Supply Chain Risk Management and Decision-making System Based on Large Model

Abstract

The present invention relates to the technical field of supply chain risk analysis, and specifically to a supply chain risk management and decision-making system based on a large model, including: a multi-layer supply chain data collection module: collecting supply chain operation data from multi-layer supply chain nodes; a multi-layer supply chain structure model construction module: constructing a multi-layer supply chain structure model between supply chain nodes through the collected operation data of each supply chain node, and automatically identifying the hidden risks in each supply chain node; a supply chain risk propagation path analysis module: after identifying the hidden risks, analyzing the hidden risk propagation path of the hidden risks in each supply chain node, and analyzing the transmission mechanism of the risks between different supply chain nodes; a decision support module: supporting enterprises to adjust supply chain strategies to reduce risks. The present invention enables enterprises to grasp the transmission law of risks in the supply chain from a global perspective, give priority to the hidden risks of key paths and nodes, and significantly improve the early warning level and risk resistance of the supply chain.

Description

Technical Field

[0001] The present invention relates to the technical field of supply chain risk analysis, and in particular to a supply chain risk management and decision-making system based on a large model. Background Art

[0002] As the complexity and uncertainty of supply chain networks continue to increase, enterprises face multi-level and multi-node potential risks in supply chain management. These risks may come from various factors such as market fluctuations and supplier defaults. In this context, traditional supply chain risk management technologies mainly rely on single-layer data monitoring and static risk assessment. However, these traditional methods are difficult to fully capture and analyze the implicit risk transmission between nodes in the supply chain. Especially in a multi-layer supply chain structure, risks are often gradually transmitted along the correlation relationships and hierarchical differences between nodes, causing local problems to evolve into systemic risks, which has a serious impact on the operations of core enterprises.

[0003] Existing supply chain risk management systems usually only focus on explicit risks, such as order delays or inventory shortages, but pay insufficient attention to the implicit risks of second- and third-tier suppliers, which makes companies lack effective early warning and decision-making support when dealing with changes in complex supply chain networks. In addition, the relationships between nodes in the supply chain network are complex, and traditional analysis methods cannot effectively capture the hierarchical structure and dependency relationships between nodes, making it difficult to implement multi-level risk transfer analysis. At the same time, traditional supply chain risk management systems usually rely on manual decision-making and fixed response strategies, with a slow response speed and unable to adjust supply chain strategies in real time to adapt to changes. Summary of the invention

[0004] The present invention provides a supply chain risk management and decision-making system based on a large model.

[0005] Supply chain risk management and decision-making system based on large models, including:

[0006] Multi-layer supply chain data collection module: collects supply chain operation data from multi-layer supply chain nodes, including order fulfillment status, delivery cycle, inventory level, and fulfillment history. Multi-layer supply chain nodes include core enterprises, first-tier suppliers, second-tier suppliers, and third-tier suppliers.

[0007] Multi-layer supply chain structure model building module: through the collected operation data of each supply chain node, a multi-layer supply chain structure model between supply chain nodes is built, the supply chain structure is dynamically updated, and based on the built multi-layer supply chain structure model, hidden risks in each supply chain node are automatically identified;

[0008] Supply chain risk transmission path analysis module: After identifying hidden risks, analyze the hidden risk transmission paths in each supply chain node, analyze the risk transmission mechanism between different supply chain nodes, and quantify the potential impact of a risk at a certain node on the entire supply chain;

[0009] Decision support module: Based on the analysis of hidden risk identification and hidden risk transmission paths, it generates decision recommendations to support enterprises in adjusting supply chain strategies to reduce risks.

[0010] Optionally, the multi-layer supply chain data collection module specifically includes:

[0011] The hierarchical node data interface submodule is used to establish data connections with nodes at different levels in the supply chain, including core enterprises, first-tier suppliers, second-tier suppliers, and third-tier suppliers, and extract supply chain operation data from nodes at all levels through interface protocols;

[0012] The data classification and extraction submodule collects order fulfillment, delivery cycle, inventory level and contract fulfillment history data from multi-layer supply chain nodes. The data includes order progress, delivery delays, inventory turnover rate, and contract fulfillment rate information to ensure that the extracted data can reflect the operating status of nodes at all levels;

[0013] The data synchronization and update submodule communicates with the supply chain nodes in real time and dynamically updates the operating data of each node according to the preset collection frequency to ensure the real-time and accuracy of data collection.

[0014] Optionally, the multi-layer supply chain structure model building module includes:

[0015] The node relationship identification submodule identifies and defines the relationship between supply chain nodes based on the operation data collected from core enterprises, first-tier suppliers, second-tier suppliers, and third-tier suppliers, and constructs the basic node diagram of the multi-layer supply chain structure;

[0016] The multi-layer structure generation submodule generates a multi-layer supply chain structure model based on the results of node relationship identification through graph network analysis technology, divides the supply chain into node groups at different levels, and forms a multi-layer supply chain structure model to ensure that the model can fully reflect the structural relationship and hierarchical division of the supply chain;

[0017] The hidden risk identification submodule uses the constructed multi-layer supply chain structure model and combines the historical and real-time data of supply chain nodes to automatically detect the hidden risks of each supply chain node in the supply chain, including supplier delays, abnormal inventory fluctuations or changes in supply cycles, and identifies the risk level of each node;

[0018] The dynamic model update submodule is used to dynamically adjust the multi-layer supply chain structure model based on real-time updates of node data so that it reflects the latest status of supply chain nodes and ensures the accuracy and real-time nature of hidden risk identification.

[0019] Optionally, the association relationship between the supply chain nodes is defined as an association matrix ,in Representation Node With Node The strength of association;

[0020] Based on the correlation matrix , generate a multi-layer supply chain structure model, build a multi-level network structure of the supply chain, and the generation of the multi-layer supply chain structure model is represented as a node set through a graph network construction algorithm and edge set ,in Representation Node With Node There is a significant correlation;

[0021] The edge generation condition is expressed as: ,in, is the state of the edge, the value is 1 or 0; is the association threshold, which is used to control whether there is a connection relationship between nodes.

[0022] Optionally, based on the multi-layer supply chain structure model, hidden risks are identified for each supply chain node, and hidden risks are detected through historical and real-time data of the node. The hidden risk probability , using a multi-factor model to calculate, weighting the multi-dimensional data of the node, the implicit risk probability is calculated as:

[0023] ,in;

[0024] Representation Node The hidden risk probability ranges from [0,1]; represents the activation function, which is used to normalize the implicit risk probability value to the range of [0,1]; The weight coefficients of the risk factors correspond to the contribution of order fulfillment, supply cycle, inventory level and fulfillment history to the risk; Representation Node The fluctuation of order fulfillment depends on the frequency of order delays; Representation Node The supply cycle fluctuation value indicates the stability of the supply cycle; Representation Node The inventory fluctuation value indicates the instability of inventory levels; Representation Node The fulfillment history score indicates the fulfillment reliability; Represents the bias term, which is used to adjust the benchmark value of the risk probability model.

[0025] Optionally, the multi-layer supply chain structure model also includes a layer parameter definition to set the layer parameters of the supply chain node. , used to identify the node Level:

[0026] Core Enterprises: ;

[0027] Tier 1 Supplier: ;

[0028] Second tier suppliers: ;

[0029] Tier 3 Suppliers: ;

[0030] The node association strength In the calculation, by introducing the hierarchical factor , consider the impact of hierarchical relationships between supply chain nodes on the correlation;

[0031] in, , indicating a node and The smaller the level difference, the stronger the node association strength. The greater the weight of is the weight coefficient of hierarchical differences, which controls the influence of hierarchical differences on the correlation degree.

[0032] Optionally, the supply chain risk propagation path analysis module specifically includes:

[0033] The risk propagation probability calculation submodule is used to calculate the risk propagation probability Pij from the occurrence node Ni to other nodes Nj based on the correlation matrix and hierarchical relationship of each node in the multi-layer supply chain structure model. The risk propagation probability Pij is affected by the correlation strength and hierarchical difference between the two nodes.

[0034] The risk impact measurement submodule calculates the risk impact value Ij of the risk propagating from node Ni to other nodes in the supply chain through the risk propagation probability and propagation path length of each node in the risk propagation path. The risk impact value reflects the potential impact of the hidden risk spreading step by step in the supply chain.

[0035] The multi-level risk propagation path identification submodule automatically identifies the main propagation path from the hidden risk node to the core node based on the risk propagation probability and risk impact value, and preferentially marks the propagation path with high risk propagation probability and high risk impact value, providing enterprises with early warning information on potential high-risk propagation links;

[0036] The overall supply chain risk quantification submodule accumulates the risk impact value of each node in the multi-level risk propagation path, quantifies the potential impact of the hidden risk of a certain node on the entire supply chain, provides quantitative analysis results of the overall supply chain risk, and calculates the total risk impact value of the supply chain.

[0037] Optionally, the risk propagation probability Pij is calculated as: ,in, For risk slave nodes To Node The probability of propagation is in the range of [0,1]. is the base coefficient of the propagation probability, which is used to adjust the size of the overall propagation probability and is a constant; Representation Node With Node The strength of association comes from the association matrix in the multi-layer supply chain structure model; exp is an exponential function, is the hierarchical influence coefficient, which is used to control the inhibitory effect of hierarchical differences on risk propagation and is a positive number; , is the level difference value, indicating that the node and nodes The difference in levels.

[0038] Optionally, the risk impact value Quantifying risk from nodes Propagate to nodes Time pair node The potential impact reflects the degree of influence of the risk spreading step by step in the supply chain. The risk impact value is calculated as:

[0039] ,in, Representation Node The hidden risk is propagated to nodes When the node The risk impact value of Representation Node The initial risk level of the node The risk intensity itself is calculated when identifying hidden risks. is the risk transmission distance attenuation factor, which reflects the attenuation degree of the risk transmission distance on the impact and is defined as the reciprocal of the shortest path distance between nodes. , Representation Node and nodes The shortest path distance between them is calculated based on the supply chain network path of the multi-layer supply chain structure model;

[0040] The identification of the main propagation path from the hidden risk node to the core node is based on the calculation result of the path risk cumulative impact value, and the path risk cumulative impact value is calculated as: ,in, Represents a slave node The cumulative risk impact value on a path propagating to the core node is used to identify the main propagation path. is the total number of nodes on the current path, is the risk propagation probability between nodes, is the individual risk impact value of the node on the path, based on The value of the cumulative risk impact value is given priority to select the propagation path with the highest cumulative risk impact value as the warning path;

[0041] The total risk impact value of the supply chain Calculated as: ,in;

[0042] is the overall risk impact value of the supply chain, indicating the potential impact of hidden risks in a single node on the entire supply chain. is the total number of initial hidden risk nodes, is the number of nodes in each propagation path, is the risk transmission probability, is the risk level of the initial hidden risk node, is the risk transmission distance attenuation factor.

[0043] Optionally, the decision support module specifically includes:

[0044] Identification and priority sorting of high-risk paths: Using the cumulative impact value of path risk Ipath calculated by the multi-level risk propagation path identification submodule, select the path with the highest cumulative risk impact value as the high-risk path of focus, sort the high-risk paths, and give priority to the paths that have a greater impact on core nodes or key supply chain nodes, so that enterprises can prioritize high-risk propagation links according to the sorting;

[0045] Node-level risk mitigation suggestions: Based on the risk impact value Ij of each node, for each high-risk node Nj, specific mitigation suggestions are generated according to its risk source. Mitigation suggestions include increasing inventory, reviewing supplier contracts, and requiring faster delivery to mitigate risk spread;

[0046] Suggestions for optimization and reorganization of the supply chain network: If the risk of a certain path or a certain supply chain node is too high, it is recommended that the company re-evaluate the supply chain relationship of the supply chain node;

[0047] Supply chain resource allocation and reserve resource allocation: using the total supply chain risk impact value , generate a global resource allocation plan.

[0048] Beneficial effects of the present invention:

[0049] The present invention integrates and updates in real time the multi-source data from core enterprises, first-tier suppliers, second-tier suppliers and third-tier suppliers through a multi-layer supply chain data acquisition module, ensuring the comprehensiveness and real-time nature of the global information of the supply chain. Based on the multi-layer supply chain structure model construction module, the correlation and hierarchical structure of nodes at each level are constructed and dynamically updated, which not only realizes the multi-level grouping of the supply chain, but also combines the multi-factor model in deep learning to identify the implicit risks of the operating data of each node, automatically discovers potential risk nodes in the supply chain, and provides enterprises with comprehensive and accurate risk perception capabilities.

[0050] The present invention adopts a risk propagation path analysis module, and through the risk propagation probability calculation based on the graph network, it clarifies the transmission mechanism of hidden risks between different nodes, and quantifies the degree of influence of the risk of a single node spreading to other nodes; by introducing the path risk impact value and the overall supply chain risk quantification value Itotal, it can identify the risk level between high-risk transmission paths and nodes, and provide enterprises with intuitive risk warnings. The multi-level risk analysis method enables enterprises to grasp the transmission rules of risks in the supply chain from a global perspective, give priority to the hidden risks of key paths and nodes, and significantly improve the early warning level and risk resistance of the supply chain.

[0051] In this invention, the decision support module generates real-time response strategies for enterprises by prioritizing high-risk paths and nodes, combining node-level risk mitigation measures, supply chain structure optimization suggestions, and resource allocation plans. It converts path analysis and global quantitative risk results into actionable supply chain adjustment suggestions, including emergency inventory configuration, backup supplier establishment, and dynamic monitoring mechanisms, to help enterprises achieve flexible allocation of supply chain resources and prepared responses. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0053] Figure 1A schematic diagram of system function modules according to an embodiment of the present invention;

[0054] Figure 2 It is a schematic diagram of a multi-layer supply chain structure model building module according to an embodiment of the present invention. DETAILED DESCRIPTION

[0055] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0056] It should be noted that the references to "one embodiment", "embodiment", "exemplary embodiments", "some embodiments" and the like in the specification indicate that the embodiments described may include specific features, structures or characteristics, but not every embodiment may include the specific features, structures or characteristics. In addition, when a specific feature, structure or characteristic is described in conjunction with an embodiment, it should be within the knowledge of a person skilled in the art to implement such feature, structure or characteristic in conjunction with other embodiments (whether or not explicitly described).

[0057] In general, a term can be understood, at least in part, from its use in context. For example, depending, at least in part, on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending, at least in part, on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0058] like Figure 1-Figure 2 As shown in the figure, the supply chain risk management and decision-making system based on the big model includes:

[0059] Multi-layer supply chain data collection module: collects supply chain operation data from multi-layer supply chain nodes. Supply chain operation data includes order fulfillment, delivery cycle, inventory level, and fulfillment history. Multi-layer supply chain nodes include core enterprises, first-tier suppliers, second-tier suppliers, and third-tier suppliers. This module uses a multi-source data interface to integrate internal data (such as enterprise ERP, supplier management system, etc.) and external data (such as market dynamics, public economic data, and financial risk data) to ensure the comprehensiveness and timeliness of the data;

[0060] Multi-layer supply chain structure model building module: through the collected operation data of each supply chain node, a multi-layer supply chain structure model between supply chain nodes is built, the supply chain structure is dynamically updated, and based on the built multi-layer supply chain structure model, hidden risks in each supply chain node are automatically identified;

[0061] Supply chain risk transmission path analysis module: After identifying hidden risks, analyze the hidden risk transmission paths in each supply chain node, analyze the risk transmission mechanism between different supply chain nodes, and quantify the potential impact of a risk at a certain node on the entire supply chain, especially the linkage effects of the bankruptcy and delivery delays of second- and third-tier suppliers on core enterprises;

[0062] Decision support module: Based on the analysis of hidden risk identification and hidden risk transmission paths, it generates decision recommendations to support enterprises in adjusting supply chain strategies to reduce risks.

[0063] The multi-layer supply chain data collection module specifically includes:

[0064] The hierarchical node data interface submodule is used to establish data connections with nodes at different levels in the supply chain, including core enterprises, first-tier suppliers, second-tier suppliers, and third-tier suppliers, and extract supply chain operation data from nodes at all levels through interface protocols;

[0065] The data classification and extraction submodule collects order fulfillment, delivery cycle, inventory level and contract fulfillment history data from multi-layer supply chain nodes. The data includes order progress, delivery delays, inventory turnover rate, and contract fulfillment rate information to ensure that the extracted data can reflect the operating status of nodes at all levels.

[0066] The data synchronization and update submodule communicates with the supply chain nodes in real time and dynamically updates the operating data of each node according to the preset collection frequency to ensure the real-time and accuracy of data collection.

[0067] The multi-layer supply chain structure model building modules include:

[0068] The node relationship identification submodule identifies and defines the relationship between supply chain nodes based on the operation data collected from core enterprises, first-tier suppliers, second-tier suppliers, and third-tier suppliers, and constructs the basic node diagram of the multi-layer supply chain structure;

[0069] The multi-layer structure generation submodule generates a multi-layer supply chain structure model based on the results of node relationship identification through graph network analysis technology, divides the supply chain into node groups at different levels, and forms a multi-layer supply chain structure model to ensure that the model can fully reflect the structural relationship and hierarchical division of the supply chain;

[0070] The hidden risk identification submodule uses the constructed multi-layer supply chain structure model and combines the historical and real-time data of supply chain nodes to automatically detect the hidden risks of each supply chain node in the supply chain, including supplier delays, abnormal inventory fluctuations or changes in supply cycles, and identifies the risk level of each node;

[0071] The dynamic model update submodule is used to dynamically adjust the multi-layer supply chain structure model based on real-time updates of node data so that it reflects the latest status of supply chain nodes and ensures the accuracy and real-time nature of hidden risk identification.

[0072] The association relationship between supply chain nodes is defined as the association matrix ,in Representation Node With Node The strength of association;

[0073] Node association strength Calculated as:

[0074] ,in;

[0075] Representation Node With Node degree of relevance; Represents the weight coefficient, which is used to adjust the weights of different correlation factors to meet ; Representation Node and The correlation of the order fulfillment status of the two is defined as the similarity of the order fulfillment status of the two, with a value range of [0,1]; Representation Node and Similarity correlation of supply cycles; Representation Node and The degree of correlation between the inventory levels of the two depends on the correlation between the inventory fluctuations of the two. Representation Node and The similarity correlation of the performance history.

[0076] The degree of association is calculated based on the similarities between different nodes in the supply chain in terms of order fulfillment, supply cycle, inventory level and fulfillment history. These factors directly reflect the closeness of collaboration and operational consistency between nodes. Specifically:

[0077] Order fulfillment: The similarity between nodes in order fulfillment can reflect the consistency of their responses to delivery time and fulfillment efficiency. If two nodes are highly similar in order fulfillment, it means that they have a close cooperative relationship in the supply chain and are dependent on each other.

[0078] Supply cycle: The similarity of supply cycles can reflect the consistency of cycles between upstream and downstream supply chains. Closer supply cycles indicate closer supply chain collaboration between nodes because they have similar demands and rhythms in time rhythm and supply frequency, which can form a more stable supply chain relationship.

[0079] Inventory level: The similarity of inventory levels between nodes often means that they have consistent inventory management needs. Nodes with similar inventories usually coordinate production or procurement to ensure an uninterrupted supply of materials or finished products, thus forming a more correlated supply chain relationship.

[0080] Fulfillment history: Similar fulfillment histories can reflect the reliability and consistency of nodes in fulfilling contracts in cooperation. The more similar the fulfillment histories between nodes are, the more stable the trust relationship and business cooperation between them are, which has a greater impact on the overall operation of the supply chain.

[0081] Therefore, the similarity of these factors is a reliable basis for calculating the degree of association, and they jointly reflect the coordination and dependency between supply chain nodes, which helps to accurately model the supply chain structure and hierarchical relationships.

[0082] Based on the correlation matrix , generate a multi-layer supply chain structure model, build a multi-level network structure of the supply chain, and the generation of the multi-layer supply chain structure model is represented as a node set through a graph network construction algorithm and edge set ,in Representation Node With Node There is a significant correlation;

[0083] The edge generation condition is expressed as: ,in, is the state of the edge, the value is 1 or 0; is the association threshold, which is used to control whether there is a connection relationship between nodes.

[0084] Based on the multi-layer supply chain structure model, the hidden risks of each supply chain node are identified, and the hidden risks are detected through the historical and real-time data of the nodes. The hidden risk probability , using a multi-factor model to calculate, weighting the multi-dimensional data of the node, the implicit risk probability is calculated as:

[0085] ,in;

[0086] Representation Node The hidden risk probability ranges from [0,1]; represents the activation function, which is used to normalize the implicit risk probability value to the range of [0,1]; The weight coefficients of the risk factors correspond to the contribution of order fulfillment, supply cycle, inventory level and fulfillment history to the risk; Representation Node The fluctuation of order fulfillment depends on the frequency of order delays; Representation Node The supply cycle fluctuation value indicates the stability of the supply cycle; Representation Node The inventory fluctuation value indicates the instability of inventory levels; Representation Node The fulfillment history score indicates the fulfillment reliability; Represents the bias term, which is used to adjust the benchmark value of the risk probability model.

[0087] The multi-layer supply chain structure model also includes the definition of hierarchical parameters, setting the hierarchical parameters of the supply chain nodes , used to identify the node Level:

[0088] Core Enterprises: ;

[0089] Tier 1 Supplier: ;

[0090] Second tier suppliers: ;

[0091] Tier 3 Suppliers: ;

[0092] Node association strength In the calculation, by introducing the hierarchical factor , considering the impact of the hierarchical relationship between supply chain nodes on the correlation, the calculation is updated as follows:

[0093] ;

[0094] in, , indicating a node and The smaller the level difference, the stronger the node association strength. The greater the weight of is the weight coefficient of hierarchical differences, which controls the influence of hierarchical differences on the correlation degree;

[0095] In the edge generation condition, by grouping nodes by level, ensure that nodes at different levels maintain a relatively tight structure. For nodes at the same level or adjacent levels, set a lower association threshold. To form a tighter connection, the edge generation condition is updated as follows:

[0096] ;in, is a lower correlation threshold, which is used to connect nodes at the same level or adjacent levels to highlight the grouping characteristics of nodes at different levels; τ is the standard correlation threshold, which is used to connect nodes across levels.

[0097] The supply chain risk propagation path analysis module specifically includes:

[0098] The risk propagation probability calculation submodule is used to calculate the risk propagation probability Pij from the occurrence node Ni to other nodes Nj based on the correlation matrix and hierarchical relationship of each node in the multi-layer supply chain structure model. The risk propagation probability Pij is affected by the correlation strength and hierarchical difference between the two nodes.

[0099] The risk impact measurement submodule calculates the risk impact value Ij of the risk propagating from node Ni to other nodes in the supply chain through the risk propagation probability and propagation path length of each node in the risk propagation path. The risk impact value reflects the potential impact of the hidden risk spreading step by step in the supply chain.

[0100] The multi-level risk propagation path identification submodule automatically identifies the main propagation path from the hidden risk node to the core node based on the risk propagation probability and risk impact value, and preferentially marks the propagation path with high risk propagation probability and high risk impact value, providing enterprises with early warning information on potential high-risk propagation links;

[0101] The overall supply chain risk quantification submodule accumulates the risk impact value of each node in the multi-level risk propagation path, quantifies the potential impact of the hidden risk of a certain node on the entire supply chain, provides quantitative analysis results of the overall supply chain risk, and calculates the total risk impact value of the supply chain.

[0102] The risk transmission probability Pij is calculated as: ,in, For risk slave nodes To Node The probability of propagation is in the range of [0,1]. is the base coefficient of the propagation probability, which is used to adjust the size of the overall propagation probability and is a constant; Representation Node With Node The strength of association comes from the association matrix in the multi-layer supply chain structure model; exp is an exponential function, is the hierarchical influence coefficient, which is used to control the inhibitory effect of hierarchical differences on risk propagation and is a positive number; , is the level difference value, indicating that the node and nodes The difference in levels.

[0103] Risk Impact Value Quantifying risk from nodes Propagate to nodes Time pair node The potential impact reflects the degree of influence of the risk spreading step by step in the supply chain. The risk impact value is calculated as:

[0104] ,in, Representation Node The hidden risk is propagated to nodes When the node The risk impact value of Representation Node The initial risk level of the node The risk intensity itself is calculated when identifying hidden risks. is the risk transmission distance attenuation factor, which reflects the attenuation degree of the risk transmission distance on the impact and is defined as the reciprocal of the shortest path distance between nodes. , Representation Node and nodes The shortest path distance between them is calculated based on the supply chain network path of the multi-layer supply chain structure model;

[0105] Identify the main propagation path from the hidden risk node to the core node based on the calculation result of the path risk cumulative impact value, which is calculated as: ,in, Represents a slave node The cumulative risk impact value on a path propagating to the core node is used to identify the main propagation path. is the total number of nodes on the current path, is the risk propagation probability between nodes, is the individual risk impact value of the node on the path, based on The value of the cumulative risk impact value is given priority to select the propagation path with the highest cumulative risk impact value as the warning path;

[0106] Quantify the potential impact of hidden risks on the entire supply chain when a certain node occurs, and the total risk impact value of the supply chain Calculated as: ,in;

[0107] is the overall risk impact value of the supply chain, indicating the potential impact of hidden risks in a single node on the entire supply chain. is the total number of initial hidden risk nodes, is the number of nodes in each propagation path, is the risk transmission probability, is the risk level of the initial hidden risk node, is the risk transmission distance attenuation factor.

[0108] and The difference between the two is that they measure the impact of risk on a single transmission path and on the entire supply chain network, respectively.

[0109] The decision support module specifically includes:

[0110] Identification and priority sorting of high-risk paths: Using the cumulative impact value of path risk Ipath calculated by the multi-level risk propagation path identification submodule, select the path with the highest cumulative risk impact value as the high-risk path of focus, sort the high-risk paths, and give priority to the paths that have a greater impact on core nodes or key supply chain nodes, so that enterprises can prioritize high-risk propagation links according to the sorting;

[0111] Node-level risk mitigation suggestions: Based on the risk impact value Ij of each node, for each high-risk node Nj, specific mitigation suggestions are generated according to its risk source (delayed order fulfillment, insufficient inventory, unstable performance history). Mitigation suggestions include increasing inventory, reviewing supplier contracts, and requiring faster delivery to mitigate risk spread;

[0112] Suggestions for optimization and reorganization of the supply chain network: If the risk of a certain path or a certain supply chain node is too high, it is recommended that the enterprise re-evaluate the supply chain relationship of the supply chain node. For example, it is recommended that the enterprise add redundant suppliers to high-risk second-tier or third-tier supplier nodes to spread the risk, or replace high-risk suppliers to reduce the risk impact value of the path;

[0113] Supply chain resource allocation and reserve resource allocation: using the total supply chain risk impact value , generate a global resource allocation plan. For nodes or paths with higher risks, it is recommended that enterprises configure reserve inventory, emergency logistics channels or backup suppliers around these nodes to enable rapid response and transfer when risks occur.

[0114] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0115] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. Supply chain risk management and decision-making system based on large models, characterized by: include: Multi-layer supply chain data collection module: collects supply chain operation data from multi-layer supply chain nodes, including order fulfillment status, delivery cycle, inventory level, and fulfillment history. Multi-layer supply chain nodes include core enterprises, first-tier suppliers, second-tier suppliers, and third-tier suppliers. Multi-layer supply chain structure model building module: Through the collected operation data of each supply chain node, a multi-layer supply chain structure model between supply chain nodes is built, the supply chain structure is dynamically updated, and based on the built multi-layer supply chain structure model, hidden risks in each supply chain node are automatically identified, including: The node relationship identification submodule identifies and defines the relationship between supply chain nodes based on the operation data collected from core enterprises, first-tier suppliers, second-tier suppliers, and third-tier suppliers, and constructs the basic node diagram of the multi-layer supply chain structure; The multi-layer structure generation submodule generates a multi-layer supply chain structure model based on the results of node relationship identification through graph network analysis technology, divides the supply chain into node groups at different levels, and forms a multi-layer supply chain structure model; The hidden risk identification submodule uses the constructed multi-layer supply chain structure model and combines the historical and real-time data of supply chain nodes to automatically detect the hidden risks of each supply chain node in the supply chain, including supplier delays, abnormal inventory fluctuations or changes in supply cycles, and identifies the risk level of each node; The dynamic model update submodule is used to dynamically adjust the multi-layer supply chain structure model based on the real-time update of node data to reflect the latest status of the supply chain nodes; The association relationship between the supply chain nodes is defined as the association matrix ,in Representation Node With Node The strength of association; Based on the correlation matrix , generate a multi-layer supply chain structure model, build a multi-level network structure of the supply chain, and the generation of the multi-layer supply chain structure model is represented as a node set through a graph network construction algorithm and edge set ,in Representation Node With Node There is a relationship of association; The edge generation condition is expressed as: ,in, is the state of the edge, the value is 1 or 0; is the association threshold, which is used to control whether there is a connection relationship between nodes; Supply chain risk transmission path analysis module: After identifying hidden risks, analyze the hidden risk transmission paths in each supply chain node, analyze the risk transmission mechanism between different supply chain nodes, and quantify the potential impact of a risk at a certain node on the entire supply chain; Decision support module: Based on the analysis of hidden risk identification and hidden risk transmission paths, it generates decision recommendations to support enterprises in adjusting supply chain strategies to reduce risks.

2. The supply chain risk management and decision-making system based on a large model according to claim 1 is characterized in that: The multi-layer supply chain data collection module specifically includes: The hierarchical node data interface submodule is used to establish data connections with nodes at different levels in the supply chain, including core enterprises, first-tier suppliers, second-tier suppliers, and third-tier suppliers, and extract supply chain operation data from nodes at all levels through interface protocols; The data classification and extraction submodule collects order fulfillment, delivery cycle, inventory level and contract fulfillment history data from multi-layer supply chain nodes. The data includes order progress, delivery delays, inventory turnover rate and contract fulfillment rate information; The data synchronization and update submodule dynamically updates the operating data of each node according to the preset collection frequency through real-time communication with the supply chain nodes.

3. The supply chain risk management and decision-making system based on a large model according to claim 1 is characterized in that: Based on the multi-layer supply chain structure model, the hidden risks of each supply chain node are identified, and the hidden risks are detected through the historical and real-time data of the node. The probability of hidden risk , using a multi-factor model to calculate, weighting the multi-dimensional data of the node, the implicit risk probability is calculated as: ,in; Representation Node The hidden risk probability ranges from [0,1]; represents the activation function, which is used to normalize the implicit risk probability value to the range of [0,1]; The weight coefficients of the risk factors correspond to the contribution of order fulfillment, supply cycle, inventory level and fulfillment history to the risk; Representation Node The fluctuation of order fulfillment depends on the frequency of order delays; Representation Node The supply cycle fluctuation value indicates the stability of the supply cycle; Representation Node The inventory fluctuation value indicates the instability of inventory levels; Representation Node The fulfillment history score indicates the fulfillment reliability; Represents the bias term, which is used to adjust the benchmark value of the risk probability model.

4. The supply chain risk management and decision-making system based on a large model according to claim 1 is characterized in that: The multi-layer supply chain structure model also includes a layer parameter definition, setting the layer parameters of the supply chain node , used to identify the node Level: Core Enterprises: ; Tier 1 Supplier: ; Second tier suppliers: ; Tier 3 Suppliers: ; The node association strength In the calculation, by introducing the hierarchical factor , consider the impact of hierarchical relationships between supply chain nodes on the correlation; in, , indicating a node and The smaller the level difference, the stronger the node association strength. The greater the weight.

5. The supply chain risk management and decision-making system based on a large model according to claim 1 is characterized in that: The supply chain risk propagation path analysis module specifically includes: The risk propagation probability calculation submodule is used to calculate the risk propagation probability Pij from the occurrence node Ni to other nodes Nj based on the correlation matrix and hierarchical relationship of each node in the multi-layer supply chain structure model. The risk propagation probability Pij is affected by the correlation strength and hierarchical difference between the two nodes. The risk impact measurement submodule calculates the risk impact value Ij of the risk propagating from node Ni to other nodes in the supply chain through the risk propagation probability and propagation path length of each node in the risk propagation path. The risk impact value reflects the potential impact of the hidden risk spreading step by step in the supply chain. The multi-level risk propagation path identification submodule automatically identifies the main propagation path from the hidden risk node to the core node based on the risk propagation probability and risk impact value, and preferentially marks the propagation path with high risk propagation probability and high risk impact value; The overall supply chain risk quantification submodule accumulates the risk impact value of each node in the multi-level risk propagation path, quantifies the potential impact of the hidden risk of a certain node on the entire supply chain, provides quantitative analysis results of the overall supply chain risk, and calculates the total risk impact value of the supply chain.

6. The supply chain risk management and decision-making system based on a large model according to claim 5 is characterized in that: The risk propagation probability Pij is calculated as: ,in, For risk slave nodes To Node The probability of propagation is in the range of [0,1]. is the base coefficient of the propagation probability, which is used to adjust the size of the overall propagation probability and is a constant; Representation Node With Node The correlation strength; exp is the exponential function, is the hierarchical influence coefficient, which is used to control the inhibitory effect of hierarchical differences on risk propagation and is a positive number; , is the level difference value, indicating that the node and nodes The difference in levels.

7. The supply chain risk management and decision-making system based on a large model according to claim 6 is characterized in that: The risk impact value Quantifying risk from nodes Propagate to nodes Time pair node The potential impact reflects the degree of influence of the risk spreading step by step in the supply chain. The risk impact value is calculated as: ,in, Representation Node The hidden risk is propagated to nodes When the node The risk impact value of Representation Node The initial risk level of the node The intensity of the risk itself, is the risk transmission distance attenuation factor, which reflects the attenuation degree of the risk transmission distance on the impact and is defined as the reciprocal of the shortest path distance between nodes. , Representation Node and nodes The shortest path distance between The identification of the main propagation path from the hidden risk node to the core node is based on the calculation result of the path risk cumulative impact value, and the path risk cumulative impact value is calculated as: ,in, Represents a slave node The cumulative risk impact value on a path propagating to the core node is used to identify the main propagation path. is the total number of nodes on the current path, is the risk propagation probability between nodes, is the individual risk impact value of the node on the path, based on The value of the cumulative risk impact value is given priority to select the propagation path with the highest cumulative risk impact value as the warning path; Total supply chain risk impact value Calculated as: ,in; is the overall risk impact value of the supply chain, indicating the potential impact of hidden risks in a single node on the entire supply chain. is the total number of initial hidden risk nodes, is the number of nodes in each propagation path, is the risk transmission probability, is the risk level of the initial hidden risk node, is the risk transmission distance attenuation factor.

8. The supply chain risk management and decision-making system based on a large model according to claim 7 is characterized in that: The decision support module specifically includes: Identification and priority sorting of high-risk paths: Using the cumulative impact value of path risk Ipath calculated by the multi-level risk propagation path identification submodule, select the path with the highest cumulative risk impact value as the high-risk path of focus, sort the high-risk paths, and enable enterprises to prioritize high-risk propagation links according to the sorting; Node-level risk mitigation suggestions: Based on the risk impact value Ij of each node, for each high-risk node Nj, specific mitigation suggestions are generated according to its risk source. Mitigation suggestions include increasing inventory, reviewing supplier contracts, and requiring faster delivery to mitigate risk spread; Suggestions for optimization and reorganization of the supply chain network: If the risk of a certain path or a certain supply chain node is too high, it is recommended that the company re-evaluate the supply chain relationship of the supply chain node; Supply chain resource allocation and reserve resource allocation: using the total supply chain risk impact value , generate a global resource allocation plan.

Images