bt, and the corresponding data values ​​are mapped to the interval [0, 1], and then the following method is used: Weight coefficient: 0≤ρ≤1, 0≤ζ≤1; sort several cloud service platforms according to reliability To, and take the cloud service platform ranked in the top three in reliability To and with the highest matching degree Pp as the target platform, and issue data processing instructions to the outside.

5. A cloud-based digital green supply chain management method according to claim 4, characterized in that: After receiving the data processing instruction, the data in the supply chain data set is identified to obtain the corresponding data features; according to the correspondence between the data features and the preprocessing strategy, the preprocessing strategy library matches the corresponding preprocessing strategy, and the data in the set is processed according to the matched preprocessing strategy; The environmental protection index Bo is constructed from the preprocessed supply chain data in the following way: the carbon emissions Tp and waste generation Fp in each sub-period are linearly normalized, and the corresponding data values ​​are mapped to the interval [0, 1], and then the following method is used: Where i = 1, 2, ..., m, m is the number of sub-periods, weight coefficient: 0 ≤ S1 ≤ 1, 0 ≤ S2 ≤ 1, and S2 + S1 = 1; Tp i is the carbon emissions in the ith sub-period, Tp avg is the mean carbon emission, is the qualified reference value of carbon emissions; Fp i The amount of waste generated in the ith sub-period, Fp avg is the mean amount of waste generated, It is the qualified reference value for waste generation.

6. A cloud-based digital green supply chain management method according to claim 5, characterized in that: After receiving the optimization instruction, the supply chain parameters are used as independent variables, and the environmental protection degree Bo is used as the dependent variable to perform multiple linear regression analysis and obtain the corresponding linear regression equation; the regression coefficient corresponding to the independent variable in the linear regression equation is used as the influence coefficient Xn; Sort each independent variable by the influence coefficient Xn, obtain the corresponding feature column, determine the position of each dependent variable in the feature column, and then obtain the closeness center Un of each dependent variable; In addition, the corresponding optimization model is obtained by training the optimization algorithm, with improving the environmental protection degree Bo of the supply chain as the optimization goal, and the supply chain is optimized according to the screened optimization features, at least including: optimizing the supply chain transportation routes and modes, using energy-saving processes and equipment in the production process, and outputting the optimized supply chain.

7. A cloud-based digital green supply chain management method according to claim 6, characterized in that: After obtaining the closeness center Un of the dependent variable, combined with the corresponding influence coefficient Xn, the optimization coefficient Ho is constructed under dimensionless conditions. If the obtained optimization coefficient Ho exceeds the optimization threshold, the corresponding dependent variable is determined as the optimization feature; The optimization coefficient Ho is constructed as follows: Among them, i is the closeness centrality of the ith dependent variable, Un avg is the mean of the closeness center, Xn i is the influence coefficient of the ith dependent variable, Xn avg is the mean of the influence coefficient, i=1, 2,…,q, q is the number of dependent variables, and the weight coefficient is: 0≤β≤1, 0≤α≤1, and α+β=1.

8. The cloud-based digital green supply chain management method according to claim 7, characterized in that: After the cloud service platform receives the prediction instruction, the collected supply chain and related data are used as sample data to train and construct the supply chain digital twin model; the trained supply chain digital twin model predicts the optimized supply chain, and obtains corresponding test data at each prediction node, and reconstructs the environmental protection degree Bo at each prediction node with the supply chain data obtained from the test; the nodes before and after optimization are sorted according to the timeline and correspond one to one, and the improvement degree Gs is constructed from the environmental protection degree Bo at the nodes before and after optimization. If the improvement degree Gs is lower than the preset improvement threshold, the supply chain is re-optimized.

9. A cloud-based digital green supply chain management method according to claim 8, characterized in that: If the improvement degree of the optimized supply chain is still lower than the improvement threshold after several optimizations, a maintenance instruction is issued to the outside. After receiving the maintenance instruction, supply chain maintenance is used as the target word to pre-build a supply chain maintenance knowledge graph. After setting the corresponding maintenance standards for each indicator parameter, the relevant data of the supply chain is identified to obtain the corresponding maintenance features. Based on the correspondence between maintenance characteristics and the supply chain, the supply chain maintenance knowledge graph outputs the corresponding maintenance plan for the supply chain, and the maintenance plan is executed to maintain the supply chain.

10. A cloud-based digital green supply chain management system, characterized by: include: The operation monitoring unit monitors and obtains the operation-related data of the supply chain during the monitoring period, and builds the supply chain operation data set after aggregation. The operation degree Yo is built from the operation data set. If the obtained operation degree Yo exceeds expectations, a cloud service management instruction is issued to the outside. The matching unit combines the supply chain data and the cloud service platform data, and recommends the corresponding cloud service platform for the supply chain through the recommendation algorithm. If the recommended cloud service platform does not match the supply chain, a new cloud service platform is selected for the supply chain based on the matching degree Pp and reliability To of the cloud service platform; The data processing unit matches the corresponding preprocessing strategy for the supply chain data, preprocesses the supply chain data, and constructs the environmental protection degree Bo from the preprocessed supply chain data. If the environmental protection degree Bo does not exceed the expectation, an optimization instruction is issued to the cloud service platform; The data analysis unit uses multiple linear regression analysis to obtain the regression coefficients of each dependent variable in the supply chain, and constructs the optimization coefficient Ho in combination with the position of each dependent variable in the feature column. The supply chain is optimized according to the optimization features selected by the optimization coefficient Ho, and a prediction instruction is issued to the outside. Optimize the maintenance unit, and use the trained supply chain digital twin model to test the optimized supply chain. Use the test data to construct the environmental protection degree Bo at each prediction node, and then generate the corresponding improvement degree Gs. If the improvement degree Gs is lower than the improvement threshold, the supply chain is optimized again. If the optimized improvement degree Gs is still lower than the improvement threshold, the supply chain maintenance knowledge graph outputs the corresponding maintenance plan for the supply chain to maintain the optimized supply chain.

Images