Supply chain partner risk management method based on block chain

Abstract

The invention relates to the technical field of block chains, in particular to a supply chain partner risk management method based on a block chain, which comprises the following steps: comprehensively and accurately evaluating supply chain partner risks by collecting multi-source dynamic data and extracting multi-dimensional risk feature vectors; the blockchain technology is utilized to link the evidence storage risk metadata, so that the transparency and credibility of risk management are enhanced; the risk assessment weight is dynamically adjusted through a reinforcement learning algorithm, so that intelligence and self-adaption of risk assessment are realized, a risk management strategy can be flexibly adjusted, and the efficiency and accuracy of risk response are improved; by generating the risk diffusion trajectory diagram with the timestamp and automatically matching the risk disposal strategy, dynamic and optimized risk management is realized, risk diffusion can be timely controlled, risk loss can be reduced, and the overall stability and competitiveness of a supply chain can be improved.

Description

Technical Field

[0001] The present invention relates to the field of blockchain technology, and particularly to a supply chain partner risk management method based on blockchain. Background Art

[0002] There are still some challenges in current supply chain partner risk management, including the following aspects: In traditional supply chain management, data among various parties is often opaque, resulting in information asymmetry and lack of trust. It is difficult for core enterprises, suppliers, manufacturers and other parties to comprehensively and accurately obtain the risk data of other parties, thus affecting the accuracy of risk assessment and the effectiveness of risk management; Traditional risk warning mechanisms may lack real-time monitoring and automatic warning functions, resulting in risks not being discovered and warned in time. Even if risks are discovered, due to untimely transmission of warning information or cumbersome processing procedures, the response may be untimely and the spread of risks cannot be effectively controlled; In the supply chain, risks may spread along transaction relationships and affect other parties. However, traditional methods are difficult to effectively simulate the risk conduction path, resulting in the inability to accurately assess the impact of risks on the overall supply chain; At the same time, traditional risk disposal processes may involve multiple links and parties, resulting in cumbersome processes and low efficiency. Due to the lack of an effective effect evaluation mechanism, it is impossible to timely understand the actual effects of risk disposal measures, and it is difficult to continuously optimize and improve the risk management process. Therefore, the present invention proposes a supply chain partner risk management method based on blockchain. Summary of the Invention

[0003] The purpose of the present invention is to solve the problems in the background art, and a supply chain partner risk management method based on blockchain is proposed.

[0004] In order to achieve the above purpose, the present invention adopts the following technical solutions: A supply chain partner risk management method based on blockchain, including: S1. Collect multi-source dynamic data, including transaction performance and compliance records; extract and integrate multi-dimensional risk feature vectors including credit risk, operational risk, and legal risk through data analysis methods, generate standardized risk metadata and store it on the blockchain for certification; S2. Deploy a risk assessment and warning intelligent contract on the consortium blockchain, extract the multi-dimensional risk feature vectors in the risk metadata certificate, and calculate the comprehensive risk score; dynamically adjust the weights of each dimension through a reinforcement learning algorithm to achieve the functions of risk assessment and warning; S3. Obtain the risk assessment and warning results, construct a supply chain topology network based on the transaction data stored on the blockchain, simulate the risk conduction process through the risk value diffusion formula, and generate a risk diffusion trajectory map with a time stamp for sharing; S4. Automatically match and execute risk disposal strategies based on the shared information in the risk diffusion trajectory diagram, and continuously evaluate the effect of risk disposal, dynamically adjust to optimize risk management.

[0005] Furthermore, the process of extracting and integrating multi-dimensional risk feature vectors including credit risk, operational risk, and legal risk through data analysis methods, generating standardized risk metadata, and storing it on the chain for evidence includes: Construct a three-dimensional risk matrix, including the credit risk dimension, the operational risk dimension, and the legal risk dimension: Calculate the credit risk feature vector through the historical payment delay rate and the order fulfillment rate. The calculation formula for the historical payment delay rate is: , where is the historical payment delay rate, is the total number of orders within the statistical period, is the order index within the statistical period, is the actual payment date of the th order, is the contractually agreed payment date of the th order; The calculation formula for the order fulfillment rate is: , where is the order fulfillment rate, is the number of orders delivered on time, is the total number of orders; Integrate the historical payment delay rate and the order fulfillment rate to obtain the credit risk feature vector ; Calculate the operational risk feature vector based on the standard deviation of logistics timeliness and inventory turnover deviation. The calculation formula for the standard deviation of logistics timeliness is: , where is the standard deviation of logistics timeliness, is the logistics index, is the actual timeliness of the th logistics, is the average logistics timeliness; The calculation formula for inventory turnover deviation is: , where is the inventory turnover deviation, is the average inventory turnover days, is the actual inventory turnover days; Integrate the standard deviation of logistics timeliness and inventory turnover deviation to obtain the operational risk feature vector ; , where is the environmental protection penalty rate, is the number of environmental protection penalties, is the total number of operating days; The calculation formula for the proportion of license expiration days is: , where is the proportion of license invalid days, is the total number of license invalid days; integrating the environmental protection penalty rate and the proportion of license invalid days to obtain the legal risk feature vector ; Adopt the JSON-LD format to serialize the multi-dimensional risk feature vector into a standardized risk metadata packet; perform a hash calculation on the serialized JSON-LD data to generate a unique data hash value; use the generated JSON-LD packet to call the deposit contract of the consortium chain; the deposit contract will record the complete JSON-LD packet and generate a risk data deposit certificate with a digital signature, where the certificate contains a timestamp and the data hash value; based on the risk data deposit certificate, use the data hash value therein as the input and write it into the Merkle tree of the consortium chain; the root node of the Merkle tree will be recorded on the blockchain; based on the record of the Merkle tree, generate a risk metadata certificate with a timestamp, where the certificate contains the data source signature and the Merkle path.

[0006] Furthermore, the process of extracting the multi-dimensional risk feature vector from the risk metadata certificate and calculating the comprehensive risk score includes: Call the generated risk metadata certificate through the on-chain verification engine, and extract the embedded JSON-LD format multi-dimensional risk feature vector and deposit certificate information from the certificate; synchronously parse the Merkle path in the certificate, extract the root node hash value, and use it as the benchmark anchor point; The risk assessment and early warning intelligent contract reads the deposited multi-dimensional risk feature vector from the temporary storage area and parses out the multi-dimensional risk feature vector; uses the multi-dimensional risk feature vector of the current cycle and the historical risk event library stored on the chain as the input state; Normalize each dimension of the risk feature vector, perform a dot product calculation in combination with the initial weight, and generate a comprehensive risk score.

[0007] Furthermore, the process of dynamically adjusting the weights of each dimension through the reinforcement learning algorithm to achieve the risk assessment and early warning function includes: Introduce the supply chain risk reinforcement learning RL agent model, and use the Q-learning algorithm to dynamically adjust the weights of each dimension: define the risk dimension set as , and the initial weight of each dimension is , satisfying ; Define the state space as , representing the matching degree between the current multi-dimensional risk feature vector and the historical event library, that is , where is the time step, representing the number of iterations for the risk assessment and early warning smart contract to learn or the discrete time points of risk events; is the L2 norm; Define the action space as , which is used to adjust the weight direction, that is, increase or decrease the dimension weight: , where ; The risk assessment and early warning smart contract dynamically updates the weight through the following Q-learning rule: , In the formula, is the updated weight; is the immediate reward, obtained based on the risk score change and disposal cost; is the learning rate; is the discounted risk factor, is the next state is the candidate weight value of; The weight adjustment result is written into the blockchain through the risk assessment and early warning smart contract, forming a learning mechanism for risk event-weight adjustment: the risk assessment and early warning smart contract calls the historical transaction behavior fingerprints stored on the chain, and recalculates the credit risk characteristics, operational risk characteristics and legal risk characteristic vectors: , , ; Compare the recalculated multi-dimensional risk characteristic vector with the original multi-dimensional risk characteristic vector generated by the three-dimensional risk matrix dimension by dimension: , In the formula, is the risk deviation rate, is the original multi-dimensional risk characteristic vector, is the recalculated multi-dimensional risk characteristic vector, is the risk characteristic vector category; Preset risk deviation thresholds β1 and β2; when the calculated risk deviation rate breaks through the risk deviation threshold, a hierarchical early warning is automatically triggered, that is, high risk ≥ β1, medium risk ∈ [β2, β1), low risk < β2: when it is identified as a high-risk level, that is, the risk deviation rate ≥ β1, a structured early warning event package including the risk level and risk characteristic categories is generated and broadcast to all associated nodes through the P2P network of the alliance chain, including the core enterprise node, the supplier node, and the manufacturer node. At the same time, the early warning timestamp, the triggering node ID, and the event hash value are recorded on the chain; when it is identified as a medium-risk level, that is, the risk deviation rate belongs to [β2, β1), an early warning notice is pushed to the risk control department of the core enterprise in the supply chain. At the same time, this node is marked as an observation object on the chain, and its priority cooperation rights and interests are restricted; when it is identified as a low-risk level, that is, the risk deviation rate < β2, no early warning is triggered; Integrate the risk deviation rate and early warning event information, and generate a new on-chain block through the risk assessment and early warning smart contract.

[0008] Furthermore, the process of obtaining the risk assessment and early warning results, constructing the supply chain topology network based on the transaction data stored on the blockchain, simulating the risk conduction process through the risk value diffusion formula, and generating a timestamped risk diffusion trajectory map for sharing includes: Each associated node in the supply chain receives the broadcast structured early warning event package through the P2P network of the alliance chain, where the associated nodes include the core enterprise node, the supplier node, and the manufacturer node; each associated node parses the received early warning event package, extracts the key information, and stores it in the local database. Using the historical transaction records stored on the blockchain, that is, the transaction amount and transaction frequency; constructing a supply chain topology network based on the transaction dependency relationship: using the identity DID identifier as the node of the supply chain topology network, that is, the core enterprise node, the supplier node, and the manufacturer node; using the weighted values of the transaction amount and transaction frequency as the edges of the supply chain topology network; in the constructed supply chain topology network, locate the warning node triggered by the early warning event, that is, the risk source node; set the initial risk value of the risk source node to , the weight set of the edges in the supply chain topology network is , where represents the node and the node The weighted value of the transaction amount and transaction frequency between them, that is, the weight of the edge between the node and the node ; along the edges in the supply chain topology network, the risk value is diffused according to the edge weight attenuation, where the risk value diffusion formula is: , In the formula, is at time the node at The risk value, is the set of neighbor nodes of node ; is the sum of the weights of the edges between all neighbor nodes of node and node ; represents the exponential function with base e, is the risk attenuation coefficient, is the shortest path length between node and node ; is the risk value of node at time ; is the current time point; during the diffusion process, the risk status of affected nodes is marked in real time, including the risk level, multi-dimensional risk characteristics, and the magnitude of the risk value; According to the information recorded during the risk value diffusion process, a risk diffusion trajectory graph with a timestamp is generated; the generated risk diffusion trajectory graph is stored in IPFS, and a corresponding CID is generated; the CID is written into the blockchain transaction log; at the same time, the trajectory graph information is shared with each associated node in the supply chain through the P2P network of the consortium chain.

[0009] Furthermore, based on the shared information of the risk diffusion trajectory graph, the risk disposal strategy is automatically matched and executed, and the risk disposal effect is continuously evaluated. The process of dynamically adjusting to optimize risk management includes: Obtain the risk level and risk characteristic type, and match the predefined risk disposal rules from the on-chain policy library; automatically generate a structured instruction package containing risk disposal measures, execution time limit, and responsible entity through the pre-set risk disposal strategy smart contract, and encrypt it with a digital signature; among them, the instruction package will be broadcast to all relevant nodes through the P2P network of the consortium chain, including core enterprise nodes, supplier nodes, and manufacturer nodes; After receiving the instruction package, each associated node will immediately execute the corresponding risk disposal measures and record the execution status through its own log; the risk disposal strategy smart contract continuously monitors the logs of each associated node, verifies whether the risk disposal measures are executed according to the instructions, and collects the execution feedback of the risk disposal measures in real time; at the same time, continuously capture the dynamic shared information in the risk diffusion trajectory graph, calculate the multi-dimensional risk feature vector and compare it with the baseline before risk disposal, and quantitatively evaluate the risk reduction amplitude. Among them, the risk reduction amplitude is quantitatively evaluated through the risk disposal efficiency dynamic evaluation formula, so as to judge the effectiveness of the risk disposal strategy: , wherein, is the risk reduction amplitude; is the integral operation with respect to the time variable t; is the integration interval, representing the scope of action or time window of risk disposal; is the risk attenuation kernel function, used to quantify the weighted influence of the risk level RH and the risk characteristic type k on the risk disposal effect; is the policy execution efficiency factor; is the risk conduction damping effect; is the indicator function. When , the value of the indicator function is 1, and the formula calculates the actual risk reduction amplitude , triggering the risk disposal mechanism; when , the value of the exponential function is 0, the overall result of the formula is 0, and the risk disposal measure is to suspend or downgrade; is the preset dynamic risk threshold.

[0010] Compared with the prior art, the beneficial effects of the present invention are as follows: By collecting multi-source dynamic data, integrating multi-dimensional feature vectors such as credit risk, operational risk, and legal risk, and storing them on the chain for certification, the comprehensiveness and accuracy of risk data are ensured, providing a solid foundation for subsequent risk assessment and improving the accuracy of risk identification; By deploying risk assessment and early warning smart contracts on the consortium chain and using reinforcement learning algorithms to dynamically adjust the weights of each dimension, the intelligence and real-time nature of risk assessment are realized, the efficiency of risk assessment is improved, and the dynamic adaptability of the assessment results is enhanced; Based on blockchain data, a supply chain topology network is constructed, the risk conduction process is simulated, and a risk diffusion trajectory map with a time stamp is generated, which is convenient to understand the risk diffusion path, take measures to control risks in a timely manner, and at the same time realize the transparent sharing of risk information; Based on the risk diffusion trajectory map, risk disposal strategies are automatically matched and executed, and the disposal effect is continuously evaluated, realizing the automation and closed-loop optimization of risk management, improving the efficiency and effect of risk response, and reducing risk losses. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 is the flow chart of the supply chain partner risk management method based on blockchain proposed by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0012] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0013] Referring to Figure 1 , the supply chain partner risk management method based on blockchain includes: S1. Collect multi-source dynamic data, including transaction performance and compliance records; extract and integrate multi-dimensional risk feature vectors containing credit risk, operational risk, and legal risk through data analysis methods, generate standardized risk metadata, and store it on the chain for certification; S2. Deploy risk assessment and early warning smart contracts on the alliance chain, extract multi-dimensional risk feature vectors from risk metadata certificates, and calculate comprehensive risk scores; dynamically adjust the weights of each dimension through reinforcement learning algorithms to achieve risk assessment and early warning functions; S3. Obtain risk assessment and early warning results, construct a supply chain topology network based on transaction data stored on the blockchain, simulate the risk conduction process through the risk value diffusion formula, and generate a risk diffusion trajectory graph with timestamps for sharing; S4. Based on the shared information of the risk diffusion trajectory graph, automatically match and execute risk disposal strategies, continuously evaluate the effectiveness of risk disposal, and dynamically adjust to optimize risk management.

[0014] It should be further noted that in the specific implementation process, the process of collecting multi-source dynamic data (including transaction performance and compliance records); extracting and integrating multi-dimensional risk feature vectors containing credit risk, operational risk, and legal risk through data analysis methods, generating standardized risk metadata, and storing it on the chain for certification is as follows: Construct a three-dimensional risk matrix, including the credit risk dimension, operational risk dimension, and legal risk dimension: Calculate the credit risk feature vector through the historical payment delay rate and order fulfillment rate: The calculation formula for the historical payment delay rate is: , where is the historical payment delay rate, is the total number of orders within the statistical period, is the order index within the statistical period, is the actual payment date of the th order, is the contractually agreed payment date of the th order; The calculation formula for the order fulfillment rate is: , where is the order fulfillment rate, is the number of orders delivered on time, is the total number of orders; Integrate the historical payment delay rate and the order fulfillment rate to obtain the credit risk feature vector ; Calculate the operational risk feature vector based on the standard deviation of logistics timeliness and inventory turnover deviation: The calculation formula for the standard deviation of logistics timeliness is: , where is the standard deviation of logistics timeliness, is the logistics index, is the number of logistics times, is the The actual timeliness of secondary logistics, is the average value of logistics timeliness; the calculation formula for inventory turnover deviation is: , where is the inventory turnover deviation, is the average inventory turnover days, is the actual inventory turnover days; integrating the standard deviation of logistics timeliness and inventory turnover deviation to obtain the operation risk feature vector ; Based on the environmental protection penalty records and the number of days of license expiration, calculate the legal risk feature vector: , where is the environmental protection penalty rate, is the number of environmental protection penalties, is the total number of operating days; the calculation formula for the proportion of license expiration days is: , where is the proportion of license expiration days, is the total number of license expiration days; integrating the environmental protection penalty rate and the proportion of license expiration days to obtain the legal risk feature vector ; Among them, the actual payment date, the contractually agreed payment date, the number of orders delivered on time, and the total number of orders are obtained by docking with the supplier's ERP system through the API gateway and extracting transaction data and relevant historical records; the actual logistics timeliness, the average logistics timeliness, the average inventory turnover days, and the actual inventory turnover days are obtained through statistical analysis of the logistics performance data captured in real time by Internet of Things devices (such as RFID, GPS) and the inventory data extracted by docking with the supplier's ERP or SCM system; the environmental protection penalty records, the total number of operating days, and the number of days of license expiration are obtained by accessing the government supervision platform and crawling legal compliance data for analysis; Adopt the JSON-LD format to serialize the multi-dimensional risk feature vector into a standardized risk metadata packet; perform a hash calculation on the serialized JSON-LD data to generate a unique data hash value; use the generated JSON-LD packet to call the deposit contract of the consortium blockchain; the deposit contract will record the complete JSON-LD packet and generate a risk data deposit certificate with a digital signature, where the certificate contains a timestamp and a data hash value to ensure the integrity and verifiability of the data; based on the risk data deposit certificate, use the data hash value in it as the input and write it into the Merkle tree of the consortium blockchain; the root node of the Merkle tree will be recorded on the blockchain to provide a lightweight verification method for the data; based on the record of the Merkle tree, generate a risk metadata certificate with a timestamp, where the certificate contains a data source signature and a Merkle path (Merkle root hash value, node hash value) to prove the integrity and authenticity of the data.

[0015] It should be further noted that in the specific implementation process, when deploying the risk assessment and early warning smart contract on the consortium blockchain, extracting the multi-dimensional risk feature vectors from the risk metadata certificate, and calculating the comprehensive risk score; the process of dynamically adjusting the weights of each dimension through the reinforcement learning algorithm to achieve the risk assessment and early warning function is as follows: Call the generated risk metadata certificate through the on-chain verification engine, extract the embedded JSON-LD format multi-dimensional risk feature vectors (credit, operation, and legal risk features) and the evidence storage voucher information (data hash value, digital signature, timestamp) from the certificate; synchronously parse the Merkle path in the certificate, extract the root node hash value (this root node hash value has been stored on the chain), and provide a benchmark anchor point for subsequent data authenticity verification; The risk assessment and early warning smart contract reads the multi-dimensional risk feature vectors that have been stored as evidence from the temporary storage area and parses out the multi-dimensional risk feature vectors; uses the multi-dimensional risk feature vectors (credit, operation, and legal risk features) of the current period and the historical risk event library stored on the chain as the input state; Perform normalization processing on the risk feature vectors of each dimension (such as mapping the historical payment delay rate from [0, 100%] to [0, 1]), perform dot product calculation in combination with the initial weights, and generate a comprehensive risk score (formula: comprehensive risk score = initial credit risk weight × credit risk eigenvalue + initial operation risk weight × operation risk eigenvalue + initial legal risk weight × legal risk eigenvalue); Introduce the supply chain risk reinforcement learning RL agent model, and use the Q-learning algorithm to dynamically adjust the weights of each dimension: Define the risk dimension set as and the initial weight of each dimension is ( is the risk dimension index), satisfying ; Define the state space as , representing the matching degree between the current multi-dimensional risk feature vector and the historical event library, that is , where is the time step, representing the number of iterations of learning by the risk assessment and early warning smart contract or the discrete time point of risk events; is the L2 norm; Define the action space as , used to adjust the weight direction, that is, increase or decrease the dimension weight: , where, is a specific action in the action space, and each action corresponds to a specific weight adjustment direction, is the set of action spaces, ; The risk assessment and early warning smart contract dynamically updates the weights through the following Q-learning rules: , where, is the updated weight; is the immediate reward, obtained based on the risk score change and the disposal cost: , is the weight factor for the comprehensive risk score change, is the weight factor for the disposal cost, is the comprehensive risk score change ( ), is the compliance cost for weight adjustment ( , is the preset cost coefficient); is the learning rate, used to control the weight allocation between new experience (immediate reward) and old knowledge (historical Q value). In supply chain risk management, can be dynamically adjusted according to the volatility of the risk environment. For example, if the supply chain stability decreases (more risk events), can be increased to adapt to the new risk pattern faster; if the environment is stable, can be decreased to smooth the weight adjustment; is the discounted risk factor, and , used to reflect the discount effect of long-term risk returns, and its value is positively correlated with the supply chain stability: , is the baseline discount rate, used to reflect the default long-term return decay rate; is the stability adjustment coefficient, used to control the amplification effect of consecutive risk-free periods on the discount rate; is the time period threshold, representing the measurement unit of consecutive risk-free periods (such as months); for example, if the supply chain has no risk for 12 consecutive months (the number of consecutive risk-free periods = 12), then , indicating full attention to future returns; if the supply chain has frequent risks recently (such as the number of consecutive risk-free periods = 0), then , paying more attention to immediate risk mitigation; is the candidate weight value for the next state , that is, after the state transition (such as a risk event is triggered or no event occurs), the new weight that the risk assessment and early warning smart contract may adopt after adjusting through the action space A (increasing or decreasing the weight); The results of weight adjustment are written into the blockchain through the risk assessment and early warning smart contract, forming a learning mechanism for risk events - weight adjustment: by calling the historical transaction behavior fingerprints stored on the chain through the risk assessment and early warning smart contract (the historical transaction behavior fingerprints are generated by extracting the root node hash values corresponding to the Merkle paths in the risk metadata certificate, and these hash values specifically include order receipt hashes, payment timestamp hashes, logistics track hashes, etc., which, as the unique identifiers of key transaction behaviors, jointly constitute the core elements of the historical transaction behavior fingerprints), recalculate the credit risk characteristics, operational risk characteristics, and legal risk characteristic vectors: , , ; Compare the recalculated multi-dimensional risk characteristic vectors dimension by dimension with the original multi-dimensional risk characteristic vectors generated by the three-dimensional risk matrix: , wherein, is the risk deviation rate, is the original multi-dimensional risk characteristic vector, is the recalculated multi-dimensional risk characteristic vector, is the risk characteristic vector category; Preset risk deviation thresholds β1 and β2; when the calculated risk deviation rate breaks through the risk deviation threshold, then automatically trigger hierarchical early warnings, that is, high risk ≥ β1, medium risk [β2, β1), low risk < β2: when identified as the high-risk level, that is, the risk deviation rate ≥ β1, then generate a structured early warning event package including the risk level and risk characteristic category, and broadcast it to all associated nodes through the P2P network of the consortium chain, including the core enterprise node, the supplier node, and the manufacturer node, and at the same time record the early warning timestamp, the triggering node ID, and the event hash value on the chain; when identified as the medium-risk level, that is, the risk deviation rate belongs to [β2, β1), then push an early warning notice (including the risk level, specific risk characteristic category, and recommended measures, such as recommending to increase the sampling ratio by 10%) to the risk control department of the core enterprise in the supply chain, and at the same time mark this node as an observation object on the chain and restrict its preferential cooperation rights (such as suspending the allocation of new orders); when identified as the low-risk level, that is, the risk deviation rate < β2, then do not trigger an early warning; Integrate the risk deviation rate and early warning event information, and generate a new on-chain block through the risk assessment and early warning smart contract; among them, the block includes the hash value of the previous block, the weight adjustment strategy version number of the RL agent, the list of receiving node IDs of the early warning event, and the delivery timestamp, ensuring the traceability and immutability of the scoring process.

[0016] It should be further noted that in the specific implementation process, the process of obtaining risk assessment and early warning results, constructing a supply chain topology network based on the transaction data stored on the blockchain, simulating the risk conduction process through the risk value diffusion formula, and generating a risk diffusion trajectory map with a timestamp for sharing is as follows: Each associated node in the supply chain receives the broadcast structured early warning event package through the P2P network of the consortium chain, where the associated nodes include the core enterprise node, the supplier node, and the manufacturer node; each associated node parses the received early warning event package, extracts the key information, and stores it in the local database for subsequent tracking of the risk conduction path; Using the historical transaction records stored on the blockchain, namely the transaction amount and the transaction frequency; constructing a supply chain topology network based on the transaction dependency relationship: using the identity DID identifier as the node of the supply chain topology network (i.e., the core enterprise node, the supplier node, and the manufacturer node); using the weighted values of the transaction amount and the transaction frequency as the edges of the supply chain topology network; in the constructed supply chain topology network, locate the warning node triggered by the early warning event, that is, the risk source node (such as the supplier node); set the initial risk value of the risk source node to , the weight set of the edges in the supply chain topology network is , where represents the node and the node the weighted value of the transaction amount and the transaction frequency between them; along the edges in the supply chain topology network, decay and diffuse the risk value according to the edge weight (the weighted value of the transaction amount and the transaction frequency), where the risk value diffusion formula is: , In the formula, is the risk value of the node at time , is the set of neighbor nodes of the node , is the sum of the weights of the edges between all the neighbor nodes of the node and the node ( is a neighbor node of the node ), is the weight of the edge between the node and the node , represents the exponential function with base e, is the risk decay coefficient, is the node and the node the shortest path length between them, is at time the node The risk value, is the current time point; the greater the edge weight, the more frequent the transaction between two nodes and the closer the cooperation, the smaller the attenuation degree of the risk value during the diffusion process, that is, the greater the possibility of risk conduction; during the diffusion process, the risk status of affected nodes (such as supplier nodes, manufacturer nodes) is marked in real time, including risk levels (high risk, medium risk, and low risk), multi-dimensional risk characteristics (credit risk, operational risk, and legal risk), and the size of the risk value. Generate a risk diffusion trajectory graph with a timestamp based on the information recorded during the risk value diffusion process; this trajectory graph is used to display the path and process of risk conduction from the warning node along the supply chain topology network to other nodes, as well as the risk status of each node at different time points; store the generated risk diffusion trajectory graph in IPFS (InterPlanetary File System) and generate the corresponding CID (Content Identifier); write this CID into the blockchain transaction log to ensure the integrity and immutability of the trajectory graph; at the same time, share the trajectory graph information with each associated node in the supply chain through the P2P network of the consortium chain.

[0017] It should be further noted that in the specific implementation process, based on the shared information of the risk diffusion trajectory graph, automatically matching and executing risk disposal strategies, and continuously evaluating the effect of risk disposal and dynamically adjusting to optimize the risk management process is as follows: Obtain the risk level (high, medium, low) and risk characteristic type (credit, operational, and legal risk characteristics), and match predefined risk disposal rules from the on-chain policy library (such as high-risk suppliers need to execute the combination strategy of order suspension + full margin); the risk disposal strategy smart contract automatically generates a structured instruction package containing risk disposal measures (such as the reduction ratio of order volume, the increase amplitude of inspection frequency), execution time limit (immediately or within 72 hours), and responsible entity (core enterprise or third-party quality inspection agency), and encrypts it through digital signature; among them, this instruction package will be broadcast to all relevant nodes through the P2P network of the consortium chain, including core enterprise nodes, supplier nodes, and manufacturer nodes. After receiving the instruction package, each associated node will immediately execute the corresponding risk disposal measures and record the execution situation through its own log; the risk disposal strategy smart contract continuously monitors the logs of each associated node, verifies whether the risk disposal measures are executed according to the instructions, and collects the execution feedback of the risk disposal measures in real time; at the same time, continuously capture the dynamic shared information in the risk diffusion trajectory graph, calculate the multi-dimensional risk feature vector and compare it with the baseline before risk disposal, and quantitatively evaluate the risk reduction amplitude. Among them, through the dynamic evaluation formula of risk disposal efficiency, quantitatively evaluate the risk reduction amplitude, so as to judge the effectiveness of the risk disposal strategy: , In the formula, is the risk reduction amplitude; is the integral operation with respect to the time variable t; is the integration interval, representing the scope of action or time window of risk disposal; is the risk attenuation kernel function, used to quantify the weighted influence of the risk level RH and the risk characteristic type k on the risk disposal effect, where , , are preset weight coefficients, is the risk level sensitivity coefficient, is the risk characteristic type influence function; is the policy execution efficiency factor, used to combine the policy matching degree sm and the execution quality fq, and punish the deviation between the policy and the execution, where , where , are preset weight coefficients of the policy matching degree sm and the execution quality fq, , are non-linear adjustment exponents, is the policy-execution deviation penalty coefficient; is the risk conduction damping effect, which simulates the suppression of risk propagation in the supply chain network through the node risk change rate and the damping coefficient, where , is the risk conduction rate constant, is the initial risk value, is the risk damping coefficient of the supply chain topology network node, is the supply chain topology network node index, is the number of supply chain topology network nodes, is the node risk change rate; is the indicator function. When , the value of the indicator function is 1, and the formula calculates the actual risk reduction amplitude , triggering the risk disposal mechanism (such as upgrading the warning level, expanding the scope of policy execution); when , the value of the exponential function is 0, and the overall result of the formula is 0, and the risk disposal measures may be suspended or downgraded; is the preset dynamic risk threshold; among them, if the risk disposal measures fail to effectively reduce the risk or the risk further escalates, a higher-level warning will be automatically triggered, and the risk disposal strategy will be re-matched.

[0018] It should be understood that in various embodiments of the present application, the magnitudes of the sequence numbers of the above processes do not mean the order of execution, and the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0019] It should be understood that determining B based on A does not mean determining B solely based on A, and B can also be determined based on A and / or other information.

[0020] As described above, this is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0021] Finally: The above is only the preferred embodiment of the present invention and is not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should all be included within the protection scope of the present invention.

Claims

1. A blockchain-based supply chain partner risk management method, characterized in that: S1. Collect multi-source dynamic data, including transaction performance and compliance records; extract and integrate multi-dimensional risk feature vectors including credit risk, operational risk, and legal risk through data analysis methods, generate standardized risk metadata, and store it on the blockchain for evidence; S2. Deploy a risk assessment and early warning smart contract on the consortium blockchain, extract the multi-dimensional risk feature vectors in the risk metadata certificate, and calculate the comprehensive risk score; Dynamically adjust the weights of each dimension through a reinforcement learning algorithm to achieve the functions of risk assessment and early warning; S3. Obtain the risk assessment and early warning results, construct a supply chain topology network based on the transaction data stored on the blockchain, simulate the risk conduction process through the risk value diffusion formula, and generate a risk diffusion trajectory map with a timestamp for sharing; S4. Based on the shared information of the risk diffusion trajectory map, automatically match and execute risk disposal strategies, continuously evaluate the risk disposal effect, and dynamically adjust to optimize risk management.

2. The blockchain-based supply chain partner risk management method according to claim 1, wherein: The process of extracting and integrating multi-dimensional risk feature vectors including credit risk, operational risk, and legal risk through data analysis methods, generating standardized risk metadata, and storing it on the blockchain for evidence includes: Construct a three-dimensional risk matrix, including the credit risk dimension, the operational risk dimension, and the legal risk dimension: Calculate the credit risk eigenvector through the historical payment delay rate and the order fulfillment rate. The calculation formula for the historical payment delay rate is: , where is the historical payment delay rate, is the total number of orders within the statistical period, is the order index within the statistical period, is the th actual payment date of the order, is the th contractually agreed payment date of the order; The calculation formula for the order fulfillment rate is: , where is the order fulfillment rate, is the number of orders delivered on time, is the total number of orders; Integrate the historical payment delay rate and the order fulfillment rate to obtain the credit risk eigenvector ; The operation risk feature vector is calculated based on the standard deviation of logistics timeliness and the deviation of inventory turnover. The formula for the standard deviation of logistics timeliness is: , where is the standard deviation of logistics timeliness, is the logistics index, is the number of logistics times, is the th actual logistics timeliness, is the average logistics timeliness; The formula for the deviation of inventory turnover is: , where is the deviation of inventory turnover, is the average inventory turnover days, is the actual inventory turnover days; Integrate the standard deviation of logistics timeliness and the deviation of inventory turnover to obtain the operation risk feature vector ; Legal risk feature vector calculated based on environmental protection penalty records and the number of days of license invalidation: , where is the environmental protection penalty rate, is the number of environmental protection penalties, is the total number of operating days; the calculation formula for the proportion of license invalidation days is: , where is the proportion of license invalidation days, is the total number of license invalidation days; integrating the environmental protection penalty rate and the proportion of license invalidation days to obtain the legal risk feature vector ; Adopt the JSON-LD format to serialize the multi-dimensional risk feature vectors into a standardized risk metadata packet; perform a hash calculation on the serialized JSON-LD data to generate a unique data hash value; use the generated JSON-LD packet to call the evidence storage contract of the consortium blockchain; the evidence storage contract will record the complete JSON-LD packet and generate a risk data evidence storage certificate with a digital signature, where the certificate contains a timestamp and the data hash value; based on the risk data evidence storage certificate, use the data hash value therein as the input and write it into the Merkle tree of the blockchain; the root node of the Merkle tree will be recorded on the blockchain; based on the record of the Merkle tree, generate a risk metadata certificate with a timestamp, where the certificate contains the data source signature and the Merkle path.

3. The blockchain-based supply chain partner risk management method according to claim 1, characterized in that: The process of extracting the multi-dimensional risk feature vectors in the risk metadata certificate and calculating the comprehensive risk score includes: Call the generated risk metadata certificate through the on-chain verification engine, extract the embedded JSON-LD format multi-dimensional risk feature vectors and evidence storage certificate information from the certificate; synchronously parse the Merkle path in the certificate, extract the root node hash value, and use it as the reference anchor point; The risk assessment and early warning smart contract reads the multi-dimensional risk feature vectors stored for evidence from the temporary storage area and parses out the multi-dimensional risk feature vectors; uses the multi-dimensional risk feature vectors in the current period and the historical risk event library stored on the blockchain as the input state; Normalize the risk feature vectors of each dimension, perform a dot product calculation in combination with the initial weights, and generate a comprehensive risk score.

4. The method for risk management of supply chain partners based on blockchain according to claim 3, characterized in that: The process of dynamically adjusting the weights of each dimension through a reinforcement learning algorithm to achieve the functions of risk assessment and early warning includes: Introduce a supply chain risk reinforcement learning RL agent model, and use the Q-learning algorithm to dynamically adjust the weights of each dimension: Define the risk dimension set as , the initial weight of each dimension is , satisfying ; Define the state space as , representing the matching degree between the current multi-dimensional risk feature vector and the historical event library, that is , where is the time step, representing the number of iterations of risk assessment and early warning smart contract learning or the discrete time point of risk events; is the L2 norm; Define the action space as , which is used to adjust the weight direction, that is, increase or decrease the dimensional weight: , where ; The risk assessment and early warning smart contract dynamically updates the weights through the following Q-learning rules: ， Wherein, is the updated weight; is the immediate reward, obtained based on the risk score change and the disposal cost; is the learning rate; is the discount risk factor, is the next state is the candidate weight value; The weight adjustment result is written into the blockchain through the risk assessment and early warning smart contract, forming a learning mechanism for risk events - weight adjustment: by invoking the historical transaction behavior fingerprints stored on the chain through the risk assessment and early warning smart contract, recalculate the credit risk characteristics, operational risk characteristics, and legal risk characteristic vectors: , , ; Compare the recalculated multi-dimensional risk feature vector dimension by dimension with the original multi-dimensional risk feature vector generated by the three-dimensional risk matrix: ， In the formula, is the risk deviation rate, is the original multi-dimensional risk feature vector, is the recalculated multi-dimensional risk feature vector, is the risk feature vector category; Preset risk deviation thresholds β1 and β2; when the calculated risk deviation rate breaks through the risk deviation threshold, a hierarchical early warning is automatically triggered, i.e., high risk ≥ β1, medium risk ∈ [β2, β1), low risk < β2: when it is identified as a high-risk level, i.e., the risk deviation rate ≥ β1, a structured early warning event package including the risk level and risk feature category is generated and broadcast to all associated nodes through the P2P network of the consortium blockchain, including the core enterprise node, supplier node, and manufacturer node, and at the same time, the early warning timestamp, triggering node ID, and event hash value are recorded on the chain; when it is identified as a medium-risk level, i.e., the risk deviation rate belongs to [β2, β1), an early warning notice is pushed to the risk control department of the supply chain core enterprise, and at the same time, this node is marked as an observation object on the chain, restricting its preferential cooperation rights; when it is identified as a low-risk level, i.e., the risk deviation rate < β2, no early warning is triggered; Integrate the risk deviation rate and early warning event information, and generate a new on-chain block through the risk assessment and early warning smart contract for packaging.

5. The blockchain-based supply chain partner risk management method according to claim 1, wherein: The process of obtaining the risk assessment and early warning results, constructing a supply chain topology network based on the transaction data stored on the blockchain, simulating the risk conduction process through the risk value diffusion formula, and generating a timestamped risk diffusion trajectory map for sharing includes: Each associated node in the supply chain receives the broadcast structured early warning event package through the P2P network of the consortium blockchain, where the associated nodes include the core enterprise node, supplier node, and manufacturer node; each associated node parses the received early warning event package, extracts key information, and stores it in the local database. Using the historical transaction records stored on the blockchain, namely the transaction amount and transaction frequency; constructing a supply chain topology network based on transaction dependency relationships: using the identity DID identifier as the nodes of the supply chain topology network, namely the core enterprise node, the supplier node, and the manufacturer node; using the weighted values of the transaction amount and transaction frequency as the edges of the supply chain topology network; in the constructed supply chain topology network, locating the warning node triggered by the warning event, namely the risk source node; setting the initial risk value of the risk source node as , the weight set of the edges in the supply chain topology network is , where represents the node and the node the weighted value of the transaction amount and transaction frequency between them, that is, the weight of the edge between the node and the node ; along the edges in the supply chain topology network, according to the edge weight attenuation to spread the risk value, where the risk value diffusion formula is: ， Wherein, is the risk value of node at time . is the set of neighbor nodes of node . is the sum of the weights of the edges between all neighbor nodes of node and node . represents the exponential function with base e is the risk attenuation coefficient is the shortest path length between node and node . is the risk value of node at time . is the current time point; during the diffusion process, the risk states of affected nodes are marked in real time, including risk levels, multi-dimensional risk characteristics, and the magnitudes of risk values; Generate a timestamped risk diffusion trajectory map according to the information recorded during the risk value diffusion process; store the generated risk diffusion trajectory map in IPFS and generate the corresponding CID; write this CID into the blockchain transaction log; at the same time, share the trajectory map information with each associated node in the supply chain through the P2P network of the consortium blockchain.

6. The blockchain-based supply chain partner risk management method according to claim 1, wherein: Based on the shared information of the risk diffusion trajectory map, automatically match and execute risk disposal strategies, and continuously evaluate the risk disposal effect and dynamically adjust to optimize the risk management process includes: Obtain the risk level and risk feature type, and match the predefined risk disposal rules from the on-chain policy library; automatically generate a structured instruction package including risk disposal measures, execution time limit, and responsible entity through the preset risk disposal strategy smart contract and encrypt it with a digital signature; among them, this instruction package will be broadcast to all relevant nodes through the P2P network of the consortium blockchain, including the core enterprise node, supplier node, and manufacturer node; After receiving the instruction packet, each associated node will immediately execute the corresponding risk disposal measures and record the execution status through its respective logs; the risk disposal strategy smart contract continuously monitors the logs of each associated node to verify whether the risk disposal measures are executed according to the instructions and collects the execution feedback of the risk disposal measures in real time; at the same time, it continuously captures the dynamic shared information in the risk diffusion trajectory map, calculates the multi-dimensional risk feature vector and compares it with the baseline before risk disposal to quantitatively evaluate the risk reduction amplitude. Among them, the risk reduction amplitude is quantitatively evaluated through the dynamic evaluation formula of risk disposal efficiency, so as to judge the effectiveness of the risk disposal strategy: ， In the formula, is the risk reduction amplitude; is the integral operation with respect to the time variable t; is the integral interval, representing the scope of action or time window of risk disposal; is the risk attenuation kernel function, used to quantify the weighted influence of the risk level RH and the risk characteristic type k on the risk disposal effect; is the policy execution efficiency factor; is the risk conduction damping effect; is an indicator function. When occurs, the value of the indicator function is 1, and the formula calculates the actual risk reduction amplitude , triggering the risk disposal mechanism; when occurs, the value of the exponential function is 0, the overall result of the formula is 0, and the risk disposal measure is to suspend or downgrade; is the preset dynamic risk threshold.

Images